Full text of "The Biological bulletin"

See other formats

Volume 201

THE

Number 1

BIOLOGICAL
BULLETIN

AUGUST 2001

Published by the Marine Biological Laboratory
http://www.biolbull.org

Made to my exact

Let's address my specs first, specifically lugh resolution,
contrast and infinity-corrected optics. They've all
reached Olympian standards thanks to Olympus. But
even more to the point, here's how the BX2's modular
design came through for me. First, the eight-position
universal condenser offers the flexibility to choose from
brightfield, darkfield and phase as well as DIC. Next,
its assortment of prisms makes it possible to match
the optical image shear to the specimen, achieving the
optimal balance of contrast and resolution. Finally,
the Plan APO objectives, with superb chromatic
correction and contrast, provide extraordinary detail.
Now let's move on.

And yours

Picture yourself sitting here, looking into your
Olympus BX2 research microscope, your fluorescence
requirements having been met. Specifically: The
aspherical collector lens produces a fluorescence
intensity that's twice as bright as others and more even
across the field. The unique excitation balancers
improve visualization of multiple labels by revealing
details that would otherwise be unseen. The six-posi-
tion filter turret makes single and multiband imaging
faster and simpler. And the rectangular field stop,
another Olympus exclusive, protects the specimen by
exposing only the precise area being imaged in addi-
tion to enhancing the S/N ratio. Time to see what's next.

OLYMPUS

FOCUS ON LIFE

Visit us at www.olympusamerica.com or call 1-800-446-5967.

specifications

And yours. And yours.

Here, imaging and automation is a must. And here, the
BX2 responds as a high-performance, highly efficient,
digital imaging machine. The motorized nosepiece,
Z-drive, condenser, illuminator and filter wheels are
fully integrated through the user-friendly software
package. It's you who commands this automated imag-
ing system with your PC, optional keypad or preset
buttons located on the microscope frame itself. Digital
images can now be acquired, processed and analyzed
faster than before. And reports and documentation
have never been this easy to generate. Which leaves
one more set of specs.

Now modularity really is in high gear as the Olympus
FLUOVIEW 500 is added, resulting in a complete confo-
cal laser scanning microscope system. It provides
five imaging channels and has an intuitive operation
that makes it readily available to everyone so that
productivity is greatly enhanced. By the way, the BX2
is the only microscope that offers a Metal Matrix
Composite frame — the ultimate in static and thermal
rigidity — making it the optimal solution for frequent
use of 3D microscopy, time-lapse observations and
high-end digital imaging. So you see, with all this mod-
ularity and flexibility, my BX2 microscope is also
your BX2 microscope.

Research Microscope Series

Cover

About 3.5 million years ago (Ma), rising sea levels
opened the Bering Strait, and the North Atlantic
Ocean was invaded by hundreds of taxa from the
North Pacific. Among the invaders was the seastar
genus Asterias. At present, two species of Asterias
are recognized in the North Atlantic: A. forbesi on
the west coast of the Atlantic, from Cape Cod south
to Cape Hatteras, and A. rubens, a European species
that ranges from southern France to Norway and
Iceland, but also occurs in the northwestern Atlan-
tic, mainly from Cape Cod north. Representatives
of these species are shown on the cover, as is a
specimen of A. amurensis, which inhabits the North
Pacific from British Columbia to Japan.

After entering the Atlantic, populations of Asterias
were separated, and speciation subsequently occurred.
The timing of the separation is critical, for it deter-
mined, in part, the mechanism involved in the specia-
tion, and it is the basis for the present geographic
distribution of Asterias species in the North Atlantic.
However, as the map on the cover illustrates, the
timetable of these events was constrained by habitat
and oceanographic instability during the Pleistocene
glaciation.1 In particular, most of the current North
American habitat of Asterias rubens was repeatedly
covered by a kilometer of ice and was unavailable to
this seastar until about 15,000 years ago — long after
the opening of the Bering Strait.

1 The map on the cover is a polar view of the North Atlantic and
Pacific Oceans during the Wisconsin glacial maximum, about 20.000
years ago. The solid blue line marks the average glacial margin; the
dashed blue lines show the extent of sea ice in summer (upper) and
winter (lower); the dotted black line illustrates how lower sea levels
during glacial maxima altered the Atlantic coastline; and the shades of
blue and green represent isotherms, highly compressed in the north-
western Atlantic, and producing a strong temperature gradient.

The speciation of Asterias in the Atlantic has been
explained by two hypotheses. Either the event oc-
curred recently, with strong natural selection pre-
cluding hybridization; or the speciation into North
American and European species occurred shortly
after Asterias entered the North Atlantic, with a
recolonization of the northwestern coast of the At-
lantic by A. rubens taking place in recent times. The
second hypothesis implies that speciation was due
to prolonged isolation and was independent of ob-
served adaptations to different water temperatures.

As reported in this issue (p. 95), John P. Wares has
collected genetic sequence data from populations of
A. forbesi, A. rubens, and A. amurensis and used
them in phylogenetic and population genetic analy-
ses to test the two hypotheses. He concludes that,
although changes in climate and ocean currents —
particularly the formation of the Labrador Cur-
rent— were concomitant with the separation of As-
terias populations in the North Atlantic 3 Ma,
permanent colonization of New England and the
Canadian Maritimes by A. rubens occurred very
recently.

(Credits: map — from B. Frenzel, M. Pecsi, and
A.A. Velichko, eds., 1992, Atlas of Paleociimates
and Paleoenvironments of the Northern Hemi-
sphere, Geographical Research Institute, Hungarian
Academy of Sciences, Budapest, p. 43; images of
Asterias forbesi and A. rubens — from the George
M. Gray Museum collection, formerly administered
by the Marine Biological Laboratory, now at the
Peabody Museum of Natural History of Yale Uni-
versity; image of A. amurensi — from a photograph
by Jan Haaga, provided online by the Alaska Fish-
eries Science Center/National Marine Fisheries Ser-
vice; cover design — by Beth Liles, MBL.)

THE

BIOLOGICAL BULLETIN

AUGUST 2001

Editor
Associate Editors

Section Editor
Online Editors

Editorial Board

Editorial Office

MICHAEL J. GREENBERG

Louis E. BURNETT
R. ANDREW CAMERON
CHARLES D. DERBY
MICHAEL LABARBERA

The Whitney Laboratory, University of Florida

Grice Marine Biological Laboratory, College of Charleston
California Institute of Technology
Georgia State University
University of Chicago

SHINYA INDUE, Imaging and Microscopy Marine Biological Laboratory

JAMES A. BLAKE, Keys to Marine
Invertebrates of the Woods Hole Region
WILLIAM D. COHEN, Marine Models
Electronic Record and Compendia

PETER B. ARMSTRONG
ERNEST S. CHANG
THOMAS H. DIETZ
RICHARD B. EMLET
DAVID EPEL
GREGORY HINKLE
MAKOTO KOBAYASHI
ESTHER M. LEISE
DONAL T. MANAHAN
MARGARET MCFALL-NGAI
MARK W. MILLER
TATSUO MOTOKAWA
YOSHITAKA NAGAHAMA
SHERRY D. PAINTER
J. HERBERT WAFTE
RICHARD K. ZIMMER

ENSR Marine & Coastal Center, Woods Hole
Hunter College, City University of New York

University of California, Davis

Bodega Marine Lab., University of California, Davis

Louisiana State University

Oregon Institute of Marine Biology, Univ. of Oregon

Hopkins Marine Station, Stanford University

Cereon Genomics, Cambridge, Massachusetts

Hiroshima University of Economics, Japan

University of North Carolina Greensboro

University of Southern California

Kewalo Marine Laboratory, University of Hawaii

Institute of Neurobiology, University of Puerto Rico

Tokyo Institute of Technology, Japan

National Institute for Basic Biology, Japan

Marine Biomed. Inst., Univ. of Texas Medical Branch

University of California, Santa Barbara

University of California, Los Angeles

PAMELA CLAPP HINKLE
VICTORIA R. GIBSON

Managing Editor

Staff Editor

~ <-, 'ceanoo _ ,. . , .

CAROL SCHACHINGER Editorial Associate

WENDY CHILD

AUG 3 1 2001

Subscription & Advertising Secretary

Published by

MARINE BIOLOGICAL LABORATORY
WOODS HOLE, MASSACHUSETTS

http://www.biolbull.org

Genomic Research Leaders Choose
Microway® Scalable Clusters

E os Biotechnology, Marine Biological Laboratory, Millennium Pharmaceu-
ticals, Mount Sinai Medical School, NIH, Pfizer, and Rockefeller University
All Choose Microway Custom Clusters and Workstations for Reliability,
Superior Technical Support and Great Pricing.

1.4 GHz Dual Athlon, 1.7 GHz Pentium 4,
or 1 GHz Dual Pentium III in 1U or 2U
Clusters

Dual Alpha 833 MHz Clusters and Towers

For maximum price/performance choose our
Alpha 1U 833 MHz, 4 MB DDR Cache CS20,
4U UP2000+ or 4V 264DP RuggedRack™

Myrinet, Gigabit Ethernet or Dolphin Wulfkit
High Speed Low Latency Interconnects

RAID and Fibre Channel Storage Solutions

Microway'1" Screamer™
Dual Alpha UP2000*
833 MHz. 4MB Cache in
RuggedRack™ Chassis
with RRR™ Redundant
Power Supply

Microway has earned an excellent reputation
since 1982. If you need a quality product that is
fine tuned and built to last, from a company that
will be around to support you for years to come,
Microway is The Number One Choice.

.Microway has delivered high-performance computing prod-
ucts since 1982, when our pioneering software made it pos-
sible to use an 8087 in the IBM-PC. In 1987 we created the
world's first PC parallel processing systems. Since then, our
QuadPuter™ architecture has migrated from Transputers to
i860s and finally to Alphas in 1995. Over the past three years,
we engineered and delivered over 300 clusters that utilized
MPI running on Linux. As a software developer and hard-
ware manufacturer, we know the value of extensive testing
and validation. We are experts at configuring and validating
the low latency interconnects we employ in our clusters.
Our technical support is legendary — the systems we sell
arrive at your site and WORK! Los Alamos chose Microway to
maintain and upgrade its 144 node Alpha Avalon Cluster
because of our reputation. Large clusters we have sold
include 400+ nodes at the University of Wisconsin and 250+
nodes at Rockefeller University.

Microway

Technology you can count on m

Microway offers three Athlon/
Pentium enclosures — 1U, 2LI and
tower, plus five Alpha configura-
tions—1 LI, 3U, 4U RuggedRack™,
QuadPuter™ and full tower. Our
264DP includes two 21264's with
up to 4 GB of memory in our cus-
tom 4U RuggedRack, which features
front accessible redundant power
supplies and hard disks. This rugged
configuration was chosen by the
U.S. Navy for onboard use. We also
offer a dual Alpha UP2000+ running
at 833MHz with 2GB of memory.
Our QuadPuter chassis holds 4
Alpha processors and up to 4GB
memory. The 1U CS20 dual Alpha
(at right) is the highest density
computational platform available.

Microway is API-Networks' Top
North American

Channel
Partner.

"Most Powerful, Highest Density
Computational Platform On the Planet"

Microway Scalable 25 Node

50 Processor Cluster Using CS20 Dual

833 MHz Alphas and Myrinet Interconnect

Yielding Peak Throughput of 82.5 Gigaflop:

I have ordered numerous Alpha and Intel-based servers and work-
stations from Microway running both Tru64 UNIX and Linux. We
have been very happy with both the performance and great value
of Microway's products. The major UNIX vendors don't come close
to Microway in this regard, and we have also found that Microway
provides better value than other Linux hardware vendors. I have
also used Microway's tech support and was pleased with their
response. We've been using their systems for over a year and
have had only a couple of minor incidents which were dealt with
promptly."

— David Kristofferson, Ph.D., MBA,
Director of Information Systems, Eos Biotechnology, Inc.

Find out why over 75% of Microway's

sales come from repeat customers.

Please call 508-746-7341 for a

technical salesperson who

speaks your language!

Visit us at www.microway.com

Research Park Box 79, Kingston, MA 0236* • 508-746-7341 • info@microway.com

AUG 3 1 2001

CONTENTS

VOLUME 201. No. 1: AUGUST 2001

RESEARCH NOTE

Seibel, Brad A., and David B. Carlini

Metabolism of pelagic cephalopods as a function of
habitat depth: a reanalysis using phylogenetically in-
dependent contrasts

NEUROBIOLOGY AND BEHAVIOR

Herberholz, Jens, and Barbara Schmitz

Signaling via water currents in behavioral interac-
tions of snapping shrimp (Alpheus heterochaelis) ....

PHYSIOLOGY AND BIOMECHANICS

Reddy, P. Sreenivasula, and B. Kishori

Methionine-enJkephalin induces hyperglycemia through
evestalk homiones in the estuarine crab Stylla sermta . . .

Mogami, Yoshihiro, Junko Ishii, and Shoji A. Baba

Theoretical and experimental dissection of gravity-
dependent mechanical orientation in gravi tactic micro-
organisms 26

SYMBIOSIS AND PARASITOLOGY

Hanten, Jeffrey J., and Sidney K. Pierce

Synthesis of several light-harvesting complex I polypep-
tides is blocked by cycloheximide in symbiotic chloro-
plasts in the sea slug, Elysia chlorotica (Gould): A case for
horizontal gene transfer between alga and animal?. . .
McCurdy, Dean G.

Asexual reproduction in Pygospio elegans Claparede
(Annelida, Polychaeta) in relation to parasitism by
Lepocreadium setiferoides (Miller and Northup) (Platy-
helminthes, Trematoda)

DEVELOPMENT AND REPRODUCTION

Stewart-Savage, J., Aimee Phillippi, and Philip O. Yund

Delayed insemination results in embryo mortality in

a brooding ascidian 52

CELL BIOLOGY

Ballarin, Loriano, Antonella Franchini, Enzo Ottaviani,
and Armando Sabbadin

Momla cells as the major immunomodulatory hemo-
cytes in ascidians: evidences from the colonial species
Botnllm schlosseri 59

ECOLOGY AND EVOLUTION

Halanych, Kenneth M.. Robert A. Feldman, and
Robert C. Vrijenhoek

Molecular evidence that Sclerolinum brattstromi is
closely related to vestimentiferans, not to frenulate
pogonophorans (Siboglinidae. Annelida) 65

Ponczek, Lawrence M., and Neil W. Blackstone

Effect of cloning rate on fitness-related traits in two
marine hydroids 76

Meidel, Susanne K., and Philip O. Yund

Egg longevity and time-integrated fertilization in a tem-
perate sea urchin (Stnmgylocenfrotus droebachiensis) .... 84

Wares, J. P.

Biogeography of Astmas: North Atlantic climate
change and speciation 95

SYSTEMATICS

Gershwin, Lisa-ann

Systematics and biogeography of the jellyfish Aurelia
labiata (Cnidaria: Scyphozoa) 104

45 Annual Report of the Marine Biological Laboratory. ... Rl

ANNOUNCEMENT

The Marine Biological Laboratory is pleased to announce that it has entered into an agreement with HighWire Press of Stanford
University to publish The Biological Bulletin electronically. The online journal will be launched on 23 August 2001. It will be
available free of charge to subscribers and the general public for the next six months. Subsequently, subscribers to The Biological
Bulletin will receive both the print and electronic versions of the journal.

We invite you to visit The Biological Bulletin online at
http://www.biolbull.org

THE BIOLOGICAL BULLETIN

THE BIOLOGICAL BULLETIN is published six times a year by the Marine Biological Laboratory, 7 MBL Street,
Woods Hole, Massachusetts 02543.

Subscriptions and similar matter should be addressed to Subscription Manager, THE BIOLOGICAL BULLETIN,
Marine Biological Laboratory, 7 MBL Street. Woods Hole, Massachusetts 02543. Subscription per year (six
issues, two volumes): $235 for libraries; $95 for individuals. Subscription per volume (three issues): $1 17.50 for
libraries; $47.50 for individuals. Back and single issues (subject to availability): $40 for libraries; $20 for
individuals.

Communications relative to manuscripts should be sent to Michael J. Greenberg. Editor-in-Chief, or Pamela
Clapp Hinkle, Managing Editor, at the Marine Biological Laboratory, 7 MBL Street, Woods Hole, Massachusetts
02543. Telephone: (508) 289-7428. FAX: 508-289-7922. E-mail: pclapp@mbl.edu.

http://www.biolbull.org

THE BIOLOGICAL BULLETIN is indexed in bibliographic services including Index Medicus and MEDLINE,
Chemical Abstracts, Current Contents, Elsevier BIOBASE/Current Awareness in Biological Sciences, and Ceo
Abstracts.

Printed on acid free paper,
effective with Volume 180, Issue 1, 1991.

POSTMASTER: Send address changes to THE BIOLOGICAL BULLETIN, Marine Biological Laboratory,

7 MBL Street, Woods Hole, MA 02543.

Periodicals postage paid at Woods Hole, MA, and additional mailing offices.

ISSN 0006-3 185

INSTRUCTIONS TO AUTHORS

The Biological Bulletin accepts outstanding original research
reports of general interest to biologists throughout the world.
Papers are usually of intermediate length (10-40 manuscript
pages). A limited number of solicited review papers may be
accepted after formal review. A paper will usually appear within
four months after its acceptance.

Very short, especially topical papers (less than 9 manuscript
pages including tables, figures, and bibliography) will be published
in a separate section entitled "Research Notes." A Research Note
in The Biological Bulletin follows the format of similar notes in
Nature. It should open with a summary paragraph of 150 to 200
words comprising the introduction and the conclusions. The rest of
the text should continue on without subheadings, and there should
be no more than 30 references. References should be referred to in
the text by number, and listed in the Literature Cited section in the
order that they appear in the text. Unlike references in Nature,
references in the Research Notes section should conform in
punctuation and arrangement to the style of recent issues of The
Biological Bulletin. Materials and Methods should be incorpo-
rated into appropriate figure legends. See the article by Loh-
mann et al. (October 1990. Vol. 179: 214-218) for sample
style. A Research Note will usually appear within two months
after its acceptance.

The Editorial Board requests that regular manuscripts con-
form to the requirements set below; those manuscripts that do
not conform will be returned to authors for correction before
review.

1 . Manuscripts. Manuscripts, including figures, should be
submitted in quadruplicate, with the originals clearly marked.
(Xerox copies of photographs are not acceptable for review pur-
poses.) The submission letter accompanying the manuscript should
include a telephone number, a FAX number, and (if possible) an
E-mail address for the corresponding author. The original manu-
script must be typed in no smaller than 12 pitch or 10 point, using
double spacing (including figure legends, footnotes, bibliography,
etc.) on one side of 16- or 20-lb. bond paper, 8 by 1 1 inches.
Please, no right justification. Manuscripts should be proofread
carefully and errors corrected legibly in black ink. Pages should be
numbered consecutively. Margins on all sides should be at least 1
inch (2.5 cm). Manuscripts should conform to the Council of
Biology Editors Style Manual, 5th Edition (Council of Biology
Editors, 1983) and to American spelling. Unusual abbreviations
should be kept to a minimum and should be spelled out on first
reference as well as defined in a footnote on the title page.
Manuscripts should be divided into the following components:
Title page. Abstract (of no more than 200 words), Introduction,
Materials and Methods, Results, Discussion, Acknowledgments.
Literature Cited, Tables, and Figure Legends. In addition, authors
should supply a list of words and phrases under which the article
should be indexed.

2. Title page. The title page consists of a condensed title or
running head of no more than 35 letters and spaces, the manuscript

title, authors' names and appropriate addresses, and footnotes
listing present addresses, acknowledgments or contribution num-
bers, and explanation of unusual abbreviations.

3. Figures. The dimensions of the printed page. 7 by 9
inches, should be kept in mind in preparing figures for publication.
We recommend that figures be about 1 times the linear dimensions
of the final printing desired, and that the ratio of the largest to the
smallest letter or number and of the thickest to the thinnest line not
exceed 1:1.5. Explanatory matter generally should be included in
legends, although axes should always be identified on the illustra-
tion itself. Figures should be prepared for reproduction as either
line cuts or halftones. Figures to be reproduced as line cuts should
be unmounted glossy photographic reproductions or drawn in
black ink on white paper, good-quality tracing cloth or plastic, or
blue-lined coordinate paper. Those to be reproduced as halftones
should be mounted on board, with both designating numbers or
letters and scale bars affixed directly to the figures. All figures
should be numbered in consecutive order, with no distinction
between text and plate figures and cited, in order, in the text. The
author's name and an arrow indicating orientation should appear
on the reverse side of all figures.

Digital art: The Biological Bulletin will accept figures sub-
mitted in electronic form; however, digital art must conform to the
following guidelines. Authors who create digital images are
wholly responsible for the quality of their material, including color
and halftone accuracy.

Format. Acceptable graphic formats are TIFF and EPS. Color
submissions must be in EPS format, saved in CMKY mode.

Software. Preferred software is Adobe Illustrator or Adobe
Photoshop for the Mac and Adobe Photoshop for Windows. Spe-
cific instructions for artwork created with various software pro-
grams are available on the Web at the Digital Art Information Site
maintained by Cadmus Professional Communications at http://
cjs. cadmus.com/da/home. html

Resolution. The minimum requirements for resolution are
1200 DPI for line art and 300 for halftones.

Size. All digital artwork must be submitted at its actual
printed size so that no scaling is necessary.

Multipanel figures. Figures consisting of individual parts
(e.g., panels A, B, C) must be assembled into final format and
submitted as one file.

Hard copy. Files must be accompanied by hard copy for use
in case the electronic version is unusable.

Disk identification. Disks must be clearly labeled with the
following information: author name and manuscript num-
ber; format (PC or Macintosh); name and version of software
used.

Color: The Biological Bulletin will publish color figures
and plates, but must bill authors for the actual additional cost of
printing in color. The process is expensive, so authors with
more than one color image should — consistent with editorial
concerns, especially citation of figures in order — combine them
into a single plate to reduce the expense. On request, when
supplied with a copy of a color illustration, the editorial staff
will provide a pre-publication estimate of the printing cost.

4. Tables, footnotes, figure legends, etc. Authors should
follow the style in a recent issue of The Biological Bulletin in
preparing table headings, figure legends, and the like. Because
of the high cost of setting tabular material in type, authors are
asked to limit such material as much as possible. Tables, with
their headings and footnotes, should be typed on separate
sheets, numbered with consecutive Roman numerals, and
placed after the Literature Cited. Figure legends should contain
enough information to make the figure intelligible separate from
the text. Legends should be typed double spaced, with consec-
utive Arabic numbers, on a separate sheet at the end of the
paper. Footnotes should be limited to authors' current ad-
dresses, acknowledgments or contribution numbers, and expla-
nation of unusual abbreviations. All such footnotes should
appear on the title page. Footnotes are not normally permitted
in the body of the text.

5. Literature cited. In the text, literature should be cited
by the Harvard system, with papers by more than two authors
cited as Jones et a/., 1980. Personal communications and ma-
terial in preparation or in press should be cited in the text only,
with author's initials and institutions, unless the material has
been formally accepted and a volume number can be supplied.
The list of references following the text should be headed
Literature Cited, and must be typed double spaced on separate
pages, conforming in punctuation and arrangement to the style
of recent issues of The Biological Bulletin. Citations should
include complete titles and inclusive pagination. Journal abbre-
viations should normally follow those of the U. S. A. Standards
Institute (USASI). as adopted by BIOLOGICAL ABSTRACTS and
CHEMICAL ABSTRACTS, with the minor differences set out below.
The most generally useful list of biological journal titles is that
published each year by BIOLOGICAL ABSTRACTS (BIOSIS List of
Serials; the most recent issue). Foreign authors, and others who
are accustomed to using THE WORLD LIST OF SCIENTIFIC PERI-
ODICALS, may find a booklet published by the Biological Coun-
cil of the U.K. (obtainable from the Institute of Biology, 41
Queen's Gate, London. S.W.7, England, U.K.) useful, since it
sets out the WORLD LIST abbreviations for most biological
journals with notes of the USASI abbreviations where these
differ. CHEMICAL ABSTRACTS publishes quarterly supplements of
additional abbreviations. The following points of reference
style for THE BIOLOGICAL BULLETIN differ from USASI (or
modified WORLD LIST) usage:

A. Journal abbreviations, and book titles, all underlined
(for italics)

B. All components of abbreviations with initial capitals
(not as European usage in WORLD LIST e.g., J. Cell. Comp.
Physiol. NOT J. cell. comp. Physiol. )

C. All abbreviated components must be followed by a
period, whole word components must not (i.e., J. Cancer Res.)

D. Space between all components (e.g., J. Cell. Comp.
Physiol.. not J. Cell. Comp. Physiol.)

E. Unusual words in journal titles should be spelled out in
full, rather than employing new abbreviations invented by the
author. For example, use Rit Vfsindafjelags Is/endinga without
abbreviation.

F. All single woid journal titles in full (e.g., Veliger,
Ecology, Brain).

G. The order of abbreviated components should be the
same as the word order of the complete title (i.e., Proc. and
Trans, placed where they appear, not transposed as in some
BIOLOGICAL ABSTRACTS listings).

H. A few well-known international journals in their pre-
ferred forms rather than WORLD LIST or USASI usage (e.g..
Nature, Science, Evolution NOT Nature, Loud.. Science, N.Y.;
Evolution, Lancaster, Pa. )

6. Reprints, page proofs, and charges. Authors may
purchase reprints in lots of 100. Forms for placing reprint orders
are sent with page proofs. Reprints normally will be delivered
about 2 to 3 months after the issue date. Authors (or delegates
for foreign authors) will receive page proofs of articles shortly
before publication. They will be charged the current cost of
printers' time for corrections to these (other than corrections of
printers' or editors' errors). Other than these charges for au-
thors' alterations, The Biological Bulletin does not have page
charges.

Reference: BiW. Bull. 201: I-?. (August 2001)

Metabolism of Pelagic Cephalopods as a Function of
Habitat Depth: A Reanalysis Using Phylogenetically

Independent Contrasts

BRAD A. SEIBEL1 * AND DAVID B. CARLINI2

1 Monterey Bay Aquarium Research Institute, 7700 Sandholdt Road, Moss Landing, California 95039;
and 'Department of Biology, 101 Hurst Hall, American University, 4400 Massachusetts Avenue, NW,

Washington, DC 20016-8007

Metabolic rates of deep-living animals have been in-
tensely studied (1 ). Within pelagic fishes, crustaceans, and
cephalopods, a strong decline in rates of mass-specific
metabolism with depth has been observed. Childress and
Mickel (2) put/onward the visual interactions hypothesis to
explain this general pattern. Their hypothesis states that
reduced metabolic rates among manv deep-sea pelagic tax-
onomic groups result from relaxed selection for strong
locomotory abilities for visual predator-prey interactions in
the light-limited deep sea. This pattern has, however, been
tested using mean metabolic rates for species as individual
data points. Felsenstein (3) warned that, because species
are descended in a hierarchical fashion from common an-
cestors, they generally cannot be considered as independent
data points in statistical analyses. Statistical methods have
recently been developed that incorporate phylogenetic in-
formation into comparative studies to create phvlogeneti-
cally independent values that can then be used in statistical
analyses. Reliable independent phylogenetic information
has only recently become available for some deep-sea or-
ganisms. The present contribution reanalyzed the metabolic
rates (4, 5) of pelagic cephalopods as a function of, for
consistency with previous studies, MDO (minimum depth of
occurrence) using phylogenetic independent contrasts de-
rived from a recent molecular phytogeny (6). This analysis
confirms the existence of a significant negative relationship
benveen metabolism and minimum habitat depth in pelagic
cephalopods but suggests that phylogenetic history also has

Received 29 August 2000; accepted 12 April 2001.
* To whom correspondence should be addressed. E-mail: bseibel@
mbari.org

considerable influence on the metabolic rates of individual
species.

Childress ( 1 ) argued against a phylogenetic basis for the
observed relationships between metabolism and depth. He
based the argument on the identification of convergence of
metabolic rates at a given depth among distantly related taxa
(fishes, crustaceans, cephalopods) as well as divergence
within closely related groups as a function of depth. This
pattern strongly suggests that species experience similar
selective regimes at any given depth and that rates of
metabolism are evolved in response to that selection. Seibel
e t al. ( 5 ) further argued, on the basis of an analysis of higher
nodes, that most of the variation in metabolic rates among
cephalopods is between families within an order, as opposed
to between genera within a family or species within a genus.
Therefore, families are more appropriate units for compar-
ison. A decline in metabolic rates with increasing habitat
depth was also observed when families were used as inde-
pendent data points (5). Nevertheless, the degrees of free-
dom used for statistical analyses in these studies are ele-
vated, to varying degrees, due to phylogenetic non-
independence of the data.

Felsenstein (3) proposed computing weighted differences
("contrasts") between the character values of pairs of sister
species nodes, or both, as indicated by phylogenetic topol-
ogy, thereby estimating an ancestral character value (e.g.,
the ancestral states of log-transformed depth and metabolic
data presented in Fig. 1 ). Insofar as the ancestral nodes are
correctly determined, each of these contrasts is independent
of the others in terms of the evolutionary changes that have
occurred to produce differences between the two members
of a single contrast (7). Felsenstein's (3) method requires
knowledge of the cladistic relationships between the species

B. A. SEIBEL AND D. B. CARLINI

1.85, -0.50 |
1.79. -0.40F
1.98, -0.36.
2.03, -0.28
2.02, -0.26_T
2.12, -0.28

2.10, -0.08.
1.91,0.10

1.72,0.27

2.14, -0.25

2.15, -0.27

1.00,0.73

1.00,0.75

1.32,0.63

2.74, -0.99

2.30, 0.07 |

2.82, -0.81 1
2.89, -0.78

2.71, -0.83

.Cranchia 1.00, -0.43
.Liocranchia 2.70, -0.57
.Leachia 1.70, -0.25
.Helicocranchia 2.48, -0.23
.Histioteuthis 2.18,0.01
.Octopoteuthis 2.00, -0.21
.Joubiniteuthis 2.70, -0.39
.Gonatus 2.00,0.82
Jllex 1.00,0.95
.L. pealei 1.00,0.81
,L. opalescens 1.00,0.68
.Sepioteuthis 1.00,0.71
.Chtenopleryx 1.70,0.37
.Bathyteulhis 2.90, -0.23
.Heteroteuthis 2.04,0.63
.J. diaphana 2.85, -0.82
.J. heathi 2.78, -0.80
.Eledonella 2.99, -0.74
.Amphitretus 2.48, -0.89
.Vampyroteuthis 2.78, -1.22
.Nautilus 2.18, -0.30

1.60, -0.38

1.85, -0.50r

1.91, -0.31

1.88,0.23

2.09, -0.23

2.15, -0.28

1.35, -0.26r

2.18, -0.25r

1.74, -0.27r

1.00, 0.72r

1.00, 0.75r

2.00, 0.33

1.70,0.27

1.70,0.47

2.76, -0.17r

2.48. -0.1 5r

2.00, 0.68r

2.00. -0.06r

2.18, -0.07 r

0.00, 0.86

2.92, -0.77

2.82, -0.81r

1.23, -0.17

2.35, -0.78

Cranchia 1.00, -0.43
Liocrancha 2.70, -0.57
L. dislocata 1.00, -0.26
L. pacifica 1.70, -0.25
Galliteuthis 2.48, -0.27
Megalocranchia 1.00, -0.27
Helicocranchia 2.48, -0.23
L. pealei 1.00,0.81
L. opalescens 1.00,0.68
Sepioteuthis 1.00,0.71
Chtenopteryx 1.70,0.37
Bathyteuthis 2.90, -0.23
Heteroteuthis 2.04, 0.63
A.felis 1.70,0.33
A. pacificus 1.70,0.21
Enoploteuthis 1.70, 0.70
Pterygioteuthis 1.70, 0.43

C. calyx 2.48, -0.17
C. imperator2A8,-Q.\2
Valbyteuthis 2.95, -0.18
G. om.'* 2.00, 0.82
G.pwos 2.00,0.53
O. deletron 2.00, 0.09
O. nielseni 2.00, -0.21
H. heteropsis 2.18, -0.14
//. hoy lei 2.18,0.01

llex 1.00,0.95
Todarodes 1.00,0.76
Onychoteuthis o.oo, 0.76

oubiniteuthis 2.70, -0.39
Mastigoteuthis 2.57, -0.23

. diaphana 2.85, -0.82
/. heathi 2.78, -0.80
Eledonella 2.99, -0.74
Amphitretus 2.48, -0.89
Oc\thoe 1.00, 0.44
Octopus 1.00, 0.44
Vampyroteuthis 2.78, -1.22
Nautilus 2.\&, -0.30

Figure 1. Phylogenetic trees used for calculating independent contrasts on metabolic rate data. Log-
transformed minimum depth of occurrence (MDO) and metabolic rates, in that order, are shown to the right of
taxon names. Ancestral states of log-transformed MDO and metabolic rate data (i.e.. weighted differences or
"contrasts." see text), calculated using the CAIC software application ( 18), are also shown at the internal nodes.
(A) A 21-taxa tree for which both COI sequences and metabolic rate data are available. Branch lengths

INDEPENDENT CONTRASTS FOR CEPHALOPOD METABOLISM

being analyzed. Several studies have attempted to construct
phylogenies for cephalopods. However, only a single reli-
able family-level phylogeny exists that includes deep-water
fauna. One previous phylogenetic analysis relied exclu-
sively on morphological characters that are associated with
buoyancy and locomotion and are thus confounded with
metabolism and depth (8). We therefore felt that analysis
was unsuitable for use in the present study. Other analyses
have been unable to obtain sufficient resolution for familial
relationships (9) or have included only shallow-living taxa
(10, 11). Carlini and Graves (6) recently analyzed the higher
level phylogenetic relationships of extant cephalopods by
using a 657-bp sequence of the mitochondrial cytochrome c
oxidase (COI) gene. The molecular sequence data from
Carlini and Graves (6) provide an opportunity to test the
visual interactions hypothesis directly, using a more valid
statistical approach. An additional analysis based on actin
gene sequences (12) was not included, primarily because
there was very little overlap between taxa for which actin
gene sequences were available and those for which meta-
bolic data are available. Furthermore, the actin study pro-
vides a more accurate reconstruction of gene family evolu-
tion within the cephalopods than of specific relationships
among taxa.

The phylogenetic trees presented here from which the
independent contrasts were calculated include only those
species for which metabolic data are available. Similar trees
were constructed including species for which enzymatic
data are available. Although it may have been preferable to
"prune" the complete COI tree rather than reconstruct trees
using only taxa for which metabolic data are available, we
decided to calculate new trees so that we could include taxa
for which COI sequences were obtained after the publica-
tion of the COI paper (6). The species we added were
Amphitretus pelagicus, Helicocranchia pfefferi, and Jape-
tella heathi. Pruning the tree would have had only a small
effect on the values of the standardized contrasts and would
not have significantly altered our conclusions.

A second requirement of Felsenstein's (3) method is
knowledge of branch lengths in units of expected variance
of change. Ideally, branch lengths should represent expected
units of evolutionary change (gradual model). For this ap-

proach to be valid, independent contrasts must be ade-
quately standardized so that they will receive equal weight-
ing in subsequent regression analyses. We plotted the
absolute value of each standardized independent contrast,
generated from the fully resolved tree (Fig. la), versus its
standard deviation (7) and found no relationship between
the two variates (data not shown). Thus, the contrasts were
adequately standardized and properly weighted in regres-
sion analysis.

However, even if a particular phylogenetic tree is well
resolved and well supported, branch lengths are always
estimates and are thus subject to error. A less optimal
approach, but one that involves fewer assumptions about the
evolutionary relationships of the taxa in question, is to
assume that every branch in the phylogeny is the same
length (punctuated model). The advantage of this approach
is that it can be used for poorly resolved trees or for data sets
where branch lengths cannot be estimated, such as those
derived from both molecular (6) and morphological (13, 14)
data. This allows more contrasts to be performed, increasing
the power of subsequent statistical tests. On the other hand,
the punctuated model is unrealistic for most data sets, as
there is likely to be significant heterogeneity with respect to
the evolutionary rates of the taxa under study. In any case,
use of a punctuated model is far superior to any method that
treats species values as independent data points.

In the present study we employed both gradual and punc-
tuated models in constructing trees for comparison. The
gradual model tree is depicted in Figure la (21 taxa, met-
abolic rates as a function of MDO). A similar tree was
constructed including species for which enzymatic data are
available (not shown, 18 taxa, enzymatic activities as a
function of MDO). A tree constructed using the punctuated
model for contrasts involving all taxa for which data are
available is depicted in Figure Ib (39 taxa, metabolic rates
versus MDO). A similar tree was constructed including
species for which enzymatic data are available (not shown.
32 taxa, enzymatic activities versus MDO).

Independent contrasts for log-transformed, normalized
mean oxygen consumption rates (4, 15-18) were produced,
for both gradual and punctuated models, using CAIC v.
2.0.0 (19), and were regressed against those produced for

(molecular clock enforced) were calculated from the strict consensus of two most-parsimonious trees (Tree
Length = 1432 steps; Consistency Index = 0.348: Retention Index = 0.334) derived from parsimony analysis
of the COI data in PAUP* (28). (B) Partially resolved 39-taxa tree representing relationships between all pelagic
taxa for which metabolic rate data are available. The conservative tree topology is based on a consensus of
molecular and morphological evidence. In this case, branch lengths are unknown and a punctuated model of
change was assumed; that is. all branches are of equal length. For example, the ancestral character state for
log-transformed metabolic rate for the Cranchia-Liocranchia node, assuming a punctuated model of change, is
calculated assuming a branch length equal to one and taking an average of the two species (—0.43 + —0.57/2 =
—0.50. corresponding to a calculated ancestral oxygen consumption rate of 0.61 /j,m O; g 'h *). Determination
of ancestral character states, assuming a gradual model of change, requires calculation of branch length using the
CAIC software.

B. A. SEIBEL AND D. B. CARLINI

c
o

a,
E

0.2-r

0--

-0.4-

6 -°-64

orj

-0.8-

-I-

-t-

-r-

•H-

0 0.2 0.4 0.6 0.8 1

Contrast: Log (Minimum Depth of Occurrence)

Figure 2. Standardized contrasts of log-transformed oxygen consump-
tion data plotted as a function of standardized contrasts of log-transformed
minimum depth of occurrence calculated from the 39-taxon tree (Fig. Ib;
punctuated model ). Contrasts for the three sister-species groupings within
the cranchiid family (Cranchia-Liocranchia; Leachia dislocata-L. paci-
fica; Galliteuthis-Megalocranchia; Fig. Ib) are indicated with open sym-
bols and are included in the plotted regression. The slope of the regression
is significant (P < 0.01 1. See Table 1 and text for equation and related
statistics.

MDO (Fig. 2, Table 1). We produced similar regressions for
contrasts of activities of citrate synthase (CS) and octopine
dehydrogena.se (ODH) (5, 20-22), indicators of aerobic and
anaerobic metabolic potential, respectively (Table 1). We
tested the validity of log transformation by using a method
suggested by Purvis and Rambaut (19). authors of the CAIC
package. Regressions of the absolute values of the contrasts
on the estimated nodal values were performed, and none had
slopes significantly different from zero. We also performed

regressions of the absolute values of the contrasts against
the standard deviations of the contrasts and detected no
relationship in any case. These two tests ensure that we did
not violate any of the assumptions of Felsenstein's (3)
model of evolution of continuous characters as a random
walk process.

Relationships between contrasts of metabolism and depth
are summarized in Table 1. A significant decline in oxygen
consumption rate with habitat depth was observed when all
taxa were included and a punctuated model was assumed
(Fig. 2: v = -0.36.Y -- 0.02, P = 0.01). A similar
relationship was observed using the gradual model ( v =
-0.59.V - 0.049. P = 0.03). but only when the Cranchia
versus Liocranchia contrast was excluded (see below). CS
and ODH activities were weakly correlated with habitat
depth when a gradual model was assumed, even with the
Cranchia versus Liocranchia contrast excluded from anal-
ysis (Table 1;CS, v = -l.Ol.v + 0.46, P = 0.06; ODH,
v = -1.26.x - 0.22, P = 0.099). Contrasts performed
using the punctuated model for the CS and ODH data
indicated a significant negative relationship between enzy-
matic activity and habitat depth with the Cranchia versus
Liocranchia contrast excluded from analysis (Table 1: CS,
v = -0.64.V + 0.08, P = 0.01; ODH, y = -1.02* -
0.04. P = 0.005).

Although these results suggest a negative trend in metab-
olism with increasing depth independent of phylogeny,
there are clearly phylogenetic influences on the data. For
example, members of the family Cranchiidae (including
Cranchia and Liocranchia, the contrast excluded from sev-
eral of the analyses) have low metabolic rates regardless of

Metabolism of pelagic cephalopods as a function of habitat depth

Table 1

Parameter

Model

All contrasts included

MO,

Punctuated

-0.36

-0.02

0.29

0.01

Gradual

n.s.

Punctuated

n.s.

Gradual

n.s.

ODH

Punctuated

n.s.

Gradual

n.s.

Cranchia vs. Liocranchia

contrast excluded

M02

Punctuated

not performed

Gradual

-0.59

-0.05

0.27

0.03

Punctuated

-0.64

0.08

0.32

0.01

Gradual

-1.01

0.46

0.26

0.06

ODH

Punctuated

-1.02

0.04

0.40

0.005

Gradual

-1.26

-0.22

0.21

0.099

Log-transformed contrasts (y) of oxygen consumption rates (MO2 = /j.mole O, g 'h ') and enzymatic activities (citrate synthase, CS. and octopine
dehydrogenase. ODH, units g~'l of pelagic cephalopods were regressed against minimum depth of occurrence (.v), expressed as y = mA" + b. Number
of taxa (n). regression coefficients (R2) and P values are also presented.

INDEPENDENT CONTRASTS FOR CEPHALOPOD METABOLISM

habitat depth. The Cranchiidae is a very diverse family, and
our data set is slightly biased toward cranchiid species (/; =
7 out of 39. MO2. punctuated model. Fig. Ib). Although
many cranchiid species undergo ontogenetic vertical migra-
tions in which successive developmental stages occupy
progressively greater depths (12, 23), some species appear
to remain near the surface until sexual maturity (24. 25).
Seibel ct ul. (4) argued that the use of transparency (26) by
the cranchiids reduces detection distances (27) at all depths
and therefore allows them to employ sit-and-wait predation
strategies, facilitating low metabolic rates, even in well-lit
epipelagic waters. With the Cranchia-Liocranchia contrast
removed, we consistently observed a much stronger rela-
tionship between metabolism and habitat depth. Several
sources of depth-related variation in metabolism, such as
buoyancy and body mass, exist in addition to phylogeny.
These have been discussed elsewhere (4. 5).

Literature Cited

Childress, J. J. 1995. Are there physiological and biochemical ad-
aptations of metabolism in deep-sea animals? Tree 10: 30-36.

2. Childress, J. J., and T. J. Mickel. 1985. Metabolic rates of animals
from the hydrothermal vents and other deep-sea habitats. Bio/. Soc.
Wash. Bull. 6: 249-260.

3. Felsenstein, J. 1985. Phylogenies and the comparative method. Am.
Nat. 126: 1-25.

4 Seibel. B. A., E. V. Thuesen, J. J. Childress. and L. A. Gorodezky.
1997. Decline in pelagic cephalopod metabolism with habitat depth
reflects differences in locomotory efficiency. Biol. Bull. 192: 262-278.

5. Seibel, B. A., E. V. Thuesen, and J. J. Childress. 2000. Light-
limitation on predator-prey interactions: consequences for metabolism
and locomotion of deep-sea cephalopods. Biol. Bull. 198: 284-298.

6. Carlini. D. B., and J. E. Graves. 1999. Phylogenetic analysis of
cytochrome c oxidase I sequences to determine higher-level relation-
ships within the coleoid cephalopods. Bull. Mar. Sci. 64: 57-76.

7. Garland. T. J., P. H. Harvey, and A. R. Ives. 1992. Procedures for
the analysis of comparative data using phylogenetically independent
contrasts. Syst. Biol. 41: 18-32.

8. Clarke, M. R. 1988. Evolution of buoyancy and locomotion in
recent cephalopods. Pp. 203-213 in The Mollusca. Vol. 12: Paleon-
tology and Neontology of Cephalopods. M. R. Clarke and E. R.
Trueman. eds. Academic Press. San Diego, CA.

9. Young, R. E., and M. Vecchione. 1996. Analysis of morphology to
determine primary sister-taxon relationships within coleoid cephalo-
pods. Am. Malacol. Bull. 12: 91-112.

10. Boucher-Rodoni, R., and L. Bonnaud. 1996. Biochemical and

molecular approach to cephalopod phylogeny. Am. Malacol. Bull. 12:
79-85.

11. Bonnaud. L., R. Bouchvr-Rodoni, and M. Monnerot. 1997. Phy-
logeny of cephalopods inferred from mitochondria] DNA sequences.
Mol. Phylogenet. Evol. 1: 44-54.

12. Carlini, D. B., K. S. Reece, and J. E. Graves. 2000. Actin family
gene evolution and the phylogeny of coleoid cephalopods (Mollusca:
Cephalopoda). Mol. Biol. Evol. 17: 1353-1370.

13. Voss, N. 1988. Evolution of the cephalopod family Cranchiidae
(Oegopsida). Pp. 293-314 in The Mollusca, Vol. 12: Paleontology and
Neontology of Cephalopods. M. R. Clarke and E. R. Trueman, eds.
Academic Press. San Diego. CA.

14. Young, R. E., and R. F. Harman. 1988. Phylogeny of the "Eno-
ploteuthid" families. Smithson. Contrib. Zool. 586: 257-270.

15. Segawa, S. 1995. Effect of temperature on oxygen consumption of
juvenile oval squid, Sepioteuthis lessoniana. Fish. Sci. 61: 743-746.

16. O'Dor, R. K., and M. J. Wells. 1987. Energy and Nutrient Flow in
Cephalopod Life Cycles, P. R. Boyle, ed. Academic Press, London. Pp.
109-133.

17. DeMont, M. E., and R. K. O'Dor. 1981. The effects of activity,
temperature and mass on the respiratory metabolism of the squid. I/lex
il/ecebrosus. J. Mar. Biol. Assoc. UK 64: 535-543.

18. Wells, M. J., and J. Wells. 1985. Ventilation and oxygen uptake by
Nautilus. J. E\p. Biol. 118: 297-312.

19. Purvis, A., and A. Rambaut. 1995. Comparative analysis by inde-
pendent contrasts (CAIC): An Apple Macintosh application for ana-
lyzing comparative data. Com/nit. Appl. Biosci. 11: 247-251.

20. Hochachka, P. W., K. B. Storey, and J. Baldwin. 1975. Squid
muscle citrate synthase: Control of carbon entry into the Krebs cycle.
Com/'. Biochem. Physiol. 52B: 193-199.

21. Baldwin, J. 1982. Correlations between enzyme profiles in cepha-
lopod muscle and swimming behavior. Pac. Sci. 36: 349-356.

22. Ballantyne, J. S., P. W. Hochachka, and T. P. Mommsen. 1981.
Studies on the metabolism of the migratory squid, Loligo opalescens:
enzymes of tissues and heart mitochondria. Mar. Biol. Lett. 2: 75-85.

23. Young, R. E. 1975. Transitory eye shapes and the vertical distribu-
tion of two midwater squids. Pac. Sci. 29: 243-255.

24. Young, R. E. 1975. Leachia pacifica (Cephalopoda, Teuthoidea):
Spawning habitat and function of the brachial photophores. Pac. Sci.
29: 19-25.

25. Voss, N. A., and R. S. Voss. 1983. Phylogenetic relationships in the
cephalopod family Cranchiidae (Oegopsida). Malacologia 23: 397-
426.

26. Seapy, R. R., and R. E. Young. 1986. Concealment in epipelagic
pterotracheid heteropods (Gastropoda) and cranchiid squids (Cepha-
lopoda). J. Zool. Lond. 210: 137-147.

27. Johnsen, S., and E. A. Widder. 1998. Transparency and visibility
of gelatinous zooplankton from the Northwestern Atlantic and Gulf of
Mexico. Biol. Bull. 195: 337-348.

28. Swofford, D. L. 1998. PAUP*: Phylogenetic Analysis Using Parsi-
mon\ ( * and Other Methods}. Version 4.0. Sinauer. Sunderland, MA.

Reference: Biol. Bull. 201: 6-16. (August 2001)

Signaling via Water Currents in Behavioral

Interactions of Snapping Shrimp

(Alpheus heterochaelis)

JENS HERBERHOLZ1 * AND BARBARA SCHMITZ2

1 Georgia State University, Department of Biology, P.O. Box 4010, Atlanta, Georgia 30302; and
2Lehrstuhl fur Zoologie, TU Miinchen, Lichtenbergstr. 4. 85747 Garching, Germany

Abstract. The snappping shrimp Alpheus heterochaelis
produces a variety of different water currents during in-
traspecific encounters and interspecific interactions with
small sympatric crabs (Eurypanopeus depressus). We stud-
ied the mechanisms of current production in tethered shrimp
and the use of the different currents in freely behaving
animals. The beating of the pleopods results in strong pos-
teriorly directed currents. Although they reach rather far,
these currents show no distinctions when directed toward
different opponents. Gill currents are produced by move-
ments of the scaphognathites (the exopodites of the second
maxillae) and can then be deflected laterally by movements
of the exopodites of the first and second maxillipeds. These
frequent but slow lateral gill currents are most probably
used to enhance chemical odor perception. The fast and
focused, anteriorly directed gill currents, however, represent
a powerful tool in intraspecific signaling, because they reach
the chemo- and mechanosensory antennules of the opponent
more often than any other currents and also because they are
produced soon after previous contacts between the animals.
They may carry chemical information about the social status
of their producers since dominant shrimp release more
anterior gill currents and more water jets than subordinate
animals in intrasexual interactions.

Introduction

Alpheus heterochaelis of the family Alpheidae (Deca-
poda, Caridea) is one of the largest snapping shrimp, reach-
ing a body length of up to 55 mm. It shows a large, modified

Received 27 November 2000; accepted 10 April 2001.
* To whom correspondence should be addressed. E-mail: biojhh@
panther.gsu.edu

snapper claw on one (left or right) side and a small pincer
claw on the other side in both sexes (Williams, 1984). The
snapper claw allows the animals to produce an extremely
fast water jet (of up to 25 m/s; Versluis et al., 2000) by rapid
claw closure after cocking the claw in the open position
(Ritzmann. 1974). The high velocity of the water jet results
in a pressure drop below vapor pressure that causes a
cavitation bubble to grow to a size of about 3.5 mm in front
of the snapper claw. The collapse of this bubble (and not as
previously supposed the mechanical contact of both claw
surfaces) causes the extremely loud (up to 215 dB re 1 ;u,Pa
at 1 m distance; Schmitz, 2001) and short (about 500 ns)
snapping sound (Versluis et al., 2000). The strong effect of
the water jet and the cavitation bubble collapse can be seen
during interspecific encounters. Small prey (e.g., worms,
goby fish, or shrimp) can be stunned or even killed by the jet
(MacGinitie, 1937; MacGinitie and MacGinitie, 1949; Mor-
ris et al., 1980; Suzuki, 1986; Downer, 1989), and interspe-
cific opponents (e.g., small sympatric crabs, Eurypanopeus
depressus) can be injured at interaction distances of on
average 3 mm (Schultz et al., 1998). Toward conspecifics
the water jet was not observed to cause any damage but
functions as a communicative signal (Herberholz and
Schmitz, 1999), both opponents ensuring an interaction
distance of on average 9 mm (Schmitz and Herberholz,
1998), which is far enough away from danger caused by
implosion of the cavitation bubble. This hydrodynamic sig-
nal is analyzed by the receiving shrimp predominantly with
the help of mechanosensory hairs on the snapper claw, and
may contain information about the strength, motivation, and
sex of the snapper (Herberholz and Schmitz, 1998; Herber-
holz, 1999).

The still rather small interaction distance of less than 1
cm in agonistic encounters between two snapping shrimp

WATER CURRENTS IN SNAPPING SHRIMP

also favors the exchange of chemical signals between the
opponents. The literature on chemical orientation and com-
munication in snapping shrimp is limited: Hazlett and Winn
( 1962) tested aggressive and defensive responses of Svnal-
pheus lu'inphilli to crushed male or female extract, and
Schein (1975) and Hughes ( 1996) investigated the choice of
Alpheus heterochaelis toward extracts of male or female
water in Y-maze experiments without clear-cut results. On
the other hand, ablation of the chemosensitive antennules in
Alpheus edwardsii strongly reduced pair formation and sex
recognition, which may be due to impeded distant or contact
chemoreception since the pairing frequency remained high
when only the antennae were ablated (Jeng, 1994).

The importance of olfactory signals during hierarchy
formation was shown in male American lobsters (Karavan-
ich and Atema. 1998a). In these experiments, the recogni-
tion of urine-carried chemical signals, which were received
by the antennules, allowed the subordinate animal to avoid
the familiar dominant shrimp, and therefore reduced the
duration and aggression of fights. The exchange of chemical
signals is also assumed to play a major role in individual
recognition and memory in male and female Homarus
americamts (Karavanich and Atema, 1998b; Berkey and
Atema, 1999). In lobsters, urine is released through a paired
set of nephropores on the ventral sides of the basal segments
of the second antennae (Parry, 1960). Agonistic behavior in
lobsters causes an increase in the probability and volume of
urine release (Breithaupt et al., 1999). The released urine is
then carried by the powerful anteriorly directed gill currents
and may therefore transfer chemical information from one
animal to another (Atema, 1985). In recent studies (Zulandt
Schneider et al., 1999; Zulandt Schneider and Moore.
2000), chemical cues were also described as an important
source for recognition of the dominance status or stress
condition of conspecifics in another crustacean, the red
swamp crayfish (Procambarus clarkii).

In light of these examples, a similar mechanism of chem-
ical signal exchange via gill currents in snapping shrimp
seems likely. We cannot, however, exclude the possibility
that the animals also exchange hydrodynamic signals. In
fact, it has been shown that the antennules of crayfish
(Mellon, 1996) and lobsters (Guenther and Atema, 1998;
Weaver and Atema, 1998) are equipped with both chemical
and mechanosensory receptors, and detailed morphological
studies of antennule sensory hairs favor the same situation
in snapping shrimp (Schmitz, unpubl. obs.). Therefore,
snapping shrimp may also perceive hydrodynamic stimuli
as well as chemical stimuli with their antennules. Previous
studies (Herberholz and Schmitz, 1998. 1999) have shown
that the transfer of hydrodynamic signals is realized by the
powerful water jet that is formed by rapid closure of the
large claw. In contrast, the much weaker gill currents appear
to be more suitable for transferring chemical information.

Suspended plastic particles were successfully used to

visualize and quantify biological flow fields in lobsters and
crayfish in a series of experiments by Breithaupt and Ayers
( 1996, 1998). Small floating particles of the same density as
seawater were added to the aquarium water and illuminated
in a horizontal or vertical plane in the vicinity of a tethered
animal. Flow fields were then analyzed by tracking individ-
ual particles. It was shown that both lobsters and crayfish
produce a great variety of flow fields by using the exopo-
dites of the maxillipeds and by fanning the pleopods. The
latter was also discussed with respect to chemical commu-
nication: male American lobsters commonly fan their pleo-
pods at the second entrance of their shelter, thus creating a
strong current that may contain chemical information about
the female positioned at the first entrance (Atema, 1985,
1988). The pleopod fanning frequencies in males correlate
with the frequencies of females checking the shelter. The
existence of pheromones that control female choice and
molting as well as male aggression was therefore assumed
(Cowan and Atema, 1990; Atema, 1995; Bushman and
Atema. 1997).

The possible exchange and use of different water currents
during agonistic encounters has rarely been studied; but see
Rohleder and Breithaupt (2000) for a preliminary study in
the crayfish Astacus leptodactylus. To test the possibility
that snapping shrimp use guided water currents as signals,
we visualized and analyzed all water currents that the
shrimp produced during their encounters with conspecifics
of the same or different sex and in encounters with sympa-
trically living mud flat crabs (Eurypanopeus depressus).

Materials and Methods

We analyzed the behavior of 12 adult specimens of
Alpheus heterochaelis. a species of snapping shrimp (6
males, 6 females; body size: 3.9 ± 0.4 cm. mean ± SD).
Each animal was tested in an encounter with a conspecific
of equal size of either the same or different sex, as well as
in an encounter with a small crab (Eurypanopeus depressus;
mean length and width of carapace: 1.6 ± 0.2 X 1.2 ± 0.2
cm, mean ± SD). All animals were caught in waters of the
Gulf coast of Florida at the Florida State University Marine
Laboratory near Panacea. Prior to the experiments the ani-
mals were labeled with small numbers designated for mark-
ing queen bees and were kept individually in perforated
plastic containers ( 1 1 X 11 X 15 cm) containing gravel and
oyster shells for shelter. The containers were placed within
a large tank (90 X 195 X 33 cm) with 330 1 of circulating
filtered seawater (salinity: 23%c^28%o; temperature: 22°-
23°C). Proteins were removed from the water, and pH.
carbonate, oxygen, CO2. and NO3 were regularly con-
trolled. The shrimp were exposed to an illumination cycle of
12 h light/ 12 h dark and fed frozen shrimp, fish, or mussels
three times a week.

For visualization of the different water currents, we pre-

J. HERBERHOLZ AND B. SCHMITZ

pared the aquarium water (temperature: 22°-24"C, water
level: 5 cm) with small, floating plastic panicles (ABS-
particles, Bayer, Leverkusen, diameter: 500-710 jum; spe-
cific weight: 1.03 kg/1). The aquarium (30 X 24 X 24 cm;
floor covered with black cloth to facilitate walking) was
positioned on a platform isolated from vibrations (Breit-
haupt et at., 1995). At the level of the interacting animals,
the seawater was illuminated from one side by a slide
projector holding a slide with a thin horizontal slit. Before
each experiment fresh seawater and particles were added,
and two animals (two snapping shrimp or one snapping
shrimp and a crab) were placed in the aquarium for 10 min
for acclimatization: the animals were separated by an
opaque divider to prevent visual, tactile, and directed-chem-
ical contact. After the partition was removed, all interac-
tions between the animals during the following 20 min were
videotaped from above (camera: Panasonic AG 455; video
recorder; Panasonic AG 7355; monitor: Sony Trinitron).
The reflexive characteristics of the suspended particles then
allowed a precise tracking using standard video-frame anal-
ysis.

Each experiment (interactions between two snapping
shrimp of the same or different sex or between a snapping
shrimp and a crab) was characterized by the number of
physical contacts between the opponents, regardless of their
duration and strength, as well as by the number of water
jets. Three different water currents were characterized, in-
cluding a lateral gill current, an anterior gill current, and a
pleopod current (Fig. la). The pleopod current was mea-
sured only when the shrimp was not in locomotion, because
this current is also likely to be used in supporting the
animal's walking. Moreover, no current was included in our
analysis unless the single-frame video analysis gave clear
evidence that it had moved two or more plastic particles.
The following parameters were evaluated for all visualized
water currents: frequency, duration (time between onset of
movement of the first floating particle and end of movement
of the last particle), range (total distance covered by an
identified particle due to a certain current: possibly under-
estimated when the current hit an opponent or an aquarium
wall), velocity and target of the currents, their potential to
transfer chemical information (i.e.. entering the area of
chemical perception at the receiver's side), the temporal
correlation between currents and previous physical contacts,
and the correlation between produced currents and water
jets in winners and losers during intrasexual interactions. To
determine a winner or loser, we counted the number of
aggressive acts and the number of submissive acts after each
physical contact between the conspecitic opponents
throughout the encounter. Aggressive acts include behav-
iors such as approach, aggressive stance, and grasping and
opening of the claws. Submissive acts include moving back-
wards and turning and tail flipping away from the opponent.
These definitions are largely adopted from Nolan and

Salmon (1970). In 11 out of 12 experiments, one animal
produced more aggressive acts and fewer submissive ones
than its opponent and was therefore determined to be the
winner while the opponent was determined to be the loser.

Statgraphics Plus 6.0 (Manugistics Group, Inc.) and
SPSS 6.0.1. (SPSS Science Software GmbH) were used for
statistics. Mean and standard deviation were calculated for
each variable of interest for each tested individual, and only
one value per individual (grand mean) is included in each
statistical test. The behavior of the respective opponents
(male and female snapping shrimp, and crabs) was not
analyzed and is not included in our results (exception: data
presented in Fig. 7). If not otherwise stated, the Friedman
rank test for repeated measurements (sample size >2) or the
Wilcoxon rank test (sample size = 2) were used, and values
with P < 0.01 and P < 0.05 are indicated in the text. We
used nonparametric statistical tests because most of the data
did not fulfill the requirements for the use of parametric
tests i.e., normality or equal variance.

To gain more insight into the mechanism of gill current
production and redirection, two snapping shrimp were teth-
ered upside down in a small petri dish filled with seawater
and floating plastic particles, and the activity of the different
mouth parts, which produced or deflected the currents, was
videotaped using a CCD camera (Sony XC-77CE) mounted
on a binocular microscope with high magnification. In ad-
dition, small drops of black ink (Brilliant Black 4001.
Pelikan) were placed between the third and fourth walking
legs of these shrimp as well as of animals tethered dorsal
side up to a vertical holder and standing on a platform so
that the gill currents could be visualized. (Fig. Ib).

Results

Visualization of water currents in tethered shrimp

A unique feature of snapping shrimp is the production of
an extremely rapid water jet by fast closure of a specialized
snapper claw. Apart from this water jet. the snapping shrimp
Alpheus heterochaelis is able to produce four kinds of water
currents (Fig. 1), which can be subdivided into two main
categories. Fanning of the pleopods causes a strong, poste-
riorly directed pleopod current, and a gill current is pro-
duced by rhythmically beating the scaphognathites as re-
vealed by our visualization experiments in two tethered
shrimp. Beating of the scaphognathites produces a depres-
sion in the gill chamber; water is therefore sucked into this
chamber and subsequently released anteriorly through two
small openings in the carapace. This "normal" gill current
can be visualized with ink in tethered animals, but it is too
slow and weak to move floating particles and was therefore
not analyzed during encounters of snapping shrimp and
their opponents. It can, however, be accelerated and de-
flected into a lateral gill current (see Fig. IB) by the
exopodites of the second and third maxillipeds. The exopo-

WATER CURRENTS IN SNAPPING SHRIMP

'normal" gill current

pleopod current

lateral gill current

antennule

anterior gill current

Figure 1. (A) Schematized drawing (lateral view) of a snapping shrimp modified after Kim and Abele
( 1988) showing four different water currents (gray arrows): the "normal" gill current, the lateral gill current, the
anterior gill current, and the pleopod current. Black arrows show the direction of water entering the gill chamber.
(B) Frontal view of an A/pheiis helerochaelis snapping shrimp, tethered to a vertical holder by means of a plastic
nut glued to the carapace and standing on a textile platform. Black ink was placed with a syringe between the
third and fourth left pereiopods (see ink trace) to visualize the gill currents. The shrimp is fanning the exopodites
of the right second and third maxillipeds. thus producing an ink-stained lateral gill current to the right.

dites of the first maxilliped do not participate in this process.
Fanning of the left exopodites results in acceleration and
deflection of the released gill current to the left side, and
fanning of the right exopodites results in deflection to the

right side. Tethered snapping shrimp never beat the exopo-
dites of both sides simultaneously, and this was also never
observed during interactions in which the illuminated par-
ticles were directed to only one side at a time. Interestingly,

J. HERBERHOLZ AND B. SCHMITZ

D 1-gc

a-gc

homo hetero

type of interaction

inter

Figure 2. Frequency of three different water currents (1-gc, lateral gill
current, a-gc, anterior gill current, pc, pleopod current) produced by Al-
pliens heterochaelis snapping shrimp in interactions with another shrimp of
the same sex (homo), of different sex (hetero), and with a Eurypanopeus
depressus crab (inter). Grand means and standard deviations for 12 snap-
ping shrimp each are shown. Significant differences within interaction
types with P < 0.01 are indicated by two asterisks (**).

a (fast) anterior gill current was restricted to encounters of
freely moving animals; it could not be elicited in tethered
shrimp. Its production obviously requires physical, chemi-
cal, or visual contact between the animals. As a result, we
were not able to analyze the producing mechanism; that is,
we did not identify the involved mouth parts.

General characteristics of released water currents

Encounters between two snapping shrimp of different sex
(hetero) are characterized by a significantly higher number
of physical contacts (23.9 ± 8.3, /; = 287; P < 0.01) than
seen in encounters between two shrimp of the same sex
(homo; 13.8 ± 6, n = 165), or between a snapping shrimp
and a crab (Eurypanopeus depressus) (interspecific; 12.7 ±
5.3. n = 157). On the other hand, snapping (water jet
production) of the tested shrimp is significantly increased
after a contact with a crab (38% ± 16<7r; P < 0.01) when
compared to snapping after hetero and homo contacts (5%
± 4% and 11% ± 11%, respectively).

These differences in mind, we first evaluated the number
of water currents (lateral gill currents, anterior gill currents,
and pleopod currents) in each experiment. Figure 2 shows
that there are no essential differences between interaction
types (homo, hetero, or interspecific). Within each interac-
tion type, however, the number of lateral gill currents sig-
nificantly (P < 0.01 ) exceeds that of anterior gill currents as
well as that of pleopod currents. In addition, in interspecific
encounters with a crab, the frequency of anterior gill cur-
rents is significantly lower than the frequency of pleopod
currents (P < 0.01).

The duration of the different water currents (Fig. 3A)
tends to be longest for lateral gill currents, with no signif-
icant differences regarding the type of the opponent. The
duration of anterior gill currents is generally shorter, with
similar values in intraspecific interactions, yet almost twice
as long as in interactions with a small crab. Anterior gill
currents in interspecific encounters are significantly shorter
in duration than lateral gill currents (P < 0.05). Pleopod
currents, in contrast, reveal very consistent values for all
types of interactions.

Figure 3B shows the range of the different currents in all
interaction types. Regardless of the opponent, the snapping
shrimp tend to produce lateral gill currents with small
ranges. Anterior gill currents generally cover larger dis-
tances in intraspecific interactions, whereas the mean value
is reduced in interactions with a crab. The most powerful
current is the pleopod current, which covers long distances
in all interaction types. Range differences within interaction
types are significant at P < 0.05 and P < 0.01, respectively.

The velocity of the water currents during the first 120 ms
(6 video frames) was evaluated for 10 examples for each
current and interaction type (Fig. 3C). There are no signif-
icant differences in the velocities within and between dif-
ferent types of interactions. The lateral gill current shows
the slowest velocities in all encounters. The anterior gill
current and the pleopod current show similar values and are
both more powerful than the lateral gill current. Initial
velocities are higher, but their analysis has not proved
satisfactory because of the standard video time resolution of
20 ms (50 frame/s).

Temporal relation of water currents to physical contact

Figure 4 compares the frequency of water currents that
were elicited within 10 s after a physical contact between
the opponents with those that were "spontaneously" pro-
duced— that is, emitted more than 10 s after a preceding
contact. As shown in Figure 4A. in all interaction types the
lateral gill current is significantly more often produced
spontaneously than following a physical contact (P < 0.01 ).
In homo interactions it occurs in only 6.2% of all cases (n =
10 of 162) shortly after a contact. During hetero interactions
this current is elicited by a contact in 11.5% of all cases
(n = 2\ of 183); in interactions with a crab, the lateral gill
currents occur within 10 s after a contact in only 8.5% of all
cases (n = 13 of 153).

The analysis of the anterior gill current reveals a com-
pletely different frequency pattern, with more elicited cur-
rents than spontaneous ones (Fig. 4B). In homo interactions
the anterior gill current is produced in 65.5% of all cases
(/; = 19 of 29) within 10 s after a preceding contact.
Similarly, in hetero interactions this gill current is elicited
by a contact in 62.5% of all cases (n = 15 of 24). Finally,
during interactions with a crab, anterior gill currents are

WATER CURRENTS IN SNAPPING SHRIMP

1
u

-a

10
8
6
4
2
0

D 1-gc

a-gc

homo hetero inter

type of interaction

Dl-gc

a-gc

*•

homo hetero inter

type of interaction

Dl-gc

a-gc

homo

hetero

inter

type of interaction

Figure 3. Duration (A), range (B), and velocity (C) of the lateral gill
current (1-gc). the anterior gill current (a-gc), and the pleopod current (pc)
in interactions of two snapping shrimp of the same sex (homo), of different

released within 10 s after a contact in 78.6% of all cases
<H = 11 of 14).

In contrast, the pleopod current, like the lateral gill cur-
rent, is significantly more often (P < 0.01) produced with-
out an immediately preceding contact in all types of inter-
actions (Fig. 4C). During homo interactions we observed
only 7.7% of pleopod currents within 10 s after the last
contact (n = 4 of 52). In hetero interactions this current is
elicited in 16.7% of all cases (n = 8 of 48) by a preceding
contact, and in interspecific interactions there are 13.0% of
pleopod currents shortly after a previous contact (n = 7 of
54).

Possible chemosensory information transfer
by water currents

If any of the water currents were used to transfer chem-
ical information, one would expect them to be directed
toward the chemoreceptive antennules of the opponent. We
therefore evaluated the number of currents that reached the
area between the opponents' claws — that is, an area mostly
covered by the flicking antennules. This was possible by
analyzing the video sequences and identifying the area of
particle dispersion with respect to the animals' position. In
fact, only the anterior gill current seems qualified to fulfill
the function of possible information transfer (Fig. 5).

In all types of interactions, the mean number of lateral gill
currents that miss the antennules is significantly higher (P <
0.01) than the mean number of those hitting the target (Fig.
5A). In homo interactions the lateral gill current reaches the
antennule area in only 0.6% of the cases (/; = 1 of 162).
During hetero interactions lateral gill currents are never
directed toward the opponent's antennules, but hit other
targets (n = 183). In interactions with a crab, the snapping
shrimp produce 0.7% (H = 1 of 153) of lateral gill currents,
which could possibly transfer chemical information.

In comparison, a higher percentage of anterior gill cur-
rents reaches the antennule area in all interaction types (Fig.
5B). During homo interactions the anteriorly projected gill
current reaches the antennules of the opponent in 35.1% of
all cases (n = 10 of 28). In hetero interactions the percent-
age (66.7%, n = 16 of 24) of anterior gill currents directed
toward the antennules is even higher than that of undirected
anterior gill currents. During interspecific interactions the
snapping shrimp projects 35.7% anterior gill currents to-
ward the antennules of the crab (// = 5 of 14).

The frequency pattern for pleopod currents is similar to

sex (hetero). and of a snapping shrimp and a crab (inter). Grand means and
standard deviations for 12 shrimp are shown in A and B; means and
standard deviations of the velocity during the first 120 ms of 10 currents
each are shown in C. A significant difference within an interaction type
with P < 0.05 is indicated by one asterisk (*) and with P < 0.01 by two
asterisks (**).

12 J. HHRBtRHOI

• l-gc(<10s) Dl-gcOlOs)

O C •

25 -

c/l
*i nA

fc z°

1 '5

**T ** **T

aj 10

s ,

3 5

, •!•

homo hetero inter

type of interaction

a-gc (< 10s) D a-gc (> 10s)

homo hetero

type of interaction

inter

i/i

1 20 -i

o 1 5

D
40

3
C

1-gc (ha)

D 1-gc (ot)

homo hetero inter

type of interaction

a-gc (ha)

D a-gc (ot)

homo hetero inter

type of interaction

pc(<10s)

D pc(>10s)

urrents

s DC C

** **

0 O

4-1

1— A

<L> 4
JO

1 9

c 2
0

J.
hoi

, » -

no hetero inter

type of interaction

4. Mean number ol' lulcriil gill currents (A), anterior gill cur-
rcnls (If), and |iloo|ioil cuiu-nls (C'l \\ithin 10 s (lilack tolunin-.) cu more
than 10 s (white columns) alter a physical contact between the opponents
in inlet, K lions ol Iwo siuppmg sin imp ol ilie same sex (homol, ot tlitTercnt
se\ (heleio). anil ol a snapping shrimp ami a crab (inter), (iraiul means and
standard deviations for 12 shrimp are shown. Significant differences with
/' < 0.01 are indicated In luo asterisks ('*).

*-•

u-
OJ

I
3
C

pc (ha)

D pc (ot)

homo hetero

type of interaction

inter

' 5. Mean number of lateral gill currents (A), anterior gill currents
(B). and pleopod currents (C) hitting the antennules of the opponent (black
columns, ha) or reaching other targets (white columns, ot) in interactions of
two snapping shrimp ol the same se\ (homo), of different sex (hetero), and of
a snapping shrimp and a crab (inter). Grand means and standard deviations for
12 shrimp are shown. A significant difference with P < 0.05 is indicated by
tine asterisk (*) and with P < 0.01 by two asterisks (**).

WATER CURRENTS IN SNAPPING SHRIMP

homo

hetero

number of a-gc

to ^ O^ 00 C

1 /

"/i

•

0 2 4 6 8 10
number of jets

•

3246
number ot jets

i i

Figure 6. Correlation between the number of water jets and the num-
ber of anterior gill currents produced in interactions of two snapping
shrimp (A) of the same sex (homo; Spearman's coefficient of rank corre-
lation r, = 0.9, P < 0.01). (B) of different sex (hetero). and (C) of a
snapping shrimp and a crab (inter). Data of \2 shrimp each — some data
points overlap.

that of lateral gill currents: the undirected currents signifi-
cantly exceed the antennule-directed ones in each interac-
tion type (P < 0.05 or 0.01, respectively; Fig. 5C). In homo
interactions an average of only 1 1.5% (n = 6 of 52) of all
pleopod currents are projected towards the chemoreceptive
antennules. and during hetero interactions 16.7% (/; = 8 of
48) of all pleopod currents reach the antennule area. Finally.
in interspecific interactions no pleopod current is aimed
towards the antennules of the crab, but all (;i = 54) are
directed elsewhere.

Anterior gill currents and water jets

In view of the prominent role of the anterior gill current
with respect to its timing after a physical contact and the
increased possibility of chemosensory information transfer,
we tested the correlation between these gill currents and
emitted water jets (Fig. 6). As mentioned before, in com-
parison to intraspecific interactions, encounters with crabs
are characterized by an increased number of water jets and
a reduced number of anterior gill currents (Fig. 6C). In
addition, more water jets are emitted in homo interactions
between snapping shrimp (Fig. 6A) than in hetero encoun-
ters (Fig. 6B). Thus, the number of anterior gill currents
significantly increases with an increasing number of water

jets only in interactions between two snapping shrimp of the
same sex (Spearman rank correlation coefficient: >\ = 0.9,
P < 0.01: Fig. 6A). This is not the case in interactions
between two shrimp of different sex d\ = 0.5, P > 0.05),
though a noticeable trend is shown and the overall low
number of water jets may have prevented a significant
result. An even lower degree of correlation is seen in
interactions with a crab (rv = 0.4, P > 0. 1 ).

As shown in Figure 7, winners of homo interactions (as
defined by aggressive and submissive acts — see Materials
and Methods) not only produce a significantly higher mean
number of water jets (N = II, P < 0.01) but also a
significantly higher mean number of anterior gill currents
than losers produce (N = 1 1; P < 0.01 ).

Discussion

Snapping shrimp (Alpheus hetemcluielis) produce two
main water currents, a strong posteriorly directed pleopod
current and an anteriorly directed gill current. We show that
the "normal" anteriorly directed gill current can be modified
and redirected into a lateral and a fast anterior gill current.
The production of the latter is restricted to social interac-
tions, in which it represents a powerful tool for chemical
signaling. Moreover, the use of the fast anterior gill currents
varies for the winners and losers of individual encounters.

Mechanisms of gill current production

Our experiments in tethered snapping shrimp show that
water is sucked into the gill chamber due to a depression
elicited by the beating scaphognathites (Fig. 1A). A "nor-
mal" gill current is then released anteriorly with low veloc-
ity through two small openings of the carapace. Once the
left or right expodites of the second and third maxillipeds
start fanning, the current is accelerated and deflected later-
ally to that side (Fig. IB). As previously described in

winner

loser

Figure 7. Frequency of water jets (jets, black columns) and anterior
gill currents (a-gc, white columns I lor winners and losers in interactions of
two snapping shrimp of the same sex. The significant differences between
winners and losers with P ^ 0.01 are indicated by two asterisks (**).

J. HERBERHOLZ AND B. SCHMITZ

lobsters (Homarus americanus), the exopodites of the first
maxillipeds do not contribute to these lateral gill currents in
snapping shrimp, whereas in crayfish (Procambarus clarkii)
these appendages are also involved (Breithaupt. 1998).

The production mechanism of the fast anterior gill current
remains unclear, since this behavior obviously requires
physical, chemical, or visual contact during intra- or inter-
specific encounters of snapping shrimp, and thus was never
seen in tethered animals. From our knowledge about the
lateral gill current, we assume that the fast anterior gill
current is created by high-frequency beating of the scapho-
gnathites without contribution of the exopodites of the sec-
ond and third maxillipeds. Since it is difficult to video-
record the mouth parts with high magnification during
social interactions, we are currently testing other methods of
monitoring scaphognathite beating frequencies during en-
counters to verify this hypothesis.

Role of the fast anterior gill current during social
interactions

The analysis of the fast anterior gill current revealed the
most surprising and interesting results. Although anterior
gill currents were observed and well described in lobsters
(Atema, 1985. 1995) and crayfish (Breithaupt, 1998), we
found decisive differences in snapping shrimp. First of all,
Alpheus heterochaelis produces different types of anterior
gill currents. The "normal" anterior current is a slow, weak
release of water, which was sucked through the gill cham-
ber, as opposed to the fast, strong, anteriorly directed gill
current, which occurs during social interactions. The pro-
duction of the fast anterior gill current is rare (Fig. 2) but
strongly linked to previous contacts with a conspecific or a
crab (Fig. 4B). Among the observed currents, only the fast
anterior current is created shortly after a preceding contact,
regardless of the type of opponent. In fact, this current never
occurred before the first contact. Moreover, we show that
only this current is suited to transfer chemical information
towards the other animal (Fig. 5B): it reaches the antennules
of the opponent in nearly 50% of all cases.

Of all analyzed currents, only the fast anterior gill current
shows some peculiarities with respect to the shrimps' op-
ponent. The number, duration, and range is smaller in en-
counters with a crab than in interactions with conspecifics
(Figs. 2, 3). We assume that the shrimp collect information
about the genus of their opponent and reduce the effort to
communicate accordingly, if it is a crab.

Role of lateral gill currents during social interactions

During social interactions between snapping shrimp and
conspecifics of the same or different sex as well as during
interactions with small crabs, the lateral gill currents are
most prominent and significantly outnumber all other ob-
served currents (i.e., pleopod currents and fast anterior gill

currents; Fig. 2). Moreover, they are produced for long
intervals but have a short range and a low velocity (Fig. 3).
They are barely elicited by physical contact (Fig. 4A) and
hardly ever reach the antennules of their opponents (Fig.
5A). These properties of the lateral gill currents do not
change with different opponents but appear to result from a
stereotyped form of production. Thus, obviously lateral gill
currents are not predestinated to play a prominent role in
active (chemical) signaling between the animals.

Still, their function needs explanation. From our obser-
vations we conclude that the lateral gill current is used to
improve the shrimps' ability to sense possible odor signals
that occur at close distance. By redirecting the "normal" gill
current, the shrimp refreshes the area around its chemical
receptors from its own smell (released by the slow and
permanent gill current) and thereby improves the detection
of the chemical surrounding. This idea is supported by our
knowledge that Alpheus heterocliaelis naturally inhabits
small, oyster-shell-covered areas with little water flow and
that individuals of the species appear to be rather stationary
within that area (Herberholz and Schmitz, pers. obs.). The
lateral gill current produced by snapping shrimp seems to be
used to remove water from the area around the antennules
and to a much lesser extent to draw water toward that region
as proposed for the posteriorly or laterally redirected gill
currents of lobsters and crayfish (Atema, 1995; Breithaupt.
1998). In contrast to lobsters and crayfish, snapping shrimp
were never observed to fan simultaneously with appendages
on both sides. Instead, they beat the exopodites of one side
at a time, and there are no obvious movements of particles
from the opposite side toward the animal's anterior region.

Role of pleopod currents during social interactions

In lobsters (Homarus americanus), pleopod currents are
used for chemical (possibly pheromonal) communication
during courtship at a shelter (Atema. 1985. 1988. 1995;
Cowan and Atema, 1990: Bushman and Atema, 1997). The
snapping shrimp Alpheus heterocliaelis, in addition to using
its pleopods for locomotion and to provide an oxygen sup-
ply for attached eggs, uses them for shelter digging, fanning
the substrate (sand or muddy-sand) backward behind it
(Nolan and Salmon, 1970). These authors also mention
(pleopod) fanning as an aggressive act, with a shrimp vig-
orously beating its pleopods and directing a water current
posteriorly quite close to another shrimp. The frequency of
pleopod fanning is not noted by Nolan and Salmon (1970),
but the behavior was described to occur between two fe-
males at the entrance of a shelter. In our experiments, we
did not provide a shelter, and all shrimp were in the middle
of their molt cycle. In view of the finding that the actual
impact of pleopod currents in lobsters depends to a high
degree on the molt state of the animals as well as on their
readiness to mate (Cowan and Atema, 1990), these condi-

WATER CURRENTS IN SNAPPING SHRIMP

tions may have affected our results. Though pleopod cur-
rents were rather often produced (Fig. 2) and (in comparison
to gill currents) show an average duration, a large range, and
high velocity (Fig. 3). there is a lack of correlation with
previous contacts (Fig. 4C) and a low precision in hitting the
antennules of the opponent (Fig. 5C). There are hardly any
differences in the characteristics of these currents towards
different opponents. All this indicates that pleopod currents
are of little relevance for (chemical) signaling or commu-
nication among snapping shrimp and between shrimp and
sympatric crabs under our conditions.

A specialized gill current for chemical .signaling and
communication ?

The transfer of chemical signals between interacting lob-
sters (see e.g., Atema. 1995; Bushmann and Atema. 1997)
and crayfish (Breithaupt et al., 1999) has been described in
detail. In lobsters these signals can evoke long-term indi-
vidual recognition (Karavanich and Atema, 1998a, b), and
in crayfish they communicate dominance status or stress
condition (Zulandt Schneider et al., 1999; Zulandt Schnei-
der and Moore, 2000). In all cases, urine-borne signals were
assumed to be the source of chemical signaling (Breithaupt
et al., 1999; Breithaupt, pers. comm.). Since the urine is
released through a paired set of nephropores on the ventral
sides of the basal segments of the second antennae (Parry.
1960). it can be carried toward an opponent by the anterior
gill current. Moreover, agonistic behavior in catheterized
lobsters increases the probability and volume of urine re-
lease (Breithaupt et al., 1999).

In the present study we show for the first time that the
pattern of water current production actually changes with
respect to the social situation of an aquatic animal. Although
snapping shrimp have the ability to produce "normal" an-
terior gill currents, they create different, more powerful,
anteriorly directed gill currents shortly after contacting their
interaction partner. These elicited currents are then more
likely to reach the opponents' area of chemical perception.
The same may hold true for lobsters and crayfish, but their
currents have not yet been quantified during social interac-
tions. On the other hand, we still have to prove that the fast
anterior gill current in snapping shrimp actually carries
chemical signals toward the opponent. Although the data
presented favor this assumption, we cannot exclude the
possibility that hydrodynamic signals transferred by the gill
currents participate in the communication between the ani-
mals. Judging by their sensory equipment, snapping
shrimp — like crayfish (Mellon. 1996) and lobsters (Guen-
ther and Atema, 1998; Weaver and Atema. 1998) — are most
likely to perceive hydrodynamic stimuli as well as chemical
stimuli with their antennules (Schmitz, unpubl.). We plan to
test this possibility by deactivating the chemical receptors
only.

In any case, the production of the fast anterior gill current
may play a critical role during hierarchy formation in snap-
ping shrimp. We show that in intrasexual encounters the
numbers of water jets and anterior gill currents are posi-
tively correlated (Fig. 6) and that both are significantly
higher in the winner than in the loser (Fig. 7). In the present
study, winner and loser met in only a single 20-min exper-
iment. Preliminary experiments show that repetitive pairing
of winners and losers reduces the number of water jets and
anterior gill currents (Obermeier and Schmitz. unpubl.).
This supports the finding that these behaviors are most
probably correlated with dominance and social status in
snapping shrimp. Although the strength of the water jet
represents the strength of the animal (see Herberholz and
Schmitz, 1999), the signal transferred by the gill current
may then allow recognition of the sender. This, in turn, can
prevent two Alpheus heterochaelis shrimp of the same sex
from engaging in more severe fighting during subsequent
encounters, thus reducing the number of the "costly" water
jets.

Acknowledgments

We would like to thank Maren Laube for help in data
analysis. The experiments comply with the current laws of
Germany. Supported by a grant of the Deutsche For-
schungsgemeinschaft (Schm 693/5-2). The work of J.H.
was additionally supported by a NIH grant (NS26457) to
Donald H. Edwards at Georgia State University.

Literature Cited

Atema, J. 1985. Chemoreception in the sea: adaptations of chemorecep-
tors and behaviour to aquatic stimulus conditions. Soc. Exp. Biol. Svm/».
Ser. 39: 387-423.

Atema, J. 1988. Distribution of chemical stimuli. Pp. 29-56 in Sensory
Bio/ogv of Aquatic Animals, J. Atema, R. R. Fay, A. N. Popper, and
W. N. Tavolga, eds. Springer Verlag, New York.

Atema, J. 1995. Chemical signals in the marine environment: dispersal,
detection and temporal signal analysis. Proc. Nail. Acad. Sci. USA 92:
62-66.

Berkey, C., and J. Atema. 1999. Individual recognition and memory in
Homarus americanus male-female interactions. Biol. Bull. 197: 253-
254.

Breithaupt, T. 1998. Flow generation by specialized appendages in
lobsters and crayfish. Pp. 185-186 in BIONA-Repori 13. R. Blickhan.
A. Wisser, and W. Nachtigall. eds. Gustav Fischer Verlag. Stuttgart.

Breithaupt. T.. and J. Avers. 1996. Visualization and quantitative anal-
ysis of biological flow fields using suspended particles. Pp. 1 17-129 in
Zooplankton: Sensory Ecology and Physiology, P. H. Lenz, D. K.
Hartline. J. E. Purcell. and D. L. Macmillan, eds. Gordon Breach
Publishers. Amsterdam.

Breithaupt, T., and J. Avers. 1998. Visualization and quantification of
biological flow fields through video-based digital motion-analysis tech-
niques. Mar. Fi~esln\'. Bt'lniv. Phvsiol. 31: 55—61.

Breithaupt, T., B. Schmitz, and J. Tautz. 1995. Hydrodynamic orien-
tation of crayfish (Procambarus clarkii) to swimming fish prey.
J. Comp. Physio/. A 177: 481-491.

Breithaupt, T., D. P. Undstrom, and J. Atema. 1999. Urine release in

J. HERBERHOLZ AND B. SCHMITZ

freely moving catheterised lobsters (Homarus ainericanus) with refer-
ence to feeding and social activities. / Exp. Biol. 202: 837-844.

Bushman, P. J., and J. Atema. 1997. Shelter sharing and chemical
courtship signals in the lobster Hinnarus americwws. Can. J. Fish.
Aqnat. Set. 54: 647-654.

Cowan, D. F., and J. Atema. 1990. Moult staggering and serial monog-
amy in American lobsters, Homarus wnericaniis. Aniin. Behav. 39:
1199-1206.

Downer, J. 1989. Mir den Augen der Tiere. Teil: Laute tier Natitr. [video
recording). Universum. BBC and WDR (Westdeutscher Rundfunk).

Guenther, C. M., and J. Atema. 1998. Distribution of setae on the
Homarus ainericanus lateral antennular rlagellum. Biol. Bull. 195:
182-183.

Hazlett. B. A., and H. E. Winn. 1962. Sound production and associated
behavior of Bermuda crustaceans (Paniiliriis, Gonodactylus, Alpheus.
and Syna/phetis}. Crustaceana 4: 25-38.

Herherholz, J. 1999. Die Bedeutung hydrodynamischer Signale in intra-
und interspezifischen Interaktionen von Pistolenkrebsen (Alpheus het-
erochaelis). Ph.D. thesis. Technische Universitat Miinchen, Germany.

Herberholz, J., and B. Schmitz. 1998. Role of mechanosensory stimuli
in intraspecific agonistic encounters of the snapping shrimp (Alpheus
hetemchae/is). Biol. Bull. 195: 156-167.

Herberholz, J., and B. Schmitz. 1999. Flow visualization and high
speed video analysis of water jets in the snapping shrimp (Alpheus
helemchaelis). J. Co/up. Phvsiol. A 185: 41-49.

Hughes, M. 1996. The function of concurrent signals: visual and chem-
ical communication in snapping shrimp. Anim. Behav. 52: 247-257.

Jeng, M.-S. 1994. Effect of antennular and antennal ablation on pairing
behavior of snapping shrimp Alpheus edwardsii (Audouin). J. Exp.
Mar. Biol. Ecol. 179: 17I-17S.

Karavanich, C., and J. Atema. 1998a. Olfactory recognition of urine
signals in dominance fights between male lobsters, Homarus america-
nits. Behaviour 135: 719-730.

Karavanich, C., and J. Atema. 1998b. Individual recognition and mem-
ory in lobster dominance. Anim. Behav. 56: 1553-1560.

Kim, W., and L. Abele. 1988. The Snapping Shrimp Genus Alpheus
from the Eastern Pacific (Decapoda: Caridea: A/pheidae). Smithso-
nian Institution Press. Washington, DC. 1 19 pp.

MacGinitie, G. E. 1937. Notes on the natural history of several marine
Crustacea. Am. Midi Nat. 18: 1031-1036.

MacGinitie, G. E., and N. MacGinitie. 1949. Natural History of Ma-
rine Animals. McGraw Hill. New York. 473 pp.

Mellon, DeK. 19%. Behavioral responses of crayfish antennules to odor-
ant and hydrodynamic stimuli. Soc. Neurosci. Abstr. 22: 1406.

Morris, R. H., D. P. Abbott, and E. C. Haderlie. 1980. Intenidal
Invertebrates of California. Stanford University Press, Stanford. 690

PP
Nolan, A. N., and M. Salmon. 1970. The behaviour and ecology of

snapping shrimp (Crustacea: Alpheus heterochelis and Alpheus nor-

manni). Forma Function 2: 289-335.
Parry, G. 1960. Excretion. Pp. 341-366 in The Physiology of Crustacea,

Vol I: Metabolism and Growth. T. H. Waterman, ed. Academic Press,

New York.
Ritzmann, R. E. 1974. Mechanisms for the snapping behavior of two

alpheid shrimp. Alpheus californiensis and Alpheus heterochelis.

J. Comp. Physio/. 95: 217-236.

Rohleder, P., and T. Breithaupt. 2000. Visualizing chemical commu-
nication in crayfish. Zoology: Analysis of Complex Systems Suppl. Ill:

103: 72.
Schein, H. 1975. Aspects of the aggressive and sexual behaviour of

Alpheus helerochaelis Say. Mar. Behav. Phvsiol. 3: 83-96.
Schmitz, B. 2001. Sound production in Crustacea with special reference

to the Alpheidae. Pp. 521-533 in The Crustacean Nervous System, K.

Wiese. ed. Springer- Verlag, Berlin.

Schmitz, B., and J. Herberholz. 1998. Snapping behaviour in intraspe-
ciric agonistic encounters in the snapping shrimp (Alpheus heterochae-

lis). J. Biosci. 23: 623-632.
Schultz, S., K. \\uppermann, and B. Schmitz. 1998. Behavioural

interactions of snapping shrimp (Alpheus heterochaelix} with conspe-

cifics and sympatric crabs (Eurypanopeus depressus). Zoology: Anal-
ysis of Complex Systems Suppl. I: 101: 85.
Suzuki, D. 1986. Stunning sounds. Series: The Nature of Things, [video

recording). CBC (Canadian Broadcasting Corporation).
Versluis, M., B. Schmitz, A. von der Heydt, and D. Lohse. 2000. How

snapping shrimp snap: through cavitating bubbles. Science 289: 2114-

2117.
Weaver, M., and J. Atema. 1998. Hydrodynamic coupling of lobster

antennule motion to oscillatory water flow. Biol. Bull. 195: 180-182.
Williams, A. B. 1984. Shrimps, Lobsters, and Crabs of the Atlantic

Coast of the Eastern United States. Maine to Florida. Smithsonian

Institution Press, Washington. DC.
/ nl. inch Schneider, R. A., and P. A. Moore. 2000. Urine as a source of

conspecific disturbance signals in the crayfish Procambarus clarkii. J.

E\p. Biol. 203: 765-771.
Zulandt Schneider, R. A., R. W. S. Schneider, and P. A. Moore. 1999.

Recognition of dominance status by chemoreception in the red swamp

crayfish, Procambarus clarkii. J. Chem. Ecol. 25(4): 781-794.

Reference: Biol. Bull. 201: 17-25. (August 201)1)

Methionine-Enkephalin Induces Hyperglycemia
Through Eyestalk Hormones in the Estuarine Crab

Scylla serrata

P. SREENIVASULA REDDY* AND B. KISHORI
Department of Biotechnology, Sri Venkateswara University, TIRUPAT! - 517 502, India

Abstract. The hypothesis is tested that methionine-en-
kephalin. a hormone produced in and released from eyestalk
of crustaceans, produces hyperglycemia indirectly by stim-
ulating the release of hyperglycemic hormone from the
eyestalks. Injection of methionine-enkephalin leads to hy-
perglycemia and hyperglucosemia in the estuarine crab
Scylla serrata in a dose-dependent manner. Decreases in
total carbohydrate (TCHO) and glycogen levels of hepato-
pancreas and muscle with an increase in phosphorylase
activity were also observed in intact crabs after methionine-
enkephalin injection. Eyestalk ablation depressed hemo-
lymph glucose (19<7r) and TCHO levels (22%), with an
elevation of levels of TCHO and glycogen of hepatopan-
creas and muscle. Tissue phosphorylase activity decreased
significantly during bilateral eyestalk ablation. Administra-
tion of methionine-enkephalin into eyestalkless crabs
caused no significant alterations in these parameters when
compared to eyestalk ablated crabs. These results support
the hypothesis that methionine-enkephalin produces hyper-
glycemia in crustaceans by triggering release of hypergly-
cemic hormone from the eyestalks.

Introduction

In decapod crustaceans, hemolymph sugar level is regu-
lated by hyperglycemic hormone. Abramowitz et al. ( 1944)
were the first to demonstrate that injection of eyestalk
extract induced hyperglycemia in Callinectes. Since then,
hyperglycemia as a response to injection of eyestalk extract

Received 14 July 2000; accepted 6 March 2001.

* To whom correspondence should be addressed. E-mail:
psreddy@vsnl.com

has been observed in almost all groups of crustaceans (see
review by Keller, 1992). This neurohormone is stored in and
released from the sinus gland. The chemical nature, mode,
and site of action of hyperglycemic hormone has been
extensively studied in a number of crustaceans (see reviews
by Keller et al., 1985; Sedlmeier, 1985). The amino acid
sequence of hyperglycemic hormones has been determined
from a large number of crustaceans (see La Combe et al.,
1999, for review). The gene for hyperglycemic hormone
was also cloned from crabs (Kegel et al., 1989), lobster
(Tensen et al., 1991), prawn (Ohira et al., 1997). isopod
(Martin et al., 1993). and crayfish (Kegel et al., 1991;
Huberman et al., 1993; Yasuda et al., 1994). Recently, we
reported the expression of hyperglycemic hormone gene at
different molt stages in Homarus americanus, the American
lobster (Reddy et al.. 1997).

Since the discovery of opioid peptides in decapod crus-
taceans by Mancillas et al. (1981). several workers have
attempted to determine the physiological function of these
peptides, but the results are fragmentary. Sarojini et al.
(1995. 1996. 1997) provided evidence that methionine-
enkephalin slowed ovarian maturation in the fiddler crab
Uca pugilator and the crayfish Procanibarus clarkii, and
suggested that methionine-enkephalin produces this effect
indirectly by stimulating the release of gonad-inhibiting
hormone from eyestalks. In Uca pugilator, methionine-
enkephalin appears to stimulate release of the concentrating
hormones for black and red pigment cells (Quackenbush
and Fingerman. 1984) and the dark-adapting hormone for
distal retinal pigment cells (Kulkarni and Fingerman. 1987).
We reported a neurotransmitter role for methionine-en-
kephalin in regulating the hemolymph sugar level of the
freshwater crab O-ioielphusa senex senex, and hypothesized
that methionine-enkephalin produces hyperglycemia indi-

P. S. REDDY AND B. KISHORI

rectly by stimulating release of hyperglycemic hormone
(Reddy, 1999).

The objectives of the present study were threefold: (a) by
extending our studies to the estuarine crab Scylla serrata, to
test our hypothesis, generated by the study of Oziotelphusa
senex senex, that methionine-enkephalin produces hyper-
glycemia in decapod crustaceans; (b) to determine the
changes in levels of tissue carbohydrates and phosphorylase
activity during methionine-enkephalin treatment; and (c) to
provide evidence that supports the triggering of release of
hyperglycemic hormone during methionine-enkephalin
treatment.

Materials and Methods

Individuals of Scylla serrata (15 ± 2 cm in carapace
width; 110 ± 5 g wet weight) were collected from the
Chennai coast, India. They were kept in large aquaria with
continuous aeration and acclimatized to laboratory condi-
tions for one week under constant salinity (25 ± 1 ppt), pH
(7.2 ± 0.1 ), and temperature (23 ± 2°C). During this period
the crabs were fed fish flesh. Feeding was stopped 24 h
before the beginning of the experiments, and no food was
given during experimentation. Only intermolt (Stage C4),
intact, male crabs were used in the present study.

Methionine-enkephalin (Sigma Chemical Co.) was dis-
solved in physiological saline (Pantin, 1934). In these ex-
periments, each of the 10 groups of crabs used consisted of
10 individuals. The first group served as normal and re-
ceived no treatment. A second group served as control, with
each crab in this group receiving an injection of 10 /il of
physiological saline (Pantin, 1934) through the base of the
coxa of the 3rd pair of the walking legs. In groups 3-5
respectively, each crab received an injection of 10~7, 10~s,
and 10~y mole methionine-enkephalin in 10 jal volume.
Both eyestalks were ablated from all the crabs in groups
6-10. The eyestalks were extirpated by cutting them off at
the base, without prior ligation but with cautery of the
wound after operation. Twenty-four hours after eyestalk
ablation, these groups were used for experimentation. Crabs
in group 6 served simply as eyestalkless animals, and crabs
in group 7 received 10 ju,l crustacean Ringer solution and
served as eyestalkless controls. In groups 8-10 respectively,
each crab was injected with 10~7, 10~x, and 10~9 mole
methionine-enkephalin in 10 /xl volume. Based on prelim-
inary kinetic studies, the crabs were sacrificed for analysis
2 h after injection (Figs. 1. 2).

Hemolymph (500 jul) was aspirated by syringe,
through the arthrodial membrane of the coxa of the 4th
pair of walking legs. The other tissues (hepatopancreas

and muscle from chela propodus) were then quickly
dissected out. weighed, and analyzed by the procedures
outlined below.

Hemolvmph total carbohydrate level. Hemolymph total
carbohydrate (TCHO) levels were estimated in trichloroace-
tic acid supernatant (10% TCA w/v) according to the
method of Carroll et al. ( 1956).

Hemolymph glucose level. For measurement of glucose,
100 /u,l of hemolymph was mixed with 300 ju.1 of 95%
ethanol. After deproteinization (4 °C, 14,000 X g, 10 min),
the sample was combined with a mixture of glucose enzyme
reagent (glucose-6-phosphate dehydrogenase and NADP)
and color reagents (phenazine methosulfate and iodo-
nitrotetrazolium chloride) (kit from Sigma). After 30 min,
the intensity of the color was measured at 490 nm and
quantified with standards.

Tissue TCHO and glycogen levels. TCHO levels in the
tissues (hepatopancreas and muscle) were estimated in the
10% TCA supernatant (5% w/v), and glycogen was esti-
mated in the ethanolic precipitate of TCA supernatant, ac-
cording to the method of Carroll et al. ( 1956).

To 0.5 ml of clear supernatant was added 5.0 ml of
anthrone reagent, and the combination was boiled for 10
min in a water bath. The samples were then immediately
cooled. A standard sample containing a known quantity of
glucose solution was always tested along with the experi-
mental samples. Absorbance was measured at 620 nm
against a reagent blank.

Tissue phosphorylase activity. Phosphorylase activity
was assayed in hepatopancreas and muscle by colorimetric
determination of inorganic phosphate released from glu-
cose-1 -phosphate by the method of Cori et al. (1955). First,
0.4 ml of the enzyme was incubated with 2.0 mg of glyco-
gen for 20 min at 35 °C, then the reaction was initiated by
the addition of 0.2 ml of 0.016 M glucose- 1 -phosphate
(G-l-P) to one tube (phosphorylase a) and a mixture of 0.2
ml of G-l-P and 0.004 M adenosine-5-monophosphate
(phosphorylase ah) to another tube.

The reaction was incubated for 15 min for determining
total phosphorylase and for 30 min for active phosphor-
ylase. The reaction was terminated by the addition of 5.0
ml of 5 N sulfuric acid. Released inorganic phosphate
was estimated by the method of Taussky and Shorr
(1953).

Protein determination. Total protein levels in the enzyme
source were estimated following the method of Lowry et al.
(1951) using bovine serum albumin as standard.

MET-ENKEPHALIN-INDUCED HYPERGLYCEMIA IN CRAB 19

Table 1

Effect of eyestalk ablation fESX) (24-h post-ablation) and injection of methionine-enkephalin into intact and ablated crabs on hemolymph total sugar
aiui glucose levels in Scylla serrata

No treatment

Ringer injection

10~7 mol/crab

10 " mol/crab

10~9 mol/crab

Dunnet's

comparison test

Total Sugar

(mg/100 ml)

Intact
(Group 1 )
ESX
(Group 2)

12.11 ± 1.01

9.41 ± 1.13"
(-22.22)

12.73 ± 1.84a
(5.12)
9.34 ± 1.03b'c
(-0.74)

28.8 ± 2.18h
(126.23)
9.41 ± 1.13b'c
(0.74)

19.64 ± I.41h
(54.28)
9.43 ± 1.01b-c
(0.96)

16.52 ± 1.94h
(29.77)
9.21 ± 1.08b'c
(-1.39)

F(4-45)= 137.160
F(4.45, = 0.099

Intact
(Group 1 )
ESX
(Group 2)

Two-way ANOVA: F, w (Between groups) = 772.002, P < 0.001; F4 9n (Among treatments) = 98.747, P < 0.001;

F490 (Interaction) = 94.552, P < 0.001.

Glucose (mg/100 ml)

6.55 ± 0.76a 12.07 ±1.34" 11.44 ±1.28" 9.13 ± 0.78" F(4 45l = 75.613

(2.16) (84.27) (74.65) (39.38)

5.52 ± 0.81Kc 5.19 ± 1.01b-c 5.21 ± 0.91bc 5.44 ± 0.77Kc F,445, = 0.387

(6.35) (-5.97) (-5.61) (-1.44)

5.19 ± 1.01h
(-19.03)

Two-way ANOVA: Fl 90 (Between groups) = 440.810. P < 0.001; F4 90 (Among treatments) = 40.092, P < 0.001:

F.,,,,, (Interaction) = 44.753, P < 0.001.

Values are mean ± SD of 10 individual crabs. Values in parentheses are percent change from control. For calculation of percent change for
eyestalk-ablated (ESX) crabs and Ringer-injected intact crabs, intact crabs served as control; for met-injected crabs. Ringer-injected crabs served as control.
a Not significant compared with intact crabs.
b/> < 0.001 compared to intact crabs.
L Not significant compared io eyestalkless crabs.

Statistical analysis. Statistical analysis of the results was
made using a two-way ANOVA test followed by Dunnet's
multiple range test (preceded by one-way ANOVA), using
SPSS version 10.0 (SPSS Inc., Chicago, ID.

the possible mobilization of glucose molecules from hepa-
topancreas and muscle to hemolymph.

Phosphorylase (both total and active) activity levels were
significantly increased in both hepatopancreas and muscle

Results

Effects of methionine-enkephalin on carbohvdrate
metabolism of intact crabs

Injection of methionine-enkephalin into intact crabs re-
sulted in significant hyperglycemia and hyperglucosemia in
a dose-dependent manner (Table 1 ). whereas injection of
physiological saline had no effect on hemolymph carbohy-
drate levels. At doses between 10~9 mol/crab (36.41%) and
lO"6 mol/crab (147.81%), the effect of methionine-en-
kephalin was statistically significant. For doses lower than
10~9 mol/crab, however, methionine-enkephalin did not
elicit a hyperglycemic response (Fig. 1 ). A time course for
methionine-enkephalin-induced hyperglycemia is shown in
Figure 2 for a IO"7 mol/crab dose, which is a nearly
saturating dose. The hemolymph glucose level increased
significantly within 30 min of methionine-enkephalin injec-
tion, reached a peak at 2 h, then declined gradually.

Hepatopancreas glycogen and TCHO levels in crabs that
received methionine-enkephalin were significantly lower
than those of control crabs (Table 2). Decreases in muscle
glycogen and TCHO levels were also significant after the
injection of methionine-enkephalin (Table 3), suggesting

~ 30

IIS MS

SALINE- _1Q

INJECTED 10

-9
'0

10
[Methioninc -Enkcphatin] (mol/crab)

-6

10*

Figure 1. Dose-dependent effect of methionine-enkephalin on the
hemolymph glucose levels in intact Scylla serrata. Two hours after injec-
tion of saline (10 /nl/animal) or methionine-enkephalin at the doses indi-
cated, hemolymph was withdrawn from crabs for glucose determination.
Each bar represents a mean ± SD (n = 10). Numbers in parentheses
indicates the percent increase from the normal values. * Significant differ-
ence from normal crabs at P < 0.001. NS Not significant.

P. S. REDDY AND B. KISHORI

Time after injection (h)

Figure 2. Time course of methionine-enkephalin-induced hyperglyce-
mia. After injection of methionine-enkephalin (10~7 moL/crab). hemo-
lymph was withdrawn from intact crabs at the time points indicated for
glucose determination. Each point represents a mean ± SD (n = 10).
Numbers in parentheses represent percent change from zero time controls.
* Significant difference from zero time control at P < 0.001. ** Signif-
icant difference from zero time control at P < 0.001. NS Not significant
from zero time control.

kephalin, indicating conversion of inactive to active phos-
phorylase.

Effects of bilateral e\estalk ablation and injection
of methionine-enkephalin into ablated crabs
on carbohydrate metabolism

Bilateral eyestalk removal caused a significant decrease
in hemolymph carbohydrate level (Table 1 ). Enhancement
of TCHO level of hepatopancreas and muscle was also
significant in eyestalk-ablated crabs (Tables 2, 3). The in-
crease was greater in muscle. Glycogen level in hepatopan-
creas increased significantly in eyestalkless crabs. A similar
pattern was observed in muscle. Tissue phosphorylase ac-
tivity levels decreased significantly in eyestalk-ablated
crabs (Tables 4, 5).

Injection of methionine-enkephalin into eyestalkless
crabs did not significantly change hemolymph carbohydrate
levels compared to Ringer-injected eyestalkless crabs (Ta-
ble 1 ). The levels of tissue TCHO and glycogen and activity
levels of total and active phosphorylase were also not sig-
nificantly altered in eyestalkless crabs after methionine-
enkephalin injection (Tables 2-5).

after the injection of methionine-enkephalin (Tables 4. 5).
The ratio of active to total phosphorylase also increased in
the tissues of crabs after the injection of methionine-en-

Discussion

The effect of eyestalk hormones on tissue carbohydrate
levels and phosphorylase activity has been extensively stud-

Table 2

Effect of eyestalk ablation (ESX) (24-h post-ablation) and injection of methionine-enkephalin into intact ami ablated crabs on hepatopancreas total
carbohydrate (TCHOl and glycogen levels in Scylla serrata

Intact
(Group 1)
ESX
(Group 2)

No treatment Ringer injection 10 7 mol/crab

10~s mol/crab

10 9 mol/crab Dunnet's comparison test

TCHO (mg/g)

13.66 ± 1.54 13.84 ± 1.6T' 8.47 ± 0.97b 9.01 ± 1.51h 9.47 ± 1.49b ^,4.45, = 38.033

(1.32) (-38.80) (-34.89) (-31.57) P < 0.001

17.87 ±1.94h 18.01 ± \.91h-' 17.44 ± 1.43hx 17.X1 ± 1.62bx 17.93 ± 1.59b'c FI44S, = 0.229

(30.96) (0.67) (-0.74) (-0.96) (-1.39)

Two-way ANOVA: F, ,,„ (Between groups) = 566.317. P < 0.001; F4 ,m (Among treatments) = 19.027. P < 0.001;

F49(1 (Interaction) = 14.896. P < 0.001.

Glycogen (mg/g)

Intact

1.22 :

t 0.10

1.23 ±0.09a

0.58 ±().14h

0.61 ± 0.13h

0.64 ± 0.2 lh

F,4.45, = 148.477

(Group 1 )

(0.82)

(-52.84)

(-50.40)

(-47.96)

P < 0.001

ESX

2.04 ± 0.29h

2.06 ±0.31

h.c

2.11 ± 1.1 8"^

2.09 ±0.21ht

2.07 ± 0.28b'c

F,44,, = 0.230

(Group 2)

(67.

21)

(0.98)

(2.42)

(1.45)

(0.48)

Two-way

ANOVA: F,

g,, (Between

groups)

= 1658.593, P

< 0.001; F490 (Among

treatments) = 24.964,

P < 0.001;

^4.90

(Interaction) =

27.016. P < 0.001.

Values are mean ± SD of 10 individual crabs. Values in parentheses are percent change from control. For calculation of percent change for ESX crabs
and Ringer-injected intact crabs, intact crabs served as control; for met-injected crabs. Ringer-injected crabs served as control.
J Not significant compared with intact crabs.
* P < 0.001 compared to intact crabs.
c Not significant compared to eyestalkless crabs.

MET-ENKEPHALIN-INDUCED HYPERGLYCEMIA IN CRAB

Table 3

Effect of eyestalk ablation <ESX) (24 h post-ablation) and injection of methionine-enkephalin into intact and ablated crabs on muscle total
nnhi<h\ilratf iTCHO) ami glycogen levels in Scylla serrata

> treatment

Ring

er injection

10~7 mol/crab

10~K mol/crab

10"' mol/crab

Dunnet's

comparison test

TCHO

(mg/g)

Intact
(Group 1 )
ESX

(Group 2)

4.39

6.26
(4

± 0.53

± 0.71h
2.59)

4.41 ± 0.49a
(0.46)
6.31 ± 0.8 lbx
(0.80)

2.94 ± 0.3 lh
(-33.33)

3.01 ± 0.37h
(-31.74)
6.25 ± O.S41"
(-0.95)

3.12 ± 0.92h
(-29.25)
6.33 ± 0.92b-c
(0.31)

M4.45) —

P < 0

^14.45) ~

30.829

.001
0.045

6.31 ± 0.76b'c

(0)

Intact
(Group 1)
ESX
(Group 2)

Two-way ANOVA: F, go (Between groups) = 579.612. P < 0.001; F4 „„ (Among treatments) = 8.707, P < 0.001;

F4 QO (Interaction) = 9.1 14, P < 0.001.

Glycogen (mg/g)

0.66 ± 0.06

F<4.45> = 45.114
P < 0.001

Ft4.45)= 0.188

0.64 ± 0.09" 0.34 ± 0.09b 0.37 ± 0.06" 0.41 ± 0.08h

(-3.03) (-46.87) (-42.18) (-35.31)

1.01 ± 0.09b 1.02 ± O.ll"-c 0.99 ± 0.14b'c 1.07 ± 0.33bx 1.03 ± 0.2 lhx

(53.03) (0.99) (-2.94) (4.90) (0.98)
Two-way ANOVA: F,gn (Between groups) = 422.031. P < 0.001; F4 9() (Among treatments) = 6.391, P < 0.001;
F41,,, (Interaction) = 6.713. P < 0.001.

Values are mean (mg glucose/g tissue) ± SD of 10 individual crabs. Values in parentheses are percent change from control. For calculation of percent
change for ESX crabs and Ringer-injected intact crabs, intact crabs served as control; for met-injected crabs. Ringer-injected crabs served as control.
a Not significant compared with intact crabs.
b/> < 0.001 compared to intact crabs.
1 Not significant compared to eyestalkless crabs.

led (Keller, 1965; Ramamurthi et al.. 1968; Sagardia, 1969).
Eyestalk removal inactivates the phosphorylase system and
activates uridine-diphosphate-glucose glycogen transglu-
cosylase (glycogen synthetase) (Keller, 1965; Ramamurthi
ct nl.. 1968). Ramamurthi et al. ( 1968) also observed stim-
ulation of uptake and incorporation of glucose I4C into the
glycogen fraction of muscle tissue after eyestalk removal;
this stimulation was accompanied by a decrease in hemo-
lymph sugar level. Injection of eyestalk extract reversed
these changes. The hyperglycemic hormone of eyestalks of
the crab Oziotelphusa senex sene.x and the prawn Penaeus
monodon enhances the activity of the phosphorylase system
(Reddy et al.. 1982, 1984; Reddy, 1992).

An increase in phosphorylase activity and a decrease in
glycogen and TCHO levels in hepatopancreas and muscle of
Scylla serrata, followed by hyperglycemia after the injec-
tion of methionine-enkephalin, indicate glycogenolysis and
mobilization of sugar molecules from tissues to hemo-
lymph. This is in agreement with other findings (see review
by Reddy and Ramamurthi, 1999). Though the hormone
that elevates hemolymph sugar is conventionally called
crustacean hyperglycemic hormone (CHH). Hohnke and
Scheer (1970) suggested that the primary function of the
CHH is not to elevate hemolymph sugar level, but to elevate
intracellular glucose through the degradation of glycogen by
activating the enzyme phosphorylase. The conversion of
phosphorylase from its inactive to active form results in

glycogenolysis, and the resultant glucose molecules leak
into the hemolymph, causing hyperglycemia. This view has
been supported by Telford (1975).

Our results clearly demonstrate that methionine-enkepha-
lin is involved in the regulation of carbohydrate metabolism
in the crab Scylla serrata. In the present study, we show that
methionine-enkephalin elicited a hyperglycemic response in
S. serrata in a dose-dependent manner (Fig. 1 ). Methionine-
enkephalin-induced hyperglycemia has been similarly dem-
onstrated in the freshwater crab Oziotelplutsa senex senex
(Reddy, 1999) and the brackish-water prawns Penaeus in-
dicus and Metapenaeits monocerus (Kishori et al., 2001).
The doses of methionine-enkephalin that induced hypergly-
cemia ranged from 10 9 to 10~6 mol/animal (Fig. 1 ), which
is comparable to those reported for O. senex senex (Reddy,
1999). Our observation that methionine-enkephalin was in-
effective in inducing hyperglycemia in eyestalk-ablated S.
serrata (Table 1) is also consistent with those obtained in
crabs (Reddy, 1999) and prawns (Kishori et al., 2001 ) and
suggests that the hyperglycemic effect of methionine-en-
kephalin results from an enhanced release of CHH (Keller,
1992; Soyez. 1997).

Injection of methionine-enkephalin into intact S. serrata
also has two other effects. It activates the phosphorylase
system, which causes degradation of glycogen. It also re-
sults in accumulation of sugar molecules in the tissues;
these molecules are ultimately mobilized to hemolymph.

P. S. REDDY AND B. KISHORI
Table 4

Effect of evestalk ablation I ESX) (24 h post-ablation) and injection of methionine-enkephalin into intact and ablated crabs on hepatopancreas
phosphorylase activity levels in Scylla serrata

No treatment Ringer injection

10~7 mol/crab

10~8 mol/crab

10 " mol/crab

Dunnet's comparison test

Phosphorylase a

3.63 ± 0.34h 3.60 ± 0.42" F,4 45, = 28.430

(35.95) (34.83) P < 0.001

1.81 ± 0.22b-c 1.84 ± 0.31b'c F,445, = 1.473

(8.38) (10.17)

Two-way ANOVA: F, w, (Between groups) = 716.848. P < 0.001; F4 91, (Among treatments) = 23.852, P < 0.001;

F4,0 (Interaction) = 18.208, P < 0.001.

Intact

2.62 ± 0.29

2.67 ± 0.33"

3.87 ± 0.46b

(Group 1)
ESX

1.72 ± 0.3 lh

(1.91)
1.67 ± 0.29b-c

(44.94)
1.69 ± O.ll"-1-

(Group 2)

(-34.35)

(-2.33)

(1.19)

Phosphorylase ab

Intact
(Group 1)
ESX
(Group 2)

4.52 ± 0.41 4.56 ± 0.44a 5.81 ± 0.67b

(0.89) (27.41)

4.06 ± 0.44b 4.08 ± 0.41hL 4.10 ± 0.39"-'

(-10.18) (0.49) (0.49)

5.69 ± 0.52"

(24.78)

4.12 ± 0.34h-c
(0.98)

5.56 ± 0.73"

(21.92)

4.09 ± 0.51b'c
(0.24)

F(4.45) = 15.846

P < 0.001
F,44,, = 0.044

Two-way ANOVA: F, uo (Between groups) = 169.103, P < 0.001; F4 w (Among treatments) = 11.291, P < 0.001:
F4,,,, (Interaction) = 9.985. P < 0.001.

Values are mean (iP released/mg protein/h) ± SD of 10 individual crabs. Values in parentheses are percent change from control. For calculation of
percent change for ESX crabs and Ringer-injected intact crabs, intact crabs served as control; for met-injected crabs. Ringer-injected crabs served as control.
a Not significant compared with intact crabs.
b P < 0.001 compared to intact crabs.
c Not significant compared to eyestalkless crabs.

causing hyperglycemia. Methionine-enkephalin might have
elevated the phosphorylase system in intact crabs in several
different ways — for example, by triggering release of hy-
perglycemic hormone or by mimicking the action of this
hormone. However, because methionine-enkephalin was
not able to produce these changes in eyestalkless crabs, it
seems most likely that methionine-enkephalin exerted its
hyperglycemic effect by triggering release of hyperglyce-
mic hormone from the sinus gland of eyestalks. This sup-
ports our earlier results that sinus glands in the eyestalks of
crabs are the main release site for hyperglycemic hormone
(Reddy and Ramamurthi, 1982).

The mechanisms whereby methionine-enkephalin causes
release of neurohormones are still uncertain. In mammals,
endogenous opioid peptides are involved in regulating the
release of neurohypophysial peptides (Bicknell et al.. 1988;
Yamada et al., 1988; Sasaki et ai, 2000). In crustaceans,
opioid-peptide-like (methionine-enkephalin-like, leucine-
enkephalin-like and /?-endorphin-like) hormones were iso-
lated and characterized from X-organ sinus gland com-
plexes of eyestalks (Fingerman et ai, 1983, 1985).
However, there is little information about the effect of
opioid peptides on release of neurohormones in crustaceans.
Sarojini et al. (1995, 1996). using highly selective opioid
antagonists, provided evidence that methionine-enkephalin
exerts its effect by acting through delta-type opioid recep-
tors in regulating ovarian maturation in Procambarus

clarkii. In vivo studies with tissues of P. clarkii showed that
methionine-enkephalin exerted its effect by at least modu-
lating the release of eyestalk peptide hormone (Sarojini et
al., 1997). Recently, we provided evidence for a neurotrans-
mitter role for methionine-enkephalin in causing hypergly-
cemia in the crab O. senex senex (Reddy, 1999). Methio-
nine-enkephalin also triggers the release of red-pigment-
concentrating hormone, black-pigment-dispersing hormone
(Quackenbush and Fingerman, 1984). and dark-pigment-
adapting hormone (Kulkarni and Fingerman, 1987). Three
facts make it seem likely that this hyperglycemic action of
methionine-enkephalin in the present study on S. serrata is
also indirect and involves stimulation of release of CHH.
Methionine-enkephalin-like material is present in the neu-
roendocrine complex of the eyestalk of crustaceans (Finger-
man et al., 1983. 1985). Methionine-enkephalin mediation
of release of neurohormones has been demonstrated
(Reddy, 1999). In cases where methionine-enkephalin has
been found to stimulate neurohormone release, it does not
act in the absence of neuroendocrine organs. As further
support for the conclusion, eyestalk extract from methio-
nine-enkephalin injected prawns showed significantly less
activity than the normal eyestalk extract in inducing hyper-
glycemia (Kishori et al., 2001 ).

Although the mechanisms that trigger release of CHH are
still unknown, it is noteworthy that 5-hydroxytryptamine
(5-HT), or serotonin, triggers CHH release in the crayfish

MET-ENKEPHALIN-INDUCED HYPERGLYCEMIA IN CRAB 23

Table 5

Effect of evesta/k ablation I ESX) (24 h post-ablation) and injection of methinine-enkephalin into intact and ablated crabs on muscle phosphor/lose
activin levels in Scylla serrala

No treatment Ringer injection 10~7 mol/crab 10 ~s mol/crab 10~" mol/crab Dunnet's comparison test

Phosphorylase a

Intact

1 .92 ± 0.09

1.94 ± 0.1 4a

3.26 ± 0.36b

3.01 ± 0.12'1

3.02 ± 0.26h

F,4.451 = 84.853

(Group 1 )

(1.04)

(68.04)

(55.15)

(55.67)

P < 0.001

ESX

0.99 ± 0.08h

1.02 ± O.llhl

1.01 ± 0.09Kc

1.04 ±O.I3hc

1.06 ± 0.2 1KC

F|44,, = 0.368

(Group 2)

(-48.44)

(3.03)

(-0.74)

(1.96)

(3.92)

Two-way ANOVA: F, 91, (Between groups) = 1711.188. P < 0.001; F4 M1, (Among treatments) = 58.745. P < 0.001;

F4 w (Interaction) = 52.927, P < 0.001.

Phosphorylase ab

Intact

2.49 ±

0.45

2.52 ± 0.49a

3.49 ± 0.4 lh

3.46 ±

0.44h

3.44 ± 0.51

44S) =

16.086

(Group 1)

(1.21)

(38.49)

(37

.30)

(36.50)

P < 0

.001

ESX

2.22 •*•

0.32b

2.18 ± 0.31bc

2.22 ± 0.34b-c

2.24 ±

0.42hc

2.25 ± 0.41

b.t

4.45) =

0.061

(Group 2)

(-11

.65)

(-0.91)

(1.83)

(2.

75)

(3.21)

Two-way

ANOVA:

F, „,, (Between groups)

= 136.048, P <

0.001; F4,

m (Among

treatments) =

10.259,

P < 0.

001;

•" 4, MO

(Interaction) =

8.734, P <

0.001.

Values are mean (iP released/mg protein/h) ± SD of 10 individual crabs. Values in parentheses are percent change from control. For calculation of
percent change for ESX crabs and Ringer-injected intact crabs, intact crabs served as control; for met-injected crabs. Ringer-injected crabs served as control.
a Not significant compared with intact crabs.
* P < 0.001 compared to intact crabs.
c Not significant compared to eyestalkless crabs.

Orconectes limosus (Keller and Bayer, 1968), Astacus lep-
todactylus (Strolenberg and Van Herp, 1977), and Procam-
barus clarkii (Lee et ai. 2000). Strolenberg and Van Herp
(1977). working with A. leptodactylus, and Martin (1978).
working with Porcellio dilatatits, found that the sinus
glands of specimens injected with 5-HT show increased
numbers of exocytotic profiles, suggestive of increased
CHH release. Exocytosis in A. leptodactylus was maximal
2 h after 5-HT was injected, and the hemolymph glucose
concentration peaked 4 h after the injection (Strolenberg
and Van Herp. 1977). In P. dilatatus, hyperglycemia in-
duced by 5-HT is mediated by 5-HT,- and 5-HT:-like
receptors in triggering release of CHH (Lee et ai, 2000).

In summary, we have shown that methionine-enkephalin
is a potent hyperglycemic regulator in the crab Scylla ser-
nita. The most likely site of action of methionine-enkepha-
lin is the eyestalks. where the X-organ-sinus glands may
respond to methionine-enkephalin stimulation by releasing
CHH. Based on these results, experiments are being con-
ducted to determine whether methionine-enkephalin en-
hances the release of CHH in crustaceans.

Acknowledgments

We thank Prof. Armugam, University of Madras. Chen-
nai, for supplying Sc\lla serrata and providing necessary
laboratory facilities, and Dr. K. V. S. Sharma, Professor,

Department of Statistics, Sri Venkateswara University, for
analyzing the data. We also thank the anonymous reviewers
whose comments improved our manuscript. We are grateful
to Prof. R. Ramamurthi. Department of Zoology, for his
encouragement. Mr. S. Umasankar and Miss B. Prema
Sheela provided skilled technical assistance. This work was
carried out with the financial assistance from Department of
Science and Technology research grant (SP/SO/CO4/96) to
Dr. PSR. We also thank the staff. Department of Biotech-
nology, Sri Venkateswara University, for their invaluable
assistance.

Literature Cited

Abramowitz, A. A., F. L. Hisaw, and D. N. Papandrea. 1944. The

occurrence of a diabetogenic factor in the eyestalks of crustaceans.
Bi»l. Bull. 86: 1-5.

Bicknell, R. J., G. Leng, J. A. Russell, R. G. Dyer, S. Mansfield, and
B. G. Zhao. 1988. Hypothalamic opioid mechanisms controlling
oxytocin neurons during parturition. Brain Res. Bull. 20: 743-749.

Carroll, N. V., N. N. Longley, and J. G. Roe. 1956. Glycogen deter-
mination in liver and muscle by use of anthrone reagent. J. Biol. Chem.
220: 583-593.

Cori, G. T., B. Illingworth, and P. G. Keller. 1955. Muscle phosphor-
ylase. Pp. 200-207 in Methods in En-ymology. Vol. I. S. P. Colowick
and N. O. Kaplan, eds. Academic Press, New York.

Fingerman. M., M. M. Hanumante. and L. L. Vacca. 1983. Enkepha-
lin-like and substance P-like immunoreactivity in the eyestalk neuroen-

P. S. REDDY AND B KISHORI

docrine complex of the fiddler crab, Uca pugilator. Soc. Neurosci.
Abstr. 9: 439.

Fingerman, M.. M. M. Hanumante. (.. K. Kulkarni, R. Ikeda, and
L. L. Vacca. 1985. Localization of substance P-like, leucine-en-
kephalin-like. methionine-enkephalin-like. and FMFR amide-like im-
munoreactivity in the eyestalk of the fiddler crab, Uca pugilator. Cell
Tissue Res. 241: 473-477.

Hohnke. L., and B. T. Scheer. 1970. Carbohydrate metabolism in
crustaceans. Pp. 147-166 in Chemical Zoology, Vol. 5. M. Florkin and
B. T. Scheer. eds. Academic Press. New York.

HII!M i in. in. A., M. B. Aguilar, K. Brew, J. Shabanowitz, and D. F.
Hunt. 1993. Primary structure of the major isomorph of the crusta-
cean hyperglycemic hormone (CHH-1) from the sinus gland of Mex-
ican crayfish Procambarus bouvieri (Ortmann): interspecies compari-
son. Peptides 14: 7-16.

Kegel, G., B. Reichwein, S. Weese, G. Gaus, J. Peter-Katalinic, and R.
Keller. 1989. Amino acid sequence of the crustacean hyperglycemic
hormone (CHH) from the shore crab, Carcinus maenas. FEBS Lett.
255: 10-14.

Kegel, G., B. Reichwein, C. P. Tensen, and R. Keller. 1991. Amino
acid sequence of crustacean hyperglycemic hormone (CHH) from the
crayfish Orconectes limosus. Emergence of a novel neuropeptide fam-
ily. Peptides 12: 909-913.

Keller, R. 1965. Uber eine hormonale Kontrolle des Polysaccharidstoff
wechsels beim Flusskiebs Cambarits maenas. Z. Vgl. Pliysiol. 51:
49-59.

Keller, R. 1992. Crustacean neuropeptides: structures, functions and
comparative aspects. Experientia 48: 439-448.

Keller, R., and J. Beyer. 1968. Zur hyperglykamischen Wirkung von
Serotonin und Augeststielextrakt beim Flusskrebs Orconectes limosus.
Z. Vgl. Physio/. 59: 78-85.

Keller, R., P. Jaros. and G. Kegel. 1985. Crustacean hyperglycemic
neuropeptides. Am. Zool. 25: 207-221.

Kishori. B., B. Premasheela, R. Ramamurthi, and P. S. Reddy. 20(11.
Evidence for hyperglycemic effect of methionine-enkephalin in prawns
Penaeus indicus and Metapenaeus monocerus. Gen. Comp. Endocri-
nnl. (in press I.

Kulkarni, G. K., and M. Fingerman. 1987. Distal retinal pigment of
the fiddler crab, Uca pugilator. Release of the dark-adapting hor-
mone by methionine-enkephalin and FMRF-amide. Pigm. Cell Res. 1:
51-56.

Lacombe, C., P. Greve, and G. Martin. 1999. Overview on the sub-
grouping of the crustacean hyperglycemic hormone family. Neuropep-
tides 33: 71-80.

Lee, C. Y., S. M. Yau, C. S. Liau, and W. J. Huang. 2000. Serotogenic
regulation of blood glucose levels in the crayfish. Procambarus clarkii:
site of action and receptor characterization. J. E.\p. Zool. 286: 596-
605.

Lowry, O. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. 1951.
Protein measurement with Folin phenol reagent. J. Biol. Chem. 193:
265-275.

Mancillas, J. R., J. F. McGinty, A. I. Selverston, H. Karten, and F. E.
Bloom. 1981. Immunocytochemical localization of enkephalin and
substance P in retina and eyestalk neurones of lobster. Nature 293:
576-578.

Martin, G. 1978. Action de la serotonine sur la glycemie et sur la
liberation des neurosecretions contenues dans la glande du sinus de
Porcellio di/atatus Bradt (Crustace. Isopode, Oniscoide). C. R. Seances
Soc. Biol. Fil. 172: 304-307.

Martin, G., O. Sorokine, and A. Van Dorsselaer. 1993. Isolation and
molecular characterization of hyperglycemic neuropeptide from the

sinus gland of the terrestrial isopod Armadillidium vulgare (Crustacea).
Em: J. Biochem. 211: 601-607.

Ohira, T., T. Watanabe, H. Nagasawa, and K. Aida. 1997. Cloning
and sequence analysis of a cDNA encoding a crustacean hyperglycemic
hormone from the Kuruma prawn Penaeus japonicus. Mot. Mar. Biol
Biotechnol. 6: 59-63.

Pantin, C. F. A. 1934. The excitation of crustacean muscle. ./. Exp. Biol.
11: 11-27.

Quackenbush, L. S., and M. Fingerman. 1984. Regulation of neuro-
hormone release in the fiddler crab. Uca pugilator: Effects of gamma-
aminobutyric acid, octopamine, met-enkephalin and beta-endorphin.
Comp. Biochem. Physiol. 79C: 77-84.

Ramamurthi, R., M. W. Mumbach. and B. T. Scheer. 1968. Endo-
crine control of glycogen synthesis in crabs. Comp. Biochem. Phvsiol.
67B: 437-445.

Reddy, P. S. 1992. Changes in carbohydrate metabolism of Pila globosa
in response to crustacean hyperglycemic hormone. Arch. Int. Phvsiol.
Biochim. Biophys. 100: 281-283.

Reddy. P. S. 1999. A neurotransmitter role for methionine-enkephalin in
causing hyperglycemia in the fresh water crab O-iotelplntsa senex
senex. Curr. Sci. 76: 1126-1128.

Reddy, P. S., and R. Ramamurthi. 1982. Neuroendocrine control of
carbohydrate metabolism in the rice field crab (Oziotelphusa senex
sene.\). J. ReprocL Biol. Comp. Endocrinol. 2: 49-57.

Reddy, P. S., and R. Ramamurthi. 1999. Recent trends in crustacean
endocrine research. Proc. Indian Natl. Sci. Acad. B65: 15-32.

Reddy, P. S., A. Bhagyalakshmi, V. Chandrasekharam, and R. Ra-
mamurthi. 1982. Hyperglycemic activity of crab and scorpion hor-
mones in grasshopper. Poecilocerus pictus. Experientia 38: 811-812.

Reddy, P. S., A. Bhagyalakshmi, V. Chandrasekharam, and R. Ra-
mamurthi. 1984. Differential responsiveness of phosphorylase sys-
tem and fat body carbohydrates of Poecilocerus pictus (Insecta) to
hyperglycemic factors of crab (Crustacea) and scorpion (Arachnida).
Gen. Comp. Endocrinol. 54: 43-45.

Reddy, P. S., G. D. Prestwich, and E. S. Chang. 1997. Crustacean
hyperglycemic hormone gene expression in the lobster Homants
americanus. Pp. 51-5 in Advances in Comparative Endocrinology,
Vol. 1, S. Kawashima and S. Kikuyama, eds. Monduzzi Editore,
Bologna. Italy.

Sagardia, F. 1969. The glycogen phosphorylase system from the muscle
of the blue crab, Callinectes danae. Comp. Biochem. Physiul. 28:
1377-1385.

Sarojini, R., R. Nagabhushanam, and M. Fingerman. 1995. Evidence
for opioid involvement in the regulation of ovarian maturation of the
fiddler crab, Uca pugilator. Comp. Biochem. Physiol. 111A: 279-282.

Sarojini, R., R. Nagabhushanam, and M. Fingerman. 1996. In vivo
assessment of opioid agonists and antagonists on ovarian maturation in
the red swamp crayfish. Procambarus clarkii. Comp. Biochem. Physiol.
115C: 149-153.

Sarojini, R., R. Nagabhushanam, and M. Fingerman. 1997. An in
vitro study of the inhibitory action of methionine enkephalin on ovarian
maturation in the red swamp crayfish, Procambarus clarkii. Comp.
Biochem. Physiol. 117C: 207-210.

Sasaki, T., K. Shimada. and N. Saito. 2000. Regulation of opioid
peptides on the release of arginine vasotocin in the hen. J. Exp. Zool.
286: 481-486.

Sedlmeier, I). 1985. Mode of action of the crustacean hyperglycemic
hormone. Am. Zoo/. 25: 223-232.

Soyez, D. 1997. Occurrence and diversity of neuropeptides from the
crustacean hyperglycemic hormone family in arthropods. Ann. NY
Acad. Sci. 814: 319-323.

MET-ENKEPHALIN-INDUCED HYPF.RGLYCEM1A IN CRAB

Slrolenberg, G. E. C.. and F. Van Herp. 1977. Mise en evidence du
phenomene d'exocytose dans la glande du sinus il'A.itacus leptodactv-
lus (Nordmannl sous I'inrluence d'injection de serotonine. C. R. Acad.
Sci. Ser. Ill Sci. Vie 284D: 57-59.

Taussky, H. M., and E. Shorr. 1953. A micro colorimetric method tor
determination of inorganic phosphate. J. Biol. Cliem. 202: 675-682.

Telford, M. 1975. Blood glucose in crayfish — III. The sources of glu-
cose and role of eyestalk factor in hyperglycemia of Cambarus robus-
tns. Catnp. Biocht'in. Pln'siol. 51: 69-73.

Jensen. C. P.. D. P. V. De Kleijn, and F. Van Herp. 1991. Cloning and
sequence analysis of cDNA encoding two crustacean hyperglycemic

hormones trom the lobster Homctrus americanus, Eur. J. Bioi'hein. 20(1:
103-106.

Vamada, T., K. Nakao, H. Itoh. N. Morii, S. Shiono, M. Sakamoto,
A. Sugawara, Y. Saito, M. Mukoyama, H. Aral, M. Eigyo, A.
Matushita, and H. Imura. 1988. Inhibitory action of leumorphin
on vasopressin secretion in conscious rats. Endocrino/ogv 122: 985-
990.

Yasuda, A., Y. Yasuda, T. Fujita, and Y. Naya. 1994. Characterization
of crustacean hyperglycemic hormone from the crayfish (Procambarus
clarkii): multiplicity of molecular forms by stereoinversion and diver-
sion functions. Gen. Comp. Endocrinol. 95: 387-398.

Reference: Biol. Bull. 201: 26-33. (August 2001)

Theoretical and Experimental Dissection

of Gravity-Dependent Mechanical Orientation

in Gravitactic Microorganisms

YOSHIHIRO MOGAMI,* JUNKO ISHII.t AND SHOJI A. BABA

Depanment of Biology, Ochanomizit University. Otsuko, Tokyo 1 12-8610, Japan

Abstract. Mechanisms of gravitactic behaviors of aquatic
microorganisms were investigated in terms of their mechan-
ical basis of gravity-dependent orientation. Two mechanical
mechanisms have been considered as possible sources of the
orientation torque generated on the inert body. One results
from the differential density within an organism (the grav-
ity-buoyancy model) and the other from the geometrical
asymmetry of an organism (the drag-gravity model). We
first introduced a simple theory that distinguishes between
these models by measuring sedimentation of immobilized
organisms in a medium of higher density than that of the
origanisms. Nr + -immobilized cells of Paramecium caitda-
tuin oriented downwards while floating upwards in the
Percoll-containing hyper-density medium but oriented up-
wards while sinking in the hypo-density control medium.
This means that the orientation of Paramecium is mechan-
ically biased by the torque generated mainly due to the
anterior location of the reaction center of hydrodynamic
stress relative to those of buoyancy and gravity; thus the
torque results from the geometrical fore-aft asymmetry and
is described by the drag-gravity model. The same mechan-
ical property was demonstrated in gastrula larvae of the sea
urchin by observing the orientation during sedimentation of
the KCN-immobilized larvae in media of different density:
like the paramecia, the gastrulae oriented upwards in hypo-
density medium and downwards in hyper-density medium.
Immobilized pluteus larvae, however, oriented upwards re-
gardless of the density of the medium. This indicates that
the orientation of the pluteus is biased by the torque gen-
erated mainly due to the posterior location of the reaction
center of gravity relative to those of buoyancy and hydro-

Received 14 July 2000; accepted 30 March 2001
* To whom correspondence should he addressed. E-mail: mogami@
cc.ocha.ac.jp

t Died on 1 March 1999.

dynamic stress; thus the torque results from the fore-aft
asymmetry of the density distribution and is described by
the gravity-buoyancy model. These observations indicate
that, during development, sea urchin larvae change the
mechanical mechanism for the gravitactic orientation. Evi-
dence presented in the present paper demonstrates a definite
relationship between the morphology and the gravitactic
behavior of microorganisms.

Introduction

Many swimming microorganisms, including ciliate and
flagellate protozoa and the planktonic larvae of some inver-
tebrates, are negatively gravitactic; that is, they tend to
swim preferentially upwards in water columns despite being
heavier than water. This behavior requires the organism to
orient upwards in relation to the gravity vector. Several
mechanisms have been postulated for the gravitactic orien-
tation of aquatic microorganisms (Chia el al., 1983; Bean,
1984; Machemer and Braucker. 1992). From a physical
point of view and taking account of the mechanical prop-
erties of these microorganisms, it has been postulated that
the interaction of gravitational and hydrodynamic forces
may cause them to orient with fore end upward. In addition
to the mechanical basis, gravitactic orientation might also
be explained on the physiological basis of gravity percep-
tion. To modulate the propulsive activity, some mechano-
sensitive devices that sense gravity (for example, statocysts)
might be needed. Although functional statocysts have been
found in some unicellular organisms (Fenchel and Finlay,
1984, 1986), a line of evidence for gravity-dependent mod-
ulation of propulsion has been accumulated for Paramecium
(Machemer et al., 1991; Ooya et al., 1992) and Eitglenu
(Machemer-Rohnisch el al., 1999), which have no stato-
cyst-like structure.

The present paper focuses on the mechanical properties of

MECHANICAL BIAS OF MICROBIAL GRAVITAXIS

microorganisms, which, irrespective of propulsion, generate
the torque to orient the organisms either upwards or down-
wards. This mechanical torque should bias the gravitactic
orientation, even if the organisms have active physiological
mechanisms of gravitaxis. According to Roberts (1970), two
mechanical mechanisms have been considered as possible
sources of the orientation torque. These are reconsidered, in the
present paper, as two mechanical models, the gravity-buoy-
ancy model and the drag-gravity model.

The gravity-buoyancy model was first postulated by Ver-
worn ( 1 889. cited in Machemer and Braucker. 1992) for the
negative gravitaxis of Paramecium. This model is based on
the differential density within an organism. If the internal
density of the organism is not homogeneous, the center of
mass (the center of gravity) does not necessarily coincide
with the centroid (the center of buoyancy). Posterior accu-
mulation of the mass would result in the upward orientation
of the organisms, and anterior accumulation would result in
the downward orientation.

The drag-gravity model was postulated by Roberts (1970)
on the basis of the low Reynolds number hydrodynamics of
the swimming of microorganisms that have a geometrical
fore-aft asymmetry. This model is characterized by a dumb-
bell with two spheres of unequal diameter but homogeneous
density, which could mimic the fore-aft asymmetry of the
microorganisms. According to Stokes' drag formula, the
larger sphere of the dumbbell can sink faster than the
smaller, at the rate of the square of the ratio of diameters.
The applicability of this model has been confirmed by
scale-model experiments (Roberts, 1970).

Organisms, in general, possess some asymmetry both in
internal density and in external geometry. It is therefore pos-
sible that these two mechanical models operate independently
to generate the gravity-induced orientation torque. Since the
mechanical properties for gravitactic orientation are indepen-
dent of propulsive thrust, we can assess the mechanical influ-
ence by measuring the orientation of immobilized organisms
sinking under gravity. Both models predict that, when immo-
bilized, an organism orients upwards when sinking in a me-
dium with a density lower than its own.

In the present paper, we show that the above two models
can be distinguished by observing what happens to an
organism placed in a medium whose density is higher than
its own. We show the results of the experiments on the
gravitactic orientation of Paramecium and sea urchin lar-
vae, both of which are known to perform typical negative
gravitaxis (Mogami et «/., 1988: Ooya et al.. 1992).

Theory

The external forces acting on the body of an aquatic micro-
organism due to gravity acceleration are gravitational (Fc) and
buoyant forces (FB), each of which is generated as the product
of the volume and density of the body or of the external fluid.
The vector sum of the forces encounters the hydrodynamic

force (FH). Since the Reynolds number of an aquatic micro-
organism in translational motion is significantly less than unity
(of the order of 10 2). FH is generated in proportion to the
velocity (Happel and Brenner, 1973; Vogel, 1994). Fc, FB, and
FH act on the center of mass (G), the centroid (B), and the
reaction center of hydrodynamic stress (//), respectively. For
an immobilized microorganism sinking in the fluid, these three
forces are balanced as

B + FH = 0.

(1)

Each term in the equation (positive in upward direction) is
described as

FG= - Vp,g,
FB = Vpg. and

(2)
(3)

(4)

where V and p, are the total volume and the average density
of the organism, p and g the density of the external fluid and
the acceleration due to gravity, and K and 5 the coefficient
of hydrodynamic drag and the sinking velocity.

We assume in the present paper that a microorganism has
a body of rotating symmetry on its fore-aft axis. The sim-
plest case of this approximation is that the body has fore-aft
symmetry, such as a prolate spheroid. When a prolate spher-
oid with uniform density is sinking in the fluid, the three
forces act on the same point and therefore do not generate
any torque to rotate the body (Fig. la).

If. however, the body of a prolate spheroid has a region of
higher density in the rear half of the body, as postulated in
the gravity-buoyancy model, G is located posterior to B and
H (Fig. Ib). This generates the torque (7\,; subscript V is
after Verworn) which is given by

TV =

sin 0,

(5)

where LG is the distance between G and B (and/or H), and
6 is the orientation angle of the fore-aft axis of the body to
the vertical.

The fore-aft asymmetry of the external geometry, as
postulated in the drag-gravity model, also separates the
reaction centers of the forces. If a microorganism of homo-
geneous density has a larger radius of revolution around the
fore-aft axis in the posterior part (Fig. Ic). H is located
anterior to B and G, according to the analogy of a fore-aft
asymmetrical dumbbell of homogeneous density (Happel
and Brenner. 1973). The torque (TR: subscript R is after
Roberts) by the anterior shift of the center of hydrodynamic
force is given by

TR = -FHLHs'm 0 = (Fc + FB)LHsin 6,

(6)

where LH is the distance between H and G (and/or B).

Provided that the Reynolds number of rotational motion
is sufficiently small, all torques should be proportional to

Y. MOGAMI ET AL.

Figure 1. Schematic drawings illustrating the mechanical (physical) basis for the generation of gravity-
dependent orientation torque. Gravity (FG), buoyancy (FB), and hvdrodynamic force (FH) are balanced in
sinking microorganisms; these forces act at the center of mass (G), the centroid (B). and the reaction center of
hydrodynamic stress (//), respectively, (a) Three forces act at the same point in the body of prolate spheroid with
uniform density, (b) The center of mass is deviated to the rear end of the body of prolate spheroid, which
generates the torque in proportion to Fa and the sine of the orientation angle to the gravity vector (W). (c) The
reaction center of hydrodynamic stress is deviated to the front end of the body with fore-aft asymmetry but with
uniform density, which generates the torque in proportion to the vector sum of FCl and FH and the sine of the
orientation angle.

the first power of rotational velocity (dQIdt). In such cases
equations of rotational motion are given by

-flTj~ = TvorTK,

(7)

where R is the coefficient of resistance for rotational motion
and T) is the viscosity of the external fluid. From these
equations the rotational velocity of each model is given as a
common form of

-=3sin0.

(8)

where the proportional factor is the instantaneous rate at
0 = 90 degrees, and given by

(9)

(10)

Rj]

V(p,-p)gLH
Rr,

for the gravity-buoyancy and drag-gravity models, respec-
tively.

Equations 9 and 10 indicate that /3r is insensitive to

changes in the density of the external medium (p),
whereas f3K reverses the sign as p exceeds the density of
organisms (p,-). This means that the two models can be
distinguished by increasing p greater than p,. When im-
mobilized organisms are immersed in the hyper-density
medium (p > p,), they would orient upwards during
floating upwards if they obeyed the gravity-buoyancy
model, whereas they would orient downwards if they
obeyed the drag-gravity model.

The gravity-buoyancy and drag-gravity models are the
two extremes of these conditions that can generate the
orientation torque depending on the different physical
mechanisms. Passive orientation of the organisms (Eq. 8),
in fact, would be explained as a result of combining the two
models, because none of three forces would necessarily
have a common reaction center. In order to extract the origin
of the mechanical bias of the orientation. Equation 8 should
be examined by measuring |3 by the sedimentation experi-
ment using media of different p. If j3 is constant independent
of p, the gravity-buoyancy model is the only mechanism for
generating the orientation torque. Otherwise, the drag-grav-
ity model may play a part in the generation of the torque. A
negative value of 0 in the hyper-density medium indicates

MECHANICAL BIAS OF MICROBIAL GRAVITAXIS

that the drag-gravity model is the major mechanism in
passive gravitactic orientation.

Materials and Methods

Microorganisms and experimental solutions

Paramecium caudatum was grown at 24 °C in a hay
infusion in Dryl's solution (2 mM sodium citrate, 1.2 mM
Na,HPO4. 1 .0 mM NaH2PO4, 1 .5 mM CaCU, pH 7.2). Cells
grown to the early stationary phase (14-20 d after incuba-
tion) were collected and adapted in the experimental solu-
tion (KCM; 1.0 mM KC1, 1.0 mM CaCU, 1.0 mM MOPS,
pH 7.2). After the adaptation, cells gravitactically accumu-
lating beneath the water surface were collected and immo-
bilized in the KCM containing 5 mM NiCU. Hyper-density
KCM (P-KCM) was prepared by substituting a colloidal
solution of Percoll (Sigma) for water up to 60% (v/v) in
KCM. At 24 °C, the specific gravity and relative viscosity of
KCM were 1.00 and 1.02, respectively; those of P-KCM
were 1.06 and 1.57. Specific gravity of the experimental
solutions was determined by weighing the known volume,
and viscosity was measured by means of an Ostwald vis-
cometer.

Larvae of the sea urchin Hemicentrotus pulcherrimus
were grown in the laboratory at 17 °C (Degawa et ai,
1986). Larvae at the mid- to late gastrula stage and the early
pluteus stage (ca. 24 and 48 h after insemination, respec-
tively) were collected by hand centrifuge and washed once
with artificial seawater (ASW; 450 mM NaCl. 10 mM KC1,
10 mM CaCl2, 25 mM MgCl2. 28 mM MgSO4, 10 mM
Tris-HCl, pH 8.0). For immobilization, larvae were im-
mersed in ASW containing 2 mM KCN. Hyper-density
ASW (P-ASW) was prepared by substituting Percoll for
water up to 22% (v/v) in ASW. At 25 °C, the specific
gravity and relative viscosity of ASW were 1.01 and 1.07,
respectively; those of P-ASW were 1.04 and 1.14.

Recordings and analyses of gravity-dependent orientation

Ni2+-immobilized Paramechtm cells and KCN-immobi-
lized sea urchin larvae were transferred, with experimental
solutions to be tested, into a chamber made of a slide and
coverslip and silicone rubber spacer (inner dimension 12 X
24 X 1 mm for Paramecium and 1 6 X 1 6 X 1 mm for sea
urchin larvae) and kept air bubble-free without any partic-
ular sealant. The chamber was set on a horizontal micro-
scope equipped with a rotating stage. After trapping immo-
bilized specimens at the bottom or the top of the chamber
(depending on the density of the medium), the chamber was
rotated upside down, and the orientation motion during
vertical movement due to gravity was recorded with a video
camera (XC-77, Sony, Tokyo) and a videotape recorder. To
avoid the hydrodynamic interactions between nearby mov-
ing objects, we chose organisms moving down (or up) far
from neighbors (>1 mm, about 5 body lengths, apart). For

measuring the orientation angle, we selected recordings in
which the orientation motion was observed in a single focal
plane.

The orientation angle as a function of time (0, t) was
measured directly on the video monitor. The rotational
velocity as a function of orientation angle (dOldt, 6) was
obtained as an average velocity ((fl,+ i •- 0,)/A?) at the
angle of geometrical average ((0, + 0,+ ,)/2) between
every successive datum of inclination angle versus time. /3
in Equation 8 was obtained by nonlinear least-squares re-
gression of the velocity data (dO/dt, 0) to the equation

d9
~dt

= /3 sin (9 + a).

(ID

where a is a factor to adjust the angle between the morpho-
logically defined fore-aft axis and the mechanically defined

axis.

Results

The drag-gravity model is the major mechanism of
Paramecium

When Paramecium was immobilized by Ni2 + , it main-
tained an anterior-thinner cell shape. This shape was pre-
served in P-KCM as well as in KCM; cells showed no
significant changes in axial length (162 ± 17 jam [/; = 30]
and 163 ± 16 jim [n = 21]. P = 0.64, for cells in KCM
and P-KCM, respectively) or in maximum width (47.2 ±
6.9 p.m and 46.5 ± 4.7 p.m, P = 0.69). Thus it is highly
likely that rotational motion of the immobilized cell occurs
with the same coefficient of resistance in both media.

Typical recordings of gravity-dependent orientation of
immobilized paramecia in the hypo- and hyper-density me-
dia are shown in Figure 2a and b. In KCM (p < p,.),
paramecia oriented upwards during sinking due to gravity,
whereas in P-KCM (p > p,) they oriented downwards
during floating up. As shown in Figure 2c. plots of orien-
tation rates (d6/dt) against orientation angle (0) fit well to
the sinusoidal function of Equation 1 1 . Values for |3 ob-
tained by least-square regression were positive in the con-
trol hypo-density medium and negative in the hyper-density
medium (Table 1 ). Negative values of (3 in the hyper-
density medium indicate that the drag-gravity model is the
major mechanism of mechanical gravitactic orientation in
Paramecium.

Sea urchin lamie change the mechanical mechanism of
gravitactic orientation during development

When sea urchin larvae were treated with KCN, their cilia
ceased beating and stood nearly perpendicular to the larval
surface. The outer morphology of the larvae was observed
to be well preserved in P-ASW as well as in ASW: for
gastrulae, axial length was 151 ± 7.6 p_m (n =~- 16) and
145 ± 6.1 jum (n = 13), P = 0.19, in ASW and P-ASW,

Y. MOGAMI ET AL.

</>
T3

0.15

0.10

0.05

-0.05

-0.10
-0.15

9 (rad)

Figure 2. Typical examples of gravity-dependent orientation of Ni2 +
immobilized Paramecium caudutiim. (a, h) Sequential images of gravity-
dependent orientation of a cell in KCM (a) and of another in P-K.CM (h),
in which recorded images are superimposed at l-s intervals and the time
sequence of the motion is illustrated by cyclic change in tone (dark — »
medium — > light). In each figure the anterior end of the cell is located to the
right, and the gravity vector is towards the bottom of the figure. Scale bar.
0.1 mm. (c) Orientation rates (iltt/ilo as a function of the inclination angle
(0). Data from the cells shown in a (KCM) and b (P-KCM) are plotted with
open and closed circles, respectively. Sinusoidal curves were obtained by
the least-squares fitting to Equation 1 1.

respectively, and the maximum width was 135 ± 3.7 JLUTI
and 132 ± 5.7 p,m, P = 0.06; for plutei, axial length was
235 ± 19 ju.m (H = 26) and 240 ± 13 /am <;i = 18), P =
0.29, in ASW and P-ASW, respectively, and the maximum
width was 175 ± 13 jam and 175 ± 12 juni, P = 0.98. This
may justify the common basis for drag coefficients in rota-
tion in the different density media, as in Parumeciiim.

The gravity-dependent orientation of immobilized larvae
is shown in Figure 3a to d, which demonstrates the clear
difference between gastrula and pluteus. In ASW (p < p,).
both gastrula and pluteus oriented upwards while sinking; in

hyper-density P-ASW, however, gastrula oriented down-
wards but pluteus upwards while floating up. As shown in
Figure 3e and f, the orientation rate appears to be a sinu-
soidal function of the orientation angle; although data from
larvae fitted less closely to Equation 1 1 than did those from
Paramecium, this was probably due to the uncertainty in
measuring the orientation angle of the larvae. We some-
times observed that larvae rotated slowly around the fore-aft
axis during sedimentation. This slow axial rotation made it
difficult to determine the fore-aft axis of the larvae.

As shown in Table 1. values of |3 obtained from gastrula
larvae were positive in the control medium and negative in
the hyper-density medium. Thus, in gastrulae as in Para-
mecium, the drag-gravity model is the major mechanism of
passive gravitactic orientation. However, pluteus larvae
have positive values of |3 both in the control and in the
hyper-density medium (Table 1). The relatively weak de-
pendency of |3 of plutei on the density of the external
medium indicates that the gravity-buoyancy model is the
major mechanism of passive gravitactic orientation in these
larvae. These results indicate that sea urchin larvae change
the mechanical mechanism of gravitactic orientation during
development.

Discussion

Estimation of the contribution of the mechanical models
in the gravitactic orientation

The Reynolds number of rotational motion (Re,) of the
microorganisms is defined as

Re, =

/:o>p

(12)

where / is a characteristic body length and w is the angular
velocity of rotation (Happel and Brenner. 1973). From the
maximum velocity of rotation (cu. 0.2 rad • s~'. Table 1),
Re, of Paramecium or sea urchin larvae is calculated to be
about 2 X 10~\ which is sufficiently smaller than unity.
This means that the linear assumption of Equation 7 (see the
Theory section) is valid to formulate the rotational motion
of these microorganisms.

The orientation torque generated as a result of the com-
bination of the torque originating from different mechanical
sources causes the passive orientation of the immobilized
organisms. It is difficult to formulate the combination, be-
cause we know little about the density distribution within an
organism and its geometrical asymmetry. The simplest as-
sumption for the combination of the rotational torque is that
G. B. and H are located on the geometrical fore-aft axis of
the organisms. This gives a sinusoidal function as a linear
summation of the sinusoidal equations, each of which is
deduced from the gravity-buoyancy and drag-gravity
model, respectively. As a result, the orientation rate is given
as

MECHANICAL BIAS OF MICROBIAL GRAVITAXIS
Table 1

Orientation rate t{5), in rad ' s . measured in different densitv media

Normal medium

Percoll-containing medium

Organism

Mean ± SD

Range

Mean

± SD

Range «

Paramecium

0.090

± 0.033

0.043

- 0.

183

-0.104

± 0.058

-0.257 -

-0.041

Sea urchin larvae

Gastrula

0.140

± 0.032

0.107

- 0.

197

-0.120

± 0.020

-0.150-

-0.090

Pluteus

0.157

± 0.03 1

0.105

- 0.

190

0.1 10

± 0.013

0.097 -

0.137

sintf.

13)

This simple linear assumption seems to be supported by the
fact that a in Equation 1 1 was calculated on average as
nearly zero (0.00 ± 0.26 rad (n = 37) for Paramecium,
0.03 ± 0.18 (;i = 15) for gastrula, and 0.06 ± 0.21 (// =
16) for pluteus). Therefore, it is likely that the morpholog-
ically defined fore-aft axis almost coincides with the me-
chanically defined axis. According to the assumption above,
|35 obtained in the different density media are given by

VpigL0 V(p, - pN)gLH
~ — ~~^ — ~

_
PP —

V(p,-pP)gLH

(15)

where /3;V is the maximum orientation velocity measured in
the normal density (pN) medium (KCM or ASW) of the
viscosity of TJ^, and fBP is that measured in the hyper-
density (pp) medium (P-KCM or P-ASW) of the viscosity
of T]P. Equations 14 and 15 give LH. the distance from B to
H, as

PP - Ps Vg '
and. thus, f3R and j8v are given by:

= /3.v -

For Paramecium, pv = 1.00, pp = 1.06 and p, = 1.03
g • cm"3 (Ooya et ai, 1992), and T)Plr\N = 1.53. For sea
urchin larvae, pN = 1.01. pp = 1.04, and p, = 1.03 and
1.03 g • cm~3, for gastrula and pluteus, respectively (values
were obtained by sedimentation equilibrium experiments:
data not shown), and TJ/./TJ^. = 1.07. Using these values
and /3V and PP in Table 1. Equations 17 and 18 can be used
to obtain values for the contribution of the two mechanisms
to negative gravitaxis in normal-density medium. The up-
ward orientation of Paramecium in KCM, corresponding to
f3N = 0.09 rad • s~ ', is the result of an upward drag-gravity

component ( J3R = 0. 1 2 rad • s ' ) combined with a smaller
downward gravity-buoyancy component (/3V = -0.03
rad • s '). The situation is similar for sea urchin gastrulae.
The upward orientation with |3A, = 0.14 rad • s~' results
from an upward drag-gravity component ({$R = 0.18 rad •
s"1) combined with a small downward gravity-buoyancy
component (j8v = -0.04 rad • s~'). However, the upward
orientation of pluteus larvae with fiN = 0.16 rad • s"1
reflects a very different situation. The gravity-buoyancy
component has reversed direction from downward to up-
ward, and has increased to j8v- = 0.13 rad • s~ ' . The upward
drag-gravity component has diminished greatly, to f3K =
( 14) 0.03 rad • s , so that it now makes only a small contribu-
tion to the upward orientation.

The mechanical property of 'Paramecium

There have been several investigations on the mechanical
basis of the passive upward orientation of Paramecium.
Most of them favored the gravity-buoyancy model as a
major mechanism of gravitactic orientation. Fukui and Asai
(1980) reported that Triton-treated immobilized cells ori-
ented mostly upwards at the sedimentation equilibrium in
sucrose density gradient. This upward orientation was evi-
dent in well-fed cells but not in starved cells. The upward-
orienting posture was found under centrifugal forces in
Ni2 + -immobilized cells in the isodensity medium (Taneda
et ai, 1987) and also in the cells swimming at isopycnic
level in the density gradient with Ficoll or Percoll (Kuroda
and Kamiya. 1989). It was also reported that upward orien-
tation was induced by centrifugal force effectively in the
cells at the early culture phase but not in those at the late
phase, which showed little or no gravitaxis. These results
appear to conform with the conclusion that the upward
orientation of Paramecium is strongly biased by the torque
resulting from the higher density of the posterior part of the
organism: the increased density is mainly due to the accu-
mulation of food vacuoles (Fukui and Asai, 1985).

It should be noted, however, that the results of the sedi-
mentation equilibrium experiments were ascribed only to
the function of the gravity-buoyancy model and not to the
contribution of the drag-gravity model, since FH = 0 with
buoyancy artificially balanced with gravity. Furthermore, it

(16)

17)

Y. MOGAMI ET AL.

1/5

0.20

0.15

010

0.05

-0.05
-0.10
-0.15
-0.20

0.25

0.20

0.15

0.10

0.05

-0.05

n/2

6 (rad)

n/2

6 (rad)

Figure 3. Typical examples of gravity-dependent orientation of KCN-
immobilized sea urchin (Hemicentrotus pulcherrimus) larvae, (a-d) Se-
quential images of gravity-dependent orientation of the single different
larvae at the gastrula (a and b) and the pluteus (c and d) stages. Movements
of a larva in ASW (a and c) and of another in P-ASW (b and d) are shown
at 3-s intervals in the same way as in Fig. 2a and b. In each figure the
animal pole of the larva (leading end in forward swimming) is located to
the right, and the gravity vector is towards the bottom of the figure. Scale
bar. 0.1 mm (e. f) orientation rates (ilti/dt) as a function of the inclination
angle (D). measured from gastrula (e) and pluteus (f). In e. data from the
gastrulea shown in a (ASW) and b (P-ASW) are plotted with open and
closed circles, respectively. In f, data from the plutei shown in c (ASW)
and d (P-ASW) are plotted with open and closed circles, respectively.
Sinusoidal curves were obtained by the least-squares fitting to Equation 1 1.

seems likely that the gravity-buoyancy component of the
orientation torque might be enhanced in these experiments.
Since the center of gravity would shift in relation to the

content and the distribution of organelles such as food
vacuoles, it is probable that in the sedimentation equilib-
rium experiments, the intracellular distribution of the or-
ganelle was reorganized by gravity during long-lasting sed-
imentation of Triton-permeabilized cells through the
sucrose density gradient (Fukui and Asai, 1980), or by a
large centrifugal acceleration ( 100 X g, Taneda et al., 1987;
300-400 x g, Kuroda and Kamiya, 1989). This may result
in accumulation of organelles in the rear part of the cell, and
may cause upward orientation, even if the cells originally
have a slightly top-heavy organelle distribution that gives a
negative j3v/ as estimated above. These facts suggest that the
results of previous experiments are still equivocal for the
contribution of the drag-gravity model in the gravitactic
orientation of Parameciiini.

The evidence presented in the Results, on the contrary,
indicate that the drag-gravity model makes a major contri-
bution to generating a torque for the gravitactic orientation.
Although the possibility of a minimal contribution cannot
be ruled out, it is clear that the gravity-buoyancy model
cannot solely explain the alteration of the sign of the rota-
tional torque in the hyper-density medium. In addition,
paramecia were observed in P-KCM to swim mostly down-
wards (data not shown). Swimming cells changed the net
direction of their helical swimming trajectory gradually
downwards and accumulated at the bottom of the chamber
against the strong floating bias. Positive gravitaxis of Par-
ameciitm in the hyper-density medium can be explained by
the drag-gravity model, not by the gravity-buoyancy model.

Developmental clmnges in the mechanical property in sea
urchin lan'ae

In the present paper we demonstrated a change in the
mechanical basis for gravitactic orientation during the de-
velopment of sea urchin larvae: from the drag-gravity model
in gastrulae to the gravity-buoyancy model in plutei. Gas-
trulae have a thicker posterior part, similar to that of Par-
ciiiieciiiin. which is required for the drag-gravity model to
function. Plutei. on the other hand, have a thicker anterior
part. Therefore they may orient the rear end upwards if the
rotational torque is generated according to the drag-gravity
model. This was not the case for plutei. Regardless of the
remarkable fore-aft asymmetry in morphology, plutei
obeyed the gravity-buoyancy model. Gravitactic orientation
by different mechanisms was also revealed in the gravitactic
swimming behavior of the larvae in P-ASW. In spite of the
strong floating bias, gastrulae swam preferentially down-
wards (positive gravitaxis) and accumulated at the bottom
of the chamber, whereas plutei swam upwards (negative
gravitaxis) and accumulated at the top of the chamber (data
not shown).

Mogami et al. ( 1 988) found that sea urchin larvae change
their gravitactic behavior during development. Larvae at the
blastula stage to the early gastrula stage swim preferentially

MECHANICAL BIAS OF MICROBIAL GRAVITAXIS

upwards. This may be explained by a major upward drag-
gravity component of orientation torque. The negative
gravitatic behavior becomes less remarkable in prism lar-
vae: they tend to swim in random directions independent of
the gravity vector. This transient disappearance of gravi-
taxis may correspond to the alteration of the orientation
mechanism revealed in the present paper. At the pluteus
stage, larvae again show negative gravitaxis as they acquire
the orientation mechanism with a major upward gravity-
buoyancy component. A strong separation between the cen-
ters of gravity and buoyancy may develop in association
with the growth of skeletal structures. Rudiments of spicules
initiated in the early gastrula fully extend to give rise to the
specific shape of the pluteus larva. The spicule is made of
magnesian calcite with a density about three times higher
than the average density (Okazaki and Inoue, 1976). As
spicules grow, they may change the density distribution to
shift the center of gravity toward the rear of the cell. If plutei
hereafter maintained the rear-end-heavy mass distribution,
they could maintain negative gravitactic behavior irrespec-
tive of pronounced morphological changes during the late
larval stages.

Although the functional role of the drag-gravity model
has been accepted in theory, it was not experimentally
demonstrated in the orientation movement of organisms. In
the present paper we present the first evidence that external
geometry is actually important to the gravitactic behavior of
aquatic microorganisms. The morphology-dependent inter-
action of the organisms with the external fluid seems to be
more complicated than hypothesized in the Theory section
of this paper. The slow axial rotation observed in sediment-
ing sea urchin larvae indicates a hydrodynamic coupling
between translational and rotational motion (Happel and
Brenner, 1973). Therefore, it is probable that the hydrody-
namic coupling secondarily functions to drift the swimming
direction upwards, as argued in previous researches (Winet
and Jahn. 1974; Nowakowska and Grebecki, 1977).

In conclusion, the present study on the mechanical prop-
erties of gravitactic orientation in the gravity field demon-
strates a relation between the morphology of microorgan-
isms and their gravitactic behavior. This relationship might
be instructive in researching cases of microbial gravitaxis
whose mechanism is still disputed.

Acknowledgments

This study was carried out as a part of "Ground Research
Announcement for Space Utilization" promoted by Japan
Space Forum.

Literature Cited

Bean, B. 1984. Microhial geotaxis. Pp. 163-198 in Membrane and
Sensory Transduction, G. Colombetti and F. Lenci, eds. Plenum Press,
New York.

Chia. F-S.. J. Buckland-Nicks, and C. M. Young. 1983. Locomotion of
marine invertebrate larvae: a review. Can. J. Zool. 62: 1205-1222.

Degawa, M., Y. Mogami, and S. A. Baba. 1986. Developmental
changes in Ca"^ sensitivity of sea-urchin embryo cilia. Comp. Bio-
chem. Physiol. 82A: 83-90.

Fenchel, T., and B. Finlay. 1984. Geotaxis in the ciliated protozoan
Loxodes. J. Exp. Biol. 110: 17-33.

Fenchel, T., and B. Finlay. 1986. The structure and function of Muller
vesicles in loxodid ciliates. J. Protozool. 33: 68-76.

Fukui, K., and H. Asai. 1980. The most probable mechanism of the
negative geotaxis of Paramecium caudatum. Pmc. Jpn. AcuJ. 56(B):
172-177.

Fukui, K., and H. Asai. 1985. Negative geotactic behavior of Parame-
ciiim caudatum is completely described by the mechanism of buoyan-
cy-oriented upward swimming. Bioph\s. J. 47: 479-482.

Happel, J., and H. Brenner. 1973. Low Reynolds Number Hydrody-
namics. Noordhoff International Publishing, Leyden.

Kuroda, K., and N. Kamiya. 1989. Propulsive force of Paramecium as
revealed by the video centrifuge microscope. Exp. Cell Res. 184:
268-272.

Machemer, H., and R. Braucker. 1992. Gravireception and gravire-
sponses in ciliates. Acta Protozool. 31: 185-214.

Machemer, H., S. Machemer-Riinisch, R. Braucker, and K. Takahashi.
1991. Gravikinesis in Paramecium: theory and isolation of a physi-
ological response to the natural gravity vector. J. Comp. Physiol. A
168: 1-12.

Machemer-Rb'hnisch, S., U. Nagel, and H. Machemer. 1999. A grav-
ity-induced regulation of swimming speed in Euglena graci/is.
J. Comp. Physiol. 185: 517-522.

Mogami, Y., C. Oobayashi, T. Yamaguchi, Y. Ogiso, and S. A. Baba.
1988. Negative geotaxis in sea urchin larvae: a possible role of
mechanoreception in the late stages of development. J. Exp. Biol. 137:
141-156.

Nowakowska, G., and A. Grehecki. 1977. On the mechanism of orien-
tation of Paramecium caudatum in the gravity field. II. Contributions to
a hydrodynamic model of geotaxis. Acta Protozool. 16: 359-370.

Okazaki, K., and S. Inoue. 1976. Crystal property of the larval sea
urchin spicule. De\: Growth Differ. 18: 413-434.

Ooya, M., Y. Mogami, A. Izumi-Kurotani, and S. A. Baba. 1992.
Gravity-induced changes in propulsion of Paramecium caudatum: a
possible role of gravireception in protozoan behaviour. J. Exp. Biol.
163: 153-167.

Roberts, A. M. 1970. Geotaxis in motile micro-organisms. J. E.xp. Biol.
53: 687-699.

Taneda, K., S. Miyata, and A. Shiota. 1987. Geotactic behavior in
Paramecium caudatum II. Geotaxis assay in a population of the spec-
imens. Zool. Sci. 4: 789-795.

Verworn, M. 1889. Psychophysiologische Protistenstudien. Gustav
Fischer, Jena. (Cited by Machemer and Braucker. 1992.)

V'ogel. S. 1994. Life in Moving Fluids: The Physical Biology of Flow.
2nd ed. Princeton University Press, Princeton. NJ.

Winet, H., and T. L. Jahn. 1974. Geotaxis in protozoa. I. A propulsion
gravity model for Tetrahymena (Ciliata). J. Theor. Biol. 46: 449-465.

Reference: Bio/. Bull. 201: 34-44. (August 2001)

Synthesis of Several Light-Harvesting Complex I

Polypeptides Is Blocked by Cycloheximide in

Symbiotic Chloroplasts in the Sea Slug, Elysia

chlorotica (Gould): A Case for Horizontal Gene
Transfer Between Alga and Animal?

JEFFREY J. HANTEN1'2 AND SIDNEY K. PIERCE2*

1 Department of Biology. University of Man-land, College Park, Maryland 20742; and
2 Department of Biology, University of South Florida. Tampa, Florida 3362G

Abstract. The chloroplast symbiosis between the asco-
glossan (=Sacoglossa) sea slug Elysia chlorotica and plas-
tids from the chromophytic alga Vaitcheria litorea is the
longest-lived relationship of its kind known, lasting up to 9
months. During this time, the plastids continue to photosyn-
thesize in the absence of the algal nucleus at rates sufficient
to meet the nutritional needs of the slugs. We have previ-
ously demonstrated that the synthesis of photosynthetic
proteins occurs while the plastids reside within the diver-
ticular cells of the slug. Here, we have identified several of
these synthesized proteins as belonging to the nuclear-
encoded family of polypeptides known as light-harvesting
complex I (LHCI). The synthesis of LHCI is blocked by the
cytosolic ribosomal inhibitor cycloheximide and proceeds
in the presence of chloramphenicol, a plastid ribosome
inhibitor, indicating that the gene encoding LHCI resides in
the nuclear DNA of the slug. These results suggest that a
horizontal transfer of the LHCI gene from the alga to the
slug has taken place.

Introduction

Most alga-animal symbioses are extracellular associa-
tions between two genetically distinct organisms. The alga
is usually located extracellularly or enclosed within vacu-

Received 22 September 2000; accepted 19 April 2001.

* To whom correspondence should be addressed. E-mail:
pierce@chumal.cas.usf.edu

Abbreviations: CAP, chloramphenicol: CHX. cycloheximide; FCPC,
fucoxanthin chlorophyll ale binding proteins; LHC. light-harvesting com-
plex; PSI; photosystem I.

oles inside the animal's cells. Rarer, but not uncommon, are
intracellular symbioses occurring with intact algal chloro-
plasts that are captured by specialized cells within the
animal. In particular, several species of ascoglossan
( = Sacoglossa) (Opistobranchia) sea slugs capture intact,
functional plastids from their algal food source and retain
them within specialized cells lining the mollusc's digestive
diverticula. This phenomenon has been termed chloroplast
symbiosis (Taylor. 1970) or kleptoplasty (Clark et al..
1990). The sequestered plastids continue to photosynthesize
for periods ranging from a few days to a few months,
depending on the species (Greene, 1970: Hinde and Smith,
1974; Graves et al., 1979; Clark et al., 1990).

The longest such association, lasting as long as 9 months,
is found in Elysia chlorotica (Gould), which obtains sym-
biotic plastids from the chromophytic alga Vaucheria lito-
rea (C. Agardh) (West, 1979; Pierce et al.. 1996). The
association begins at metamorphosis of the slug from plank-
tonic veliger to juvenile. In laboratory cultures, filaments of
V. litorea must be present for metamorphosis to take place
(West et al.. 1984). Veligers home in, attach to the fila-
ments, and metamorphose into juvenile slugs over the next
24 h. The juveniles eat the algal filaments and sequester the
chloroplasts within one of at least two morphologically
distinct types of epithelial cells lining the walls of the
digestive diverticula (West et al., 1984). Once the plastids
are sequestered, the slugs can sustain photosynthesis at rates
sufficient to satisfy the nutritional needs for the complete
life cycle of the slug, when provided with direct light and
carbon dioxide (Mujer et al., 1996; Pierce et al., 1996).

SYMBIOTIC PLASTID GENES IN SLUGS

Even in nature the slugs obtain most of their energy from
photosynthesis (West. 1979).

The longevity of this relationship in E. chlorotica makes
it especially interesting. Photosynthesis requires the contin-
uous synthesis of a variety of chloroplast proteins because
many of them, including those used in light harvesting, are
rapidly degraded and must be replaced (Greenberg et cil..
1989; Mattoo et ai, 1989; Barber and Andersson, 1992;
Wollman et ai. 1999). Furthermore, photosynthesis re-
quires the interaction of as many as 1000 proteins, only
about 13% of which are coded in the plastid genome (Mar-
tin and Herrmann, 1998). In the plant cell, substantial nu-
clear input is required to sustain photosynthetic function, in
the form of direct coding of the proteins as well as providing
the means for their intracellular transport and regulation
(Berry-Lowe and Schmidt. 1991; Wollman et ui, 1999).
Considering the level of nuclear and extra-plastid input
required, it is not surprising that the longevity of the plastids
in most kleptoplastic slugs is relatively short. However,
several photosynthetic proteins are synthesized in the se-
questered plastids of E. chlorotica (Pierce et ai, 1996),
including the large subunit of RuBisCO, Dl, D2, CP43, cyt
/and others (Pierce et ai, 1996; Mujer et ai, 1996; Green
et ai, 2000). Although all of the synthesized plastid proteins
identified to date are plastid encoded (Mujer et ai, 1996;
Pierce et ai, 1996; Green et ai, 2000). two groups of
synthesized plastid proteins can be distinguished pharma-
cologically: those inhibited by cycloheximide (CHX). an
SOS cytosolic ribosome inhibitor (Obrig et ai, 1971). and
those inhibited by chloramphenicol (CAP), which inhibits
protein synthesis on 70S plastid and mitochondrial ribo-
somes (Lamb et ai, 1968: Stone and Wilke. 1975).

Because the inhibition by CHX suggests that the genes
for several plastid proteins must reside in the nuclear DNA,
we have done some experiments to identify these proteins
and test that possibility. Our present study reports the iden-
tification of several of the CHX-blocked proteins as mem-
bers of the light-harvesting complex 1 (LHCI). a family of
pigment-binding proteins responsible for collecting radia-
tion energy from sunlight and transferring it to photosystem
I (PSI). LHCI proteins are encoded by the Lhca genes in the
nuclear genome of all the plants and algae whose genomes
have been examined (Jansson, 1994. 1999; Green and Durn-
ford, 1996; Durnford et ai, 1999; Wollman et ai, 1999).
This result suggests that the LHCI genes have been some-
how transferred from the algal nucleus to the slug's DNA.

Materials and Methods

Animals and alga

Specimens of Elysici chlorotica were collected in both the
spring and fall from an intertidal marsh near Menemsha
Pond on the island of Martha's Vineyard, Massachusetts.
The slugs were maintained in 10-gallon aquaria at 10 °C in

aerated, artificial seawater (ASW: Instant Ocean. 925-1000
mosm) on a 16/8-h light/dark cycle (GE cool-white fluores-
cent tubes. 15 W).

Sterile cultures of Vaucheria litorea were maintained in
enriched ASW (400 mosm) [modified from the F/2 medium
(Bidwell and Spotte. 1985)]. The alga was grown at 20 °C
on a 16/8-h light/dark cycle (GE cool-white fluorescent
tubes; 40 W). and the medium was changed weekly.

Inhibitor treatments and plastid protein labeling

All reagents used were molecular bio-grade (DNase-,
RNase-. and protease-free) purchased from Sigma unless
otherwise noted. Effective concentrations of CHX and CAP
were determined empirically with initial dose-response
curves (Pierce et ai, 1996). CHX (2 mg ml"1) was used to
inhibit protein synthesis on SOS cytosolic ribosomes; CAP
(160 /o,g ml , stock concentration 50 mg ml"1 in absolute
ethanol) was used to inhibit translation on 70S plastid and
mitochondrial ribosomes. Two to four slugs, total wet
weight about 1.25 g, were placed into glass scintillation
vials containing ASW (1000 mosm) and the appropriate
inhibitor, and incubated under intense light (150 W, GE
Cool Beam incandescent indoor flood lamp) at 20 °C in a
gently agitating water bath. After 1 h. 20 /iCi ml"1 [35S]-
methionine (0.7 MBq ml"1, trans-[35S]-methionine, ICN)
was added, and the slugs were incubated for an additional
6 h. previously demonstrated to provide ample time to
incorporate radioactive label into the plastid proteins (Pierce
et ai, 1996). Additional slugs were incubated in 0.025%
ethanol/ASW (v/v) solution plus [35S]-methionine to serve
as a control for the carrier in CAP treatments.

Chloroplast isolation and protein separation

Chloroplasts were isolated from slugs by using a centrif-
ugation protocol. The slugs were homogenized in the pres-
ence of the mucolytic agent N-acetyl-cysteine (500 mM).
and the homogenate was filtered successively through
cheesecloth, Miracloth (Calbiochem), and then nylon mesh
(60 ;um to 10 jitm) to remove large debris and the copious
amount of mucus the animals produce. The plastids were
purified on a pre-formed. 25% Percoll (v/v) gradient, which
provides a very pure fraction containing large numbers of
intact plastids (Pierce et ai, 1996). In this experiment, the
lowest green band containing labeled plastids was isolated
from the gradient by using a flamed Pasteur pipette, and
residual Percoll was removed by centrifugation. The puri-
fied chloroplast pellets were resuspended, lysed by freeze-
thawing. and stored at — 20°C until use. The incorporation
of radioactive label was determined by a liquid scintillation
counter (Beckman LS60001C). and the protein content was
determined using the modified Lowry assay (Peterson,
1977). The resulting specific activity was calculated as
counts per minute (cpm) (jug protein)^'. Chlorophyll con-

J. J. HANTEN AND S. K. PIERCE

tent was determined by extracting the pigment in 80%
acetone, then measuring the extract absorbance spectropho-
tometrically at 652 nm. The results were calculated as
micrograms per microliter according to standard equations
(Joyard et al., 1987).

Sodium dodecyl sulfate-polyacrylamide gel electro-
phoresis (SDS-PAGE) autoradiography was used to assess
the effects of CHX and CAP on the pattern of protein
synthesis. The plastid lysates obtained from the above pro-
cedure were boiled for 2 min in Tris-HCl (pH 6.8)-10%
SDS (w/v) buffer containing 5% /3-mercaptoethanol (|3-
ME) (v/v). The solubilized proteins were loaded in equal
amounts onto 15% SDS-polyacrylamide gels and separated
by electrophoresis (Laemlli, 1970). The gels were stained
with Coomassie brilliant blue, dried, and exposed to film
(Kodak Biomax MR) for 2 to 30 days at -80°C, depending
on the level of radioactive label present. Approximate mo-
lecular masses of the proteins were determined by compar-
ison to the migration distances of known molecular weight
standards (BioRad, broad-range kaleidoscope) run in adja-
cent lanes on each gel.

Immunoblot identification of plastid proteins

After the plastid isolation and protein separation via
SDS-PAGE as described above, the proteins were electro-
phoretically transferred (30 V, 4 °C, overnight) to PVDF
membranes (Immobilon-P; Millipore) (Towbin et til..
1979). As additional controls, V. litorea chloroplasts [iso-
lated and purified using a 30% to 75% Percoll step gradient
as previously described (Pierce et al., 1996)] and thylakoids
from the red alga Porphyridium cmentum (generously do-
nated by Professor Elisabeth Gantt, University of Mary-
land), were lysed. and the proteins were separated electro-
phoretically and transferred to membranes as above. The
membranes were blocked with 5% (w/v) dehydrated milk
dissolved in Tris-buffered saline (TBS) (Tris-base 50 mM.
NaCl 0.9%, pH 7.5) for 1 h at room temperature, washed
twice in TBS for 10 min, and treated with primary antibody
for 1 h. In this case, the primary antibody was a polyclonal
antibody to LHCI which was produced in a rabbit using a
22-kDa, recombinant LHCI polypeptide produced from a
clone of the LhcaRI gene of P. cmentum (Grabowski et al.,
2000) (also provided by Professor Gantt) ["/?/" indicating it
is a rhodophyte gene (Tan et al., 1997a)] as the antigen
combined with Freund's adjuvant in a standard immuniza-
tion procedure. After binding of the primary antibody, the
membranes were washed twice as above and incubated with
secondary antibody, anti-rabbit conjugated hydrogen perox-
idase, for 1 h. After washing, the bands were visualized with
a 4-chloro-l-napthol and hydrogen peroxide reaction ac-
cording to manufacturer's instructions. The immunolabeled
western blots were exposed to film as described above to

identify the coincidence of antibody binding and radioactive
incorporation in the presence of each inhibitor.

As a control to confirm that the CAP was blocking
plastid-directed protein synthesis and that CHX was not,
parallel measurements were run to monitor cytochrome /
(cyt/ ) synthesis. Earlier experiments conducted on E. clilo-
rotica have demonstrated that cyt / is synthesized in the
slugs and is encoded in the plastid DNA (Green et al.,
2000). Thus, if CHX and CAP are working as expected,
their effect on cyt / and any nuclear-encoded proteins
should be opposite. Anti-cyt/, raised to P. cmentum cyt/,
was also a gift of Professor Gantt.

Immunoprecipitations

Immunoprecipitations were conducted to confirm the
identity of the radioactive immunolabeled bands on the
western blots, using a modified version of the protocol
previously used to precipitate proteins from isolated E.
chlorotica plastids (Pierce et al., 1996). Plastid proteins
were solubilized in lysing buffer ( 10 mM Tris-HCl, 10 mM
EDTA, 150 mM NaCl, 1 mM PMSF, 1% (v/v) Nonidet
P-40, pH 8.0), using equal amounts of chlorophyll per
sample, mixed with a small amount of Protein-A Sepharose
beads to eliminate nonspecific binding, and incubated on ice
with occasional agitation. The beads were removed by cen-
trifugation and discarded, the supernatant was saved, and
the appropriate antibody was added to the lysate and rotated
overnight (4 °C). Protein-A beads, swelled in washing
buffer (50 mM Tris-HCl, 5 mM EDTA, 150 mM NaCl, 1
mM PMSF, 0.1% (v/v) Nonidet P-40, pH 8.0), were added
the following morning and rotated (3 h, room temperature).
The antigen-antibody-protein-A Sepharose bead com-
plexes were washed several times in washing buffer and
removed by centrifugation. In the case of cyt/, the antigen-
antibody-protein-A Sepharose bead complexes were resus-
pended in 10.0 M urea. 10% SDS (w/v), 5% |3-ME (v/v), pH
12.5, and boiled for 10 min to liberate the cyt/ antigen. The
solution was centrifuged, the supernatant was removed, and
the beads were discarded. The supernatant proteins were
separated by SDS-PAGE as described above, and the gel
was autoradiographed.

The LHCI antibody-antigen complex could not be broken
efficiently with any treatment, which prevented the visual-
ization of the labeled LHCI proteins via SDS-PAGE. Al-
though this was unexpected, it is not unusual and may have
been caused by a number of factors. The presence of several
different LHCI polypeptides with varying isoelectric points,
ranging between 4.5 and 9.5 (De Martino et al., 2000).
makes it very difficult to create optimal reaction conditions
for each one. The polyclonal antibody molecules bind to all
the LHCI polypeptides as well as to each other, creating a
large antigen-antibody complex with a core inaccessible to
the chemicals necessary to liberate the antigen. Very few

SYMBIOTIC PLASTID GENES IN SLUGS

researchers have attempted LHC immunoprecipitations be-
cause of the pitfalls involved in precipitating inner-mem-
brane proteins (Anderson and Blobel, 1983). Instead, other
protocols have been designed using mild detergents to ex-
tract intact photosystem holocomplexes from the thyla-
koids. followed by protein separation on sucrose density
gradients (Fawley and Grossman, 1986; Buchel and Wil-
helm, 1993: Wolfe end., 1994; Schmid et id.. 1997). These
isolations require large amounts of starting material
(Schmid et id.. 1997) that greatly exceed what is available to
us in the slugs. So, instead, we used the LHCI antibody to
demonstrate that LHCI had incorporated radioactivity.

Following the procedure described above, the protein A
Sepharose beads were reacted with anti-LHCI and then with
a radiolabeled plastid protein extract. The antigen-anti-
body-protein-A Sepharose bead complexes were repeatedly
washed by centrifugation until the radioactivity in the su-
pernatant was reduced to background. The washed antigen-
antibody-protein-A Sepharose bead complexes were resus-
pended in optifluor (Packard), and radioactivity was
determined by a scintillation counter. Controls for nonspe-
cific binding to protein-A Sepharose beads were conducted
with the same procedure, but without the addition of the
LHCI antibody. Counts per minute resulting from nonspe-
cific binding were subtracted from experimental values for
each inhibitor treatment and controls, and the final data were
converted to cpm (jug chlorophyll)" ' (ju,g protein)"1. The
normalized data were averaged and expressed in terms of
percent of control for each inhibitor.

Results

Plastid protein synthesis and identification

The Coomassie-stained SDS-PAGE gels of protein ex-
tracts from isolated slug plastids were similar to controls
regardless of the inhibitor present, either CHX or CAP,
indicating no difference in the protein composition of the
plastids after treatment (Fig. 1). However, autoradiograms
of SDS-PAGE gels of plastid proteins extracted from slugs
incubated in the presence of [35S]-methionine indicate that
very different patterns of protein synthesis occur in the slugs
between controls and inhibitors as well as between inhibi-
tors (Fig. 2). CHX has a profound effect on protein synthe-
sis, preventing synthesis of the majority of the protein bands
labeled in the absence of inhibitor (Fig. 2, CON), whereas
the synthesis of many more labeled bands occurs in the
presence of CAP. Furthermore, these protein bands differ
from those visualized in the CHX treatments (Fig. 2).

Verification of inhibitor effects

Cyt/ antibodies reacted with a protein band synthesized
in the presence of CHX on western blots at approximately
36 kDa (Fig. 3). Immunoprecipitations using anti-cyt /

(kDa) CON CHX CAP
218_l

43.5

33.9 _

17.4_

7.6_

Figure 1. Coomassie brilliant blue-stained 15% SDS-PAGE gel of
proteins extracted from isolated Elysia chloroplasts. The protein bands
visualized are identical regardless of the inhibitor treatment, CHX or CAP
(CON refers to control). Approximate molecular weights are indicated to
the left.

identify a band with a molecular weight corresponding to
cyt/, confirming its identity (Fig. 4). Autoradiograms of the
same gels show [15S]-methionine incorporation into cyt/ in
the presence of CHX. but not in the presence of CAP (Fig.
5).

The anti-LHCI we made to Porphyridium cruentum re-
combinant LHCI recognized both the recombinant LHCI
antigen (Fig. 5 A, lane 1 ) and the LHCI polypeptides from P.
cruentum thylakoids (Fig. 5B, lane 2). Six polypeptide
bands were identified in P. cruentum, ranging in approxi-
mate molecular weights from 19 to 24 kDa (Fig. 5B, lane 2),
sizes consistent with those previously described for the
LHCI polypeptides in this species (Tan et al, 1995). The
antibody bound onto western blots of plastid proteins from
Vciucheria litorea and Elysia chlorotica, with or without the
CHX and CAP treatments (Fig. 5C, lanes V. lit.. CON,

J. J. HANTEN AND S. K. PIERCE

(kDa) CON CHX CAP

126-
90.

43.5

Discussion

LHCI, a family of plastid polypeptides essential for pho-
tosynthesis, is synthesized while Vaucheria litorea chloro-
plasts reside within the cells of the digestive diverticula of
Elysia chlorotica. In addition, our data indicate the LHCI
polypeptides are probably the products of genes located in
the host-cell nuclear genome because their synthesis is
inhibited by the cytosolic ribosome inhibitor, CHX, but not
by the presence of the plastid ribosome inhibitor, CAP. This
remarkable result would not be surprising in a plant or algal
species since the LHCI polypeptide family's genes, Lhcal-
Lhca6, reside in the nuclear DNA of all plants and algae
examined to date (Jansson, 1994; Green and Durnford,

33.9_

(kDa)
126_

17.4 —

7.6

Figure 2. Autoradiograph of plastid proteins separated by SDS-PAGE
gel run under the same conditions as those depicted in Figure 2. The plastid
proteins incorporating [35S]-methionine label differ following treatment
with CHX or CAP. The control (CON) represents chloroplast proteins
isolated from slugs without inhibitor treatment. Arrows identify the ap-
proximate positions of cyt/ (large arrow) and the LHCI (small arrows)
proteins.

CHX, CAP). As expected, the six polypeptide bands bound
by the anti-LHCI in V. litomi and E. chlorotica plastids
have a slightly greater size range — 18 to 32 kDa — than
those identified in P. cnicntiiin. These same antibody-la-
beled bands from E. chlorotica plastid proteins incorporate
radioactive label in the presence of CAP. but incorporation
is blocked by the presence of CHX (Fig. 6).

The amount of radiolabel precipitated by anti-LHCI from
the slug plastid extracts following CHX treatment is only
2% of the control level, indicating a reduction in LHCI
synthesis (Fig. 7). In contrast, the LHCI proteins in CAP-
treated slugs incorporated [35S]-methionine at 92% of con-
trol rates, more than 40-fold higher than the level found in
CHX treated animals (Fig. 7).

43. 5 _

33.9 _

17.4_

7.6 _

Figure 3. Immunoblot labeled with antibody to cyt / (A), and its
corresponding autoradiograph (B). The slugs were exposed to CHX and the
proteins were labeled as described in the methods. Anti-cyt / binds at
approximately 36 kDa, coincident with a radiolabeled protein. The arrow
indicates the autorudiograph band corresponding to the position of cyt /.

SYMBIOTIC PLASTID GENES IN SLUGS

(kDa)

CONTROL^

CHX

CAP

16.8_

CBB Auto

CBB

Auto

CBB

Auto

Figure 4. Immunoprecipitation of cyt/. Coomassie brilliant blue (CBB)-stained gels of proteins precipitated
with anti-cyt / from chloroplast extracts from slugs subjected to no inhibitor (Control), to CHX. or to CAP. and
their corresponding autoradiographs (Auto). The arrow indicates the position of cyt/. Large bands above and
below cyt/ are the heavy and light chains of the antibody, respectively. The radioactivity corresponding to the
antibody bands in control and CHX is probably undissociated cyt /.

1996: Durnford et at., 1999: Jansson. 1999; Wollman et a/..
1999). However, the synthesis of LHCI directed by an
animal's genome indicates that genes have been transferred
into the slug DNA.

Although surprising, the site of synthesis and the identi-
fication of LHCI seem to be without question as long as
inhibitor and antibody specificity are not problems. Both
CHX and CAP have been used in a wide array of studies,
and their sites of action are well established. In fact, they
have been used, exactly as we have done here, to establish
that the site of synthesis of the "light harvesting chlorophyll
protein" (=LHCI) occurs on 80s cytoplasmic ribosomes in
Phaeodactyliini tricomutum (Fawley and Grossman, 1986).

There are several reasons to conclude that our antibody is
specific. We raised the antibody against the red alga LHCI
not only because it was available, but also because the
chromophytes, the taxonomic group of V. litorea, probably
arose through a secondary symbiosis from a red alga (Rieth,
1995; Green and Durnford, 1996; Palmer and Delwiche,
1996; Martin and Herrmann, 1998: Delwiche, 1999). Fur-
thermore. Porphyridium cnientum LHCI possesses both
sequence homologies and immunological relatedness to the
chromophytic light-harvesting proteins (Wolfe et ai, 1994;
Rieth, 1995; Tan et al.. 1997b). Thus, a polyclonal antibody
raised to a rhodophyte LHCI should have a good chance of
specifically recognizing the LHCI polypeptides in V. lito-
rea. Our results indicate that the anti-LHCI binds the P.
cnientum recombinant LHCI, the antieenic source for the

antibody, as well as all six of the native P. cnientum LHCI
proteins (Tan et al., 1995; Grabowski et al., 2000) in control
immunoblots of extracted thylakoids. The anti-LHCI immu-
noblots of E. chlorotica and V. litorea also identified six
protein bands with a greater size range than the LHCI
proteins identified in P. cnientum. Those bands are consis-
tent with the sizes of LHCI polypeptides from many species
(Gantt. 1996; Jansson. 1999: Wollman et ai, 1999), and no
other bands were labeled by the antibody. Seeing six LHCI
proteins is not surprising, because LHCI is typically found
in multiple homologs in algae, ranging from two in one
species of Xanthophyceae (Buchel and Wilhelm, 1993) to at
least six paralogs in some rhodophytes (Tan et al., 1995),
and as many as eight in the chromophyte Heterosigma
carterae (Durnford and Green, 1994). With few exceptions
[such as in Euglena gracilis (Jansson, 1994)], each is en-
coded by a separate, nuclear gene belonging to the Lhc
super-gene family (Jansson, 1999). Thus, location of the
gene aside, the presence of six LHCI proteins in the endo-
symbiotic plastids in E. chlorotica is not surprising.

It seems clear that each of the bands immunodecorated by
anti-LHCI corresponds to a single LHCI polypeptide and
not a dimer. LHCI dimers can result from their association
with other LHC proteins and their respective photosystems
/;; situ, and they do not always readily dissociate under the
denaturing conditions of SDS-PAGE (Tan et al., 1995). If
LHCI dimers were present here, they should have minimum
molecular weights of about 36 kDa, corresponding to dou-

J. J. HANTEN AND S. K. PIERCE

A B C

(kDa) 1 (kDa) 2 (kDa) V. lit CON CHX CAP

33.9—

29.0—

33.9—

17.4

18.2 —

17.4—

Figure 5. Immunoblots testing the antibody raised to Porphyridium inientum LHCI. (A) Anti-LHCI binds
the recombinant 22 kDa Llica RI product from P. cruentwn (lane i ). Its appearance as a 28-30 kDa protein in
SDS-PAGE and subsequent immunoblots results from the addition of a 33 amino acid N-termina! fusion in the
recombinant protein (Grabowski el at., 2000). (B) Anti-LHCI binds LHCI polypeptides extracted from P.
cruentum thylakoids (lane 2). (C) Vauclieria litorea (lane V. lit.) and Ely\ia chlorotica plastid proteins have six
bands binding the anti-LHCI identical in size to each other. All six proteins are present in the slugs regardless
of the inhibitor treatment [lanes CON (control). CHX and CAP]. Molecular weights are indicated to the left of
(A). (B). and (C).

ble the molecular weight of the smallest immunolabeled
band. However, the largest of the six immunolabeled bands
present in the gels is about 32 kDa, seemingly too small to
be an LHCI dimer.

Other dimers might form with a number of photosystem
I (PSD proteins due to the close association of LHCI with
the PSI subunits that compose the PSI-LHCI holocomplex
(Wollman et al.. 1999; Jansson. 1999). This also does not
seem to be the case here. Anti-PSI. raised against the
cyanobacteria PSI holocomplex (again, courtesy of Profes-
sor Gantt), binds a single 10-kDa protein band on western
blots of E. chlorotica plastid proteins (data not shown). The
combination of this PSI polypeptide with any of the three
smaller bands (18-20 kDa) that react with the anti-LHCI
could form a dimer with molecular weights comparable to
each of the three larger polypeptides (28-32 kDa). How-
ever, since anti-PSI and anti-LHCI do not co-label any
bands, an LHCI-PSI dimer is unlikely.

An additional possibility might be that one of the bands
could be another LHC-type protein possessing immunolog-
ical similarities to LHCI, such as the fucoxanthin chloro-
phyll ale binding proteins (FCPC) found in chromophytes
or light-harvesting complex II (LHCII) proteins. In fact, our
previous work has demonstrated the presence of FCPC in
plastids of both E. chlorotica and V. litorea. However, the
size of the FCPC protein identified there does not corre-
spond to the weights of the proteins bound by the anti-LHCI

used here (Pierce et al.. 1996; Green et til., 2000). Further-
more, previous attempts to demonstrate FCPC synthesis
with radioactive labels in the slugs have not yielded positive
results (Pierce et ai. 1996).

The LHCII family of polypeptides is closely related to
LHCI, performing similar functions in photosystem II to
those performed by LHCI in PSI. The LHC II genes are in
the same nuclear-encoded Lhc super-gene family (Jansson,
1999) and share sequence homologies with those genes
encoding LHCI (Durnford et al., 1999; Jansson. 1999;
Wollman et al., 1999). There is, however, a clear separation
in the phylogenies of LHCI and LHCII (Durnford et al..
1999), indicating some degree of dissimilarity between the
two proteins. Nevertheless, the possibility seems to remain
that the proteins bound by our antibody could be from
LHCII.

Of the LHCII components, CP24. CP26, and CP29 con-
tain the most sequence similarities to the LHCIs (Green and
Durnford. 1996) and have molecular weights. 25-30 kDa
(Wollman et al., 1999). that roughly correspond to these of
the three largest polypeptides identified in our anti-LHCI
immunoblots of E. chlorotica and V. litorea plastid proteins
(28-32 kDa), which appear to be slightly larger than most
LHC proteins in chromophytes (Green and Durnford, 1996).
An LHCII antibody derived from pea (generously donated
by Dr. Kenneth Cline, University of Florida) was unreactive
in our iinmunoblotting protocol (data not shown). This

SYMBIOTIC PLASTID GENES IN SLUGS

CAP

CHX

Figure 6. Immunoblot (IB) of LHCI synthesized in the presence of
CAP and 35[S]-methionine, and its corresponding autoradiograph (CAP).
The arrows indicate radiolabeled bands coinciding to LHCI immunola-
beled bands shown in (IB). The bands in (CAP) are not labeled in the
presence of CHX (CHX).

result seems to indicate that the polypeptides are not LHCII,
but since the similarity between the green plant and chro-
mophyte LHC proteins is relatively low (Green and Durn-
ford, 1996; Durnford et al., 1999), we probably cannot
completely eliminate the possibility that the anti-LHCI is
binding LHCII polypeptides. However, just like LHCI, all
of the LHCII genes are nuclear encoded in the plants and
algae where they have been found (Jansson, 1994, 1999;
Wollman et al.. 1999), and even if we have identified
LHCII, the conclusion is still the same: that an algal LHC
gene has been transferred to the DNA of the slug.

The immunoprecipitations provide additional evidence
that the LHCI polypeptides are being synthesized on the
cytoplasmic ribosomes in the slug. The high amount of
radioactivity precipitated by the antibody in the presence of
CAP compared to that precipitated in the presence of CHX
demonstrates that the proteins recognized by the anti-LHCI
are indeed synthesized in the slugs. Since the amount of
radioactivity incorporated varied from slug to slug and from
experiment to experiment, we had to normalize the immu-
noprecipitation data as percent of control in order to com-

pare them. However, in a typical experiment, the values for
the amount of radioactive material incorporated into the
precipitate in the presence of CAP ranged from 5000 to
25,000 cpm, whereas those in the presence of CHX ran from
150 to 400 cpm, which may give a clearer picture of the
level of material bound by the antibody.

The results of the pharmacological experiments, the im-
munoblots, and the immunoprecipitations. taken together,
provide substantial evidence that LHCI is the identity of
some of the plastid proteins that are synthesized in the
presence of CAP. The inhibition of LHCI synthesis by CHX
suggests that the algal Lhca genes have somehow been
transferred to the slug.

To be certain that a gene transfer has occurred, direct
evidence of the gene in the genomic DNA of the slug must
be found, and we are pursuing this confirmation. However,
in addition to the results presented here, other circumstantial
evidence for the transfer of the LHCI genes between alga
and slug is available in several characteristics of the asso-
ciation. First, although the turnover rate of LHCI in E.
chlomtica is unknown, the fact that it is synthesized indi-
cates that it is not an unusually robust protein — LHCI
replacement is necessary for plastid function to proceed.
Second, Lhca genes have not been found in the plastid
genomes of any organism (Durnford et «/., 1999), including
other Vaucheria species (Linne von Berg and Kowallik,
1992). Of course, if LHCI were present in the plastid
genome, it would be synthesized with CHX present, as is
the case with the cyt / controls; but it is not. Third, the V.
litorea plastid genome is 119.1 kb (Green et al., 2000),
which is similar in size to those of other algae, including V.

125-

••— •

0 75 -i

0) 50-

0
0)

°- 25-

, i

CONTROL

CHX

CAP

Inhibitor-Treatment

Figure 7. CHX inhibits synthesis of LHCI. In the presence of CHX.
anti-LHCI precipitated only 2% of control radioactivity incorporated into
LHCI compared to 92% of control in the presence of CAP. Control rates
were defined as 100%. and inhibitor rates were calculated as a mean
percent of control (>i = 6).

J. J. HANTEN AND S. K. PIERCE

sessilis and V. hursata (Linne von Berg and Kowallik, 1988,
1992), hut small relative to those of other plants (Martin and
Herrmann. 1998). Even though the plastid genomes of chro-
mophytic algae have a greater coding capacity, relative to
their size, than other algae because of fewer introns and
inverted repeats (Rieth, 1995). they are too small to carry
sufficient genetic information to encode all of the enzymes
required for photosynthesis and plastid protein targeting.
Fourth, transfer of algal DNA remnants or a nucleomorph-
type structure during plastid capture seems unlikely. To
date, nucleomorphs have been found only in the Crypto-
phyta and Chlorarachniophyta (Delwiche. 1999; Zauner et
til.. 2000) and have not been identified in any chromophyte
(Maier et a!., 1991; Delwiche, 1999). Although DNA of this
type would probably be transcribed on nucleomorph SOS
ribosomes (Douglas et al, 1991) and blocked by CHX.
neither substantial electron microscopy (Kawaguti and Ya-
masu, 1965; Graves et til., 1979; Mujer et al., 1996) nor
molecular testing (Green et al., 2000) has so far produced
evidence for either nucleomorphs or algal nuclear remnants
in E. chlorotica. Furthermore, if algal DNA remnants were
present somewhere in the slug cells, the likelihood is remote
of their containing the correct genes and being present in all
of the plastid-containing cells in all of the slugs in the
populations year after year. Finally, others have suggested
that some of the proteins necessary to maintain photosyn-
thesis may be encoded in the mitochondria! genome and are
redirected to the chloroplast (Rumpho et al.. 2000). Al-
though dual targeting of proteins has been demonstrated in
Arabidopsis (Chow et al.. 1997: Menand et al.. 1998), it
seems highly unlikely with LHC1. LHCI has never been
found associated with mitochondria in any organism; and
CAP, which inhibits the mitochondria! ribosomes in addi-
tion to those associated with the plastids. would prevent its
synthesis anyway.

The horizontal transfer of DNA from the endosymbiont
to the nucleus of the host cell provides the basis for the
theory of the endosymbiotic origin of eukaryotic organelles.
This movement of the symbiont's genes to the host enabled
the host to incorporate the organelle's function into its own
biochemistry and to faithfully replicate it in subsequent
generations. The remnants of eubacterial genes in the mi-
tochondria! and plastid genomes of modern eukaryotes
probably resulted from such events (Martin and Herrmann.
1998). Most of the discussions regarding the evolution of
plastids focus on the horizontal gene transfer resulting from
the primary endosymbiotic event in which a primitive pro-
karyote engulfed a cyanobacteria (Palmer. 1993; Reith.
1995; Palmer and Delwiche, 1996; Martin et al., 1998;
Tengs et al., 2000). Other hypotheses propose a secondary
endosymbiosis, probably involving a eukaryote that en-
gulfed a red or green alga (Gibbs, 1981; Palmer and Del-
wiche, 1996; Martin et al.. 1998: Zhang et al., 1999; Del-
wiche, 1999; Tengs et al.. 2000), that produced the plastids

of the chromophytic algae and their relatives. In many of
these cases, the identity of the initial host, symbiont, or both
is unknown. In the case of E. chlorotica and V. litorea, the
origin of LHCI is known; if the gene has been transferred,
the transfer occurred between two multicellular eukaryotes
and represents a case of tertiary endosymbiosis.

Finally, the mechanism by which such a gene transfer
could occur may be found in the viruses that appear in each
generation of the slugs at the end of their life cycle. The
viruses have several features in common with Retroviridae
and seem to be endogenous (Pierce et al.. 1999). Retrovi-
ruses are capable of transferring genes between organisms;
if they are incorporated in the germ cells, they are trans-
ferred to the subsequent generations as Mendelian genes
(Scharfman et al.. 1991). Thus, resolving the relationships
between the slugs, alga, plastids, and viruses may have
profound implications for both cell and evolutionary biol-
ogy.

Acknowledgments

Research support was provided by a National Science
Foundation award (IBN-9604679) to SKP. We thank Elisa-
beth Gantt and Beatrice Grabowski for their helpful sug-
gestions.

Literature Cited

Anderson, I). J., and G. Blobel. 1983. Immunoprccipitations of proteins
from cell-free translations. Methods Enzvmol. 96: I 1 1-120.

Barber, J., and B. Andersson. 1992. Too much of a good thing: Light
can be bad for photosynthesis. Trends Biochem. Sei. 17: 61-66.

Berry-Lowe, S. L., and G. W. Schmidt. 1991. Chioroplast protein
transport. Pp. 257-302 in The Molecular Biology of Plastids: Cell
Culture mill Somatic Cell Genetics of Plants. Vol. 7A, L. Bogorad and
I. K. Vasil. eds. Academic Press. New York.

Bidui'll. J. P.. and S. Spotte. 1985. Artificial Seawaters: Formulas and
Method*. Jones and Bartlett. Boston. Pp. 305-306.

Buchel, C., and C. YVilhelm. 1993. Isolation and characterization of a
photosystem I-associated antenna (LHCI I and a photosystem I-core
complex from the chlorophyll i-containing alga Pleurochloris ineirin-
xcnsis (Xanthophyceae). ./. Photochem. Photohiol. B Bio/. 20: 87-93.

Chow, K-S., D. P. Singh. J. M. Roper, and A. G. Smith. 1997. A single
precursor protein for ferrochelatase-I from Arabiilopsis is imported in
vitro into both chloroplasts and mitochondria. J. Biol. Cliem. 272:
27565-2757 1 .

Clark, K. B., K. R. Jensen, and H. M. Strits. 199(1. Survey of func-
tional kleploplasty among West Atlantic Ascoglossa ( =Sacoglossa)
(Mollusca: Opistobranchia). Veliger 33: 339-345.

Delwiche, C'. F. 1999. Tracing the thread of plastid diversity through the
tapestry of life. Am. Nat. 154: S164-S177.

I)c Martino, A., D. Douady, M. Quinet-S/ely, B. Rousseau, F. Crep-
ineau, K. Apt, and L. Caron. 2000. The light-harvesting antenna of
brown alga: highly homologous proteins encoded by a multigene
family. Eur. J. Biochem. 267: 5540-5549.

Douglas, S. E., C. A. Murphy, D. F. Spencer, and M. W. Gray. 1991.
Cryptomonad algae are evolutionary chimaeras of two phylogenetically
distinct unicellular eukaryotes. Nature 350: 148-151.

Durnford. D. G., and B. R. Green. 1994. Characterization of the

SYMBIOTIC PLASTID GENES IN SLUGS

light-harvesting proteins of the chromophytic alga. Olisthodiscus luteus
(Ht'ierosigma carterae). Biochim. Bio/thy*. Ada 1184: 118-126.

Durnfiird, I). (;., J. A. Deane, S. Tan, G. I. McFadden, E. Gantt, and
B. R. Green. 1999. A phylogenetic assessment of the eukaryotic
light-harvesting antenna proteins, with implications for plastid evolu-
tion. J. M»l. t'.vol. 48: 59-68.

Fawley, M. \V., and A. R. Grossman. 1986. Polypeptides of a light-
harvesting complex of the diatom. Phaeodactylum tricornutum are
synthesized in the cytoplasm of the cell as precursors. Plant Pli\\iot.
81: 144-155.

Gantl. E. 1996. Pigment protein complexes and the concept of the
photosynthetic unit: chlorophyll complexes and phycobilisomes. Pho-
tosyn. Re.',. 48: 47-53.

Gihbs. S. P. 1981. The chloroplasts of some algal groups may have
evolved from endosymbiotic eukaryotic algae. Ann. NY Acad. Sci. 361:
193-207.

Grabowski, B., S. Tan, F. X. Cunningham, Jr., and E. Gantt. 200(1.
Characterization ot the Porphyridium cntentum chl (/-binding LHC by
in vitro reconstitution: LlicaRI binds 8 chi-o molecules and propor-
tionately more carotenoids than CAB proteins. Photosyn. Res. 63:
85-96.

Graves, D. A., M. A. Gibson, and J. S. Bleakney. 1979. The digestive
diverticula of Aldcria modesta and Elvsiu chlorotica. Veliger 21:
415-422.

Green, B. J., W-Y. Li. J. R. Manhart, T. C. Fox, E. L. Summer, R. A.
Kennedy, S. K. Pierce, and M. E. Rumphn. 2000. Mollusc-algal
chloroplast endosymbiosis. Photosynthesis, thylakoid protein mainte-
nance, and chloroplast gene expression continue for many months in
the absence of the algal nucleus. Plant Physiol. 124: 331-342.

Green, B. R., and D. G. Durnford. 1996. The chlorophyll carotenoid
proteins of oxygenic photosynthesis. Annit. Rev. Plant. Physiol. Plant.
Mol. Bini 47: 6S5-7I4.

Greenberg, B. M., V. Gaba, O. Cananni, S. Malkin, A. K. MaHoo, and
M. Edelman. 1989. Separate photosensitizers mediate degradation of
the 32 kDa photosystem II reaction center protein in the visible and UV
spectral regions. Proc. Natl. Acad. Sci. USA 86: 6617-6620.

Greene, R. \V. 1970. Symbiosis in sacoglossan opistobranchs: functional
capacity of symbiotic chloroplasts. Mar. Biol. 7: 138-142.

Hinde. R., and I). C. Smith. 1974. "Chloroplast symbiosis" and the
extent to which it occurs in Sacoglossa (Gastropoda: mollusca). Biol. J.
Li in,. Soc. 6: 349-356.

Jansson, S. 1994. The light-harvesting chlorophyll «//>-binding proteins.
Biochim. Biopliys. Ada 1184: 1-19.

Jansson, S. 1999. A guide to the Lhc genes and their relatives in
Arahidopsis. Trends Plant Sci. 4: 236-240.

Joyard, J., A-J. Dome, and R. Douce. 1987. Use of thermolysin to
probe the cytosolic surface of the outer envelope membranes from
plastids. Pp. 195-206 in Methods in Enzymo/agy. Vol. 148, L. Packer
and R. Douce, eds. Academic Press. San Diego, CA.

Kawaguti. S., and T. Vamasu. 1965. Electron microscopy on the sym-
biosis between an elysioid gastropod and chloroplasts of a green alga.
Biol. J. Okayama Univ. 11: 57-65.

I .M mill. I'. K. 1970. Cleavage of structural proteins during assembly of
the head of bacteriophage T4. Nature 227: 608-685.

Lamb, A. J., G. D. Clark-Walker, and A. VV. Linnane. 1968. The
biogenesis of mitochondria: the differentiation of mitochondria! and
cytoplasmic protein synthesizing systems in vitro by antibiotics. Bio-
chim. Biophys. Ada 161: 415-427.

Linne von Berg, K.-H., and K. V. Kovvallik. 1988. Structural organi-
zation and evolution of the plastid genome of Vaucheria sessilis (Xan-
thophyceae). Biosystems 21: 239-247.

Linne von Berg. K.-H., and K. V. Kowallik. 1992. Structural organi-

sation of the chloroplast genome of the chromophytic alga Vaucheria
hursata. Plant Mol. Biol. 18: 83-95.

Maier, U.-G., C. J. B. Hoffman, E. Eschbach. J. Wolters, and G. L.
Igloi. 1991. Demonstration of nucleomorph-encoded eukaryotic
small subunit ribosomal RNA in cryptomonads. Mol. Gen. Genet. 230:
155-160.

Martin, W., and R. G. Herrmann. 1998. Gene transfer from organelles
to nucleus: How much, what happens and why? Plant Physiol. 118:
9-17.

Martin, VV., B. Stoebe, V. Goremykin. S. Hansmann, M. Hasegawa,
and K. V. Ko\\allik. 1998. Gene transfer to the nucleus and the
evolution of chloroplasts. Nature 393: 162-165.

Mattoo, A. K., J. B. Marder, and M. Edelman. 1989. Dynamics of the
photosystem II reactions center. Cell 5: 241-246.

Menand, B.. L. Marechal-Drouard, W. Sakamoto, A. Dietrich, and H.
Wintz. 1998. A single gene of chloroplast origin codes for mitochon-
drial and chloroplastic methionyl-tRNA synthetase in Arabidopsn
thaliana. Proc. Null. Acad. Sci. USA 95: 11014-11019.

Mujer, C. V., D. L. Andrews, J. R. Manhart, S. K. Pierce, and M. E.
Rumpho. 1996. Chloroplast genes are expressed during intracellular
symbiotic association of Vaucheria litorea plastids with the sea slug
Elysia chlorotica. Proc. Natl. Acad. Sci. USA 93: 12333-12338.

Obrig, T. G., W. J. Culp, W. L. McKeehan, and B. Hardesty. 1971.
The mechanism by which cycloheximide and related glutarimide anti-
biotics inhibit peptide synthesis on reticulocyte ribosomes. J. Biol.
diem. 246: 174-1X1.

Palmer. J. D. 1993. A genetic rainbow of plastids. Nature 364: 762-763.

Palmer, J. D., and C. F. Delwiche. 1996. Second-hand chloroplasts and
the case of the disappearing nucleus. Proc. Natl. Acad. Sci. USA 93:
7432-7435.

Peterson, G. L. 1977. A modification of the protein assay method of
Lowry et al. which is more generally applicable. Anal. Biochem. 83:
346-356.

Pierce, S. K., R. W. Biron. and M. E. Rumpho. 1996. Endosymbiotic
chloroplasts in molluscan cells contain proteins synthesized after plas-
tid capture. J. Exp. Biol. 199: 2323-2330.

Pierce, S. K.. T. K. Maugel, M. E. Rumpho, J. J. Hanten, and W. L.
Mondy. 1999. Annual viral expression in a sea slug population: life
cycle control and symbiotic chloroplast maintenance. Biol. Bull. 197:
1-6.

Rieth, M. 1995. Molecular biology of rhodophyte and chromophyte
plastids. Annu. Rev. Plan! Physiol. Plant Mol. Biol. 46: 549-575.

Rumpho, M. E., E. J. Summer, and J. R. Manhart. 2000. Solar-
powered sea slugs. Mollusc/algal chloroplast symbiosis. Plant Physiol.
123: 29-38.

Scharfmann, R., J. H. Alexrod, and I. M. Verma. 1991. Long-term in
vivo expression of retrovirus-mediated gene transfer in mouse fibro-
blast implants. Proc. Natl. Acad. Sci. USA 88: 4626-4630.

Schmid, V. H. R., K. V. Cammarata, B. U. Bruns, and G. W. Schmidt.
1997. In vitro reconstitution of the photosystem I light-harvesting
complex LHCI-730: Heterodimerization is required for antenna pig-
ment organization. Proc. Natl. Acad. Sci. USA 94: 7667-7672.

Stone, A. B., and D. Wilkie. 1975. Loss of cytochrome oxidase in
Sacchromvces cerevisiae during inhibition of mitochondrial proteins
synthesis by erythromycin and chloramphenicol. J. Gen. Microbiol. 91:
150-156.

Tan, S., G. R. Wolfe, F. X. Cunningham, Jr., and E. Gantt. 1995.
Decrease in the PSI antenna complex with increasing growth irradiance
in the red alga Porphyridiwn t rucntitm. Photosyn, Res. 45: 1-10.

Tan, S., F. X. Cunningham, Jr., and E. Gantt. 1997a. LhcaRl of the
red alga Porph\ridin»i cnientiim encodes a polypeptide of the LHC I
complex with seven potential chlorophyll a-binding residues that are
conserved in most LHCs. Plant Mol. Biol. 33: 157-167.

J. J. HANTEN AND S. K. PIERCE

Tan, S., A. Ducret, R. Aebersold, and E. Gantt. 1997b. Red algal LHC
I genes have similarities with both Chi ale- and a/obinding proteins:
A 21 kDa polypeptide encoded by LhcaR2 is one of the six LHC I
polypeptides. Photosvn. Res. 53: 129-140.

Taylor, D. L. 1970. Chloroplasts as symbiotic organelles. Int. Rev. C\tol.
27: 29-64.

Tengs. T., O. J. Dahlberg, K. Shalchian-Tabrizi, D. Klaveness, K.
Rudi, C. F. Delwiche, and K. S. Jakobson. 2000. Phylogenetic
analyses indicate that the 19' hexanoloxy-fucoxanthin-containing
dinoflagellates have tertiary plastids of haptophyte origin. Mol. Bio/.
Evol. 17: 718-729.

Towbin, H., T. Staehelin. and J. Gordon. 1979. Electrophoretic trans-
fer of proteins from polyacrylamide gels to nitrocellulose sheets: pro-
cedure and some applications. Proc. Nail. Acad. Sci. USA 76: 4350-
4354.

West, H. H. 1979. Chloroplast symbiosis and the development of the
ascoglossan opistobranch Elysia chlomticii. Ph.D. dissertation. North-
eastern University. Boston. MA.

West, H. H., J. Harrigan, and S. K. Pierce. 1984. Hybridization of two
populations of marine opistobranch with different developmental pat-
terns. Veliger 26(3): 199-206.

Wolfe, G. R., F. X. Cunningham, D. Durnford, B. R. Green, and E.
Gantt. 1994. Evidence for a common origin of chloroplasts with
light-harvesting complexes of different pigmentation. Nature 367:
566-568.

Wollman, F.-A., L. Minai, and R. Nechushtai. 1999. The biogenesis
and assembly of photosynthetic proteins in thylakoid membranes. Bio-
cliiin. Bi,i/>li\s. Acta 1411: 21-85.

Zauner, S., M. Fraunholz, J. Wastl. S. Penny, M. Beaton, T. Cavalier-
Smith, U.-G. Maier, and S. Douglas. 2000. Chloroplast protein and
centrosomal genes, a tRNA intron, and odd telomeres in an unusually
compact eukaryotic genome, the cryptomonad nucleomorph. Proc.
Nail. Acad. Sci. USA 97: 200-205.

Zhang, Z., B. R. Green, and T. Cavalier-Smith. 1999. Single
gene circles in dinoflagellate Chloroplast genomes. Nature 400: 155-
159.

Reference: Bin/. Bull. 201: 45-51. (August 2001)

Asexual Reproduction in Pygospio elegans Claparede

(Annelida, Polychaeta) in Relation to Parasitism by

Lepocreadium setiferoides (Miller and Northup)

(Platyhelminthes, Trematoda)

DEAN G. McCURDY*
Coastal Studies Center, 6775 College Station, Bowdoin College. Brunswick, Maine 04011-8465

Abstract. Life-history theory predicts that parasitized hosts
should alter their investment in reproduction in ways that
maximize host reproductive success. I examined the timing of
asexual reproduction (fragmentation and regeneration) in the
polychaete annelid Pygospio elegans experimentally exposed
to cercariae of the trematode Lepocreadium setiferoides. Con-
sistent with adaptive host response, polychaetes that became
infected by metacercariae of trematodes fragmented sooner
than unexposed controls. Parasites were not directly associated
with fission in that exposed polychaetes that did not become
infected also fragmented earlier than controls. For specimens
of P. elegans that were not exposed to trematodes, new frag-
ments that contained original heads were larger than those that
contained original tails, whereas original head and tail frag-
ments did not differ in size for infected polychaetes. In infected
specimens, metacercariae were equally represented in original
head and tail fragments and were more likely to be found in
whichever fragment was larger. Despite early reproduction,
parasitism was still costly because populations of P. elegans
exposed to parasites were smaller than controls when mea-
sured 8 weeks later and because exposure to cercariae reduced
survivorship of newly divided polychaetes. Taken together, my
results suggest that early fragmentation is a host response to
minimize costs associated with parasitism.

Introduction

Hosts respond to parasitism in a number of ways, which
include avoidance of parasites in space or time (e.g.. mi-

Received 19 October 2000; accepted 10 April 2001.
* Current address: Department of Biology, Albion College, Albion,
Michigan 49224.

E-mail: dmccurdy<s1 albion.edu

gration; Folstad et ai, 1991). removal of parasites before
they cause damage (e.g.. grooming; Leonard et ai, 1999),
and immunological defense (e.g., encapsulation; Kraaije-
veld and Godfray. 1997). There is increasing evidence that
hosts may also exhibit life-history adaptations to minimize
the impacts of parasites on reproductive success (Minchella
and LoVerde, 1981; Polak and Starmer, 1998; McCurdy et
ai, 1999, 2000a). Life-history responses of hosts after ex-
posure to parasites represent reallocations of energy in ways
that increase reproductive success relative to non-responses.
This type of reaction has been termed an adaptive host
response (Minchella, 1985; Forbes, 1993). Unlike avoid-
ance or resistance to parasites, life-history responses pose
little or no maintenance costs to hosts (i.e., no cost when
parasites are absent) because the hosts do not alter their life
histories until they come into contact with parasites (Min-
chella, 1985). Specifically, in systems where parasites pose
greater costs to host energy budgets over time (decreasing
future reproductive potential to a greater extent than current
reproduction), hosts are expected to respond to infections by
hastening their onset of reproduction. This response occurs,
for example, in intertidal amphipods infected by trematodes
(McCurdy et ai, 1999, 2()()0a). Although the reproductive
success of hosts that respond through life-history variation
is lower than that of hosts not exposed to parasites, it is
greater than that of infected hosts that fail to respond (Mc-
Curdy et ai, 2001).

To date, tests of the hypothesis of the adaptive host
response have been confined to hosts that reproduce sexu-
ally (Minchella and LoVerde, 1981; Polak and Starmer,
1998; McCurdy et ai, 1999, 2000a). However, the re-
sponses of sexual hosts can be difficult to interpret because
selection may act differently on males and females to max-

D. G. McCURDY

imize reproductive success (Zuk, 1990), and because of
factors specific to sexual mating systems (e.g., mate avail-
ability and choice: McCurdy et til.. 2000b: but see Min-
chella and LoVerde, 1981, for an exception). As a result,
host investment in reproduction (reproductive effort) is as-
sessed, but the actual consequences of this investment (re-
productive success) are difficult to quantify (Perrin et ai,
1996: McCurdy et ai. 2000a).

I tested for adaptive host response in an asexual spionid
polychaete, Pygospio elegans. This species is common in
intertidal mudflats and sandflats throughout the Northern
Hemisphere (Anger, 1984). Adults of this species construct
tubes in the sediment and feed on detritus and phytoplank-
ton (Anger et ai. 1986). Asexual reproduction in P. elegans
is accomplished through transverse fragmentation, followed
by rapid regeneration of missing components (Rasmussen,
1953; Hobson and Green, 1968; Gibson and Harvey, 2000).
Wilson ( 1985) found that asexual reproduction in P. elegans
is density- and resource-dependent in that populations grew
larger when polychaetes were housed at low densities or
provided with augmented levels of food. The impacts of
parasitism on asexual reproduction and regeneration, how-
ever, have not been investigated in this or other species of
polychaetes.

I investigated life-history responses of P. elegans to par-
asitism by exposing polychaetes to cercariae of the trema-
tode Lepocreadiwn setiferoides. During the spring and sum-
mer, cercariae emerge from mud snails, Ilyanassa obsoleta,
and infect spionid polychaetes as second-intermediate hosts;
winter flounder, Pseudopleuronectes americanus, serve as
final hosts of the parasite (Martin. 1938; McCurdy et ai.
2000c). In their polychaete host, trematodes do not repro-
duce. However, unlike many species of trematodes that
emerge from /. obsoleta, metacercariae of L. setiferoides do
not simply encyst within their second-intermediate hosts,
but continue to grow and develop for several weeks (Martin.
1938; McCurdy, pers. obs.). Thus, the costs of parasitism to
the energy budgets of polychaete hosts are expected to
increase over time after infection.

Predictions

If Pygospio elegans responds to parasitism through life-
history variation. I predicted that polychaetes would frag-
ment soon after infection, before parasitism becomes costly.
P. elegans also exhibits flexibility in asexual reproduction,
as individual polychaetes may fragment into more than five
pieces (Rasmussen, 1953; Gibson and Harvey. 2000). In
light of this fact, I also expected that newly infected
polychaetes might minimize the impacts of parasitism by
isolating infection in small fragments or even lose infections
by dividing across infected segments. In addition, P. el-
egans may also reproduce sexually (including poecilogo-
nous development, with planktotrophic and adelophagic

larvae; Morgan et ai. 1999), so I examined polychaetes for
evidence of sexual reproduction as a possible response to
parasitism. For infected polychaetes, the advantages of sex-
ual reproduction might include enhanced dispersal of off-
spring (Chia et ui. 1996) — possibly away from infected
snails — and increased genetic variation (Lively, 1996). In
fact, the evolution and maintenance of sexual reproduction
have been explained as a host response to parasitism be-
cause sex is more likely to produce individuals that are able
to escape parasitism over evolutionary timescales (reviewed
by Hurst and Peck, 1996).

In addition to investigating host responses to parasitism,
I assessed the impact of parasitism by Lepocreadium setif-
eroides on the asexual reproductive success of P. elegans
over an 8-week period. I also assessed the costs of parasit-
ism to survivorship and regeneration of polychaetes that had
previously been cut into two fragments, mimicking the
fragmentation that results from asexual reproduction or sub-
lethal predation (Woodin, 1982: Zajac, 1995). In all cases, I
considered two additional possibilities, other than adaptive
host response, to explain observed changes in host behavior
and development in relation to parasitism. First, such
changes might have been due to adaptations of parasites to
increase transmission rates (parasite manipulation; Poulin et
ai. 1994). This possibility is particularly relevant to the
parasite-host system I studied because there is evidence for
parasite manipulation by cercariae and metacercariae of
another trematode that parasitizes Ilyanassa obsoleta (Cur-
tis, 1987; McCurdy et ai. 1999, 2000a). Second, observed
changes in behavior might have been due to side effects of
infection that are not adaptive for the host or parasite
(Poulin, 1995).

Materials and Methods

Collections and infection protocols

I collected specimens of Pygospio elegans from a mudflat
between Wyer and Orr's Islands, Harpswell, Maine
(43°47'N, 69°58'W). This mudflat is located in Casco Bay,
Gulf of Maine, and has semidiurnal tides that range from 2
to 4 in (Born. 1999). I chose to sample at the Wyer-Orr's
mudflat because densities of P. elegans were high there
(>20,000 m~2), but Ilyanassa obsoleta and its associated
cercarial parasites were rare (<0.25 snails m"2), minimiz-
ing the likelihood that polychaetes used in experiments were
already infected. I collected polychaete tubes in the mid-
intertidal zone by sieving the top 5 cm of mud (500-jum
mesh) and transported tubes to the nearby running-seawater
laboratory at the Coastal Studies Center of Bowdoin Col-
lege for sorting. I retained only undamaged, entire adult
polychaetes (>2 mm) that were not about to fragment
(detectable because P. elegans constricts just prior to fis-
sion; Gibson and Harvey. 2000).

To obtain cercarial trematodes for experiments. I col-

PARASITISM AND ASEXUAL REPRODUCTION

lected specimens of /. obsoleta from throughout the inter-
tidal zone at Strawberry Creek. Great Island, Maine
(43°49'N. 69°58'W). This mudflat is located 2.5 km from
the Wyer-Orr's mudflat and supports high densities of /.
obsoleta (>10 m~2). In the laboratory. I housed 550 mud
snails in separate 9-oz plastic cups with 125 ml of filtered
seawater (55 ^im, 31 ppt, 23 °C). I retained only large snails
(>15 mm. tip of apex to lip of siphonal canal) because
previous studies have shown that the prevalence of Lep o-
creadium setiferoides increases with shell height of snails
(Curtis. 1997; McCurdy el ai. 2000c). After 30 h. I exam-
ined each cup for cercariae of L. setiferoides (identified
using McDermott, 1951). combined cercarial-infested sea-
water from cups of six snails that had shed cercariae, and
pipetted 20 ml of the solution into each dish that contained
a polychaete that was to be exposed. Unexposed
polychaetes each received 20 ml of seawater from six cups
that contained snails that did not shed cercariae (confirmed
by dissection, as cercarial release is a poor indicator of
infection status; Curtis and Hubbard. 1990).

Experiments

To investigate the impact of parasites on the timing of
asexual reproduction, I individually housed 52 adult speci-
mens of P. elegans in 150-ml custard dishes filled with
unfiltered seawater with or without cercariae (18 °C. 16 h
light day"1). After 24 h. I transferred each polychaete to a
new dish filled with seawater and lined with defaunated
mud (prepared by passing mud through a 425-ju.m sieve and
heating it to 70 °C). Every 24 h, I suspended each dish from
a harness and determined the status of each polychaete by
observing its tube (or tubes) through the bottom of its dish
with the aid of a fiber-optic illuminator and 10X magnifying
loupe. Polychaetes could easily be observed because they
constructed tubes that opened against the bottoms of their
dishes. Polychaetes were fed the pea-flower- based supple-
ment Liquifry Marine (Interpet Inc.; Brown el ai, 1999)
every 3 days (concentration = 1 drop 1 ~ ' ) following a
complete change of water. I removed polychaetes from the
experiment when they died or fragmented, and I measured
the relaxed length of all fragments with an ocular microme-
ter (nearest 0.1 mm; Gudmundsson. 1985). I then dissected
each fragment to determine if it was infected by trematode
metacercariae and compared median time-to-fragmentation
among exposed but uninfected. exposed and infected, and
unexposed polychaetes. In making this comparison. I sepa-
rated exposed but uninfected polychaetes from unexposed
ones because of the possibility that host response might be
associated with indirect cues associated with parasitism
(i.e., response might not require an actual infection to oc-
cur). To compare time-to-fragmentation. I applied a non-
parametric Kruskal-Wallis ANOVA because the residuals

for all groups were non-normal. I then applied Dunn's
method to compare differences among medians (Zar, 1996).

To investigate how exposure to parasites affected the
reproductive success of P. elegans, I randomly housed 18
sets of 10 polychaetes (hereafter referred to as populations
of polychaetes) in separate dishes and exposed half of the
sets to cercariae of trematodes (housing conditions for
polychaetes were as described above). Because the infection
status of polychaetes that died during this experiment could
not be determined without disturbing surviving polychaetes,
I assessed rates of experimental and background infection
by randomly removing two sentinel populations after 3
days: a population of polychaetes that had been exposed to
cercariae, and a population of unexposed polychaetes. Rates
of infection at that time represented maximum levels that
could occur because cercariae of L. setiferoides survive for
less than 48 h outside a host (Stunkard, 1972). After 8
weeks. I removed the remaining dishes and processed each
population by counting the number of polychaetes retained
after sieving (425-jam mesh) and dissecting each polychaete
to determine its infection status.

To assess survivorship and regenerative ability of newly
divided polychaetes in relation to parasitism. I cut 59
polychaetes into two fragments and exposed 30 pairs of
fragments to cercariae. Cutting each polychaete resulted in
a smooth, clean blastema similar to that resulting from
sublethal predation or asexual fragmentation (Gibson and
Harvey. 2000; pers. obs.). To mimic conditions in nature,
where newly fragmented polychaetes generally remain in
the same burrow during regeneration (Gudmundsson, 1985;
Gibson and Harvey, 2000). I individually housed original
head and tail fragments together in a dish with seawater and
mud (housing conditions as described above). To avoid
disturbing fragments (as above), I assessed initial rates of
infection at 3 days after exposure or non-exposure by re-
moving and dissecting randomly chosen sentinel pairs of
exposed fragments (/; =: 10 polychaetes) and unexposed
fragments (n = 10 polychaetes). At 10 days after exposure
or non-exposure. I removed all remaining fragments, mea-
sured their lengths, and determined their infection status.

Results

Parasitism and host fragmentation

In the experiment investigating the impact of trematodes
on the timing of asexual reproduction in Pygospio elegans,
parasite prevalence was low (42.3% of polychaetes exposed
became infected; n = 26). Asexual fragmentation always
yielded two fragments; one containing the original head and
thorax and a second containing the original tail (see Gibson
and Harvey, 2000, for a description of body components). In
all cases, polychaetes fragmented within 24 h of observable
constrictions. Time-to-fragmentation differed between ex-
posed and infected, exposed but uninfected. and unexposed

D. G. McCURDY

-o

F 2-1

l b

-= 1 •<

1 6

i i

Unexposed !x posed/ Exposed/

LtninlcLlcd Infected

Treatment

Figure 1. Median ( ± quartiles) numbers of days for asexual reproduc-
tion to occur in individuals of Pygospio elegans that were experimentally
infected, exposed but not infected, and not exposed to cercariae of the
trematode Lepocreadiwn setiferoides. Polychaetes and parasites were col-
lected from mudflats in Harpswell, Maine, and housed in the laboratory.
Median-, with the same letter do not differ significantly from each other.

polychaetes (//|2.52) = 10.56. P < 0.01: Fig. 1). Specif-
ically, polychaetes that were exposed to cercariae but did
not become infected fragmented earlier than unexposed
polychaetes (Q = 2.99, P < 0.005), as did polychaetes
that were exposed and became infected (Q = 2.16, P •
0.05). Of all polychaetes that were exposed to cercariae,
however, infection status did not affect time-to-fragmenta-
tion (Q = 0.49, NS).

For unexposed polychaetes and exposed polychaetes that
remained uninfected. fragments that contained original
heads were larger than those that contained original tails,
whereas lengths of original head and tail fragments did not
differ for infected polychaetes (Table 1 ). In infected
polychaetes. parasites were just as likely to be found in
fragments that contained original heads (n = 5) as those
that contained original tails (n 5) (an additional
polychaete harbored a metacercaria in each new fragment).
For infected polychaetes, infected fragments were signifi-
cantly larger than uninfected fragments (infected fragments:
x ± s = 2.0 ± 0.2 mm; uninfected fragments: x ± s =
1.4 ± 0.2 mm; paired r(l)) = 2.28. P < 0.05). and in 9
of 10 cases, metacercariae were found in the larger fragment
(Xf, > = 6.4, P = 0.01). Cercariae were not observed to
penetrate segments that comprised, or were adjacent to,
planes of fission.

Parasitism and host asexual reproductive success

At 3 days post exposure, 17 of 20 fragments (8.5 of the
original 10 polychaetes) were alive in the sentinel popula-
tion that was exposed to cercariae. Only one fragment in this
population was infected by trematodes — a living tail frag-
ment infected with a single metacercaria. In the sentinel
population that was not exposed to cercariae, 18 of 20

fragments were alive after 3 days and no parasites were
found (one fragment, containing an original head, was lost
during processing). At 8 weeks after exposure or non-
exposure, I saw no evidence of recent fission in polychaetes
as all fragments had complete or nearly complete heads and
tails. Therefore. I considered all fragments equally when
measuring population sizes at that time. Populations of
polychaetes that were exposed to cercariae were smaller
than those that were not exposed (exposed populations: A ±
5 = 17.3 ± 2.4 polychaetes; unexposed populations: x ±
s = 29.8 ± 3.7 polychaetes; /(14) = 2.84, P = 0.01).
When dissected, only seven polychaetes in exposed popu-
lations were infected (one polychaete in each of three pop-
ulations and two polychaetes in each of two populations),
and none of the polychaetes in any of the unexposed pop-
ulations was infected.

Considering sentinel polychaetes that had been cut into
two pieces, 2 of 10 polychaetes exposed to cercariae were
infected at 3 days post-exposure. In each case, the infection
was in the original head fragment and by a single metacer-
caria. None of the 10 unexposed polychaetes was infected.
When examining the remaining polychaetes 7 days later, I
found that both head and tail fragments of exposed
polychaetes were less likely to be alive than the respective
fragments of unexposed polychaetes (head fragments:

= 8.07, P < 0.005; tail fragments:

, ,

= 12.22, P <

0.001 : Fig. 2). Only two exposed polychaetes were infected
by metacercariae (one polychaete had an infected tail frag-
ment and another an infected head fragment; /; = 20). and
no unexposed polychaetes were infected (n == 19). In all
cases, regeneration of "lost" components was nearly com-
plete by 10 days, and lengths of original head and tail
fragments did not differ in relation to exposure (unexposed
heads: A ± SE = 2.65 ± 0.15; exposed heads: x ± SE =
2.56 ± 0.26; r(26) = 0.32. NS: unexposed tails: x ± SE =

Table 1

Si;cs af fraxiiicnt* produced hy uxe.\iial fission of Pygospio elegans
in relation to panisitism

Fragment length (mm|

Heads

Tails

Paired / test

Unexposed 2.1 ± 0.2 1.57 ± 0.1 /,2,, = 2.7. P = 0.01

Exposed but

uninfected 2.4 ± 0.2 1.57 ±0.2 f,,4l = 2.3, P = 0.04

Exposed and

infected 1.9 ± 0.2 1.71 ± 0.2 ?,,„, = 0.8. P = 0.44

Data are means and standard errors for lengths of fragments containing
original heads and those containing original tails of polychaetes that were
experimentally infected, exposed but not infected, and not exposed to
cercariae of the trematode Lepocreadiwn setiferoides. The last column
shows results from paired t tests for lengths of original head versm, tail
fragments.

PARASITISM AND ASEXUAL REPRODUCTION

Heads Tails

Original fragments

Figure 2. Proportions (±95% confidence intervals) of original head
and tail fragments of individuals of Pygospio elegans that survived for 10
days in the laboratory following exposure or non-exposure to cercariae of
the trematode Lepocreatl/iuii .fciiti'i-niilfs. Sample sizes are shown above
the bars.

2.71 ± 0.19: exposed tails: x ± SE = 2.49 ± 0.28; f(22) =
0.63. NS).

Discussion

Parasitism and host fragmentation

In support of the hypothesis of adaptive host response I
found that specimens of Pygospio elegans infected by meta-
cercariae of Lepocreadium setiferoides hastened their onset
of asexual reproduction relative to unexposed controls. By
doing so, polychaetes may be expected to achieve greater
reproductive success than if they had failed to respond
because of increasing costs associated with parasitism over
time (Forbes, 1993). However, my observation that early
fragmentation also occurred in exposed polychaetes that
remained uninfected complicates this interpretation. In a
study that separated hosts by exposure and infection status,
Minchella and Loverde ( 1981 ) found that freshwater snails
of the species Biomphalaria glahrata increased their rates
of early egg laying when infected by Schistosoma mansoni,
but that the rates for exposed but uninfected individuals and
unexposed controls did not differ. These authors argued that
only infected snails responded because successful parasit-
ism was associated with a high cost to future reproduction
(castration).

For individuals of P. elegans exposed to, but not infected
by, cercariae. early reproduction could still be an adaptive
host response if exposure to cercariae in nature is a reliable
indicator that costly infections will soon result (Minchella.
1985). Support for this idea comes from the observation that
Ilyanassa obsoleta infected by L. setiferoides, although
uncommon across mudflats, can remain for several months
in small patches where some P. elegans are found (Mc-

Curdy et til., 20()0c). As a result, thousands of cercariae are
shed in areas where infections are most likely to occur.
Additional information on the infection process of L. setif-
crnitles is necessary to determine whether polychaetes de-
tect cercariae, and whether the exposure-related response
resulted from the presence of cercariae or from failed at-
tempts at penetration. There is evidence from other parasite-
host systems that invertebrates can detect and exhibit anti-
parasite behaviors to minimize the likelihood of infection
(e.g.. Leonard et al.. 1999).

Early fragmentation of P. elegans is unlikely to be a
parasite adaptation, because it apparently does not increase
transmission rates for cercariae or metacercariae. Specifi-
cally, fragmentation was not associated with increased sus-
ceptibility to parasitism: most polychaetes fragmented after
free-living cercariae would have (>48 h; Stunkard, 1972).
For metacercariae. residing in small fragments would not
appear to benefit transmission to final hosts, because floun-
der select prey at larger sizes relative to conspecifics, and
even small differences in prey size preference can pro-
foundly influence the energy budgets of predators foraging
on mudflats (MacDonald and Green. 1986: Boates and
Smith, 1989; Keats. 1990). To assess whether early frag-
mentation is actually adaptive for parasites or hosts, the
consequences of early fragmentation could be further ex-
plored by constructing a model derived from empirical
observations of parasites, their intermediate hosts, and the
predators that are their final hosts. This approach was used
recently to show that the early onset of receptivity to mating
observed in females of the amphipod Corophium volntator
infected by the trematode G\naecotyla adunca resulted in
greater reproductive success for the amphipods than if they
had waited to become receptive at the optimal time for
uninfected females (McCurdy et al., 2001).

I found no evidence that fragmentation of P. elegans
served to isolate or remove metacercariae, in that fission
produced only two fragments, the smaller of which almost
never contained metacercariae. It is unclear whether the
greater presence of metacercariae in larger fragments is
adaptive for the parasite or its host or whether larger frag-
ments merely represent larger targets for parasites. Meta-
cercariae might benefit from residing in larger fragments
because of the availability of additional resources for para-
site development or the possibility of a greater transmission
rate to final hosts (as stated above, flounder tend to select
larger prey). If residing in larger fragments is parasite-
mediated, the observation that metacercariae develop near
the site of initial penetration (Stunkard. 1972; pers. obs)
indicates that the mechanism does not involve movements
by metacercariae through the host coelom and into larger
fragments. Fragmentation could also be interpreted as a host
response: If larger fragments are better able to tolerate
stresses associated with parasitism, the result would be a net
reproductive benefit to hosts. In fact, host response need not

D. G. McCURDY

be exclusive of benefits to parasites, depending on the
timing of altered behavior of infected hosts (McCurdy et ai.
1999). Simulated parasites such as Sephadex beads (Suwan-
chaichinda and Paskewitz. 1998) could be used to help
separate effects mediated by the parasite from those medi-
ated by the host. Experiments with simulated parasites
would provide cues to the host that it has become infected
while removing the possibility of parasite manipulation.

Across all experiments, I found no evidence for onset of
sexual reproduction, observing neither eggs nor spermato-
phores. Seasonal constraints may have precluded sexual
reproduction, which usually occurs only during the winter
in P. elegans (Rasmussen, 1953; Gudmundsson, 1985; Wil-
son, 1985). However, even if the polychaetes had shown
evidence of sexual reproduction, this tactic might be ex-
pected to increase reproductive success only if mates were
available; an unlikely event given the rarity of parasites in
natural populations of P. elegans (above).

Parasitism and host asexual reproductive success

I found that even a low level of exposure to cercariae (on
average, 8% of cercariae that a single snail sheds in 30 h)
reduced the asexual reproductive success of P. elegans
(45%, measured in populations 8 weeks after exposure). In
a related finding from another experiment, both head and
tail fragments were less likely to survive to complete regen-
eration than were unexposed fragments. Direct effects of
parasitism are not sufficient to account for these results
given that few exposed polychaetes actually became in-
fected in either experiment. One possibility is to explain the
reduced reproductive success of exposed but uninfected
hosts as the result of a trade-off between host reproductive
effort and costly activities associated with defenses against
parasites. Recent work has shown that hosts exposed to
parasites may trade off energy used in reproduction for
behaviors or immune responses to resist parasites (Sheldon
and Verhulst, 1996; Leonard et ai, 1999).

Regardless of the underlying causes, the dramatic reduc-
tion in reproductive success of P. elegans after exposure to
cercariae has implications for natural populations of this
species and for soft-bottom intertidal communities. Pygos-
pio elegans often dominates such communities, and thus
can directly affect the distribution and abundance of other
infauna (Wilson, 1983: Brey, 1991; Kube and Powilleit,
1997). In addition, it is possible that parasitism of P. elegans
may influence the structure of intertidal communities by
altering or creating engineering functions in hosts. Engi-
neering functions are those that produce new habitat as a
result of changes in behaviors or life history associated with
parasitism (Thomas et ai, 1999). Clearly, researchers
should consider the impacts of parasites on the ecology and
evolution of the reproductive strategies of marine inverte-
brates and on the structure of infaunal communities.

Acknowledgments

I thank Glenys Gibson for her suggestions on experimen-
tal design and Mark Forbes for our many discussions about
life-history theory. Funding was provided by postdoctoral
fellowships from the Natural Sciences and Engineering Re-
search Council of Canada and the Coastal Studies Center,
Bowdoin College.

Literature Cited

Anger, V. 1984. Reproduction in Pygospio elegans (Spionidae) in rela-
tion to its geographical origin and to environmental conditions: a
preliminary report. Fortschr. Zoo/. 29: 45-51.

Anger, K.. V. Anger, and E. Hagmeier. 1986. Laboratory studies on
larval growth of Polydora ligni, Polydora ci/iata, and Pygospio elegans
(Polychaeta: Spionidae). Helgol. Meeresunters. 40: 377-395.

Boates, J. S., and P. C. Smith. 1989. Crawling behaviour of the am-
phipod. Corophium vohitator and foraging by semipalmated sandpip-
ers. Culidri.\ pus ilia. Can. J. Zoo/. 67: 457-462.

Born, M. A. 1999. Tidelog: Northern New England Edition. Pacific
Publishers. Bolinas. CA.

Brey, T. 1991. Interactions in soft bottom benthic communities: quanti-
tative aspects of behaviour in the surface deposit feeders Pygospio
elegans (Polychaeta) and Macoma balthica (Bivalvia). Helgol. Meere-
sunters. 45: 301-316.

Brown, R. J.. M. Conradi, and M. H. Depledge. 1999. Long-term
exposure to 4-nonylphenol affects sexual differentiation and growth of
the amphipod Corophium volutator (Pallas, 1766). Sri'. Total Environ.
233: 77- X8.

Chia, F. S., G. Gibson, and P. V. Qian. 1996. Poecilogony as a
reproductive strategy of marine invertebrates. Oceanol. Acta 19: 203-
208.

Curtis, L. A. 1987. Vertical distribution of an estuarine snail altered by
a parasite. Science 235: 1509-151 1.

Curtis, L. A. 1997. Ilvanassa obsoleta (Gastropoda) as a host for trema-
todes in Delaware estuaries. J. Parasitol. 83: 793-803.

Curtis, L. A., and K. M. K. Hubhard. 1990. Trematode infections in a
gastropod host misrepresented by observing shed cercariae. J. Exp.
Mar. Bwl. Ecol. 143: 131-137.

Folstad, I., A. C. Nilssen, O. Halvorsen, and J. Andersen. 1991. Par-
asite avoidance: the cause of post-calving migrations in Rangifer? Can.
J. Z<«>/. 69: 2423-2429.

Forbes, M. R. L. 1993. Parasitism and host reproductive effort. Oikos
67: 444-450.

Gibson, G. I)., and J. M. L. Harvey. 2000. Morphogenesis during
asexual reproduction in Pygospio elegans Claparede (Annelida.
Polychaeta). Biol. Bull. 199: 41-49.

Gudmundsson. H. 1985. Life history patterns of polychaete species of
the Family Spionidae. J. Mar. Biol. Assoc. UK 65: 93-1 11.

Hobson, K. D.. and R. H. Green. 1968. Asexual and sexual reproduc-
tion of Pygospio elegans (Polychaeta) in Barnstable Harbor. Massa-
chusetts. Biol. Bull. 135: 410 (abstract).

Hurst, L. D., and J. R. Peck. 1996. Recent advances in understanding of
the evolution and maintenance of sex. Trends Ecol. Evol. 11: 46-52.

Keats, D. VV. 1990. Food of winter flounder Pseudopleuronectes ameri-
ciinii.i in a sea urchin dominated community in eastern Newfoundland
( Canada 1. Mar. Ecol. Prog. Ser. 60: 13-22.

Kraaijeveld, A. R., and C. J. Godfray. 1997. Trade-off between para-
sitoid resistance and larval competitive ability in Drosophi/a melano-
t>astcr. Nature 389: 278-280.

Kube. J., and M. Powilleit. 1997. Factors controlling the distribution of
Maren;.elleria cf. viridis. Pvgospio e/egans and Strehlospio shrubsoli

PARASITISM AND ASEXUAL REPRODUCTION

(Polychaeta: Spionidae) in the southern Baltic Sea, with special atten-
tion for the response to an event of hypoxia. Aquat. Ecol. 31: 1 87-198.

Leonard, N., M. R. Forbes, and R. L. Baker. 1999. Effects of Limno-
chares americana (Hydrachnidia: Limnochandae) mites on life history
traits and grooming behaviour of its damselfly host, Enallagma ebrium
(Odonata: Coenagrionidae). Can. J. Zool. 77: 1615-1622.

Lively, C. M. 1996. Host-parasite coevolution and sex: Do interactions
between biological enemies maintain genetic variation and cross-fer-
tilization? Bioscience 46: 107-1 14.

Macdonald, J. S., and R. H. Green. 1986. Food resource utilization by
live henthic feeding fish in Passamaquoddy Bay, New Brunswick
(Canada). Can. J. Fish. Aquat. Sci. 43: 1534-1546.

Martin. \V. E. 1938. Studies on trematodes of Woods Hole: the life
cycle of Lepocreadium setiferoides (Miller and Northup). Allocreadi-
idae. and the description of Ct-rcuria cumingiae N. Sp. Biol. Bull. 75:
463-474.

McCurdy. D. G., M. R. Forbes, and J. S. Boates. 1999. Testing
alternative hypotheses for variation in amphipod behaviour and life
history in relation to parasitism. Int. J. Parasitol. 29: 1001-1009.

McCurdy, D. G., M. R. Forbes, and J. S. Boates. 2000a. Male amphi-
pods increase their mating effort before behavioural manipulation by
trematodes. Can. J. Zool. 78: 606-612.

McCurdy, D. G., J. S. Boates, and M. R. Forbes. 2000b. Reproductive
synchrony in the intertidal amphipod Corophium volutator. Oikos 88:
301-308.

McCurdy, D. G., J. S. Boates. and M. R. Forbes. 2000c. Spatial
distribution of the intertidal snail llyanassa obsolete! in relation to
parasitism by two species of trematodes. Can. J. Zoo/. 78: 1 1 37-1 143.

McCurdy, D. G., J. S. Boates, and M. R. Forbes. 2001. An empirical
model of the optimal timing of reproduction for female amphipods
infected by trematodes. J. Parasitol. 87: 24-31.

McDermott, J. J. 1951. Larval trematode infection in Nassa obsoleta
(Say), from New Jersey waters. Ph.D. dissertation. Rutgers University.
Newark, NJ.

Minchella. D. J. 1985. Host life-history variation in response to parasit-
ism. Parasitology 90: 205-216.

Minchella, D. J., and C. M. LoVerde. 1981. A cost of increased early
reproductive effort in the snail Biomphalaria glabrata. Am. Nat- 118:
876-881.

Morgan, T. S., A. D. Rogers, G. L. J. Paterson, L. E. Hawkins, and M.

Sheader. 1999. Evidence for poecilogony in Pygospio elegans

(Polychaeta: Spionidae). Mar. Ecol. Prog. Ser. 178: 121-132.
Perrin, N., P. Christie, and H. Richner. 1996. On host lite-history

response to parasitism. Oikos 75: 317-210.
Polak, M., and W. T. Starmer. 1998. Parasite-induced risk of mortality

ele\ ates reproductive effort in Drosophila. Proc. R. Soc. Land. B 265:

2197-2201.

Poulin. R. 1995. "Adaptive" changes in behaviour of parasitized ani-
mals: a critical review. Int. J. Parasitol. 25: 1371-1383.
Poulin. R.. J. Brodeur, and J. Moore. 1994. Parasite manipulation of

behaviour: should hosts always lose? Oikos 70: 479 — 1-84.
Rasmussen. E. 1953. Asexual reproduction in Pygospio elegans Clapa-

rede (Polychaeta sedentaria). Nature 171: 1 161-1 162.
Sheldon, B. C., and S. Verhulst. 1996. Ecological immunology: costly

parasite defenses and trade-offs in evolutionary ecology. Trends Ecol.

Evfil. 11: 317-321.
Stunkard, H. VV. 1972. Observations on the morphology and life-history

of the digenetic trematode, Lepocreadium setiferoides (Miller and

Northup. 1926) Martin. 1938. Biol. Bull. 142: 326-334.
Suwanchaichinda. C., and S. M. Paskewitz. 1998. Effects of larval

nutrition, adult body size, and adult temperature on the ability of

Anopheles gambiae (Diptera: Culicidae) to melanize Sephadex® beads.

J. Med. Entomol. 35: 157-161.
Thomas, F., R. Poulin. T. de Meeiis, J. Guegan, and F. Renaud. 1999.

Parasites and ecosystem engineering: what roles could they play? Oikos

84: 167-171.
Wilson. \V. H., Jr. 1983. The role of density dependence in a marine

infaunal community. Ecologv 64: 295-306.
Wilson. W. H., Jr. 1985. Food limitation of asexual reproduction in a

spionid polychaete. Int. J. Invertcbr. Reprod. Dev. 8: 61-65.
Woodin, S. A. 1982. Browsing: important in marine sedimentary envi-
ronments? Spionid polychaete examples. J. Mar. Biol. Assoc. UK 60:

35-45.
Zajac. R. N. 1995. Sublethal predation on Polydora comuta

(Polychaeta: Spionidae): patterns of tissue loss in a field population.

predator functional response and potential demographic impacts. Mar.

Biol. 123: 531-541.
Zar, J. H. 1996. Biostatistical Analysis. Prentice-Hall, Upper Saddle

River. NJ.
Zuk, M. 1990. Reproductive strategies and disease susceptibility: an

evolutionary viewpoint. Pamsitol. Today 6: 231-233.

Reference: Bio/. Bull. 201: 52-58. (August 2001)

Delayed Insemination Results in Embryo Mortality

in a Brooding Ascidian

J. STEWART-SAVAGE1 *, AIMEE PHILLIPPI2, AND PHILIP O. YUND2

'Department of Biological Sciences, University' of New Orleans, New Orleans. Louisiana 70148; and
2 Darling Marine Center, School of Marine Sciences, University of Maine, Walpole, Maine 04573

Abstract. We explored the effects of temporal variation
in sperm availability on fertilization and subsequent larval
development in the colonial ascidian Botryllus schlosseri. a
brooding hermaphrodite that has a sexual cycle linked to an
asexual zooid replacement cycle. We developed a method to
quantify the timing of events early in this cycle, and then
isolated colonies before the start of the cycle and insemi-
nated them at various times. Colony-wide fertilization lev-
els (assayed by early cleavage) increased from zero to 100%
during the period when the siphons of a new generation of
zooids were first opening, and remained high for 24 h before
slowly declining over the next 48 h. Because embryos are
brooded until just before the zooids degenerate at the end of
a cycle, delayed fertilization might also affect whether em-
bryos can complete development within the cycle. Conse-
quently, we also determined the effect of delayed insemi-
nation on successful embryo development through larval
release and metamorphosis. When fertilization was delayed
beyond the completion of siphon opening, there was an
exponential decline in the percentage of eggs that ultimately
produced a metamorphosed larva at the end of the cycle.
Thus, even though the majority of oocytes can be fertilized
when insemination is delayed for up to 48 h, the resulting
embryos cannot complete development before the brooding
zooids degenerate.

Introduction

Field experiments have contributed greatly to current
understanding of fertilization processes in free-spawning
marine invertebrates (reviewed by Levitan and Petersen,
1995; Yund, 2000). In response to the evidence of potential

Received 20 October 2000; accepted 8 March 2001.
* To whom correspondence should he addressed. E-mail: jssavage@
uno.edu

sperm limitation reported in some field studies, many lab-
oratory studies have started to explore diverse related as-
pects of invertebrate reproductive biology such as gamete
viscosity (Thomas, 1994a,b). egg size and sperm swimming
speed (Levitan, 1998), egg longevity (Meidel and Yund,
2001 ), sperm morphology (Eckelbarger et at., 1989a,b), and
the kinetics of fertilization (Young. 1994; Levitan. 1998;
Powell et a/., 2001 ). However, results from laboratory stud-
ies have in turn led some authors to question the extent to
which simple field fertilization experiments adequately
mimic the details of fertilization processes in nature (e.g.,
Thomas, 1994a,b; Meidel and Yund, 2001). Field experi-
ments may often circumvent aspects of reproductive strat-
egies that have evolved to mitigate sperm limitation (Yund,
2000). Hence laboratory experiments still play a vital role in
understanding reproductive strategies, and field fertilization
studies should endeavor to incorporate the details of the
fertilization process gleaned from laboratory work.

Performing realistic field experiments with marine inver-
tebrates that brood embryos presents challenges that are
very different from those faced when dealing with broadcast
spawners. The biggest challenge with field fertilization
studies of broadcasters is interpreting results obtained by
artificially holding eggs in a concentrated group (e.g., Levi-
tan and Young. 1995; Wahle and Peckham, 1999) or by
removing them from the water column after only a brief
interval (Levitan, 1991; Coma and Lasker, 1997). This issue
is moot with brooders, who by definition retain eggs and
have internal fertilization. However, a different set of prob-
lems merits further consideration. The precise timing of egg
viability, sperm release, and fertilization itself is often less
well understood than in broadcasters. Sperm function may
be regulated by the female through sperm chemotaxis
(Miller, 1985), activation (Bolton and Havenhand, 1996), or
storage (Bishop and Ryland, 1991). In the latter case, the

EFFECTS OF DELAYED INSEMINATION

temporal pattern of fertilization within a female may be
uncoupled from the pattern of sperm release by males. In
hermaphrodites, the potential for self-fertilization is a con-
cern, and genetic analyses of paternity may be required to
conclusively exclude selting in some taxa (Yund and Mc-
Cartney, 1994). For some brooders, the actual path of sperm
access to eggs is poorly understood. Information on all of
these topics is critical both to the design of more realistic
field fertilization studies and to the interpretation of existing
studies.

The colonial ascidian Botryllus schlosseri is a useful
model for field fertilization studies (Grosberg, 1991; Yund
and McCartney, 1994; Yund, 1995, 1998). Fertilization is
internal, and embryos are brooded until released as tadpole
larvae (Milkman, 1967). When colonies are grown on glass
surfaces, egg production can be quantified non-destructively
(Yund et cil.. 1997), thus permitting estimation of fertiliza-
tion levels by comparing egg and embryo counts (Yund,
1995, 1998). Although the general time of fertilization
within the life cycle (i.e., temporal resolution on the order of
a day) has long been known (Milkman, 1967), the finer-
scale timing (temporal resolution on the order of hours) has
not been explored. Many authors have assumed that the
apparent temporal separation of fertilization and sperm re-
lease prevents self-fertilization (e.g., Milkman, 1967; Gros-
berg, 1987; Yund and McCartney, 1994), but we have
recently shown (Stewart-Savage and Yund, 1997) that
sperm release commences several days earlier than previ-
ously thought. Although sperm storage has been demon-
strated in another colonial ascidian (Bishop and Ryland.
1991; Bishop and Sommerfeldt. 1996), past workers have
implicitly assumed that storage is unlikely in B. schlosseri
(Milkman, 1967; Grosberg. 1991; Yund, 1995, 1998). To
the best of our knowledge, this assumption has never been
explicitly tested. To address this interrelated set of issues,
this paper explores the effect of variation in the timing of
fertilization on fertilization levels and subsequent larval
development in B. schlosseri. and compares those results
with published information on the timing of sperm release.

Materials and Methods

Study organism

Colonies of Botryllus schlosseri are composed of asexu-
ally produced zooids arranged in clusters, or systems, with
all zooids in a system sharing a common exhalant siphon.
Throughout the life of a colony, all zooids periodically
undergo a synchronous asexual zooid replacement cycle in
which a new generation of zooids, termed buds, forms
between the existing zooids (Berrill, 1941; Izzard. 1973). At
the end of the life span of adult zooids (about 8 days at
16°C; cycle length is temperature dependent), the buds
expand, take over the function of the previous generation of
zooids (which are quickly resorbed). and then commence

their sexual reproductive cycle. The sexual cycle includes
the internal fertilization of the mature eggs soon after the
inhalant siphons open (Milkman, 1967); the continuous
release of sperm starting 16 h later (Stewart-Savage and
Yund. 1997); and the brooding of developing embryos,
which are released just before the zooids degenerate at the
end of the cycle (Milkman, 1967).

Standard methods

The colonies of B. schlosseri that were employed in this
study were collected from the Damariscotta River. Maine.
Animals were grown on glass microscope slides in the
flowing seawater system at the University of Maine's Dar-
ling Marine Center. Field-collected colonies that had been
established in laboratory culture were divided to provide
clonal replicates (ramets) of genotypes. Colonies employed
in all experiments were monitored for the approach of
takeover (the transition between zooid generations). When
colonies were about to commence takeover (late stage 5
through early stage 6 by the criteria of Milkman, 1967), they
were isolated in 50 ml of sperm-free (aged >24 h) seawater.
Isolated colonies were housed in an incubator at 16°C
(range: 14-18 °C) and fed phytoplankton (Duniella sp.) at
densities of approximately 105 cells/ml. Water and food
were changed twice daily. Colonies were monitored for
siphon opening and then isolated in individual 250-ml con-
tainers with algae (water and food were changed daily) until
exposed to sperm. Sperm exposure was accomplished by
placing colonies in a flowing seawater tank in proximity to
numerous male-phase colonies (>24 h after siphon open-
ing; Stewart-Savage and Yund, 1997) for 1 h. After insem-
ination, colonies were rinsed with aged seawater and re-
turned to isolation.

Experimental protocols

To standardize insemination times, we first had to accu-
rately quantify the start of the reproductive cycle (i.e., the
functional opening of siphons). Inhalant siphons are formed
early in the takeover process, but the common exhalant
siphon of a system generally does not form until near the
end. However, it is difficult to ascertain functional siphon
opening on morphological criteria alone. In the course of
other work, we observed that the consumption of green
algae immediately turned the digestive systems of actively
feeding zooids (i.e., those that must have open siphons)
green. Consequently, we used algal uptake as an assay for
siphon opening. To establish the temporal pattern of siphon
opening, we isolated 14 colonies and briefly exposed them
to algae three to four times during the process of takeover.
At each sample interval we recorded the percentage of
siphons that were open (% of zooids with green digestive
systems). From these data we calculated an average rate of
siphon opening. This approach subsequently allowed us to

J. STEWART-SAVAGE ET AL.

make single observations of the percentage of siphons that
were open and back-calculate the time of the first siphon
opening. Both of our other experiments use this approach to
estimate the time of initial siphon opening, and the timing of
insemination is expressed relative to this event.

To examine the effect of the timing of fertilization on
fertilization levels, we exposed colonies to sperm through a
range of different times after siphon opening (0.5 to 96 h;
n =- 79). Colonies with about 20 eggs (mean of 20.0 ±
standard error of 11.6) were utilized throughout, and all
eggs and embryos in a colony were surgically removed
10-18 h after insemination and scored for successful de-
velopment. Initial studies indicated that embryos should be
in the 8-cell to the 32-cell stages during this time range.
Uncleaved eggs were scored as unfertilized, as were em-
bryos with an abnormal cleavage pattern (arrested cleavage,
abnormal cell number or shape). A few embryos at ad-
vanced developmental stages (e.g., gastrula) were excluded
from the data set since fertilization was by either contami-
nating or self sperm.

To examine the effect of timing of fertilization on sub-
sequent development and metamorphosis, colonies were
initially fertilized in sets of multiple ramets per genotype.
For each genotype, one ramet was left unfertilized (to assess
the level of sperm contamination or self-fertilization), one
ramet was fertilized about 22 (±2) h after the beginning of
siphon opening (when results from the previous experiment
indicated that all siphons should be open), and remaining
ramets (2-3) were fertilized at various times up to 85 h after
initial siphon opening. Because fertilization was consis-
tently minimal in unfertilized controls and the availability of
genotypes with multiple egg-bearing ramets was often lim-
ited, later trials were conducted without the control treat-
ment. Before takeover, we counted the number of eggs
produced by each colony (minimum egg production was set
at 25 eggs). After insemination, colonies were returned to
isolation until all ramets of a genotype had been fertilized
and at least 24 h had elapsed since the last insemination.
Colonies were subsequently housed in a flowing seawater
table with an independent seawater supply while embryonic
development proceeded: they were re-isolated at stage tour
(Milkman. 1967). After each isolated colony had started the
next reproductive cycle, all metamorphosed juveniles in the
isolation container were counted. Data from colonies that
died or became visibly unhealthy during the experiment
were discarded.

Results

Timing of siphon opening

Feeding did not begin until after the organization of
zooids into new systems and formation of the common
exhalant siphon. Although the rate of siphon opening varied
among colonies (Fig. 1: range of 3.0%/h— 17.8%/h), the

100 n

o
o
N
so

H 40 -

-o
u
u

20 -

0 4

Time from Initial Observation (h)

Figure 1. Rate of siphon opening in colonies of Botryllus schlosseri as
assayed by the presence of algae in the digestive system. Colonies were
isolated in 50 ml aged seawater with 2 x 105 algae/ml and monitored at
intervals of from 1 to 12 h. Zero time is the first observation of algae in the
gut. Temporal patterns for 14 individual colonies are shown. Differences in
the v-intercept simply reflect how far the takeover process had proceeded
when colonies were first observed; slopes indicate the rate of siphon
opening.

average rate of siphon opening of the colonies was 7.8%/
h ± 4.5%/h (X ± SD). We used the average rate of siphon
opening to normalize the time of sperm exposure to the start
of siphon opening for colonies in the other two experiments.

Effect of timing of insemination on fertilization levels

To determine the time frame during which eggs can be
fertilized within the female, we exposed virgin females to a
1-h pulse of sperm at various times after the beginning of
siphon opening and assayed successful fertilization by the
percentage of normally cleaved embryos present (Fig. 2).
When virgin females were exposed to sperm during the
period in which their siphons were opening (first 24 h), the
level of fertilization increased with time (Fig. 2B). In col-
onies fertilized during siphon opening, there was no spatial
relationship between fertilized and unfertilized eggs either
within or among systems; it was common to find both in the
same zooid. Because the rate of increasing fertilization
(5.4%/h) is similar to the rate of siphon opening (7.8%/h ±
4.5%/h), we conclude that fertilization of the eggs within a
zooid occurs shortly after the opening of the siphon.

After the completion of siphon opening, fertilization suc-
cess remained high (>90%) for 24 h and then declined over
the next 48 h with a 7"500, of 72 h (Fig. 2B). In a subset of
genotypes where multiple ramets were inseminated at dif-
ferent times in the same reproductive cycle, thus controlling
for potential genotype and cycle effects, the effect of

EFFECTS OF DELAYED INSEMINATION

0 24 48 72 96

Insemination Time
(h after start siphon opening)

Figure 2. Effect of insemination pulse timing on fertilization levels.
Colonies were isolated before the start of siphon opening, monitored for the
timing of siphon opening, and exposed to sperm for 1 h: the number of
cleaving embryos was determined 10-18 h later. (A) Fertilization levels in
different ramets of seven genotypes fertilized at different points in the same
reproductive cycle. (Bl Overall effect of insemination time on fertilization
success in ramets from 25 genotypes. The line represents a polynomial
regression of the data (R2 = 0.580).

delayed insemination on fertilization varied by genotype
(Fig. 2 A). Of the seven genotypes in which different ramets
were inseminated at different times, five genotypes had a
decline in fertilization that mirrored the population data. In
the other two genotypes, fertilization levels declined rapidly
in one. but remained relatively stable over 60 h in the other.
Excluding the genotype that exhibited little decline in fer-
tilization, the average T50Vf for the reduction of fertilization
was 62 ± 15 h, a value similar to the population-wide
regression.

Effect of liming of insemination on embr\o development
and metamorphosis

The maximum duration of gestation is fixed by the length
of the asexual zooid replacement cycle. Since eggs could be

fertilized well after siphon opening, but the time of embryo
release is fixed, we examined the effect of delayed insem-
ination on reproductive success. Successful embryo meta-
morphosis was selected as an assay of reproductive success
because it integrates possible effects on fertilization, devel-
opment, larval behavior, and settlement. In five trials that
included unfertilized (low control), insemination at 22 h
(high control), and ramets inseminated at different times
after siphon opening, the percentage of eggs that success-
fully developed through metamorphosis consistently de-
creased with the time of insemination (Fig. 3A). The unfer-
tilized controls resulted in either zero or very low (<5%)
levels of larval metamorphosis (Fig. 3 A). However, the
percent of eggs developing through metamorphosis varied
substantially among 22-h insemination controls (Fig. 3A).
Because of the low levels of successful metamorphosis in
two genotypes fertilized at 22 h. we calculated the T50%
relative to the maximum value for each genotype. The
relative T50C7c for the reduction of metamorphosis success
was 41 ± 6 h after the start of siphon opening (about 19 h
after the completion of siphon opening). When data from all
12 trials were combined (Fig. 3B), larval metamorphosis
exhibited an exponential decline with fertilization time be-
yond 22 h. No larval metamorphosis occurred when colo-
nies were fertilized more than 78 h after the start of siphon
opening.

Two outliers (both ramets of the same genotype) had
disproportionately high levels of metamorphosis when fer-
tilized about 48 h after siphon opening (Fig. 3B. open
squares). Independent evidence (i.e.. observations of suc-
cessful embryo development in isolated colonies) suggested
that this genotype may sometimes be able to self-fertilize.
Alternatively, the high fertilization levels in these two col-
onies may be the result of sperm contamination. Because
these inconsistent values are limited to one genotype, we
have excluded these values from the regression in Figure
3B. Inclusion of the two points in the regression has little
effect on the equation parameters, but it substantially re-
duces the coefficient of determination. Note that many other
ramets of this genotype were employed in this experiment
(Fig. 3B, open squares) and produced results consistent with
those of the other genotypes.

Discussion

Although more than 50% of Boti-yllus schlosseri eggs can
be fertilized 38 to 48 h after the completion of siphon
opening (Fig. 2), few viable larvae are produced unless
fertilization occurs within the first 19 h (Fig. 3). The de-
crease in embryo production after delayed fertilization
could be caused by either egg aging or limitations on the
duration of brooding. As in most invertebrates, the time
required to complete development is a function of temper-
ature in B. schlosseri. Since the asexual zooid replacement

J. STEWART-SAVAGE ET AL.

.5 o
§" p.

•I o

g E

Q -

100 r

75 -

~^^T! • \ .

" Tj HI

3 24

72 Unfert

o - — •

.£ o

.2 e-

o o
§ |

Q 2

100 r

75 -

0 -

Insemination Time
(h after start siphon opening)

Figure }. Effect of insemination pulse timing on embryo development
and larval metamorphosis. Colonies with quantified egg production were
isolated before the start of siphon opening, monitored for the timing of
siphon opening, and exposed to sperm for 1 h: the number of settled
juveniles was determined 5-7 days later. (A) Developmental success of
different ramets from five genets. In three of the genets, one ramet was
never exposed to sperm (unfertilized, solid symbols). (B) Overall effect of
insemination time on successful development. The open squares are the
ramets from the putative self-fertilizing genotype; closed symbols repre-
sent the other 1 1 genotypes. The line is an exponential regression of the
data except for two outliers at 48 h (R2 = 0.713).

cycle is also a function of temperature (Grosberg, 1982).
delayed fertilization could cause the brooding zooids to
degenerate before the embryos have become competent to
undergo metamorphosis. The deleterious effects of egg ag-
ing have been demonstrated in mammals (Juetten and
Bavister. 1983; Xu et <(/., 1997), but such effects are usually
manifested early in development. Since early development
was normal in all but one colony with delayed fertilization
(pers. obs.), the decreased gestational duration caused by

delayed fertilization is more likely to be responsible. Nev-
ertheless, additional work on the mechanism by which de-
layed fertilization decreases larval production could more
fully resolve this issue.

In spite of the narrow temporal window in which both
fertilization and development are likely to be successful
(Figs. 2 and 3), field experiments indicate that colonies of B.
schlosseri are very adept at acquiring sperm. A single male-
phase colony can fertilize most eggs of a nearby female-
phase colony with very few sperm (Yund, 1998). If several
males are present, they compete to fertilize eggs (Yund.
1995. 1998), and closer males can be successful at the
expense of more distant males (Yund and McCartney,
1994). Although sperm transfer usually occurs among
nearby colonies (Yund. 1995). sperm can also be obtained
from very distant locations when insufficient local sperm
are available (Yund, 1998). Even eggs of colonies isolated
from the nearest natural populations by tens of meters can
be fertilized at appreciable levels (Yund and McCartney,
1994). The apparent ease of fertilization under field condi-
tions, in spite of a very limited temporal window for suc-
cessful fertilization and development, suggests that the pro-
cess of sperm capture by colonies must be extremely
efficient. Nevertheless, in low-density populations where
sperm may be in short supply (Yund. 1998). or in marginal
habitats in which sperm production is suppressed (Stewart-
Savage et a/.. 2001). our work suggests that reproductive
failure may occur in spite of successful fertilization if fer-
tilization occurs too late in the reproductive cycle. Recent
field sampling has demonstrated this phenomenon in natural
populations near the end of the annual reproductive season
(Yund and Phillippi, unpubl. data).

Unlike the colonial ascidian Diplosoma listerianum, in
which fertilization can be temporally disassociated from
sperm exposure and colonies can store sperm for up to one
month (Bishop and Ryland. 1991 ; Bishop and Sommerfeldt,
1996). B. schlosseri colonies apparently cannot store sperm.
The evidence for this conclusion is, first, that colonies
isolated in sperm-free seawater were not fertilized until we
experimentally supplied a sperm pulse, indicating that
sperm are not stored and transferred from one asexual
generation of zooids to the next. The apparently complete
resorption of all zooid tissue at the end of the cycle further
suggests that transmission between cycles is unlikely. Sec-
ond, the tight temporal relationship between siphon opening
and fertilization (Fig. 2B) suggests that sperm cannot enter
until the new generation of zooids opens its siphons and
starts to feed. Third, the narrow window of time in which
fertilization is both possible (Fig. 2) and results in viable
offspring (Fig. 3) eliminates any apparent fitness advantage
to sperm storage within a single asexual generation.

The route of sperm access to eggs in B. schlosseri is
unknown, but there are at least two possible points of entry
(Ryland and Bishop. 1993): sperm enter through the

EFFECTS OF DELAYED INSEMINATION

inhalant siphon and cross the pharyngeal basket to reach the
eggs, or sperm enter through the exhalant siphon and then
swim to the eggs. During takeover in B. schlosseri, the
exhalant siphon of each system is formed before the inhal-
ant siphons of all of the component zooids open, and the
precise timing of exhalant siphon formation varies among
systems (pers. obs.). If sperm enter via the exhalant siphon,
fertilization levels in the early time intervals of our fertili-
zation timing experiment should have varied among sys-
tems, but should not have varied within a system. However,
we routinely found mixtures of fertilized and unfertilized
eggs within the same system, suggesting that sperm entry to
each zooid required an open inhalant as well as exhalant
siphon. Although further work is required to determine the
route of sperm entry into Botryllus colonies, we think it is
unlikely that sperm enter via the exhalant siphon.

Hermaphroditism creates another challenge for success-
ful reproduction in B. schlosseri. Inbreeding depression
(Sabbadin, 1971) is likely to exert selective pressure to
prevent self-fertilization, even though selting would be a
possible mechanism to assure fertilization in the narrow
time window in which fertilization can produce functional
embryos. When the data in this paper are combined with
previous data on the timing of sperm release (Stewart-
Savage and Yund. 1997). it is apparent that the male and
female phases of the reproductive cycle overlap in B.
schlosseri (Fig. 4). Sperm release overlaps for about 48 h
with the window for successful fertilization, but there is
substantially less overlap with the narrower window in
which fertilization results in viable embryos (Fig. 4). Con-
sequently, B. schlosseri is not a true sequential hermaphro-
dite (Milkman, 1967), but the male and female phases are
functionally separated in time. This functional segregation
of the reproductive phases probably plays some role in

0 24 48 7: 96 120 144 168 192 216 240
Time From Completion of Siphon Opening (h)

Figure 4. Relationship between male and female reproductive phases
in Botryllus schlosseri. Data collected at different temperatures have been
normalized to a 10-day cycle length. The zero time point is the completion,
rather than the initiation (as in Figs. 2 and 3), of siphon opening. The sperm
release curve is redrawn from Stewart-Savage and Yund (1997) with
permission.

ensuring that few metamorphosing embryos result from
self-fertilization. However, the very success of our experi-
mental protocols indicates that one or more additional
mechanisms to prevent self-fertilization must exist. Eggs of
colonies isolated in small volumes of water until points in
the reproductive cycle at which substantial self-sperm
should have been present (Fig. 4) nevertheless remained
unfertilized until we introduced a pulse of sperm (with the
possible exception of the two outliers in Fig. 3B). Conse-
quently, some form of self-incompatibility, as described in
other colonial and solitary ascidians (Rosati and De Sands.
1978: Bishop, 1996), appears likely in B. schlosseri (see
also Scofield et «/., 1982).

Acknowledgments

Financial support was provided by the National Science
Foundation (OCE-97-30354). This is contribution number
366 from the Darling Marine Center.

Literature Cited

Berrill, N. J. 1941. The development of the bud in Bolryllus. Biol. Bull.
80: 169-1X4.

Bishop, J. D. D. 1996. Female control of paternity in the internally
fertilizing compound ascidian Diplosoma listeriannm. I. Autoradio-
graph investigation of sperm movements in the female reproductive
tract. Proc. R. Soc. Land. B 263: 369-376.

Bishop, J. D. D.. and J. S. Ryland. 1991. Storage of exogenous sperm
by the compound ascidian Diplosoma listeriannm. Mar. Biol. 108:
111-118.

Bishop. J. D. D.. and A. D. Sommerfeldt. 1996. Autoradiographic
investigation of uptake and storage of exogenous sperm by the ovary of
the compound ascidian Diplosoma listerianum. Mar. Biol. 125: 663-
670.

Bolton. T. F.. and J. N. Havenhand. 1996. Chemical mediation of
sperm activity and longevity in the solitary ascidians Ciona intestinalis
and Ascitlit'lla aspersa. Biol. Bull. 190: 329-335.

Coma, R., and H. R. Lasker. 1997. Effects of spatial distribution and
reproductive biology on in situ fertilization rates of a broadcast-spawn-
ing invertebrate. Biol. Bull. 193: 20-29.

Kckelbarger. K. J., C. M. Young, and J. L. Cameron. 1989a. Modified
sperm ultrastructure in four species of soft-bodied echinoids (Echino-
dermata: Echinothuriidae) from the bathyal zone of the deep sea. Biol.
Bull. 177: 230-236.

F.ckelharger, K. J., C. M. Young, and J. L. Cameron. 1989b. Ultra
structure and development of dimorphic sperm in the abyssal echinoid
Phrissoc\sti.t multispina (Echinodermata: Echinoidea): implications for
deep sea reproductive biology. Biol. Bull. 176: 257-271.

Grosberg, R. K. 1982. Ecological, genetical, and developmental factors
regulating life history variation within a population of the colonial
ascidian Botryllus schlosseri (Palas) Savigny. Ph.D. dissertation. Yale
University. New Haven, CT.

Grosberg. R. K. 1987. Limited dispersal and proximity-dependent mat-
ing success in the colonial ascidian Botryllus schlosseri. Evolution 41:
372-384.

Grosberg, R. K. 1991. Sperm-mediated gene flow and the genetic struc-
ture of a population of the colonial ascidian Botryllus schlosseri.
Evolution 45: 130-142.

Izzard. C. S. 1973. Development of polarity and bilateral asymmetry in
the palleal bud of Botryllus schlosseri (Pallas). ./. Morphol. 139: 1-26.

J. STEWART-SAVAGE ET AL

Juetten, J., and B. D. Bavister. 1983. Effects of egg aging on in vitro
fertilization and first cleavage division in the hamster. Gamete Res. 8:
219-230.

Levitan, D. R. 1991. Influence of body size and population density on
fertilization success and reproductive output in a free-spawning inver-
tebrate. Biol. Bull. 181: 261-268.

Levitan. D. R. 1998. Does Bateman's principle apply to broadcast-
spawning organisms? Egg traits influence in situ fertilization rates
among congeneric sea urchins. Evolution 52: 1043-1056.

Levitan, D. R., and C. Petersen. 1995. Sperm limitation in the sea.
Trends Ecol. Evol. 10: 228-231.

Levitan, D. R., and C. M. Young. 1995. Reproductive success in large
populations: experimental measures and theoretical predictions of fer-
tilization in the sea biscuit Clypeasler rosaceus. J. Exp. Mar. Biol.
Ecol 190: 221-241.

Meidel, S. K., and P. O. Yund. 2001. Egg longevity and time-integrated
fertilization in a temperate sea urchin (Strongylocentrotus droebachien-
sis). Biol. Bull. 201: 000-000.

Milkman. R. 1967. Genetic and developmental studies on Botrylhis
schlosseri. Biol. Bull. 132: 229-243.

Miller, R. L. 1985. Sperm chemo-oriemation in the metazoa. Pp. 275-
337 in Biology of Fertilization. Vol. 2, C. B. Metz and A. Monroy. eds.
Academic Press, Orlando. FL.

Powell, I). K., P. A. Tyler, and L. S. Peek. 2001. Effect of sperm
concentration and sperm ageing on fertilization success in the Antarctic
soft-shelled clam Laternula elliptica and the Antarctic limpet Nucella
concinna. Mar. Ecol. Prog. Ser. 215: 191-200.

Rosati, F., and R. De Santis. 1978. Studies on fertilization in the
ascidians. I. Self-sterility and specific recognition between gametes of
dona intestinalis. Ev/>. Cell Res. 112: 1 1 1-1 19.

Ryland, J. S., and J. D. D. Bishop. 1993. Internal fertilization in
hermaphroditic colonial invertebrates. Oceanogr. Mar. Biol. Annit.
Rev. 31: 445-477.

Sabbadin, A. 1971. Self- and cross-fertilization in the compound ascid-
ian Botryllus schlosseri. Dev. Biol. 24: 379-391.

Scofield, V. L., J. M. Schlumpberger, and I. L. Weissman. 1982.
Colony specificity in the colonial tunicate Bony/Ins and the origins of
vertebrate immunity. Am. Zoo/. 22: 783-794.

Stewart-Savage, J., and P. O. Yund. 1997. Temporal pattern of sperm

release from the colonial ascidian. Botryllus schlosseri. J. Exp. Zooi

279: 620-625.
Stewart-Savage, J.. A. Stires, and P. O. Yund. 2001. Environmental

effect on the reproductive effort of Botrylhis schlosseri. Pp. 3 1 1-3 14 in

Biology of Ascidians. H. Sawada. H. Yokosawa, and C. C. Lambert,

eds. Springer-Verlag, Tokyo, Japan.
Thomas, K. I. M. 1994a. Physical properties of gametes in three sea

urchin species. J. E.\p. Biol. 194: 263-284.
Thomas. F. I. M. 1994b. Transport and mixing of gametes in three

free-spawning polychaete annelids, Phragmatopoma californica

(Fewkes), Sabellaria cementarium (Moore), and Schi-obranchia insig-

nis (Bush). J. Exp. Mar. Biol. Ecol. 179: 1 1-27.
Wahle, R. A., and S. H. Peckham. 1999. Density-related reproductive

trade-offs in the green sea urchin. Strongylocentrotus droebachiensis.

Mar. Biol. 134: 127-137.
Xu, Z., A. Abbott, G. S. Kopf, R. M. Schultz. and T. Ducibella. 1997.

Spontaneous activation of ovulated mouse eggs: time dependent effects

on M-phase exit, cortical granule exocytosis, maternal messenger ri-

bonucleic acid recruitment and inositol 1,4,5-trisphosphate sensitivity.

Biol. Reprod. 57: 743-750.
Young, C. M. 1994. The biology of external fertilization in deep-sea

echinoderms. Pp. 179-200 in Reproduction. Lan-al Biology and Re-
cruitment of the Deep-sea Benthos. C. M. Young and K. J. Eckelbarger,

eds. Columbia University Press. New York.
Yund, P. O. 1995. Gene flow via the dispersal of fertilizing sperm in a

colonial ascidian (Botryllus schlosseri): the effect of male density. Mar.

Biol. 122: 649-654.
Yund, P. O. 1998. The effect of sperm competition on male gain curves

in a colonial marine invertebrate. Ecology 79: 328-339.
Yund, P. O. 2000. How severe is sperm limitation in natural populations

of marine free-spawners? Trends Ecol. Evol. 15: 10-13.
Y'und, P. O., and M. A. McCartney. 1994. Male reproductive success

in sessile invertebrates: competition for fertilizations. Ecologv 75:

2151-2167.
Yund, P. O., Y. Marcum, and J. Stewart-Savage. 1997. Life-history

variation in a colonial ascidian: broad-sense heritabilities and tradeoffs

in allocation to asexual growth and male and female reproduction. Biol.

Bull. 192: 290-299.

Reference: Biol. Bull. 201: 59-64. (August 20(11)

Morula Cells as the Major Immunomodulatory

Hemocytes in Ascidians: Evidences From

the Colonial Species Botryllus schlosseri

LORIANO BALLARIN1-*, ANTONELLA FRANCHINI2, ENZO OTTAVIANI2, AND

ARMANDO SABBADIN1

1 Department of Biologv, University of Padova, via U. Bassi 58/B, 35100 Padova. Italy: and

^Department of Biologv, University of Modena and Reggio Emilia,

via Campi 213/D, 41100 Modena, Italy

Abstract. Immunocytochemical methods were used to
study the presence and distribution of IL-1 -a- and TNF-a-
like molecules in the hemocytes of the colonial ascidian
Botryllus schlosseri. Only a few unstimulated hemocytes
were positive to both the antibodies used. When the hemo-
cytes were stimulated with either mannan or phorbol 12-
mono-myristate. the phagocytes were not significantly
changed in their number, staining intensity, or cell morphol-
ogy. In contrast, stimulated morula cells were intensely
labeled, indicating that these cells play an important immu-
nomodulatory role.

Introduction

Phagocytes and morula cells are two types of circulating
hemocytes that play a key role in ascidian immunobiology.
Phagocytes can easily recognize and ingest non-self cells
and particles (Smith, 1970; Anderson. 1971; Fuke and Fu-
kumoto. 1993: Ballarin et ai. 1994: Ohtake et ai, 1994;
Dan-Sohkawa et ai, 1995; Cima et ai, 1996) and are able
to synthesize and release opsonic agglutinins (Coombe et
ai, 1984; Kelly et ai, 1992; Ballarin et ai. 1999). Morula
cells, a ubiquitous hemocyte type among ascidians. take part
in a variety of biological functions of irnmunological rele-
vance, such as hemolymph clotting, tunic synthesis, and

Received 18 July 2000; accepted 10 May 2001.

* To whom correspondence should be addressed. E-mail:
ballarin@civ.bio.unipd.it

Abbreviations: FSW. filtered seawater; HA. hyaline amoebocytes: IL.
imerleukin; MLC. macrophage-like cells: PMM. phorbol 12-mono-myris-
tate; TNF. tumor necrosis factor.

encapsulation of foreign bodies (Endean, 1955b; Smith.
1970; Anderson, 1971; Chaga, 1980; Wright. 1981: Za-
niolo. 1981). They are by far the most frequent circulating
ascidian cell-type (Endean. 1955a: Andrew. 1961; Smith.
1970; Kustin et ai, 1976; Ballarin et ai, 1995). and their
abundance suggests direct involvement in other important
defense reactions. Although most of their roles in ascidian
immune responses still remain unclear, morula cells can
induce cytotoxicity after recognition of foreign molecules or
cells (Parrinello. 1996; Cammarata et ai. 1997: Ballarin et
til.. 1998), and they are also required for phagocytosis
(Smith and Peddie. 1992).

Cytokines are soluble molecules that mediate communi-
cation among various immunocyte types in vertebrate im-
mune systems. In the last decade, much evidence has accu-
mulated indicating that cytokine-like molecules are also
involved in invertebrate immune responses, and their pres-
ence has been demonstrated in hemocytes of molluscs,
annelids, arthropods, echinoderms, and tunicates (Beck and
Habicht. 1991; Ottaviani et ai. 1995a.b. 1996: Franchini et
iii. 1996). Cytokine-like molecules stimulate cell prolifer-
ation, increase hemocyte motility and phagocytic activity,
and induce nitric oxide synthase (Raftos et ai, 1991: Otta-
viani ft ai, 1995b). As regards ascidians, the activities of
interleukin-l (IL-1 )- and IL-2- but not tumor necrosis factor
(TNF)-like molecules have been revealed in various spe-
cies, either solitary or colonial (Beck et ai. 1989). Tunicate
IL-1 -like molecules modulate immune responses and are
secreted by hemocytes in response to exogenous stimuli
(Raftos et ai. 1991. 1992. 1998: Beck et ai, 1993; Kelly et
ai, 1993).

L. BALLARIN ET AL.

We have studied — in hemocytes of the colonial ascid-
ian Botryllus schlosseri — the presence and distribution of
molecules that are immunoreactive to antibodies raised to
human IL-l-a and TNF-a. The results indicate that these
immunoreactive molecules are mainly detectable in stim-
ulated morula cells, suggesting that these cells have a role
in immunomodulation. Moreover, previous results in
other ascidian species are supported (Smith and Peddie,
1992).

Materials and Methods

Animals

Wild colonies of Botryllus schlosseri from the lagoon of
Venice, Italy, were used. They were kept in aerated aquaria,
attached to glass slides, and fed with Liquifry Marine
(Liquifry Co., England) and algae.

Hemocyte monolayers

Colonies were rinsed in filtered seawater (FSW), pH 7.5,
containing 10 mM L-cysteine as anticoagulant. The tunic
marginal vessels were then punctured with a fine tungsten
needle, and hemolymph was collected with a glass micropi-
pette. Hemolymph was centrifuged at 780 X g for 10 min,
and pellets were resuspended in FSW to a final hemocyte
concentration of 8-10 X 106 cells/ml. Samples of the he-
mocyte suspension (50-100 /xl) were cytocentrifuged onto
slides with a Shandon Instrument Cytospin II running at 500
rpm for 2 min. Hemocytes were then stained with May
Griinwald-Giemsa for morphological examination with a
Leitz Dialux 22 light microscope.

Hemocyte stimulation

Cell suspensions were placed in 1-ml tubes on a revolv-
ing mixer, and hemocytes were stimulated by incubation for
5, 15, 30, and 60 min with mannan at 5 mg/ml or phorbol
12-mono-myristate (PMM) at 20 nM in FSW containing 10
mM L-cysteine to prevent cell clotting. Mannan. a quite
common microbial polysaccharide, is easily recognized by
mannose receptors, the presence of which has been indi-
rectly interred on the surface of Botryllus phagocytes (Bal-
larin et al., 1994). PMM is a well-known activator of protein
kinase C that mimics the action of diacylglycerol (Wolfe,
1993). The above-reported concentrations of the two com-
pounds were previously demonstrated as the most effective
in stimulating Botryllus phagocytes and the related respira-
tory burst (Ballarin et al.. 1994; Cima et al.. 1996). FSW

was used for controls. The viability of hemocytes, after the
incubation, was assessed by the trypan blue exclusion assay
(Gorman et al., 1996).

Immunocytochemistry

The immunocytochemical procedure described by Otta-
viani et al. (1990) was performed. The following two pri-
mary antibodies were used: polyclonal anti-human IL-l-a
(1:250, 1:500, 1:1000) (Santa Cruz Biotech., USA) and
monoclonal anti-human TNF-a (1:25, 1:50, 1:100) (Neo-
Markers, USA). Cells were incubated with primary antibod-
ies overnight at 4°C, and reactivity was revealed by immu-
noperoxidase staining using avidin-biotin-peroxidase
complex (Hsu et al.. 1981). The best results were obtained
with anti-IL-1-a and anti-TNF-a diluted 1:500 and 1:25,
respectively. In control preparations, the primary antibodies
were either substituted with non-immune sera or absorbed
with homologous antigen (i.e., human IL-l-a and TNF-a)
before addition to hemocyte monolayers. Moreover, a poly-
clonal antibody raised against Botryllus agglutinin (BA)
(Ballarin et nl., 2000) was also assayed as a control for
specificity. Nuclei were counterstained with hematoxylin.
The frequency of positive hemocytes, phagocytes, and
morula cells was reported as the percentage of the total
hemocyte number, which was determined by counting at
least 600 cells in 10 fields under the light microscope.

Statistical analysis

All experiments were repeated in triplicate, and statistical
analysis was performed using the chi-square test (^2).

Results

Morphology of cytocentrifuged Botryllus hemocytes

The main hemocyte types present in B. schlosseri hemo-
lymph were identifiable under the light microscope after
cytocentrifugation. Lymphocyte-like cells, representing
2%-4% of circulating hemocytes. contain a large round
nucleus surrounded by a thin layer of basophilic cytoplasm.
Phagocytes, which include hyaline amoebocytes (HA; ac-
tively phagocytosing cells) and macrophage-like cells
(MLC) (Ballarin et al.. 1994). have roundish nuclei and
neutrophilic cytoplasm which, in the case of MLC, sur-
rounds one or more vacuoles containing ingested material
(Fig. la, b). Phagocytes constitute 30%-40% of circulating
blood cells. Morula cells, the frequency of which is 30%-
50% of total hemocytes, are characterized by the presence
of several yellowish-green vacuoles (Fig. 2a, c). Nephro-

CYTOKINE-LIKE MOLECULES IN BOTRYLLUS

. HA

MLC

Figure 1. Cytocentrifuged Botry/liis schlosseri hemocytes stained with May Griinwald-Giemsa solution, (a)
Lymphocyte-like cell (LL) and hyaline amebocyte (HA); (b) macrophage-like cell (MLC; n: nucleus; v:
vacuole); (cl nephrocyte (N) with several empty vacuoles (arrowheads). Bar = 10 /xm.

cytes and pigment cells (6%-10% of circulating hemocytes)
were not well preserved after cytocentrifugation; they ap-
peared as giant cells with empty vacuoles (Fig. Ic).

Response of unstimulated hemocytes to anti-cytokine
antibodies

Using anti-IL-1-a and anti-TNF-a, only some phago-
cytes and a few morula cells were labeled after immuno-
peroxidase staining (Table 1). Thus, most HA, MLC, and
morula cells were not immunoreactive with either antibody
(Fig. 3). Moreover, no other cell-types stained positively for

stimulated

Figure 2. Unstimulated (a, c) and stimulated (h. d) morula cells after
immunoperoxidase staining with anti-cytokine antibodies, (a. b) Incubation
with anti-IL-1-a antibody; (c. d) treatment with the TNF-a antibody.
Bar = 15 jum.

the two cytokines. No labeling was observed when non-
immune sera were used.

Response of stimulated hemocytes to anti-cytokine
antibodies

When monolayers of hemocytes were activated with ei-
ther mannan or PMM, the number of immunoreactive
morula cells and the intensity of their immunoreactivity
were progressively augmented with increasing incubation
times (Figs. 2, 4). The difference in the number of unstimu-
lated and stimulated reactive morula cells was always sig-
nificant (P < 0.001 ). In contrast, no significant changes with
respect to unstimulated hemocytes were observed in the
number, morphology, or stain intensity of positive phago-
cytes for all the incubation times. In each preparation, more
than 95% of hemocytes were viable. Unstimulated and
stimulated hemocytes always showed negative results with
either non-immune sera or absorbed antibodies. The
anti-BA antibody, as previously reported (Ballarin et al.,
2000), only recognized amebocytic phagocytes and no
morula cells (Fig. 3), supporting the specificity of the anti-
cytokine antibodies used.

Table 1

Immunoreactivity of unstimulated Botryllus hein(>cvtes to antibodies
raised to human cytokines

Antibodies'1

Cell type

Anti-iL-1-a

Anti-TNF-a

Phagocytes'"
Morula cells

0.4 ± 0.3
1.1 ±0.9

0.9 ± 0.4
4.5 ± 1.2

a Values are percentage of total hemocytes plus or minus the standard
deviation.

h Phagocytes include hyaline amoebocytes and macrophai -li;

L. BALLARIN ET AL.

anti-BA

l&r

anti-cytokine

Figure 3. Immunocytochemistry on Botryllus schlosseri hemocytcs with anti-BA (a, h), and anti-cytokine
(c-e) antibodies, (a) Positive HA; (hi negative morula cells; (c) unlabeled. unstimulated HA; (d) stimulated HA
positive for IL-l-a; (e) stimulated MLC positive for TNF-a. Bar = 1? /xm.

Discussion

In the present work, we demonstrate that molecules rec-
ognized by antibodies raised to human IL-l-o and TNF-o
are present in immunocytes of the compound ascidian Bot-
tyllus schlosseri. After stimulation, only morula cells,
among all hemocytes, show a marked and significant in-
crease in immunoreactivity. The increase in the number of
immunoreactive cells depends on the length of the time of
hemocyte incubation with the stimulating agents. In con-
trast, among unstimulated hemocytes, only some morula
cells and a few phagocytes are immunoreactive. Therefore,
although the ligands recognized by the antibodies used are
unknown and notwithstanding that serological cross-reac-
tivity is not sufficient proof of evolutionary homology be-
tween those ligands and vertebrate cytokines. still our data
indicate that the morula cells have an important immuno-
modulatory role in ascidian blood.

We hypothesize that morula cells are the main source of
cytokine-like molecules in Botryllus hemolymph, which can
better explain their abundance in the circulation. Indeed,
these cells are able to encapsulate foreign bodies (Anderson,
1971; Wright, 1981; De Leo el al, 1996) and are involved
in clotting after blood vessel damage (Vallee, reported by
Wright, 19X1). In many ascidian species, they can also
induce cytotoxicity after recognition of foreign molecules or
cells (Parrinello, 1996; Cammarata el al.. 1997: Ballarin ct
uL. 1998). All these events can be modulated by cytokine-

like molecules produced by activated cells. In agreement
with this view, TNF-a-like molecules are involved in insect
encapsulation (Franchini et ui, 1996), and IL-1-like mole-
cules have been shown to stimulate echinoderm coelomo-
cyte aggregation, which occurs in encapsulation (Beck and
Habicht, 1991 ). Moreover, in vertebrates, both TNF-a and
IL-l-n stimulate immune and inflammatory responses, and
TNF-a is required for blood coagulation (Abbas et al..
1991).

The induction of cytokine-like molecules in hemocytes
after stimulation has already been reported in bivalve mol-
luscs and insects: in all these cases, phagocytes are the
immunoreactive cells (Hughes et al.. 1990; Franchini et al.,
1996). Analogously, in vertebrates, mononuclear phago-
cytes are the main source of both IL-l-a and TNF-a (Abbas
ct al.. 1991 ). Nevertheless, the situation in Botryllus appears
peculiar in that positivity to anti-cytokine antibodies is
absent from the majority of phagocytes without significant
differences in its distribution between unstimulated and
stimulated cells.

Although morula cells have no phagocytic activity, they
are reported to promote phagocytosis by ascidian phago-
cytes (Smith and Peddie, 1992). Thus, the stimulatory effect
on phagocytes and the enhancement of phagocytosis by
morula cell lysates (Smith and Peddie, 1992) may easily be
explained by the immunomodulatory role of the cytokines
they produce. This idea is strongly supported by the obser-

CYTOK1NE-L1K.E MOLECULES IN BOTRYLLUS

35-

25-

15 •

time (min)

Figure 4. Morula cells positive to anti-IL-1-a and anti-TNF-a, ex-
pressed as percentage of total hemocytes. after stimulation with either
mannan at 5 mg/ml (circles) or PMM at 20 nM (triangles) for 5, 15, 30, and
60 min. *P < 0.001 vs. control (unstimulated hemocytes, t = 0).

vation that the time-dependent increase of immunoreactive
morula cells closely resembles the time-dependent increase
in the frequency of phagocytizing hemocytes in in vitro
assays (Ballarin et al., 1997). The opsonic role of tunicate
IL-1-like molecules reported by Kelly et at. (1993) is in
agreement with this view.

Acknowledgments

The authors wish to thank Mr. M. Del Favero, Mr. R.
Mazzaro, and Mr. C. Friso for their technical assistance.
This work was supported by a grant from the University of
Padova to one of us (L.B.).

Literature Cited

Abbas, A. K., A. H. Lichtman, and J. S. Pober. 1991. Cellular and
Molecular Immunology. W. B. Saunders, Philadelphia.

Anderson. R. S. 1971. Cellular responses to foreign bodies in the tuni-
cate Molgula manhatlensis (DeKay). Biol Bull. 141: 91-98.

Andrew, \V. 1961. Phase microscope studies of living blood-cells of the
tunicates under normal and experimental conditions, with a description
of a new type of motile cell appendage. Q. J. Microsc. Sci. 102:
89-105.

Ballarin. I... F. Cima, and A. Sabbadin. 1994. Phagocytosis in the
colonial ascidian Botryllus schlosseri. Dev. Comp. Immunol. 18: 467-
481.

Ballarin. L., F. Cima. and A. Sabbadin. 1995. Morula cells and histo-
compatibility in the colonial ascidian Botryllus schlosseri. Zoo/. Sci.
12: 757-764.

Ballarin, L., F. Cima, and A. Sabhadin. 1997. Calcium homeostasis
and yeast phagocytosis in the compound ascidian Botryllus schlosseri.
Comp. Biochem. Physiol. 118A: 153-158.

Ballarin, L., F. Cima, and A. Sabbadin. 1998. Phenoloxidase and
cytotoxicity in the compound ascidian Botryllus schlosseri. Dev. Comp.
Immunol. 22: 479-492.

Ballarin, L.. C. Tonello, L. Guidolin, and A. Sabbadin. 1999. Purifi-
cation and characterization of a humoral opsonin. with specificity for
D-galactose, in the colonial ascidian Botryllus schlosseri. Comp. Bio-
chem. Phy.iiol. 123B: 115-123.

Ballarin, L., C. Tonello, and A. Sabbadin. 2000. Humoral opsonin
from the colonial ascidian Botryllus schlosseri as a member of the
galectin family. Mar. Biol. 136: 823-827.

Beck. G., and G. S. Habicht. 1991. Primitive cytokines: harbingers of
vertebrate defense. Immunol. Toda\ 12: 180-183.

Beck, G., G. R. Vasta, J. J. Marchalonis, and G. S. Habicht. 1989.
Characterization of interleukin- 1 activity in tunicates. Comp. Biochem.
Physiol. 92B: 93-98.

Beck, G., R. F. O'Brien, G. S. Habicht, D. L. Stillman, E. L. Cooper,
and D. A. Raftos. 1993. Invertebrate cytokines III: Invertebrate
interleukin- 1 -like molecules stimulate phagocytosis by tunicate and
echinoderm cells. Cell. Immunol. 146: 284-299.

Cammarata. M., V. Arizza, N. Parrinello, G. Candore, and C. Caruso.
1997. Phenoloxidase dependent cytotoxic mechanism in ascidian
(Sryela plicata) hemocytes against erythrocyte and K562 tumor cells.
Eur. J. Cell. Biol. 74: 302-307.

Chaga, O. Y. 1980. Ortho-diphenoloxidase system of ascidians. Tsi-
tologiya 22: 619-625 (in Russian).

Cima, F., L. Ballarin, and A. Sabbadin. 1996. New data on phagocytes
and phagocytosis in the compound ascidian Botryllus schlosseri (Tu-
nicata: Ascidiacea). Ital. J. Zoo/. 63: 357-364.

Coombe, D. R., P. L. Ey, and C. R. Jenkin. 1984. Particle recognition
by haemocytes from the colonial ascidian Botrylloides leachi: evidence
that the B. leachi HA-2 agglutinin is opsonic. J. Comp. Phvsiol. 154:
509-521.

Dan-Sohkawa, M., M. Morimoto, and H. Kaneko. 1995. In vitro
reactions of coelomocytes against sheep red blood cells in the solitary
ascidian Halocynthia roret-i. Zoo/. Sci. 12: 411 — 117.

De Leo, G.. N. Parrinello, D. Parrinello. G. Cassara, and M. A. Di
Bella. 1996. Encapsulation response of dona intestinalis (Ascidia-
cea) to intratunical erythrocyte injection. I. The inner capsular archi-
tecture. J. Invertebr. Pcithol. 67: 205-212.

Endean, R. 1955a. Studies of the blood and tests of some Australian
ascidians. I. The blood of Pyitrci stolnnifera (Heller). Aust. J. Mar.
Freshwater Res. 6: 35-59.

Endean, R. 1955b. Studies of the blood and tests of some Australian
ascidians. III. The formation of the test of Pyura stolonifera (Hellerl.
Aust. J. Mar. Freshwater Res. 6: 157-164.

Franchini, A.. J. A. Mivan. and E. Ottaviani. 1996. Induction of
ACTH- and TNF-a-like molecules in the hemocytes of Calliphora
vomitoria (Insecta. Diptera). Tissue Cell 28: 587-592.

Fuke, M., and M. Fukumoto. 1993. Correlative fine structural, behav-
ioral, and histochemical analysis of ascidian blood cells. Ada Zoo/.
IStockh.) 74: 61-71.

Gorman. A., J. McCarthy, D. Finucane. \V. Reville, and T. Cotter.
1996. Morphological assessment of apoptosis. Pp. 1-20 in Tech-
niques in Apoptosis: A User's Guide. T. G. Cotter and S. J. Martin, eds.
Portland Press, London.

Hsu, S. M., L. Raine, and H. Fanger. 1981. Use of avidin-biotin-
peroxidase complex (ABC) immunoperoxidase techniques: a compar-
ison between ABC and unlabeled antibody (PAP) procedures. J. His-
tochem. C\tochem. 29: 577-580.

L. BALLARIN ET AL.

Hughes, T. K., E. M. Smith, R. Chin, P. Cadet, J. Sinisterra, M. K.
Leung, M. A. Shipp, B. Scharrer, and G. B. Stefano. 1990.

Interactions of immunoreactive monokines (interleukine-1 and tumor
necrosis factor) in the bivalve mollusc Mytilu.i ednlis. Proc. Natl. Acail.
Sci. USA 87: 4426-4429.

Kelly, K. L., E. L. Cooper, and D. A. Raftos. 1992. Purification and
characterization of a humoral opsonin from the solitary urochordate
Styela clava. Comp. Biochem. Physiol. 103B: 749-753.

Kelly, K. L., E. L. Cooper, and D. A. Raftos. 1993. Cytokine-like
activity of a humoral opsonin from the solitary urochordate Sryela
clava. Zoo/. Sci. 10: 57-64.

Kustin, K., D. S. Levine, G. C. McLeod, and W. A. Curby. 1976. The
blood of Ascidia nigra: blood cell frequency distribution, morphology,
and the distribution and valence of vanadium in living blood cells. Bio/.
Bull. 150: 426-441.

Ohtake, S.-I., T. Abe, F. Shishikura, and K. Tanaka. 1994. The
phagocytes in hemolymph of Halncynthia roretzi and their phagocytic
activity. Zoo/. Sri. 11: 681-691.

Ottaviani, E., F. Petraglia, G. IVlontagnani, A. Cossarizza, D. Monti,
and C. Franceschi. 1990. Presence of ACTH and /3-endorphin im-
munoreactive molecules in the freshwater snail Planorbarius corneus
(L.) (Gastropoda. Pulmonatal and their possible role in phagocytosis.
RcKiil. Pent. 27: 1-9.

Ottaviani, E., E. Caselgrandi, and C. Franceschi. 1995a. Cytokines
and evolution: in vitro effects of IL-la. IL-1/3. TNF-a and TNF-/3 on
an ancestral type of stress response. Biochem. Biophys. Res. Commun.
207: 288-292.

Ottaviani, E., A. Franchini, S. Cassanelli, and S. Genedani. 19951).
Cytokines and invertebrate immune responses. Biol. Cell. 85: 87-91.

Ottaviani, E., A. Franchini, D. Kletsas, and C. Franceschi. 1996.

Presence and role of Cytokines and growth factors in invertebrates, /fa/.

J. Zoo/. 63: 317-323.
Parrinello, N. 1996. Cytotoxic activity of tunicate hemocytes. Pp. 190-

217 in Invertebrate Immunology. B. Rinkevich and W. E. G. Muller,

eds. Springer-Verlag. Berlin.
Raftos, D. A., E. L. Cooper, G. S. Habicht, and G. Beck. 1991.

Invertebrate cytokines: tunicate cell proliferation stimulated by an

interleukin l-like molecule. Proc. Natl. Acad. Sci. USA 88: 9518-

9522.
Raftos, D. A., E. L. Cooper, D. L. Stillman, G. S. Habicht, and G. Beck.

1992. Invertebrate cytokines II: release of interleukin-1-like mole-
cules from tunicate hemocytes stimulated with zymosan. Lymphakine

Cytokine Rex. 11: 235-240.
Raftos, D. A., D. L. Stillman, and E. L. Cooper. 1998. Chemotactic

responses of tunicate ( Urochordata. Ascidiacea) hemocytes in vi'fro.

J. Invertebr. Patliol. 72: 44-49.
Smith, M. J. 1970. The blood cells and tunic of the ascidian Halocynthia

auranlium (Pallas). I. Hematology, tunic morphology, and partition of

cells between blood and tunic. Biol. Bull. 138: 354-378.
Smith, V. J., and C. M. Peddie. 1992. Cell cooperation during host

defense in the solitary tunicate Ciona intestinalis (L). Biol. Bull. 183:

211-219.
Wolfe, S. L. 1993. Molecular and Cellular Biology. Wadsworth, Bel-

mont, CA.
Wright, R. K. 1981. Urochordates. Pp. 565-626 in Invertebrate Blood

Cells. Vol. 2. N. A. Ratcliffe and A. F. Rowley, eds. Academic Press.

London.
Zaniolo, G. 1981. Histology of the ascidian Botryllus schlosseri tunic: in

particular, the test cells. Boll. Zoo/. 48: 169-178.

Reference: Bio/. Bull. 201: 65-75. (August 2001)

Molecular Evidence that Sclerolinum brattstromi Is

Closely Related to Vestimentiferans, not to Frenulate

Pogonophorans (Siboglinidae, Annelida)

KENNETH M. HALANYCH1 *. ROBERT A. FELDMAN2, AND ROBERT C. VRIJENHOEK1

1 Biology Department MS 33, Woods Hole Oceanographic Institution. Woods Hole, Massachusetts

02543; : Molecular Dynamics, Inc.. part of Amersham Pharmacia Biotech, 928 East Arques Ave.,

Sunnyvale, California 94086-4250; and 3 Monterey Bay Aquarium Research Institute.

7700 Sandholdt Road. Moss Landing, California 95039

Abstract. Siboglinids. previously referred to as pogono-
phorans, have typically been divided into two groups, frenu-
lates and vestimentiferans. Adults of these marine proto-
stome worms lack a functional gut and harbor
endosymbiotic bacteria. Frenulates usually live in deep,
sedimented reducing environments, and vestimentiferans
inhabit hydrothermal vents and sulfide-rich hydrocarbon
seeps. Taxonomic literature has often treated frenulates and
vestimentiferans as sister taxa. Sclerolinum has traditionally
been thought to be a basal siboglinid that was originally
regarded as a frenulate and later as a third lineage of
siboglinids. Monilifera. Evidence from the 18S nuclear
rDNA gene and the 16S mitochondria! rDNA gene pre-
sented here shows that Sclerolinum is the sister clade to
vestimentiferans although it lacks the characteristic mor-
phology (i.e.. a vestimentum). The rDNA data confirm the
contention that Sclerolinum is different from frenulates, and
further supports the idea that siboglinid evolution has been
driven by a trend toward increased habitat specialization.
The evidence now available indicates that vestimentiferans
lack the molecular diversity expected of a group that has
been argued to have Silurian or possibly Cambrian origins.

Introduction

Siboglinids were formerly called pogonophorans and in-
clude two groups of marine protostomes, frenulates and
vestimentiferans, that are commonly referred to as beard-

Received 22 November 2000: accepted 1 1 April 2001.
* To whom correspondence should be addressed. E-mail: khalanvch@
whoi.edu

worms and tubeworms, respectively. Both groups lack a
functional gut as adults and rely on endosymbiotic bacteria
for nutrition. They have a closed circulatory system and
possess a metamerized tail region called the opisthosoma.
Vestimentiferans are distinguished from frenulates by the
presence of a vestimentum, a winged region near the ante-
rior of the organism. Both taxa occur in reducing environ-
ments and typically are found at depths below several
hundred meters. Due to the limited availability of samples
and the difficulty of retrieving live specimens, several as-
pects of their biology (e.g.. reproduction, physiology) are
still poorly understood. Vestimentiferans, in general, have
been better studied than frenulates because they are key-
stone species in eastern Pacific hydrothermal vent habitats
and in Pacific and Caribbean seeps.

The taxonomic literature concerning frenulate and vesti-
mentiferan siboglinids has a colorful and confusing history.
One taxonomic scheme recognizes frenulates (aka pogono-
phorans sensu stricto) and vestimentiferans as distinct phyla
(Jones, 1985). Alternatively, vestimentiferans have also
been recognized as a class within the phylum Pogonophora
(Jones, 1981; Ivanov, 1994). Others place frenulates and
vestimentiferans within the phylum Annelida (Land and
N0rrevang, 1977; Kojima et al.. 1993; Bartolomaeus. 1995;
McHugh, 1997; Rouse and Fauchald. 1997; also see South-
ward, 1988). The latter hypothesis has been supported by
recent morphological (Rouse and Fauchald. 1995. 1997).
embryological (Young et al., 1996; Southward, 1999), and
molecular analyses (Kojima et al., 1993: McHugh, 1997;
Blacker*-//., 1997; Kojima, 1998; Halanych et al.. 1998). To
further complicate matters, a ranked classification scheme
has produced different names for the same clade of organ-

K. M. HALANYCH. R. A. FELDMAN. AND R. C. VRUENHOEK

isms. Vestimentiferans have been called Vestimentifera
(Jones, 1981). Obturata (Jones, 1981; Southward, 1988;
Southward and Galkin, 1997). and Afrenulata (Webb.
1969). Frenulates have been called Pogonophora (Jones.
1985), Frenulata (Webb, 1969), Perviata (Southward,
1988), and originally Siboglinidae (Caullery, 1914).

Hereafter we apply the following nomenclature: (1) Ves-
timentifera are equated with Obturata and Afrenulata; (2)
Frenulata are equated with Perviata and Pogonophora
(sensu Jones, 1985); (3) Monilifera is a third monogeneric
clade that includes Sclerolinum; and (4) Siboglinidae refers
to the clade that includes Vestimentifera, Frenulata, and
Monilifera. We recognize that the term "Pogonophora" is
more commonly used and that rules of priority for nomen-
clature do not apply to higher taxa. However, we have opted
to use the term "Siboglinidae" throughout this manuscript to
emphasize that this group of organisms represents derived
annelids (McHugh, 1997; Rouse and Fauchald. 1997). We
restrict the term "pogonophoran" to common usage.

Even among siboglinids, there has been one group.
Sclerolinum, that has been particularly problematic in terms
of phylogenetic position. Unlike most frenulates that live in
the mud, Sclerolinum species can live on decaying organic
material like wood or rope made from natural fibers (Webb,
1964a; Southward, 1972). This taxon was originally con-
sidered a member of the frenulate family Polybrachiidue
(Southward. 1961 ). but Webb ( 1964b), mainly citing differ-
ences in the postannular region, argued that Sclerolinum
could not be ascribed to either of the two orders (Theca-
nephria and Athecanephria) of siboglinids recognized at the
time (vestimentiferans had not been discovered yet). He
erected a new family, Sclerolinidae, that he states should
"have order rank." Ivanov ( 1991 ) more formally recognized
the unique nature of Sclerolinum, and in 1 994 he proposed
that Frenulata (= Perviata), Monilifera (= Sclerolinidae),
and the Vestimentifera be regarded as three taxa with equal
rank (i.e.. classes within the phylum Pogonophora). Addi-
tionally, Ivanov ( 1994) further suggested that Monilifera are
allied to the Vestimentifera on the basis of the common
absence of several characters (e.g.. spermatophores, teloso-
mal diaphragm, metasoma preannular and postannular re-
gions) relative to the Frenulata. Southward (1999) sug-
gested that Monilifera might be similar to the ancestral
siboglinid form, thus predicting that it should occupy a basal
position in siboglinid phylogeny. Distinguishing between
these hypotheses on the placement of Sclerolinum will
allow us to test the notion of Black el al. (1997) that habitat
preference or specificity may be an important factor in
siboglinid evolution. If Black et al. are correct, Sclerolinum
is expected to occupy a position between frenulates and
vestimentiferans (which may be consistent with Ivanov's
ideas), and not a position basal to the frenulate-vestimen-
tiferan clade.

To date, molecular studies that include siboglinids have
either focused on vestimentiferans (Williams et al.. 1993;
Black et al.. 1997; Kojima et al., 1997; Halanych et al.,
1998) or have addressed siboglinid origins (Winnepen-
ninckx et al., 1995a; Kojima et al., 1993; Kojima, 1998;
McHugh, 1997). Most studies have included only one frenu-
late representative. Although Black et al. (1997) included
two "frenulate" siboglinids, one of these, the Loihi worm,
was undescribed. Additionally, several 18S sequences were
reported in a symposium contribution (Halanych et al.,
1998) for which page limitations did not permit detailed
analyses or explanation. Herein we extend these previous
analyses by increasing the sampling of frenulates. including
Sclerolinum, and using novel 18S rDNA and 16S rDNA
data. The present findings support the notion that habitat
requirements have been important in siboglinid evolution.
Additionally, frenulates are sister to a Sclerolinum-vesti-
mentiferan clade, the latter of which showed limited diver-
sity suggestive of a recent radiation within the clade.

Materials and Methods

Taxa employed

Table 1 lists the species analyzed and GenBank accession
numbers for the rDNA sequences used in this study. The
frenulate and vestimentiferan operational taxonomic units
(OTUs) included in this study represent all of the currently
recognized genera available to the authors. The addition of
closely related species within a genus would have increased
OTUs without increasing the phylogenetic signal for the
issues under examination and were therefore excluded. For
example, there are no nucleotide differences observed in the
18S rDNA of Escarpia spicata (Guaymas Basin) and E.
laminata (Florida Escarpment). Limiting the number of
OTUs also reduced computation time, allowing for more
thorough analyses. Unless otherwise noted, collection local-
ities correspond to those given in Black et al. (1997).
Siboglinum ekmani, S. fiordicum. and Sclerolinum
brattstromi were collected near Bergen, Norway, and iden-
tified by Eve Southward. Marine Biological Association of
the United Kingdom. Identification of the frenulates Spiro-
brachia and Polybrachia were made by Eve Southward on
the basis of animal and tube morphology. Both specimens
were collected by TVGrab from the Aleutian Trench
(57°27.394'N, 148°00.013'W) at a depth of 4890 m on the
German research vessel Sonne.

The non-siboglinid annelid OTUs for the 18S data were
chosen to represent a diversity of lineages for which se-
quences were available. The arthropod (Anemia) sequence
was designated as the most distant outgroup for rooting
purposes. Based on both morphology (e.g., Eernisse et al.,
1992) and molecular studies (e.g.. Halanych et al.. 1995;
Winnepenninckx et al., 1995a; Aguinaldo et al., 1997;
Eernisse, 1997), arthropods are clearly outside of the proto-

Timi used in rDNA anal\ses

SIBOGLINID EVOLUTIONARY HISTORY
TABLE 1

Organism

GenBank Accession'1

GenBank Accession11

18S rDNA

I6S rDNA

Organism

18S rDNA

16S rDNA

Pogonophora

Frenulata

Galalheiiliiuiin brachiosum AF168738

Polybrachia sp. AF 168739

Siboglinum fiordicinn GB X79876h

Siboglinum fiordicwn AF3 15060

Siboglinum ekmani AF3 15062

Spirobriit-liiti sp. AF 168740

Vestimentifera

Escarpia spicata AF 168 741

Escarpiid n. sp.

Lumellihriichia barhami AF168742

Oiisisia alvinae AF168743

Ridgeia piscesae AF 168 744

Ridgeia piscesae GB X79877h

Chaetopterida

Chaetopterus variopedatus U67324C

AF3 15040 Hirudinea

AF3 15037 Haemopis sanguisuga X91401J

Hinulu nit'dk-iniilix AF3 15058

AF3 15039 Oligochaete

AF3 15038 Enchytraeus sp. Z83750d

AF315036 Phyllodocida

Glycera americiina U19519e

AF3 15041 Polynoidea

AF3 1 5053 Lepidonotopodium fimbriutum AF3 1 5056

AF3 15043 Branchipolynoe symmytilida AF3 15055

AF3 15044 Sabellida

AF315045 Sabella piminniti U67144'

AF3 15047 Tubificidae

AF3 15052 Tubifex sp. AF3 15057

AF3 15048 Echiura

AF3 15051 Ochetostoma erythrogrammon X79875h

AF3 15054 Urechis sp. AF3 15059

Sipuncula

Riftia pachyptila

AF 168745

AF3 15049

Phascolosoma gnnuilaiiini

X79874b

AF3 15050

Nemertea

Tevnia jerichonana

AF168746

AF3 15042

Linens sp.

X79878"

Monilifera

Mollusc

Sclerolinum brattstromi

AF3 15061

AF3 15046

Scutopux ventrolineatus

X91977'

Annelida

Priapulida

Alvinellidae

Priiipuliix caudatus

X802341

Puralvinella pabniformis

AF 168747

Arthropod

Anemia salina

X01723h

a Unless otherwise noted, sequences were obtained in this study.

b Sequence from Winnepenninckx et al. ( 1995a).

c Sequence from Nadot and Grant (unpublished).

J Sequence from Kim el al. ( 1996).

° Sequence from Halanych et al. ( 1995).

' Sequence from Winnepenninckx et al. ( 1996).

g Sequence from Winnepenninckx et al. ( 1995b).

h Sequence from Nelles et al. ( 1984).

stome worm radiation. Because siboglinids are not closely
related to molluscs and because of rate heterogeneity prob-
lems within the Mollusca, only a single representative (the
aplacophoran Scutopus) was used. Due to alignment limi-
tations, outgroups employed in the 16S analyses — a leech,
an oligochaete, two polynoid polychaetes, and an echiu-
rid — were more limited (see Table 1). Because different
investigators collected the data at different times, there was
not a 1:1 correspondence in OTUs between data sets. We
felt it more important to present all the relevant data rather
than trim taxa from the data sets. The aligned data sets are
available at the journal's Supplement's page (http://
www.mbl.edu/BiologicalBulletin/VIDEO/BB.video.html)
and at TREEBASE (http://phylogeny.harvard.edu/treebase).

Data collection

Total genomic DNA was extracted using a modified
hexadecyl-trimethyl-ammonium bromide (CTAB) protocol
(Doyle and Dickson. 1987). The entire 18S nuclear rDNA
gene was amplified via PCR (polymerase chain reaction),
using the universal metazoan oligonucleotide primers 18e
and 18P (Halanych et al.. 1998). A region of the 16S
mitochondria! rDNA was amplified using 16Sar-5' and
16Sbr-3' primers (Palumbi, 1996). Each 50 /xl reaction
consisted of about 50 ng of template DNA, 0.5 /u,A/ of each
primer, 2.5 mM MgCl2, 200 pM dNTPs, 5 ju.1 of manufac-
turer's 10X reaction buffer, and 1.5 U Tag polymerase
(Promega Inc.. Wisconsin). Cycling profiles were as fol-

K. M. HALANYCH. R. A. FELDMAN, AND R. C. VRIJENHOEK

lows: 18S — initial denaturation at 95 °C for 3 min, 35
cycles of amplification (denaturation at 95 °C for 1 min,
annealing at 50 °C for 2 min, extension at 72 °C for 2 min
30 s), and a final extension at 72 °C for 5 min: 16S — initial
denaturation at 94 °C for 2 min, 40 cycles of amplification
(denaturation at 94 °C for 30 s, annealing at 46 °C for 30 s,
extension at 72 °C for 1 min), and a final extension at 72 °C
for 7 min. PCR products were purified using the QIAEX II
gel extraction kit (Qiagen Inc., California). Approximately
60 ng of purified PCR product was used in sequencing
reactions according to the manufacturer's instructions (FS
Dye Termination Mix or Big Dye, Applied Biosystems Inc.,
California). The reaction profile was 25 repetitions of de-
naturation at 94 °C for 30 s, annealing at 50 °C for 15 s, and
extension at 64 °C for 4 min. Dye-labeled fragments were
separated by electrophoresis on a Perkin Elmer ABI 373A
or 377 DNA sequencer. Both strands of the PCR product
were sequenced. In addition to the PCR primers, the oligo-
nucleotide primers used for sequencing are given in
Halanych el al. (1998) or Hillis and Dixon (1991). The
sequences were assembled and verified using the AutoAs-
sembler and Sequence Navigator programs (Applied Bio-
systems Inc., California). The terminal primer regions were
not included in the sequences submitted to GenBank or in
the phylogenetic analyses.

Phylogenelic analyses

Sequence alignment was produced with a Clustal W
program (Thompson el al., 1994) and subsequently cor-
rected by hand using the protostome secondary structure
models available through the Ribosomal Database project
(http://rdp.cme.msu.edu/html/). Regions that could not be
unambiguously aligned (e.g., divergent loop domains) were
excluded from analyses. Tree reconstructions were imple-
mented with the PAUP* 4.0b4b2 program (Swofford,
2000), and MacClade 3.06 (Maddison and Maddison, 1992)
was used for character and tree analyses. Neighbor-joining
(NJ), parsimony, and maximum likelihood (ML) analyses
were performed and yielded similar results. In the interest of
brevity, results and discussion will focus on ML analyses.

NJ trees were reconstructed under Jukes-Cantor, Kimura-
2-parameter, Tamura-Nei. general-time-reversible, and log/
det models. All except log/det were examined under equal
rates of among-site rate variation using the empirically
derived gamma shape parameter, a, of 0.3 (see Swofford el
al.. 1996, for summary of different assumptions used in
these models). A Kishino-Hasegawa ( 1989) likelihood eval-
uation of the resulting topologies revealed no significant
differences between models for either the 16S or the 18S
data. Kishino-Hasegawa evaluations estimated a six-substi-
tution-type rate matrix for which nucleotide base frequen-
cies were set to empirical values and a was estimated. NJ
bootstraps consisted of a log/det correction (model was

arbitrarily chosen) with 1000 iterations. Parsimony analyses
consisted of heuristic searches with 100 random sequence
additions and tree-bisection-reconnection (TBR) branch
swapping. Transitions (Ti) and transversions (Tv) were
given equal weighting. ML evaluation of parsimony topol-
ogies was the same as for NJ topologies. One thousand
iterations were used for parsimony bootstrap analyses.
When using likelihood to search for the "best" tree (as
opposed to evaluating given trees), computation time was
limiting. Therefore, we used a nucleotide model with two
substitution types where the Ti/Tv ratio was set to the value
estimated for the best parsimony tree (empirical base fre-
quencies were used). ML searches were heuristic with 10
random sequence-addition replicates. ML bootstraps em-
ployed the "Faststep" option with 100 iterations.

Results

The 18S rDNA data set consisted of 26 OTUs and 1935
nucleotide positions. Of the 1614 nucleotide positions that
could be unambiguously aligned. 34.6% (559 positions)
were variable and 18.7% (303 positions) were parsimony
informative. Figure 1 shows the single best likelihood tree
(Ln likelihood = -8260.55148) recovered. All search
methods in all analyses found a monophyletic siboglinid
clade (bootstrap support was >98% for all methods). Res-
olution within the vestimentiferan clade, as well as between
annelid groups, was poor, however. The moniliferan
Sclerolinum brallslromi falls out with the vestimentiferan
taxa in all analyses (bootstrap S 98%). The remaining
trenulates form a distinct sister-clade to the Sclerolinwn-
vestimentiferan clade with >99%> bootstrap support.

Resolution among annelid taxa and within the vestimen-
tiferans was poor due to the lack of phylogenetic signal.
Because this paper does not focus on the annelid radiation,
we did not try to enhance resolution among all annelid taxa.
However, we did attempt to boost the signal within the
vestimentiferan clade by employing a less inclusive taxo-
nomic alignment. For metazoan 18S sequences, inclusion of
broader taxonomic diversity can often create larger regions
of ambiguous alignment that should not be included in
analyses, due to poor assumptions about positional homol-
ogy. Thus by reducing the taxonomic breadth examined, the
phylogenetic signal can potentially be increased by a "bet-
ter" alignment (Halanych, 1998). Unfortunately, even when
just the siboglinids were aligned, little genetic diversity was
observed, and the vestimentiferan taxa were still poorly
resolved (not shown). The exception was Lamellibrachia
harhami, which was consistently placed as the most basal
vestimentiferan. Table 2 shows the logdet/paralinear dis-
tances (below diagonal) and absolute distances (above di-
agonal) for this less-inclusive, siboglinid-only alignment (in
which most divergent domains could be unambiguously
aligned). Even though the distance values for the siboglinid-

SIBOGLINID EVOLUTIONARY HISTORY

100
100

100

ipirooracnia
I Polybrachia

_l g1 Galathealmum
^— Siboglinumekmani
100 r Siboghnum fiordicum GB

Siboglinum fiordicum

Escarpia
Ridgeia
RidgeiaGB
Oasisia
— Riftia
Tevnia

Lamellibrachia
Sderolinum

^— Enchytraeus -Oligochaete
^^^^^^— ^^— Haemopis -Leech

Moniliferan

Sabella - Polychaete
Paralvinella - Polychaete

Phascolosoma
-Sipunculid

Ochetostoma - Echiund

Chaetopterus - Polychaete
Glycera Polychaete
— ^^— — Lineus Nemertean

Scutopus Mollusk

Artemia - Arthropod

Priapulus- Priapulid

0-01 substitutions/site

Figure 1. Results of 18S rDNA phylogenetic analyses. The single best likelihood tree (Ln likelihood =
— 8260.551481 found. Analysis details are given in the text. Maximum likelihood bootstrap values of >50% are
given in bold. Parsimony (italicl and neighbor joining (underlined) values are also given for the major nodes of
interest (values for other nodes were omitted in the interest of space). Branch lengths are drawn proportional to
the inferred amount of change along the branch (scale shown).

only alignment are only slightly greater than the full align-
ment values, the greatest distance within vestimentiferans
was only 0.02 (with a maximum of 25 nucleotide differ-
ences), revealing that there was very little 18S genetic
diversity within this group.

The 16S rDNA data set consisted of 24 OTUs, each with
497 nucleotide positions. Of the 465 nucleotide positions
that could be unambiguously aligned, 60.4% (281 positions)
were variable and 47.7% (222 positions) were parsimony
informative. The reconstructed topology (Ln likelihood =
-3967.21062). Figure 2, was qualitatively similar to the
18S topology. Siboglinids are divided into two major lin-

eages: vestimentiferans plus the moniliferan Sderolinum
brattstromi (bootstrap support 83% for ML and 100% for
NJ and parsimony) and a frenulate sister-clade (bootstrap
support >94% in all analyses). Again. 51. brattstromi was
basal to the vestimentiferans. In a departure from the 18S
analyses, Riftia pachyptila, not Lamellibrachia barhami.
often fell out as the most basal vestimentiferan. However,
this was never supported by >54% bootstrap support; ML
analyses that excluded the non-siboglinid outgroups re-
vealed that the base of the Vestimentifera was poorly re-
solved with 16S data. A comparison of genetic divergence
values (Table 3) indicates that there was limited genetic

K. M. HALANYCH. R. A. FELDMAN, AND R. C. VRIJENHOEK

TABLE 2

Paim'ise distances for the siboglinid-only IKS rDNA data set: absolute distances above diagonal and log/det distances below diagonal

1 Spirobrachia

109

113

132

131

124

122

125

124

131

125

121

120

2 Polybrachia

0.07

—

104

138

137

139

140

147

142

140

136

3 Galathealinum

0.07

0.01

—

106

142

141

142

143

150

145

143

139

4 Siboglinum ekniani

0.05

0.06

—

116

117

113

1 III

113

110

121

112

107

112

5 Siboglinum fiordicum

0.08

0.09

0.07

—

14(1

136

140

143

138

134

139

6 Siboglinum fiordicum GB

0.08

0.09

0.07

0.00

—

143

139

143

146

141

137

140

7 Escarpia

0.08

0.09

0.07

0.09

—

8 Ridgeia

0.08

0.09

0.07

0.09

0.00

—

9 Ridgeia GB

0.08

0.09

0.07

0.09

0.01

0.00

—

10 Oasisia

0.08

0.09

0.07

0.09

0.01

0.00

0.01

—

1 1 Riftia

0.08

0.09

0.07

0.09

0.01

—

12 Tevnia

0.08

0.09

0.07

0.09

0.00

0.01

0.00

0.01

—

1 1

1 3 Lamellibrachia

0.07

0.09

0.07

0.08

0.09

0.01

—

14 Sclerolintiin

0.08

0.09

0.07

0.09

0.02

—

variation within vestimentiferans (<0. 1 1 log/del distance; a
maximum of 47 nucleotide differences).

As for the frenulate clade, neither 18S or 16S supported a
monophyletic Siboglinum: but because only two Siboglinuin
species were examined, additional taxa are needed to verify the
status of this frenulate taxon. Additionally, we performed
Kishino-Hasegawa ( 1989) likelihood evaluation for both genes
to test the monophyly of the frenulate and vestimentiferan-
Sclerolinum clades. To this end, we used the constraints option
in PAUP* 4.0b4b2 to conduct parsimony heuristic searches
(specifics same as above) to find the best trees that were
consistent and inconsistent with the monophyly of these
clades. Both the 16S and the 18S data significantly support the
monophyly of both groups ( 1 8S frenulates — average ML score
supporting monophyly = -8244.69, non-monophyly score =
-8278.135. P value < 0.01; 16S frenulates— monophyly =
-3894.889. non-monophyly = -3927.49, P value < 0.005;
18S vestimentiferan-Sc/ero/znMW! — monophyly = —8244.69,
non-monophyly = -8271.922, P value < 0.05; 16S vestimen-
tiferan-Sclerolinum — monophyly = —3894.889, non-mono-
phyly = -391 1.802, P value < 0.05).

Discussion

The monophyly of siboglinids (aka, Pogonophora sensu
hit 11) is supported by morphological (Southward, 1988,
1993; Rouse and Fauchald, 1995; Rouse. 2001). embryo-
logical (Southward, 1999), and molecular (Winnepenninckx
et ui. 1995a; Black el ai. 1997; McHugh. 1997; Halanych
c/ ai. 1998. this study) evidence. Thus, in agreement with
others (Southward, 1988, 1999; Ivanov, 1994; McHugh,
1997), we see no support for the recognition of vestimen-
tiferuns and frenulates as having fundamentally different
body plans (i.e.. "phyla" sensu Jones. 1985). The assertion
made by Webb !l964b) and later by Ivanov (1991. 1994)
that Sclerolinitm was notably different from frenulates is

validated by the present data. Moreover, we found that
Sclerolinitm brtittstroini is closely allied to the vestimenti-
ferans. and does not occupy a position basal to a frenulate-
vestimentiferan clade, confirming Ivanov's (1991; 1994;
Ivanov and Selivanova, 1992) ideas that moniliferans oc-
cupy a position intermediate between vestimentiferans and
frenulates.

Southward (1993) also suggested a possible evolutionary
link between Sclerolinum and vestimentiferans. This con-
tention is confirmed by the present analysis, as well as a
recent morphological cladistic analysis (Rouse, 2001). Us-
ing 44 morphological characters coded for all recognized
siboglinid genera. Rouse found support for the monophyly
of Frenulata. Vestimentifera, and the Sclerolinum-vestimen-
tiferan clade. However, our use of nomenclature differs
from Rouse with regard to the term Monilifera, which he
applies to the Sr/e>w/i'»«w-vestimentiferan clade. Because
this term was originally (Ivanov and Selivanova, 1992)
applied to only Sclerolinitm, and because of the morpho-
logical differences from vestimentiferans. Rouse's use of
the term will inject confusion into the literature. Although
we acknowledge that Monilifera, as defined here, is redun-
dant with the generic name Sclerolinum, several aspects of
siboglinid evolution and taxonomy are in need of additional
study. Thus, we have chosen not to name this clade until
more is understood about siboglinid evolution.

The placement of Sclerolinum was especially interesting
in the context of the evolution of habitat preference. Previ-
ous studies of vestimentiferans (Black et ui. 1997). clams
(Peek et til.. 1997), mussels (Craddock et ai. 1995). and
shrimp (Shank et ai, 1999) reveal that vent-endemic organ-
isms are related to. and possibly derived from, species
associated with hydrocarbon seeps that occur near subduc-
tion zones and continental margins. Furthermore, recent
observations (Feldman et ai. 1998; Baco et ai. 1999; Distel

SIBOGLINID EVOLUTIONARY HISTORY

94 £

i Spirobrachia
! 100 r Polybrachia

cum

*— Galathealmum

700

100

79 1 — Escarpia

1- Escarpiid n. sp.

- Tevnia

• Ridgeia 1 \

- Ridgeia 2 53
Ridgeia 3 /

,
'

— Oasisia

80 r Lamellibrachia 1

*~ Lamellibrachia 2

67 • Lamellibrachia 3

Lamellibrachia 4

83 1

99 I- Riftia 1

;

iXY 1 "• Riftia2

1 'II Monilife*'01"1

100

1 Lepidonotopodium - Polychaete

ifex - Oligochaete

^^— 0.05 substitutions/site

Figure 2. Results of 16S rDNA phylogenetic analyses. The best likelihood tree (Ln likelihood =
-3967.21062) found. Another tree with a Ln likelihood score of -3967.25739 was found in the same search.
The trees differed in relationships within the Ridgeia clade. Analysis details are given in the text. Maximum
likelihood (ML) bootstrap values of >50% are given in bold. Parsimony (italic) and neighbor joining (under-
lines) values are also given for the major nodes of interest (values for other nodes were omitted in the interest
of space). In the ML bootstrap analysis, Lamellibrachia and Sclerolinum formed a clade in 55% of the iterations.
That is not shown above because it is incompatible with the "best" ML tree. Branch lengths are drawn
proportional to the inferred amount of change along the branch (scale shown).

c/ ill.. 2000) reveal that several symbiont-bearing clams,
vestimentiferan tubeworms, and mussels can survive on
rotting organic material, such as wood or a whale carcass.
The moniliferan S. brattstromi and related species (e.g., S.
javanicum, S. minor, and 5. major) are typically found
growing on decaying organic material such as wood or rope
(Webb, 1964a, b; Southward, 1972; Ivanov and Selivanova,
1992). Other members of the genus, (e.g., S. sibogae and S.
magdalenae) lived buried in mud (Southward, 1972). These

habitat preferences suggest that affinity for a mud or silt
habitat was ancestral in siboglinids, allowing us to speculate
that a pattern of evolution from low-oxygen, sedimented
habitats to decaying organic material to hydrocarbon seeps
to hydrothermal vents has occurred within the Sclerolinum-
vestimentiferan clade.

Although neither the 18S nor the 16S data clearly resolve
relationships within the Vestimentifera, the cytochrome c
oxidase subunit I (COD data of Black et al. (1997) show

K. M. HALANYCH, R. A. FELDMAN, AND R. C. VRIJENHOEK

-t

-t r-

I- so CN m -^
. -f -t rr. ~

• P*-, pr, P*-, -t pr, P*-, -

3 OC O OO ON -

t- -t >r, -f -f i/

• r*~i ON sO C

?S |

r-i

- pr, — \G P

- r-i ON m OsC-fsI

3 OC OC vC ^C —

- ON -^- OC C

N 1 d

-t r

-t r-i c: C

3 I/-, _ ON in r*

NO VO

r*~i t

d d

—

'/-, -

HI I =

,___

~ sC oc \o r-j p

Oi
- 0 CS | C

3 IO rl
1 r*-i ^f

3 d O

ON ^
ri r

; -t oc so si

3 Tj- r-j rr, p- -r p*-. r

- p- O oc ri C

NO | d C

3 -f rn

3 d d

~ r

i doc

3 NO f)

3 d d

?! c

- ri ri ri i/
; ri ci O P

d d d c

1 sC OC

3 d d

oc N£

-' £ ^ ^ S
3 d d d c

3 NO 1^

3 d d

-t -r -t

r^ >

3 rj ON rj- C

3 ifi v~.

l^ O

N P- in ON -

- 1- ON "-, NO r-l

1 ° c
- r-l -f | O C

3 O O O C

3 O O

— -

- — — —

— OC 3

j co o in -

- *c *c

-t

- -

: £ ^ Tj ^

; gj § S S S 3 :

- -t doc

SSSS

3 d d d c

3 r*-< oo -T o

3 d d

3 vO m
I r*"i t

- =

- — r*-, ri p*

, co ^ en ¥ en rr,

ri rr, m ON C

3 rn Cfv T C

3 10 "^

ON -

r ON OC r*-j r

i c «-*•< v-- r- <r-, oc

d d d d c

3 d d d c

3 d d

FN ON O O O -

- OO OO fN C

3 ^ g

^ -f

-t "

: :2 H L <

? ^ r=, ^ - £ =

:• o o o o c

3 O O O C

3 O O

•T o

D OO OC P- O C

3 r-- oo o -

•• >s~> r\

•< 5

— C

; -t \c »n -^

; — — — r*

t- p- p- rj in O C
-, r i —

i o d d d c

3 o' d d c

3 O O

H .g

•-" in c

P; 0 2C g 0 C

N CfN CfN f^l ^

- r~ -f

O a

3 in — ON «/

-. ., ^ „ | 2 S =

3 O O O C

3 d d

— -

- — — —

883 S

FN OC OC OO ON C
2 O' O O O C

- r-- f,

' diagonal and

<N V

— C

-1- y

3 r- sO so r

; O _ __ r,

D r-". O O C
5 — ri ri r

1 ON sO d O O C
- ri —

3-t -r ^c c
o o o -
NO d d d d c

P- I — OO sC OC O

c; o o o o c

3 O O O O C

- — 6 o o c
i d d d d c

C OO P- I — OC C

3 0 0 0 0 C

3 0 0 0 C

3 OO r-- — r
3 — rf, m -^

3 d d o' c

N <^, CC f<-, C

300

J NO r*~i
t m •*

3 d d
N -t rl

1
2

^ -

- r~~ m p~- P

- — r | — —

j d d d d d c

r~r. r~~ 'OC OC OO ON 0

O O O O O O C

D O O O O C

0 ON OC OO OO C

3 O O O O C

3 O O O C
3 — t^-i p-} T

3 O O

ON r*

-, OC ON —
ri

d d d d d d c

•( — r-i rr, ri — • r] -

3 O O O O C
f -t rr, rj r] r

3 O O O C
1 ri oc CfN ^

3 O O

-.to

£ S

I £ i c

pr, i/

I <-; "

3OOOOOOC
} -t P*^ p^ m — ri u

5 O O O O C
j sO in Tf ri <••

3 O O' O C

-, t oo ON r
", nl f^i r*^ T

3 O O
1 t t

PI 0-

<r-.

i rl -- Z
D 1 O O C

r 1 P~ sO C

-OONQOOO — <"

icncscNcnmcnc

:dddddo"c

;ONOOONONOCOC

1 — — — p- C

3 d d d d c

5 ON O OO OC 0-
1 fN m ri ri r

N ri — oo C
1 <•*-) ^t m T-

3 d d d c

3 — ON Tf a

3 O NC
f t -f

3 d d
3 »/-. rl

-t

~ C: O C

3000000C

3 0 0 0 0 C
- r-i rt O rl r

3 O O O C
J — oo r- r

3 O O
(• m t

•a,

• '^

1 Spil'obrachia

1 P^/,J,,-,,^/,,V,

3 Galatheatinum 0.
4 Siboglinum fiordicum 0.

5 Siboglinum ekmani 0.

A CV^x,

3OOOOOOC

o-
c

_ ^ ^

s. -g. .2 -a -a | .3 =

juj(-.c£i5c£O-

:P-OCONO — ri P-

4 Lamellibrachia 1 0.
5 Lamellibrachia 3 0.
6 Lamellibrachui 4 0.
7 RiftUi 1 0.
v j?,vv;,. i

9 Sclerolinum 0.
20 Branchipolynoe 0.
21 Lepidonotopodium 0.
^T ii*-n^l«;.

3 O O
= £ c

: '^~Z
5 K £

SIBOGLINID EVOLUTIONARY HISTORY

TABLE 4

Pt'rci'ni o/ significant fc.v/.Y \\~hen comparing relative substitution ruie\
between the two major .\ihnKlinitl chides

Test type*

Number
of tests

Significant
results

Percent
significant

Between frenulates and

vestimentiferan-

Sclerolinum clade

18S rDNA

480

331

69.0

16S rDNA

350

13.4

Within frenulates

18S rDNA

150

35.3

16S rDNA

2.0

Within vestimentiferan-

Sclerolinum clade

18S rDNA

280

28.6

16S rDNA

455

2.9

Results of relative rates tests based on an HKY plus gamma model in the
HyPhy program. The program is distributed by S. Muse, Department of
Statistics. North Carolina State University.

* The 18S comparisons employed all Lophotrochozoan outgroups.

seep tube worms to be basal to vent tube worms (but see
Williams el cii, 1993). This pattern in the evolution of
habitat preference roughly proceeds from less reducing to
more reducing (greater sulfide and methane availability)
environments. A similar evolutionary trend was observed in
bathymodiolid mussels (Craddock el al.. 1995; Distel el al.,
2000). Examination of additional taxa is needed to verify
whether this is a general trend in the evolution of vent and
seep taxa.

All molecular studies to date (Williams et ill., 1993:
Black et al.. 1997; and Tables 2 and 3) reveal that vesti-
mentiferans exhibit very limited molecular diversity for a
group suggested to be several hundred million years old.
This lack of diversity may be due to a slowdown in the rate
of molecular change (i.e., nucleotide substitution) in vesti-
mentiferans. a recent common origin for extant vestimen-
tiferans, or possibly both. For the present 18S rDNA se-
quences, vestimentiferans appear to have experienced a
significant molecular slowdown relative to the frenulates or
other protostome taxa (Table 4; as judged using an HKY
plus gamma correction model in the HyPhy software pack-
age distributed by S. Muse, Department of Statistics. North
Carolina State University). With 16S data, only 13.4% of
tests between frenulates and members of the vestimenti-
fexan-Sclerolinum clade were significant. Although this
value is not statistically significant, it is a greater percentage
than is found within either group ( — 3%), suggesting that a
limited rate discrepancy may exist. Similar rate disparities
were not observed for COI data (Black et al.. 1997). but
only one frenulate was used in the comparison. Nonetheless,
we concluded that present-day vestimentiferans constitute a
young evolutionary group.

In contrast, previous interpretation of Silurian tubeworm
fossils (Little et al.. 1997) as vestimentiferans suggested
that these worms constitute an ancient animal lineage. It is
possible that the Silurian tubeworm fossils represent an
earlier offshoot from an ancient siboglinid lineage, but this
will be impossible to test as the fossils lack the necessary
soft-tissue preservation. Additionally, we note that many
wormlike invertebrates make tubes. For example, some
alvinellid polychaetes observed during our recent expedi-
tion to vents along the Southern Eastern Pacific Rise (32°S,
100°W) occupied tubes with diameters comparable to the
tubes of mature Riftia pachyptila. Many of the alvinellid
tubes were partially overgrown by sulfide chimneys, and
thus were effectively "fossilized." Although we are not
convinced of the interpretation of Silurian fossils as repre-
sentative of an extant lineage of vestimentiferans, we should
point out that specimens from the Cretaceous are convinc-
ing (Little et al., 1999). In contrast, all the hydrothermal
vent-endemic taxa that have been examined with appropri-
ate molecular tools appear to be from relatively recent
radiations (i.e., <100 MY; Black et al.. 1997; Peek et al..
1997; Shank et al., 1999; McArthur and Koop. 1999; but see
McArthur and Tunnicliffe, 1998, for possible exceptions).

Acknowledgments

We appreciate thoughtful interactions and support of our
colleagues at Rutgers University. We wish to thank the
crews and staff of the R/V Altantis/Alvin, the German re-
search vessel Sonne. and the Bergen Marine Station in
Espegrend for their help in obtaining organisms. Samples of
the Spirobrachia and Polybrachia were provided by R. Lutz
(with help from Gyongyver Levai) and identified by Eve
Southward, who has been especially generous with infor-
mation and guidance. The Escarpiid n. sp. was kindly made
available by Verena Tunnicliffe and Eve Southward. Mate-
rial from Norway was collected with aid from the Training
and Mobility of Researchers Programme of the European
Union, through Contract NO. ERBFMGECT950013 to Eve
Southward. Research was supported by an NSF grant.
OCE96-33131 to RCV and R. Lutz. The Richard B. Sellars
Endowed Research Fund and The Andrew W. Mellon Foun-
dation Endowed Fund for Innovative Research provided
partial support to KMH. This is WHOI contribution number
10443.

Literature Cited

Aguinaldo, A. M. A., J. M. Turboville, L. S. Linford, M. C. Rivera,
J. R. Garey, R. A. Raff, and J. A. Lake. 1997. Evidence for a clade
of nematodes. arthropods and other moulting animals. Nature 387:
489-493.

Baco, A. R., C. R. Smith, G. K. Roderick, A. S. Peek, and R. C.
Vrijenhoek. 1999. Molecular identification of vesicomyid clams as-
sociated with whale-tails on the California Slope. Mar. Ecol. Prog. Ser.
182: 137-147.

K. M. HALANYCH. R. A. FELDMAN, AND R. C. VRIJENHOEK

Bartolomaeus, T. 1995. Structure and formation of the uncini in Pecti-
naria koreni, Pectinaria auricoma (Terebellida) and Spirorbis spiror-
bis (Sabellida): implications for annelid phylogeny and the position of
the Pogonophora. Zoomorphology 115: 161-177.

Black, M. B., K. M. Halanych, P. A. Y. Maas, W. R. Hoeh, J. Hashi-
moto, D. Desbruyeres, R. A. Lutz, and R. C. Vrijenhoek. 1997.
Molecular systematics of vestimentiferan tubeworms from hydrother-
mal vents and cold-water seeps. Mai: Biol. 130: 141-149.

Caullery, M. 1914. Sur les Siboglimdae, type nouveau d'invertebres
recueilli par I'expe'dition du Siboga. C. R. Acad. Sci. 158: 2014-2017.

Craddock, C., W. R. Hoeh, R. G. Gustafson, R. A. Lutz, J. Hashimoto,
and R. C. Vrijenhoek. 1995. Evolutionary relationships among
deep-sea mytilids (Bivalvia: Mytilidae) from hydrothermal vents and
cold-water methane/sulfide seeps. Mar. Biol. 121: 477-485.

Distel, D. L., A. R. Baco, E. Chuang, W. Morrill, C. Cavanaugh, and

C. R. Smith. 2000. Marine ecology: Do mussels take wooden steps
to deep-sea vents? Nature 403: 725.

Doyle, J. J., and E. Dickson. 1987. Preservation of plant samples for

DNA restriction endonuclease analysis. Taxon 36: 715-722.
Eernisse, D. J. 1997. Arthropod and annelid relationships re-examined.

Pp. 43-56 in Arthropod Relationships, R. A. Fortey and R. H. Thomas,

eds. Chapman and Hall. London.
Eernisse, D. J., J. S. Albert, and F. E. Anderson. 1992. Annelida and

Arthropoda are not sister taxa: a phylogenetic analysis of spiralian

metazoan phylogeny. Syst. Biol. 41: 305-330.
Feldman, R. A., T. M. Shank, M. B. Black, A. R. Baco, C. R. Smith,

and R. C. Vrijenhoek. 1998. Vestimentiferan on a whale fall. Biol.

Bull. 194: 116-119.
Halanvch, K. 1998. Considerations for reconstructing metazoan history:

signal, resolution and hypothesis testing. Am. Zool. 38: 929-941
Halanvch. K. M., J. D. Bacheller, A. M. A. Aguinaldo, S. M. Liva,

D. M. Hillis, and J. A. Lake. 1995. Evidence from 18S ribosomal
DNA that the lophophorates are protostome animals. Science 267:
1641-1643.

Halanych, K. M., R. A. Lutz, and R. C. Vrijenhoek. 1998. Evolution-
ary origins and age of vestimentiferan tube-worms. Cult. Biol. Mm: 39:
355-358.

Hillis, D. M., and M. T. Dixon. 1991. Ribosomal DNA: molecular
evolution and phylogenetic inference. Q. Rev. Biol. 66: 41 I — 153.

Ivanov, A. V. 1991. Monilifera — a new subclass of Pogonophora. Dokl.
Akad. Nauk. S.S.S.R. 319: 505-507.

Ivanov, A. V. 1994. On the systematic position of Vestimentifera. Zool.
Jalu-h. Abt. Syst. Okol. Geogr. Tiere 121: 409-456.

Ivanov. A. V., and R. V. Selivanova. 1992. Sclerolinum javanicum sp.
n., a new pogonophoran living on rotten wood. A contribution to the
classification of Pogonophora. Biol. Moryu (VLadivost.) 1—2: 27-33.

Jones, M. L. 1981. Riftia pachyptila, new genus, new species, the
vestimentiferan worm from the Galapagos rift geothermal vents
(Pogonophora). Pmc. Biol. Soc. Wash. 93: 1295-1313.

Jones, M. L. 1985. On the Vestimentifera, new phylum: six new species,
and other taxa, from hydrothermal vents and elsewhere. Bull. Biol. Soc.
Wash. 6: 117-158.

Kim, C. B., S. Y. Moon, S. R. Gelder, and W. Kim. 1996. Phylogenetic
relationships of annelids, molluscs, and arthropods evidenced from
molecules and morphology. J. Mol. Evol. 43: 207-215.

Kishino, H., and M. Hasegawa. 1989. Evaluation of the maximum
likelihood estimate of the evolutionary tree topologies from DNA
sequence data, and the branching order in Hominoidea. J. Mol. Evol.
29: 170-179.

Kojima, S. 1998. Paraphyletic status of Polychaeta suggested by phylo-
genetic analysis based on the amino acid sequences of elongation
factor- 1 -alpha. Mol. Pliyhgenet. Evol. 9: 255-261.

Kojima, S., T. Hashimoto, M. Hasegawa, S. Murata, S. Ohta, H. Seki,
and N. Okada. 1993. Close phylogenetic relationship between Ves-
timentifera (tube worms) and Annelida revealed by amino acid se-
quence of elongation factor- la. J. Mol. Evol. 37: 66-70.

Kojima, S., R. Segawa, J. Hashimoto, and S. Ohta. 1997. Molecular
phylogeny of vestimentiferans collected around Japan, revealed by the
nucleotide sequences of mitochondria! DNA. Mar. Biol. 127: 507-513.

Land, J. v. d., and A. Narrevang. 1977. The systematic position of
Lamellibrachia (Annelida. Vestimentifera). Z. Zool. Syst. Evolu-
tionsforsch. 1975: 85-101.

Little, C. T. S., R. J. Herrington, V. V. Maslennikov, N. J. Morris, and
V. V. Zaykov. 1997. Silurian hydrothermal-vent community from
the southern Urals. Russia. Nature 385: 146-148.

Little, C., J. Cann, R. Herrington, and M. Morrisseau. 1999. Late
Cretaceous hydrothermal vent communities form the Troodos ophio-
lite. Cyprus. Geology 27: 1027-1030.

Maddison, W. P., and D. R. Maddison. 1992. MacClade: Analysis of
Phylogeny ami Character Evolution. Sinauer Associates, Sunderland,
MA.

McArthur, A. G., and B. F. Koop. 1999. Partial 28S rDNA sequences
and the antiquity of hydrothermal vent endemic gastropods. Mol.
Phylogenet. Evol. 13: 255-274.

McArthur, A. G., and V. Tunnicliffe. 1998. Relics and antiquity revis-
ited in modern vent fauna. Pp. 271-291 in Modem Ocean Floor
Processes and the Geological Record, R. A. Mills and K. Harrison, eds.
Geological Society, London.

McHugh, I). 1997. Molecular evidence that echiurans and pogono-
phorans are derived annelids. Proc. Null. Acad. Sci. USA 94: 8006-
8009.

Nelles, L., B. L. Fang, G. Volckaert, A. Vandenberghe, and R. De
Wachter. 1984. Nucleotide sequence of a crustacean 18S ribosomal
RNA gene and secondary structure of eukaryotic small subunit ribo-
somal RNAs. Nucleic Acids Res. 12: 8749-8768.

Paiumbi, S. R. 1996. Nucleic acids II: the polymerase chain reaction. Pp.
205-248 in Molecular Systematics, D. M. Hillis, C. Mortiz. and B. K,
Mable. eds. Sinauer Associates, Sunderland. MA.

Peek, A. S., R. G. Gustafson, R. A. Lutz, and R. C. Vrijenhoek. 1997.
Evolutionary relationships of deep-sea hydrothermal vent and cold-
water seep clams (Bivalvia: Vesicomyidae ): results from mitochondria!
cytochrome oxidase subunit I. Mar. Biol. 130: 151-161.

Rouse, G. 2001. A cladistic analysis of Siboglinidae Caullery. 1914
(Polychaeta, Annelida): formerly the phyla Pogonophora and Vesti-
mentifera. Zool. ./. Linn. Soc. 132: 55-80.

Rouse, G. W., and K. Fauchald. 1995. The articulation of annelids.
Zool Sci: 24: 269-301.

Rouse, G. W., and K. Fauchald. 1997. Cladistics and polychaetes. Zool.
Sci: 26: 139-204.

Shank, T. M., M. B. Black, K. M. Halanych, R. A. Lutz, and R. C.
Vrijenhoek. 1999. Miocene radiation of deep-sea hydrothermal vent
shrimp (Caridea: Bresiliidae): evidence from mitochondrial cyto-
chrome oxidase subunit I. Mol. Phylogenet. Evol. 13: 244-254.

Southward, E. C. 1961. Siboga-Expeditie Pogonophora. Siboga-Expe-
ditie series, vol. 25. E. J. Brill, Leiden.

Southward, E. C. 1972. On some Pogonophora from the Caribbean and
the Gulf of Mexico. Bull. Mai: Sci. 22: 739-776.

Southward, E. C. 1988. Development of the gut and segmentation of
newly settled stages of Ridgeia (Vestimentifera): implications for re-
lationship between Vestimentifera and Pogonophora. ./. Mai: Biol.
Assoc. UK 68: 465-4X7.

Southward, E. C. 1993. Pogonophora. Pp. 327-369 in Microscopic
Anatomy of Invertebrates. Wiley-Liss. New York.

Southward, E. C. 1999. Development of Perviata and Vestimentifera
(Pogonophora). Hydrobiologia 402: 185-202.

Southward. E. C., and S. V. Galkin. 1997. A new vestimentiferan

SIBOGLINID EVOLUTIONARY HISTORY

(Pogonophora: Obturatu) from hydrothermal vent fields in the Manus
Back-Arc Basin (Bismarck Sea. Papua New Guinea, southwest Pacific
Ocean). J. Nut. Hist. 31: 43-55.

Swofford, D. L. 2000. PAUP* 4.0 (Phylogenelic Analysis Using Pursi-
mony). Sinauer Associates. Sunderland. MA.

Swofford. D. L., G. J. Olsen, P. J. VVaddell, and I). M. Hillis. 1996.
Phylogenetic inference. Pp. 407-514 in Molecular Systematics, D. M.
Hillis. C. Mortiz, and B. K. Mable, eds. Sinauer Associates. Sunder-
land. MA.

Thompson, J. D., D. G. Higgins, and T. J. Gibson. 1994. CLUSTAL
W: improving the sensitivity ot progressive multiple sequence align-
ment through sequence weighting, position specific gap penalties and
weight matrix choice. Nucleic Acids Res. 22: 4673-4680.

Webb, M. 1964a. A new hitentaculate pogonophoran from Hardanger-
fjorden. Norway. Sarsia 15: 49-55.

Webb, M. 1964b. Additional notes on Sclerolinum brattstromi (Pogono-
phora) and the establishment of a new family. Sclerolinidae. Sarsia 16:
47-58.

Webb, M. 1969. Lumellibrachia barhami, gen. nov. sp. nov. (Pogono-
phora). from the northeast Pacific. Bull. Mar. Sci. 19: 18-47.

Williams, N. A., D. R. Dixon, E. C. Southward, and P. W. H. Holland.
1993. Molecular evolution and diversification of the vestimentiferan
tube worms. J. Mar. Bioi Assnc. U.K. 73: 437-452.

Winnepenninckx, B., T. Backeljau, and R. De Wachter. 1995a. Phy-
logeny of protostome worms derived from 18S rRNA sequences. Mol.
Bio/. Evol. 12: 641-649.

Winnepenninckx, B., T. Backeljau, L. Y. Mackey, J. M. Brooks, R. De
Wachter, S. Kumar, and J. R. Garey. 1995b. 18S rRNA data
indicate that Aschelminthes are polyphyletic in origin and consist of at
least three distinct clades. Mol. Bio/. Evol. 12: 1 132-1137.

Winnepenninckx, B., T. Backeljau, and R. De Wachter. 1996. Inves-
tigation of molluscan phylogeny on the basis of 18S rRNA sequences.
Mol. Bid. Evol. 13: 1306-1317.

Young. C. M., E. Vazquez, A. Metaxas, and P. A. Tyler. 1996. Em-
bryology of vestimentiferan tube worms from deep-sea methane/sul-
phide seeps. Nature 381: 514-516.

Reference: Bio/. Bull. 201: 76-83. (August 2001)

Effect of Cloning Rate on Fitness-Related Traits
in Two Marine Hydroids

LAWRENCE M. PONCZEK* AND NEIL W. BLACKSTONE
Department of Biological Sciences, Northern Illinois University: DeKulh, Illinois 601 15

Abstract. Hydractinia symbiolongicarpus and Poilo-
coryna cornea are colonial marine hydroids capable of
reproducing both sexually and asexually. Asexual reproduc-
tion, by colony fragmentation, produces a genetic clone of
the parent colony. This study examines the effect of very
different cloning rates on colony growth rate, oxygen uptake
rate, and colony morphology. Colonies of one clone of each
species were maintained for an extended time in two treat-
ments: in a state of constant vegetative growth by repeated
cloning, and in a state restricted from vegetative growth (no
cloning). For both species, tissue explants taken from the
growing colonies grew more slowly than similar explants
taken from the restricted colonies. For one species, tissue
explants from the growing colonies used oxygen at a higher
rate than similar explants from restricted colonies: for the
other species, no difference was detected, although the
sample size was small. For both species, tissue explants
from restricted colonies formed more circular, "sheet-like"
shapes, whereas those from their growing counterparts
formed more irregular, "runner-like" shapes. After these
experiments, in the third winter of treatment, all colonies
experienced a severe tissue regression. Within 6 months
after this event, the colonies had regrown to their former
sizes. A growth assay at this point revealed no difference in
growth rate, possibly suggesting an epigenetic basis for
these results. Changes in clonal growth rates and morphol-
ogy correlated with variation in fragmentation rate might
affect the ecology of these and other clonal organisms.

Introduction

Clonal (or "modular") organisms differ from unitary,
sexually reproducing ones in that asexual reproduction pro-

Received 10 July 2000; accepted 26 March 2001.
* To whom correspondence should be addressed. E-mail: plankton9@
prodigy.net

duces identical genetic copies of the parent rather than
genetically unique offspring. In many organisms, episodes
of clonal reproduction may be intercalated with periods of
gamete production. This clonal life cycle has obviously
been successful; two-thirds of metazoan phyla contain
clonal species (Bell, 1982), and most of the earth's sessile
biotic covering is composed of clonal life forms (Jackson et
ai, 1985). During the past few decades, studies in evolu-
tionary biology have delineated various differences between
clonal and unitary organisms. In terms of ecology, for
instance, clonal organisms typically outcompete aclonal
ones in marine hard substratum environments, where space
is commonly a limiting resource (Jackson, 1977; Larwood
and Rosen, 1979). Because its modules are functionally
independent, a colony has great regenerative powers and
can recover from a substantial colony mortality (Hughes
and Cancino, 1985). In addition, a comparison of life his-
tories between the two reproductive modes reveals some
inherent and fundamental differences. Fecundity of clones
or colonies is indeterminate, because iteration of vegeta-
tively produced modules can yield an indefinite number of
reproductive units. This contrasts with unitary organisms,
whose fecundity typically levels off or declines with age
(Hall and Hughes, 1996). Consequently senescence, a de-
rived property of the unitary soma (Medawar, 1952), may
be negligible in clonal genets (Hughes, 1989), which may
be very large and comprise a number of unconnected ramets
(genetically identical but physiologically separate units).

Despite this considerable attention, one attribute of
clonal organisms that has not been studied is the effect on
fitness-related traits of the rate of cloning — that is, the
number of episodes of asexual reproduction prior to the
sexual phase of the life cycle. The rates might differ, for
example, between corals inhabiting a turbulent shallow-
water habitat and those occupying a deeper, more phys-
ically stable environment (e.g., Wulff, 1985). How might

EFFECT OF CLONING RATE IN HYDROIDS

variability in this attribute affect the functioning and
fitness of an organism?

This question was addressed experimentally using marine
hydroids as a model system. This is an appropriate use,
since hydractiniid hydroids are colonial organisms com-
monly used in laboratory manipulations. The species used
here. Hydractinia symbiolongicarpus and Podocoryna
(=Podocoryne camea), reproduce both asexually (by col-
ony fragmentation) and sexually (Brusca and Brusca. 1990).
The rate of cloning can be precisely controlled in the lab-
oratory, since a colony fragment can be surgically excised
from a parent colony and cultured as an independent, yet
genetically identical, colony. For a genotype of each of
these species, rate of cloning was varied, and growth rate
and colony morphology were measured. These are fitness-
related traits (Larwood and Rosen. 1979; McFadden el ui,
1984; Jackson et a!.. 1985; Yund. 1991; Brazeau and
Lasker, 1992). Fitness has been defined as the expected
contribution of a phenotype, genotype, or allele to future
generations, relative to other organisms and genes in the
environment; and thus it may be measured as numerical
dominance over time (Stearns, 1992). For hydroids, which
typically inhabit space-limited habitats, the fitness advan-
tage of a high relative growth rate is manifest. Colony
morphology is another factor that may affect competitive
ability, and therefore it also is important to the success of a
clone in a particular environment (Larwood and Rosen.
1979; Jackson et al., 1985). Although laboratory experi-
ments such as these cannot measure actual fitness in nature,
we attempt to gain insight into what might happen when
similar colony fragments, one from a rapidly fragmenting
clone and one from a relatively unfragmented clone, meet in
the same natural environment. Do such fragments grow at
different rates and thus have different fitnesses? A differ-
ence in growth rate could be correlated with a difference in
metabolic efficiency. After finding a difference in colony
growth rate between these treatments, we measured oxygen
consumption rate as an indicator of overall metabolic rate.
The implications of these results are discussed in the general
context of clonal biology.

Materials and Methods

Studv species

Hydractiniid hydroids (phylum Cnidaria) are marine an-
imals that live as encrusting colonies consisting of repeated
modular units (polyps) specialized for feeding or reproduc-
tion. The polyps are interconnected by tubular stolons that
house gastrovascular canals, forming a net-like structure.
Each colony thus comprises a single integrated physiolog-
ical unit. A colony grows onto suitable available substratum
by extending peripheral stolons and developing erect feed-
ing polyps at intervals along them. Colony growth ceases
when space is no longer available; for instance, in H.

symbiolongicarpus, which typically encrusts the shells of
hermit crabs, colony growth is limited by the size of the
shell. P. cornea also grows on hermit crab shells, but is
found on other hard substrata as well (Edwards, 1972). The
two species differ further in that H. symbiolongicarpus
forms a relatively dense mat of stolonal tissue as the colony
enlarges, but P. carnea does not — its stolons are separated
by areas free of tissue. Clonal reproduction occurs when a
fragment that is separated from a parent colony (e.g., by
physical abrasion) is situated on a surface suitable for at-
tachment and growth (Jackson et al., 1985). An entire new
colony, genetically identical to the parent colony, can grow,
limited in size by available space. Sexual reproduction is
accomplished by gamete formation and release into the
surrounding seawater where syngamy may occur, leading to
the development of a motile planula larva. (H. svmbiolon-
gicarpits produces gametes directly from specialized repro-
ductive polyps; the life cycle off. camea includes a motile
gametogenic medusa stage.) The planula larva may then
attach to a hermit crab shell or other hard substratum suit-
able for growth, where a genetically novel colony develops.

Culture methods

Colonies of both H. symbiolongicarpus and P. carnea
were collected from the shells of hermit crabs near the Yale
Peabody Museum Field Station in Connecticut in 1994.
Explants, consisting of a small portion of a colony made up
of a few feeding polyps along with interconnecting stolons,
were surgically removed from the field-collected colonies
and secured with nylon thread to rectangular (3 in X 1 in)
glass microscope slides to create stock colonies from which
samples could be removed. The slides were then suspended
from floating racks in 1 20-1 aquaria filled with Reef Crystals
artificial seawater (salinity 35%c), and maintained at a tem-
perature of 20.5° ± 0.5°C. The aquaria used undergravel
filtration, and 50% of the water was changed each week.
Ammonia, nitrites, and nitrates were maintained below de-
tectable levels (Aquarium Systems test kits). The colonies
were fed brine shrimp nauplii 3 times per week. No attempt
was made to control the amount of food ingested: observa-
tion indicates that generally all polyps in all colonies feed to
repletion. Thus, colonies with more or larger polyps are
capable of consuming more food. An artificial light cycle of
14:10, L:D, was provided, supplemented by natural light
coming in through windows. The colonies were allowed to
grow over the slides until large enough to permit removal of
a sufficient number of small explants for use as experimen-
tal replicates. All replicates used in the experiment were
maintained in the same conditions as the stock colonies.

Experimental manipulations

For each hydroid species, 25 clonal replicates from a
single parent colony were created on 12-mm rounu

L. M. PONCZEK AND N. W. BLACKSTONE

coverslips by surgical explanting from the stock colonies.
New explants were secured to the coverslips with nylon
thread. Within about one day, the colony attaches itself to
the glass of the coverslip. Five of these (for each species)
were treated in the following way: a colony was allowed to
grow until it had either nearly covered the coverslip or until
it began to produce reproductive polyps (in preparation for
gamete production). Then a small piece of the colony,
consisting of two feeding polyps together with the intercon-
necting stolon, was explanted onto a fresh coverslip. The
new colony was cultured as before, and the old colony was
discarded. In this way, the growing replicates were main-
tained in a state of constant growth and purely clonal
reproduction; these colonies are herein referred to as "grow-
ing." The remaining 20 replicates (for each species) were
allowed to grow completely over the 12-mm cover slips and
to produce gametes or medusae in an unrestricted manner.
They were left undisturbed for the duration of the experi-
ment, except for removal of explants for the purpose of the
various assays. These colonies are referred to as "re-
stricted." The restricted replicates were maintained in higher
numbers because of the impossibility of regenerating them
(without cloning) in the event of colony mortality. The
number of colonies produced ensured that sufficient colony
tissue was available for the assays.

Measures of growth rate

Explants consisting of exactly two intermediate-sized
feeding polyps and a minimal amount of interconnecting
stolon were taken from the experimental colonies, with
multiple explants from the same replicate kept to a mini-
mum; that is, an effort was made to take suitably sized
fragments from all of the 5 growing replicate colonies and
as many of the 20 restricted colonies as possible to obtain
the 1 2 replicates of each treatment for the growth assays.
These were then attached synchronously to fresh 12-mm
coverslips. Although 1 2 replicates per treatment per species
were initiated, some explants failed to attach to the cover-
slips, so actual sample sizes per treatment were smaller.
Explants were allowed to grow for a period of 3 weeks;
none exhausted the available space during the assays. None
of the colonies assayed entered a gamete- or medusa-pro-
ducing phase, so all polyps present during the assays were
feeding polyps. Colony size was measured as number of
polyps produced and, in two of the three growth assays
performed, by total protein content of the colony. For the
3-week polyp counts and total protein measures, between-
treatment comparisons were made for each species using
analysis of variance.

To ensure that the experimental colonies did not inadver-
tently get replaced by any vagrant colonies (of different
genotype) that might have found their way into the aquar-
ium, clonal identity was tested. This was also done to

support the assumption that significant genetic divergence
was not occurring in the colonies during the experiments.
To test clonal identity, explants were made from all five of
the growing colonies for each species onto clean micro-
scope slides (one slide per explant). One explant from a
randomly selected restricted colony was then placed on each
slide, and the pair of colonies was allowed to grow until
stolonal contact was made. When meeting in this way,
colonies from the same clone will merge to produce a single
physiological entity having interconnected stolons (Hughes,
1989; Mokady and Buss, 1996). When unrelated clones
meet, tissue rejection rather than fusion occurs.

Measures of total protein

Subsequent to polyp counts, the colony to be measured
was first macerated in ultrapure water (200-450 /xl, de-
pending on colony size) using a Teflon pestle driven by an
electric drill. Then a small sample of the resulting fluid was
assayed with the Bio-Rad protein assay kit #500-001. which
uses a bovine gamma globulin protein standard and the
Bradford method of protein staining (with Coomassie bril-
liant blue G-250 dye). Binding of the dye to proteins causes
a maximum absorbance shift from 465 nm to 595 nm.
Absorbance at this wavelength was measured in a Beckman
DU-64 spectrophotometer and compared to a standard curve
to determine protein amounts. Although questions arise in
using a bovine standard for assays of cnidarians (Zamer et
al., 1989), these concerns are mitigated in this case because
the same genotypes are being compared, and thus relative,
not absolute, comparisons are sufficient.

Measures of oxygen uptake rate

Colonies to be assayed were obtained by explanting two-
polyp fragments from all five growing colonies and several
restricted colonies onto fresh coverslips. These were al-
lowed to grow until they nearly reached the edge of the
coverslip. All assays were performed 24 h after feeding: at
that time polyps are generally not contracting (Dudgeon et
al., 1999), and the colonies could be considered to be in a
resting state. Measures of oxygen uptake made at these
times can be used as an indication of standard metabolic rate
(Schmidt-Nielsen, 1997; Lowell and Spiegelman, 2000).
For each assay, a colony of each treatment type was se-
lected, matched as closely as possible in size to minimize
any size effects. The two colonies were then assayed se-
quentially. The assays were done in this pairwise fashion so
that any ambient conditions that might affect oxygen uptake
rate (variation in atmospheric pressure, etc.) would not
introduce a sampling bias into the data for either treatment.

Colonies were assayed for rate of oxygen uptake with a
Strathkelvin Instruments oxygen meter, model 781. The
temperature of the oxygen measurement chamber was con-
trolled with a Neslab Instruments model RTE-IOOD exter-

EFFECT OF CLONING RATE IN HYDRO1DS

nal circulation waterbuth at 20.5° ± 0.02°C. The colony
was attached, with a small amount of grease, to a 12-mm
glass coverslip to which a small stir bar had been affixed.
After instrument calibration, the measurement chamber was
loaded with 1.0 ml of seawater filtered to 0.2 p.m and
saturated with oxygen by stirring. Oxygen uptake was mea-
sured every 3 min for a period of at least 30 min with
stirring. Shortly after each individual assay was begun, the
rate of oxygen uptake by the sample colony stabilized and
remained linear for the entire 30-min period. The rate thus
obtained from each sample provided an observation to be
used in the data analysis. Data from the oxygen uptake rate
assays were analyzed using analysis of covariance, begin-
ning initially with a test of heterogeneity of slopes. When
the slopes were found to not differ, between-treatment dif-
ferences in elevation were compared.

Characterization of colony morphology

A hydroid colony can be described as tending towards
having a more "sheet-like" or "runner-like" morphology
(McFadden et al., 1984). Sheet-like colonies, typical of H.
synibiolongicarpus, are characterized by a relatively circu-
lar central stolonal area whose periphery has few projecting
stolons with free ends. Runner-like colonies, characteristic
of P. cornea, have a relatively large number of projecting
free-ended stolons and a small enclosed central stolonal
area. A size-free shape measure that may be used to com-
pare colony morphologies is given by (colony perimeter)/
V(colony area) (Blackstone and Buss, 1991). A minimum
value of 2\/-n describes a circular colony with no projecting
peripheral stolons; this is the quintessential sheet. As the
value of the metric increases, the colony appears more
runner-like. Using this shape metric, colonies of both spe-
cies were tested for a treatment effect. Colonies to be
analyzed were explanted onto fresh 12-mm glass coverslips
and allowed to grow until a stolon reached the edge of the
coverslip, at which time shape analysis was begun. Colony
perimeters and areas were quantified by first imaging the
colony, then performing image analysis with OPTIMAS 5.0
software (Media Cybernetics) for the Windows operating
system. Data gathered in this way were analyzed using
analysis of variance, the F statistic being computed to
compare treatments for each species.

Time course of experiments

The initial experimental explants were made at the be-
ginning of August 1996 (H. synibiolongicarpus) and in
mid-September 1996 (P. carnea). The first growth assays
were performed 12 months later. Shape analyses were done
in December 1997. The second growth assays for H. syni-
biolongicarpus were performed in July 1998, nearly 24
months after initial explants. At this same time, oxygen
uptake assays of P. carnea were done. In November 1998.

oxygen uptake assays of//, symbiolongicarpus were begun.
Shortly after this time, when six pairs of H. .svmbiolongi-
carpus colonies had been assayed for oxygen uptake, all of
the colonies in the experiment underwent a severe tissue
regression. This event truncated the H. synibiolongicarpus
oxygen uptake assays and precluded a planned second
growth assay for P. carnea. Similar midwinter regressions
generally occur in field-collected hydroid colonies exposed
to natural light (pers. obs.). In the case of the manipulated
colonies, this regression was especially severe, with all
colonies experiencing almost complete tissue death. How-
ever, enough living tissue remained in the colonies so that
within 6 months they had regained their previous size. A
final growth assay was done for P. carnea after 32 months
from the initiation of the experiment (beginning of May).
and for H. synibiolongicarpus after 35 months (beginning of
July). Also after 35 months, fusion tests between restricted
and growing colonies of each species were begun.

Results

Measures of growth rate

The first growth-rate assay was performed about 12
months after the initial explants of the experimental colo-
nies were made. For both species, the growing colonies
grew more slowly than the restricted ones (Fig. 1; ANOVA
of log-transformed 3-week polyp counts; H. synibiolongi-
carpus, F = 9.27. df = 1, 20, P < 0.007; P. carnea, F =
13.21, df = 1. 22. P < 0.002). A second growth assay was
begun for H. synibiolongicarpus after 24 months, entailing
polyp counts as well as measures of total protein. Again, the
restricted colonies grew at a faster rate than their growing
counterparts, this time to a more pronounced degree (Fig. 2;
ANOVA of log-transformed polyp counts. F = 65.02, df =
I. 17. P <K 0.001: ANOVA of log-transformed total pro-

H symbiolongicarpus

P camea

Figure 1. Growth rate comparisons of growing and restricted
tinia symbiolongicarpus and Podocoiyiui carnea colonies from the ;i
performed after 12 months of experimental treatment. Means and -.kmdard
errors of the number of polyps in a colony are represented.

L. M. PONCZEK AND N. W. BLACKSTONE

Polyps

Protein

Figure 2. Growth rate comparison of growing and restricted Hydrac-
linia symbiolongicarpus colonies from the assay performed after 24
months of experimental treatment. The left v-axis shows the number of
polyps in a colony; the right v-axis shows total colony protein. Means and
standard errors are represented.

tein, F = 227.70, df = 1, 17, P « 0.001 ). Note that colony
size, measured as number of polyps, and total colony pro-
tein are highly correlated (Fig. 3). A second growth assay
for the P. cornea colonies was precluded by the widespread
midwinter tissue regression that occurred in early 1999.

The fusion tests resulted in the colonies fusing, suggest-
ing that significant genetic divergence, at least at histocom-
patibility loci, had not occurred. This result also strongly
supports the assumption that experimental colonies were not
replaced by other genotypes during the experiments.

Measures of oxygen uptake rale

At all sizes, growing colonies of P. cornea consumed
oxygen at a higher rate than did restricted colonies (Fig. 4a).

£• 150 -

S. 100

(/>

0. 50 -

•
o

Growing
Restricted

n=9

n=10

O O

0 100 200 300 400 500

Total protein per colony (ng)

Figure 3. Bivariate scatter plots of the number of polyps in a colony
and its total protein content. Linear regression using combined data from
both treatments yields the equation v = 0.507.x - 2.44 (^-squared = 0.98).
This intercept is not significantly different from zero (T = —0.568. P >
0.58). Regression lines for growing and restricted colonies do not differ in
slope (ANCOVA, F = 0.84. df = 1. 15, P > 0.37) or elevation (F = 0.97.
df = 1. 16. P > 0.34).

c
E

o
co

_i£

D.
CXI

(a) P carnea

o o o

o o

•

Growing
Restricted

n=12
n=13

(b) H. symbiolongicarpus

o 0

•
o

Growing n=6
Restricted n=6

0 500 1000 1500 2000 2500 3000

Colony size (ug protein)

Figure 4. Bivariate scatter plots of oxygen uptake rate of growing and
restricted colonies, (a) Data for Podocoryna cornea. The slopes of the
regression lines for the growing and restricted treatments do not differ
(ANCOVA, F = 1.75. df = 1. 21, P > 0.20), but an elevation difference
was found (F = 20.54, df = 1, 22, P < 0.0002). These relationships were
strengthened by omission of a single outlying data point from the growing
data set (slope: F = 0.10. df = 1, 20, P > 0.76; intercept: F = 41.06, df =
I. 21. P < 0.0001). (b) Data for Hydractinia svmbiolongicarpus. The
slopes of the regression lines for the two treatments were not significantly
different (ANCOVA. F = 0.83. df = 1. 8, P > 0.39), and neither were the
intercepts (F = 0.98, df = 1. 9. P > 0.35).

Although no significant difference in oxygen consumption
rate was found between treatments for H. symbiolongicar-
pus (Fig. 4b), a trend may be discerned in the data that
would indicate agreement with the result found for P. car-
nea. The sample size is too small to render this trend
statistically significant, however.

Characterization of colonv morpholog\

Growing colonies of both species had a more runner-like
morphology than their restricted counterparts (Fig. 5; H.
symbiolongicarpus, F = 12.56, df = 1, 20, P < 0.002; P.
carnea. F = 6.16, df = 1, 22, P < 0.0212).

Growth rate after regression

A growth assay was performed 4 to 6 months after the
pronounced winter regression. At this time, no significant

EFFECT OF CLONING RATE IN HYDROIDS

(Runner-like)

o
o

(Sheet-like)

n=12
T

H symb/olongicarpus

P camea

Figure 5. Comparison of growing and restricted colonies after 18
months of experimental treatment in terms of colony morphology as given
by the shape metric (colony perimeter)/\ (colony area). Means and stan-
dard errors are represented.

difference was detected between treatments in either species
for growth as measured by total colony polyp counts (Fig.
6a; H. symbiolongicarpus, F = 0.06, df = 1, 16, P > 0.806;
P. carnea, F = 0.44, df = 1, 18, P > 0.516. data for both
analyses log-transformed) or by total colony protein (Fig.
6b; H. symbiolongicarpus, F = 0.17. df = 1. 16. P > 0.689;
P. carnea, F = 1.04. df = 1, 18, P > 0.321; data for both
analyses log-transformed).

Discussion

Two experimental treatments were used in this study of
hydroid colonies. One group of replicates was allowed to
completely overgrow and remain undisturbed on 12-mm
coverslips ("restricted" colonies); a second group was re-
peatedly cloned as vegetative growth continued, without
being allowed to enter into a gamete-producing sexual
phase ("growing" colonies). A clear difference in growth
rate was found between treatments in both species studied.
with restricted colonies exceeding growing colonies in
growth rate during controlled assays. Since only one clone
was used per species, this result is not replicated at the level
of the species. Nevertheless, at a higher level (i.e., species
within family), the two clones provide replication of this
primary result.

Assays of the oxygen uptake rate between treatments
revealed that the growing colonies of Podocoryna carnea
exceeded the restricted ones in oxygen consumption. Al-
though no significant statistical difference was found for
Hydractiniu symbiolongicarpus, the sample size was small,
and a trend seems to be discernible in the data that would
suggest agreement with the result for P. ciirneu. Such a
result may seem counterintuitive: the colony that uses more
oxygen might also be expected to grow faster. On the other
hand, higher oxygen uptake may be correlated with lower
growth rate if the former indicates greater metabolic expen-

diture on, for instance, somatic maintenance. Such a hy-
pothesis is not entirely implausible. These hydroid colonies
are ecologically space-limited, typically inhabiting small
hermit crab shells. It is likely that selection favors rapid
sequestration of available space to prevent the settlement of
competitors; colonies may maximally allocate energy re-
sources to growth until the available space is covered.
Under such conditions of intense metabolic demand, cellu-
lar metabolism may generate high levels of reactive oxygen
species (Allen, 1996; Chiueh, 2000). These reactive species
can cause various defects in macromolecules. so continu-
ously growing colonies might experience defects in the
mechanisms of oxidative phosphorylation or allocate
greater resources to production of anti-oxidant enzymes
(e.g., Blackstone, 2001). Thus the data are consistent with
the hypothesis that growing colonies expend more energy
on functions other than somatic growth, although further
study of this issue is needed. Our interpretation of these
results is that the restricted colonies are metabolically more
efficient and so can allocate more energy to growth (Lowell
and Spiegelman, 2000).

g- 40-,

| 30-
to

(a)

H symbiolongicarpus P camea

3 60-

o 30 -

"ro

•5 20 -

10 •

(b)

H symbiolongicarpus

P camea

Figure 6. Growth rate comparisons of growing and restricted Hydrac-
tinia symbiolongicarpus and Pntlt><i>ntiti curnea colonies from the ass;iy
performed after 32-35 months of experimental treatment. Means and
standard errors are represented, (a) Number of polyps per colony (M Total
colony protein content.

L. M. PONCZEK AND N. W. BLACKSTONE

The widespread tissue regression that occurred appar-
ently reset to zero the growth rate difference that had been
entrained by the experimental treatments. By this view, the
physiological basis of the difference prior to regression was
transmitted to the clonal fragments of the growing colonies,
becoming enhanced over time as shown by the decreasing
colony growth rate. This may suggest an epigenetic basis for
the phenomenon, wherein a particular state of gene activity
underlies the increased rate of oxygen consumption coupled
with the reduced growth rate. During the regression event.
all colonies lost most of their living tissue, effecting a cell
population bottleneck. The elimination of the growth rate
difference could perhaps be due to sampling error in the
cells that escaped death during the regression, or to some
dedifferentiation process involving a return to a metabolic
ground state. In any case, cells of similar condition and gene
activity seem to have survived the regression. Periodic
regressions of this kind have been observed in some clonal
taxa and are possibly related to senescence (Bayer and
Todd, 1997; Gardner and Mangel. 1997). The life span of
the modules (polyps) that make up a colony may be ex-
tended through cycles of degeneration and regeneration
(Hughes. 1989).

Comparing absolute growth rates of colonies undergo-
ing both treatments early in the experiment (Fig. 1) with
those measured some two years later (Fig. 3) reveals a
consistent decline. Furthermore, the growth rate equal-
ization after regression occurred not by the growing
colonies recovering a rapid growth rate but by the faster
growing restricted ones assuming a similarly diminished
rate. This reduction in growth rate over time may be
considered to be a manifestation of colony senescence
(Bell, 1988). By this criterion, growing colonies senesced
more rapidly than restricted ones prior to the tissue
regression event, suggesting that a high cloning rate
accelerates colony senescence relative to uncloned colo-
nies. After regression, the degree of clonal senescence
(measured by growth rate) became equalized.

Hydractiniid hydroid colonies fragment to produce po-
tentially viable clonal modules, thus enlarging and dis-
persing the genet asexually (Cerrano et al.. 1998). The
colony fragmentation rate (equivalent to the cloning rate
considered in this study) presumably could vary with the
physical environment in which the hydroids are found. In
aquaria. Cerrano et al. ( 1998) found that clonal colonies
arising from fragments of Podocoryna exigna colonies
can grow on a sandy-bottom substratum and that hermit
crabs with naked shells placed into this environment were
colonized within a few days. If such a process occurs
naturally in P. exigna and other hydractiniid hydroids.
such as the species used in this study, a genet might
extend itself naturally by fragmentation. Clonal lineages
may vary in fragmentation rate and growth rate of colo-
nial ramets. This study shows that cloning rate could

possibly affect the growth rate of a ramet within a lineage
through negative feedback, since variation in growth rate
may be passed on through some epigenetic mechanism
such as cytosine methylation (but see Tweedie and Bird,
2000; and Amedeo et al., 2000). Nevertheless, histocom-
patibility data (Grosberg et al.. 1996; Mokady and Buss,
1996) suggest that in at least some populations of H.
symbiolongicarpus the rate of fragmentation is low rela-
tive to the rate of sexual recruitment.

The alteration in morphology with variation in cloning
rate might have a bearing on the ecological functioning of a
hydroid colony (McFadden et al., 1984; Yund, 1991;
Brazeau and Lasker, 1992). Intraspecific competition is
common between Hydractinia colonies (Buss and Black-
stone, 1991). The present study has shown that a high
cloning rate can produce a more runner-like colony mor-
phology, thus tending towards a form associated with a
"guerrilla" ecological strategy (Jackson et al., 1985). Such a
clone might have more limited direct competitive ability,
but might also be dispersed to more locations due to its
greater rate of fragmentation.

Asexual reproduction is an essential part of the life his-
tory of all clonal organisms and is thus an important factor
in their evolution and ecology. In some taxa, fragmentation
rate depends on morphological characters, which are at least
in part genetic and thus subject to selection. The fragmen-
tation rate of clones of branching coral reef demosponges
was found to depend on branch thickness (Wulff, 1985). A
coral of the genus Plexaura has evidently evolved morpho-
logical characters that make fragmentation more common in
this species than in its congeners and produce some popu-
lations in which more than 90% of the individuals are
clonemates (Lasker, 1990). A possible difference in growth
rate dependent on cloning rate would have to be taken into
account when considering the demographic impact of frag-
mentation.

The effects of the two experimental treatments on the
clonal replicates of both hydroid species indicate that fre-
quently fragmenting colonies exhibit reduced colony
growth rates, hence diminished reproductive potential and
compromised competitive ability in the space-limited hab-
itats in which they are typically found. Moreover, a within-
species difference in colony morphology was found be-
tween unfragmented colonies and those maintained in a
constant state of vegetative growth by repeated cloning
(fragmenting); this difference could affect the ecological
functioning of the colonies in nature. However, these dis-
crepancies may disappear if a large-scale regression of
colony tissue occurs. Regardless of the specific physiolog-
ical mechanisms producing these differential effects, frag-
mentation rate can be important to various aspects of the
biology of clonal organisms.

EFFECT OF CLONING RATE IN HYDROIDS

Acknowledgments

Comments were provided by K. Gasser, B. Johnson-
Wint, and P. Meserve. The National Science Foundation
(IBN-94-07049 and IBN-00-90580) provided support.

Literature Cited

Allen, J. F. 19%. Separate sexes and the mitochoncirial theory of aging.

./. Thcor. Binl. 180: 135-140.
Amedeo, P.. V. Habu. K. Afsar, O. Mittelstein Scheid. and J. Pasz-

ko« ski. 2000. Disruption of the plant gene MOM releases transcrip-

tional silencing of methylated genes. Nature 405: 203-206.
Bayer, M., and C. D. Todd. 1997. Evidence for zooid senescence in the

marine bryozoan Electro pilosa. Invertebr. Bio/. 116: 331-340.
Bell. G. 1982. The Masterpiece of Nature: The Evolution and Genetics of

Sexuality: Croom Helm. London and Canberra.
Bell. G. 1988. Sex and Death in Protozoa: The History of an Obsession.

Cambridge University Press. Cambridge.
Blackstune. N. \V. 2001. Redox state, reactive oxygen species, and

adaptive growth in colonial hydroids. J. Exp. Biol. 204: 1845-1853.
Blackstone, N. VV., and L. \V. Buss. 1991. Shape variation in hydrac-

tmiid hydroids. Biol. Bull. 180: 344-405.
Brazeau. D. A., and H. R. Lasker. 1992. Growth rates and growth

strategy in a clonal marine invertebrate, the Caribbean octocoral Bria-

reum asbestiiiitm. Biol. Bull. 183: 269-277.

Brusca, R., and G. Brusca. 1990. Invertebrate Biologv. Sinauer Asso-
ciates. Sunderland, MA. 922 pp.
Buss. L. VV., and N. W. Blackstone. 1991. An experimental exploration

of Waddington's epigenetic landscape. Philos. Trans. R. Sue. Lond.

332: 49-58.
Cerrano, C., G. Bavestrello, S. Puce, and M. Sara. 1998. Biological

cycle ofPodocoryna exigua (Cnidaria: Hydrozoa) from a sandy bottom

of the Ligurian sea. J. Mar. Biol. Assoc. U.K. 78: 1101-11 1 1.
Chiueh, C. C., ed. 2000. Reactive Oxygen Species. Ann. N. Y. Acad. Sci.

899: 425 pp.
Dudgeon, S., A. Wagner, J. R. Vaisnys, and L. W. Buss. 1999. Dy-

namics of gastrovascular circulation in the hydrozoan Podocoryne

camea: the one-polyp case. Biol. Bull. 196: 1-17.
Edwards. C". 1972. The hydroids and the medusae Podocoryne areolata.

P. borealis. and P. camea. J. Mar. Biol. Assoc. U. K. 52: 97-144.
Gardner, S. N., and M. Mangel. 1997. When can a clonal organism

escape senescence' Am. \at. 150: 462-490.
Grosberg, R. K., D. R. Levitan, and B. B. Cameron. 1996. Evolution-

ary genetics of allorecognition in the colonial hydroid Hydractinia
symbiolongicafpus. Evolution 50: 2221-2241).

Hall, V. R., and T. P. Hughes. 1996. Reproductive strategies of modular
organisms: comparative studies of reef-building corals. Ecology 77:
950-963.

Hughes, R. N. 1989. ,4 Functional Biology of Clonal Animals. Chapman
and Hall. New York.

Hughes, R. N., and J. M. Cancino. 1985. Ecological overview of
cloning in metazoa. Pp. 153-186 in Population Biology and Evolution
of Clonal Organisms. J. B. C. Jackson. L. W. Buss, and R. E. Cook,
eds. Yale University Press, New Haven.

Jackson, J. B. C. 1977. Competition on marine hard substrata: the
adaptive significance of solitary and colonial strategies. Am. Nat. Ill:
743-767.

Jackson, J. B. C., L. W. Buss, and R. E. Cook, eds. 1985. Population
Biology and Evolution of Clonal Organisms. Yale University Press.
New Haven.

Larwood, G., and B. Rosen, eds. 1979. Biology and Systematics of
Colonial Organisms. Academic Press. London.

Lasker. H. R. 1990. Clonal propagation and population dynamics of a
gorgonian coral. Ecology 71: 1578-1589.

Lowell, B. B., and B. M. Spiegelman. 2000. Towards a molecular
understanding of adaptive thermogenesis. Nature 404: 652-660.

McFadden. C. S.. M. J. McFarland, and L. VV. Buss. 1984. Biology of
hydractiniid hydroids. 1. Colony ontogeny in Hydractinia echinata
(Flemming). Biol. Bull. 166: 54-67.

Medawar, P. B. 1952. An Unsolved Problem of Biology. H. K. Lewis.
London.

Mokady, O., and L. VV. Buss. 1996. Transmission genetics of allorecog-
nition in Hydractinia symbiolongicarpus (Cnidaria: Hydrozoa). Genet-
ics 143: 823-827.

Schmidt-Nielsen, K. 1997. Animal Physiology: Adaptation and Envi-
ronment. Cambridge University Press, Cambridge.

Stearns, S. C. 1992. The Evolution of Life Histories. Oxford University
Press, Oxford.

Tweedie. S.. and A. Bird. 2000. Mutant weed breaks silence. Nature
405: 137-138.

Wulff, J. L. 1985. Variation in clone structure of fragmenting coral reef
sponges. Binl. J. Linn. Sac. 27: 311-330.

Yund, P. O. 1991. Natural selection on hydroid colony morphology by
intraspeciric competition. Evolution 45: 1564-1573.

Zamer, W. E., J. M. Shick, and D. W. Tapley. 1989. Protein measure-
ment and energetic considerations: comparisons of biochemical and
stoichiometric methods using bovine serum albumin and protein iso-
lated from sea anemones. Limnol. Oceanogr. 34: 256-263.

Reference: Biol. Bui!. 201: 84-94. (August 2001)

Egg Longevity and Time-Integrated Fertilization in a

Temperate Sea Urchin (Strongylocentrotm

droebachiensis)

SUSANNE K. MEIDEL* AND PHILIP O. YUND
School of Marine Sciences, Darling Marine Center, University of Maine, Walpole. Maine 04573

Abstract. Recent tield experiments have suggested that
fertilization levels in sea urchins (and other broadcast
spawners that release their gametes into the water column I
may often be far below 100%. However, past experiments
have not considered the potentially positive combined ef-
fects of an extended period of egg longevity and the release
of gametes in viscous fluids (which reduces dilution rates).
In a laboratory experiment, we found that eggs of the sea
urchin Strongylocentrotus droebachiensis had high viability
for 2 to 3 d. Fertilization levels of eggs held in sperm-
permeable egg baskets in the field and exposed to sperm
slowly diffusing off a spawning male increased significantly
with exposure from 15 min to 3 h. In a tield survey of
time-integrated fertilizations (over 24, 48, and 72 h) during
natural sperm release events, eggs held in baskets accrued
fertilizations over as much as 48 h and attained fairly high
fertilization levels. Our results suggest that an extended
period of egg longevity and the release of gametes in
viscous fluids may result in higher natural fertilization lev-
els than currently expected from short-term field experi-
ments.

Introduction

Recent work has started to explore the fertilization dy-
namics of free-spawning marine organisms that release one
or both gametes into the water column (e.g., algae: Pearson
and Brawley, 1996; corals: Lasker et al., 1996; starfish:
Babcock et al.. 1994; sea urchins: Levitan et al.. 1992;
ascidians: Yund, 1998; fish: Petersen etui, 1992). Although
the details of scientific approaches vary, studies can be

Received 22 September 2000; accepted 26 April 2001.
* To whom correspondence should be addressed. E-mail:
meidel@maine.edu

broadly grouped into experiments in which a limited num-
ber of manipulated organisms are induced to spawn, and
surveys of natural spawning events (Levitan. 1995; Yund,
2000). Experimental studies that control spawning syn-
chrony and spatial relationships to test specific mechanistic
hypotheses generally suggest that fertilization levels may be
limited by sperm availability unless males and females
spawn simultaneously, at close range, or under nearly ideal
flow conditions (see Levitan and Petersen, 1995; and Yund,
2000, for reviews). In contrast, many surveys of natural
spawns report fairly high fertilization levels, at least at the
times and places in which most members of a population
spawn (Yund, 2000). However, comparisons between exist-
ing experiments and surveys are complicated by two major
factors. First, results from experimental studies can success-
fully predict fertilization levels in natural spawns only if
experimental conditions (both biotic and abiotic) accurately
mimic natural spawning conditions; however, experiments
often circumvent reproductive strategies that may have
evolved to enhance fertilization (Yund, 2000). Second, ex-
periments and surveys are rarely conducted with the same
species, so it is virtually impossible to distinguish between
taxonomic and methodological effects in existing studies.

Echinoderms have proven to be a particularly valuable
model system for short-term field experiments, and experi-
mental fertilization data from echinoderms generally sup-
port the paradigm of severe sperm limitation under a wide
range of flow and population conditions (e.g., Pennington,
1985; Levitan, 1991; Levitan et al., 1992; Wahle and Peck-
ham, 1999; but see Babcock et a!., 1994). However, there
are no published surveys of fertilization levels in natural
spawns of echinoderms. The absence of survey data is
probably due in part to a lack of information on temporal
spawning patterns and the proximate environmental cues
that initiate spawning (though multiple cues have been

SEA URCHIN FERTILIZATION DYNAMICS

proposed and investigated: Hirnmelman, 1975; Starr et al.,
1990. 1992. 1993).

Two interrelated adaptations that have been largely by-
passed in previous experimental studies may have consid-
erable effects on fertilization levels in natural spawns of
temperate echinoderms. The first is an extended period of
egg viability, which potentially allows fertilizations to ac-
crue over time. Short-term experiments make one or both of
the following assumptions: that most eggs are fertilized
within the first few seconds of release (Denny and Shibata.
1989; Levitan et al., 1991) and that gametes are quickly
diluted to concentrations below which fertilization can oc-
cur. Consequently, extended egg viability has implicitly
been presumed to have little influence on fertilization levels
in the field. Meanwhile, recent estimates of egg longevity
have steadily extended what was presumed to be a relatively
short period of viability. Pennington ( 1985) reported a min-
imum viability period of 24 h for eggs of the temperate sea
urchin Strongylocentrotiis droebachiensis (Muller), and
eggs of a West coast sea urchin are now known to be viable
for up to 2 wk when stored under axenic conditions (Epel et
ai. 1998). If eggs can be fertilized for a long period of time,
extended or repeated exposure of eggs to sperm during
long-duration spawning events (or events in which multiple
males spawn successively) could result in high time-inte-
grated levels of fertilization, even if sperm are limiting in
the short term.

A second adaptation that may interact with extended egg
longevity to increase fertilization levels is the release of
gametes in viscous fluids, which reduces gamete dilution
rates and potentially increases the duration of egg exposure
to sperm. Thomas ( 1994) has shown that three species of sea
urchins (Tripneustes gratilla, Echinometra miitliaei, and
Colobocentrotus at rants) release gametes in such viscous
fluids that eggs and sperm remain on the test and spines at
current speeds less than 0.13 m • s~'. When the current
speed increases, gametes are transported away from this
reservoir in long (3-4 cm) strings or clumps, which led
Thomas (1994) to hypothesize that sea urchins may achieve
high fertilization levels if gametes encounter each other in
these structures. Sperm concentrated in clumps presumably
also have greater longevity because of a reduction in the
respiratory dilution effect (Chia and Bickell, 1983). In con-
trast to natural sperm release, fertilization experiments often
mimic "males" with syringes from which diluted gametes
are extruded at a fixed (and fast) rate, thus circumventing
the potentially beneficial effect of "sticky" sperm that cling
to the test and spines and slowly diffuse away.

In this study, we investigate the effects of these two
aspects of sea urchin reproductive biology on fertilization
levels in Strongylocentrotiis droebachiensis. We initially
determine the duration of egg viability at two points during
the reproductive season. We then explore whether extended
(3 h) exposure of eggs to sperm diffusing off a male sea

urchin enhances fertilization levels relative to short-term
( 15 min) contact at various downstream distances. Finally,
we use the full period of egg viability to assay time-inte-
grated fertilization levels during natural sperm release
events in small populations and use the distribution of
developmental stages in these field samples to evaluate the
temporal distribution of fertilization events.

Materials and Methods

General procedures

To obtain fresh eggs and sperm for use in experiments
and field sampling, sea urchins (Strongylocentrotus droe-
bachiensis) were injected through the peristomial mem-
brane with 0.2-2.0 ml of 0.5 M KC1. Females spawned into
50-ml glass beakers containing chilled seawater that had
been aged (~ 15-20 h; hereafter referred to as aged seawa-
ter) to eliminate ambient sperm. Female spawn was checked
to confirm the absence of immature oocytes (as indicated by
the presence of a large nucleus and nucleolus) and then
washed three times with aged seawater. Dry sperm was
pipetted directly from the aboral surface of spawning males
and kept refrigerated until use (maximum 2 h).

To assay fertilization levels in the field, unfertilized eggs
were deployed in sperm-permeable containers. These egg
baskets consisted of a 0. 1-m-long frame of PVC pipe (in-
ternal diameter 0.05 m) with the sides (—90% of circum-
ference) cut away, covered with 35-jim Nitex mesh (after
Wahle and Peckham, 1999, as modified from Levitan et ai.
1992). and two Styrofoam floats attached for positive buoy-
ancy. Baskets were suspended from the surface or deployed
on the bottom in different spatial arrangements as described
in the following sections.

Egg longevity

To determine the viability period of eggs of Strongylo-
centrotus droebachiensis, we performed laboratory experi-
ments at the beginning (experiment 1 : February 28 to March
2. 2000) and in the middle (experiment 2: March 28 to April
1, 2000) of the spawning season along the coast of Maine
(March to May, Cocanour and Allen, 1967). In each exper-
iment. 120 JJL\ of freshly spawned eggs (mean ± SE of
1651 ± 69 eggs) from each of four females were added to
10 ml aged seawater (aerated for 1 h prior to use) in 20-ml
glass scintillation vials. At the start of each experiment (0 h)
and after 24. 48. 72. and 96 h (experiment 2 only), eggs in
each of four replicate vials per female (only one replicate
per female at 0 h in experiment 1 ) were fertilized with 20 /xl
of a 10-fold sperm dilution ( 10 jul fresh dry sperm from 3
males. 90 /u,l aged seawater). Vials were gently agitated
three times during a 15-min period, following which the
fertilization process was stopped with the addition of 2.5 ml
37% formaldehyde. At each time point, one additional via!

S. K. MEIDEL AND P. O. YUND

per female was fixed without fertilization, as a control for
false fertilization envelopes (from causes such as egg dam-
age or low egg quality). Vials were kept at ambient seawater
temperature ( 1°-3°C) during both experiments. Fertilization
levels were calculated as the percentage of a random sub-
sample of 300 eggs with a fertilization envelope.

Two-way analyses of variance (ANOVA) with the fixed
factors Female (four levels) and Time (three levels in ex-
periment 1; five in experiment 2) were used to analyze
variation in fertilization levels (% fertilization). To achieve
homogeneity of variances, percent fertilization values were
arcsine transformed for experiment 1 (O'Brien's test, F =
1.20. P > 0.32) but not transformed for experiment 2
(O'Brien's test, F 1.35, P > 0.19). The Student-
Newman-Keuls (SNK) test was used for post-hoc compar-
isons of levels within main effects in the absence of a
significant interaction effect.

Cumulative fertilization in the field: 15 min vs 3 h

In this experiment, we determined whether extended (3 h)
exposure of eggs in baskets to sperm from a spawning male
enhanced fertilization levels relative to short-term (15 min)
exposure. We constructed a fertilization platform that was
mounted on a concrete block (L X W X H: 0.36 m X
0.33 m X 0.14 m) deployed by a rope. The platform
consisted of a pine board (1.59 m X 0.24 m X 0.02 m)
bolted to the concrete block so that it extended 0.31 m
upstream of the block and 0.92 m downstream. The board
housed one male and two female stations. The male station
was simply a surface-mounted PVC plate (0.08 m X 0. 12 m
X 0.003 m), located 0.30 cm from the upstream end of the
board, to which a spawning male could be fastened. Female
stations consisted of eyebolts anchoring ropes that extended
to the surface and were located 0.3 and 1.0 m downstream
of the male station.

Experiments were performed on a sandy substratum be-
low the dock of the University of Maine's Darling Marine
Center in the Damariscotta River estuary (ME. 43°50'N,
69°33'W) at a depth of 4.30 m at mean low water (MLW).
For each trial (n = 8), four egg baskets (two side by side
—0.05 m above the platform at each of two female stations)
containing 500 ju.1 freshly spawned eggs (mean ± SE:
7613 ± 455 eggs) from one female were attached to the
eyebolts. A male was induced to spawn by injection of
2.5-4.5 ml 0.5 M KC1 and then attached to the male station
with rubber bands. The fertilization platform was then im-
mediately deployed. In addition to the platform, two mobile
female stations (baskets on weighted lines with the lower
basket 0.35 m above the substratum) were deployed 2 m
upstream (control for ambient sperm: one basket) and
— 2.60 m downstream (two baskets spaced 0.1 m apart
vertically, omitted from trial 1 ) from the male station. After
15 min, one egg basket from each of the three downstream

female stations was retrieved without disturbing the remain-
der of the array, by pulling it to the surface on its own line.
The remaining baskets were retrieved after 3 h, and the
presence or absence of sperm on the aboral surface of the
male was recorded. Eggs were immediately collected and
fixed with formaldehyde. To determine fertilization levels,
300 eggs per vial (200-300 in five cases, 154 in one case)
were randomly sampled and scored for the presence or
absence of a fertilization envelope. Where sufficient num-
bers of eggs were retrieved (82% of baskets), small sub-
samples were taken before fixation and scored after about
15-20 h for the presence or absence of later developmental
stages.

During trials 2 through 8, current velocity was recorded
with a 3D- ACM acoustic-doppler current meter (Falmouth
Scientific). Each trial took place around mid-tide (i.e., com-
menced — 1.5 h after high [or low] water and ended — 1.5 h
before low [or high] water) to minimize variation in the flow
regime.

Three laboratory controls (held at — 3°C), consisting of
200 ju.1 freshly spawned eggs in 10 ml aged seawater, were
assayed for ( 1 ) fertilization at the start of each trial; (2)
fertilization at the end of each trial; and (3) the presence of
false fertilization envelopes, scored twice (after retrieval of
15 min and 3 h samples). Laboratory controls were scored
in the same manner as field samples.

A two-way ANOVA with the fixed factors Time (two
levels) and Distance (three levels) was used to determine
differences in fertilization levels (%) in field samples. Per-
cent fertilization values were arcsine transformed prior to
analysis to achieve homogeneity of variances (O'Brien's
test. F = 0.94, P > 0.47).

Sperm availability in nature

We measured cumulative (over 24, 48, or 72 h) fertiliza-
tion levels of eggs retained in baskets during natural spawn-
ing events of Strongylocentrotus droebachiensis. This sam-
pling design is a hybrid between an experiment and a true
survey of natural spawns, because any sperm present were
naturally released, but egg locations were under experimen-
tal control. Sampling started in mid-February and ended in
early April in 1999 and 2000 but varied in intensity (both
spatial and temporal) during the two years. In 1999, samples
were collected at a single station at Christmas Cove (ChC,
mouth of the Damariscotta River estuary); in 2000, samples
were collected from three stations at ChC and four stations
at Clarks Cove (C1C. 1 km seaward of the Darling Marine
Center and -9 km from the ChC site). Both sites were
relatively sheltered with a sandy substratum, and surveys of
the immediate surroundings indicated the absence of sea
urchin populations other than those sampled (pers. obs.). A
small population of 5. droebachiensis ( — 150 animals in
1999, -60 in 2000) occurred naturally at ChC. At C1C, we

SEA URCHIN FERTILIZATION DYNAMICS

released about 350 sea urchins on a rock ledge around the
lower low water line on January 29, 2000, but this popula-
tion appeared to have declined to about 30 animals by April
7, 2000.

At each site, multiple stations were positioned to provide
samples at different nominal distances from the sea urchins.
At ChC, station 1 was within 1 m of a rock wall that was
inhabited by sea urchins during the autumn months; station
2 was on the shoreward end of a floating dock, 5 m straight
offshore of the wall: and station 3 was on the seaward end
of the same dock, about 13 m from the wall. The shallow
depth of station I ( 1 .4 m at MLW) allowed sampling at only
one depth (0.15 to 0.35 m above the substratum). At stations

2 and 3, we sampled the surface waters during each interval
(1.4 to 6.2 m above the substratum, depending on the tidally
variable water depth): at times of anticipated sperm pres-
ence (based on 1999 results) we also sampled the bottom
water 0.15 to 0.35 m above the substratum. During 1999.
only station 3 was sampled, and egg baskets were deployed
only near the surface. Because the sea urchins were free to
move, the positions of our stations relative to spawning
males could not be known precisely. However, likely loca-
tions can be inferred from sea urchin movement patterns. In
1999, sea urchins mainly remained on the rock wall or
wandered between stations 1 and 2. whereas in 2000 many
animals spent the spawning season on a piling adjacent to
station 2.

We employed a similar sampling scheme at C1C. with
minor modifications to accommodate local dock structures.
Station 1 was within 1 m of the rock ledge to which sea
urchins were transplanted: station 2 was 1 m straight off-
shore of station 1 (along a fixed wooden dock): and stations

3 and 4 were on floating docks about 12 m from station 1,
at 45° angles to either side of the transect from stations 1 to
2. Because of minimal water depth ( 1 .0 to 1 .4 m at MLW).
all stations were sampled at only a single depth (stations 1
and 2: 0.15 to 0.35 m above the substratum; stations 3 and
4: 0.4 to 3.5 m above the substratum, depending on the
tidally variable water depth). Stations 3 and 4 were sampled
only when sperm were expected to be present.

At each site, sets of three replicate egg baskets (spaced
— 0.1 m apart vertically) were deployed at each station and
depth and retrieved 24 h (1999 only), 48 h. or (on only three
occasions) 72 h later. In 1999. baskets contained 500 fil of
eggs ( — 7600 eggs) from one female, and in 2000 they
contained 800 /j.1 of eggs (mean number ± SE: 11216 ±
787 eggs) pooled from two to three females. Laboratory
controls (200 /xl of eggs in 10 ml aged seawater) were used
to determine the incidence of fertilization membranes prior
to basket deployment (presumably reflecting sperm contam-
ination) and at the time of retrieval (presumably reflecting
false membranes). To determine fertilization levels. 300
eggs per basket or vial were randomly subsampled and
scored in three categories: unfertilized, presence of a fertil-

ization envelope, or development through a later stage
(2-64 cells, unhatched/hatched blastula. gastrula). Eggs
with fertilization envelopes present were judged to have
been fertilized only if the sample also contained later de-
velopmental stages. From 41% of baskets (181 out of 441 ).
fewer than 300 eggs were retrieved; in these cases, all
retrieved eggs were scored. For the calculation of mean
fertilization levels, only baskets with more than 50 retrieved
eggs were used, resulting in a loss of replicates at some sites
and times.

We estimated the approximate distribution of fertilization
events during a sample interval from the distribution of
developmental stages in a sample and the known rate of
development to each stage. We used Stephens' (1972) de-
velopmental times for S. droebachiensis at 4°C from fertil-
ization to 32-cell stage (2-cell: 5 h: 4-cell: 8 h: 8-cell:
10.5 h; 16-cell: 14 h; 32-cell: 18 h). From the 64-cell stage
to gastrulation, we used our own observations of develop-
mental times (64-cell: 21 h; blastula: 24 h; hatching: 40 h;
early gastrula: 48 h). We calculated the distribution of
fertilizations (%) in time as the percent at each stage (i.e., of
a certain age, in h) of all embryos detected (pooled from
three replicate baskets).

To establish the extent to which spawning had occurred
during the 2000 sampling period, we collected sea urchins
for analysis of gonad index (wet weight of gonads as a
percentage of total wet body weight) from ChC l/i = 10)
and C1C (;; = 1 1 ) on April 7 and 11. 2000, respectively.

Results

Egg longevity

Egg viability in aged seawater in the laboratory (as as-
sayed by fertilization with fresh sperm) varied significantly
among time intervals and females in both experiments (Fig.
1). In experiment 1 (February 28 to March 2, 2000). the
effects of both Female (F3.3f) = 5.68, P = 0.003) and
Time (F2 36 = 8.94, P < 0.001 ) were significant, but the
interaction between the two main factors was not (F6 36 =
1.74. P = 0.14). Post-hoc comparisons revealed that
fertilization levels were significantly lower for female 2. but
similar for females 1. 3. and 4 (SNK-test. P < 0.05: Fig.
1A). Fertilization levels were highest at 0 h, similar at 24
and 48 h (SNK. P > 0.05 ). and significantly lower by 72 h
(SNK, P < 0.05). In experiment 2 (March 28 to April 1,
2000), there were again significant Female (F3 60 = 18.0.
P < 0.001 ) and Time (F4 60 = 273. P < 0.001 ) effects,
as well as a significant interaction between the two main
factors (F,,60 = 32.9, P < 0.001). Fertilization of eggs
from females 1 and 4 remained relatively high at 72 h. while
levels declined markedly for females 2 and 3 (Fig. IB). For
females 1 and 2. fertilizations dropped to very low levels by
96 h. while fertilizations for females 3 and 4 were higher at
96 h than at 72 h (Fig. IB). Of a total of 36 control sample-

S. K. MEIDEL AND P. O. YUND

A) February 28 - March 3. 2000

loo-,. — ,

ED"

1148

D :

B) March 28 - Apnl 1.2000

Female

Figure 1. Mean ( +SE) fertilization levels (%) over time of eggs from
four female sea urchins (A) at the beginning (experiment 1 ) and (B) in (he

middle (experiment 2) of the spawning season. Replication is four vials for
eac

I).

me expermen o e spawnng season. epcaon s

each female/time combination (except experiment 1 at 0 h: replication =

( 16 and 20 in experiments 1 and 2. respectively), 5 had 0.3%
false fertilization envelopes and 1 had 0.7%.

In spite of the significant variation among sample times
and females in both experiments, egg viability was basically
quite high for 48 to 72 h (Fig. 1). With the exception of
female 2 in experiment 1. more than 75% of eggs held in
aged seawater in the laboratory were viable for 48 h (Fig. 1 ).
At 72 h, viability was in the 50%-75% range for eggs from
6 of the 8 females (Fig. 1 ).

Cumulative fertilization level (15 min vs 3 li)

When eggs in baskets were exposed to a continuous
sperm supply from a spawning male, fertilization levels
increased from 15 min to 3 h at distances of 0.3 and 1.0 m
downstream from the male, but remained similar over time
at 2.6 m (Fig. 2). In the 15-min samples, fertilization de-
creased with distance from 0.3 to 1.0 m. but remained
similar between 1 and 2.6 m (Fig. 2). In the 3-h samples,
fertilization decreased monotonically with distance. The
two-way ANOVA indicated significant Time (F, ,9 =
31.3, P < 0.001) and Distance (F2 39 - 40.1, ' P <
0.001 ) effects, as well as a significant interaction between
the two main factors (F2__,9 = 4.87, P = 0.013). In 5 out

of 8 trials, the male still had sperm on its test at the end of
the 3-h deployment, suggesting that fertilization would have
continued well beyond the end of our sample interval.

Upstream controls for ambient sperm levels (Fig. 2) gen-
erally had SO. 3% fertilization except in trials 1, 6, and 7
when fertilization levels reached 5.3%, 9.0%, and 2.0%,
respectively. We attribute fertilizations in trial 6 to a large
boat wake that probably created oscillatory water motion
and transported sperm towards the upstream control sample
immediately before retrieval of the 15-min samples, and we
attribute fertilizations in trial 7 to false envelopes (see
below). Fertilizations in trial 1 could not be attributed to any
obvious cause, and the recorded value was subtracted from
the fertilization levels recorded in experimental baskets for
that trial.

The apparent absence of a decline in fertilization between
the 1- and 2.6-m samples at 15 min and the lack of an
increase in fertilization between the 15-min and 3-h samples
at 2.6 m are both attributable to one exceptional sample.
During trial 5. we recorded a fertilization level of 48% at
2.6 m at 15 min. while values in other trials ranged only
from 0.0%' to 3.3% (mean ± SE %: 1.4% ± 0.5%; n = 6)
at 15 min and from 3.7%- to 15.3% (6.9% ± 1.8%; n = 6)
at 3 h. If this outlier is excluded, fertilization declines from
1 to 2.6 m at 15 min and increases from 15 min to 3 h at
2.6 m.

In laboratory controls, fertilization levels were always
very high at the beginning (mean ± SE: 94.6% ± 1.7%;
/; = 8) and the end (94.8% ± 1.6%; n = 8) of a trial.
Controls for sperm contamination or false fertilization en-
velopes mostly indicated 0% envelopes (15 min, 0.3% ±
0.2%; 3 h, 0.5%. ± 0.4%; n = 8) except in trial 7 where

Distance from male (m)

*•

Current direction

Figure 2. Fertilization as a function of distance and duration of sperm
exposure in the field experiment. Mean ( +SE) fertilization levels (%) are
reported for each time/distance combination. Spawning male is located at
0.0 m mark. Upstream basket was retrieved after 3 h (hatched bar);
downstream baskets after 15 min (stippled bars) or 3 h. Replication is 8
trials, except 7 trials for 2.6 m after 15 min. and 6 trials for 2.6 m after 3 h.

SEA URCHIN FERTILIZATION DYNAMICS

\.T7c and 3.3% envelopes were found after 15 min and 3 h,
respectively. These percentages were subtracted from the
fertilization levels recorded in the Held for that trial.

Current velocities varied widely during trials 2 through 7
and ranged mainly from 0.08 to 0.20 m • s~ ' (Fig. 3). Mean
velocities varied 5-fold among trials during the initial 15-
min period (from 0.026 to 0.130 m • s~') but were quite
similar over 3 h (from 0.121 to 0.155 m • s~").

Sperm availability in nature

In both years of the survey (1999, 2000) and at both sites
(ChC, C1C), no fertilizations were recorded during most of
the sample intervals. However, in both years several sperm-
release events of variable magnitude were detected. In 1999
at ChC (only station 3 surface was sampled), fertilizations
occurred on March 5 (mean time-integrated fertilization
level 4.7% ). March 23 (57.3%). March 31 (6.6%). and April
1 (24.69r ). In 2000 at ChC, fertilizations occurred on Feb-
ruary 19 (station 1 only. 39.5%), March 10 (station 1.
10.3%; station 2, 9.3% surface; no bottom samples were
deployed and no fertilization was detected at station 3),
March 19 (station 1, 62.3%; station 2. 34.3% surface and
11.3% bottom; station 3. 30.4% surface and 5.3% bottom),
and March 29 (station 1, 3.4%-; station 2, 4.6% surface;
station 3, 4.5% surface; no bottom samples were deployed).
At C1C (sampled only in 2000), fertilizations were detected
on March 10 (station 1, 24.1%; no fertilization was detected

at station 2; stations 3 and 4 were not sampled), March 17
(station 1, 27.7%: station 2. 10.4%; station 3. 26.2%; station
4, 3.7%), and April 3 (station 1, 6.9%; station 2, 3.3%;
stations 3 and 4 were not sampled).

In laboratory controls, fertilization levels were always
very high at the start of each sample interval (mean ± SE
%: 1999, 96.7% ± 0.7%, // --•• 19; 2000, 93.8% ±
0.9%, n = 20). Controls for false fertilization envelopes
(stored in the laboratory and fixed upon retrieval of the
corresponding field sample) had very low levels of false
envelopes (1999, 0.8% ± 0.5%, n = 16; 2000, 0.2% ±
0.1%; n = 20).

Based on the distribution of developmental stages (two-
cell to early gastrula) at the time of collection, we estimated
that the temporal fertilization pattern varied markedly
among the major sperm release events that we detected
(Figs. 4-6). Because the discrete developmental stages that
we scored are separated by longer time intervals later in
development, the 24-h sample interval utilized in 1999 at
ChC produced far better resolution of the time of fertiliza-
tion (~3 h) than did the 48- to 72-h intervals employed in
2000 ( — 3-h resolution for the 24 h immediately preceding
sample collection, but —10 h for the portion of the interval
>24 h prior to collection). In 1999. fertilizations occurred in
fairly continuous trickles over about 48 h (March 3-5; Fig.
4A) or 24 h (March 22-24; Fig. 4B) or in two distinct pulses
of similar magnitude about 24 h apart (March 30-April 1;

0.25 -,

0.20-

0.15 -

0.10

0.05 -

0.00

Time (mm)

150

180

Trial

4
5

Figure 3. Current velocity (m • s ') during seven trials of the field fertilization experiment. Vertical dashed
line indicates 15-min interval. Trial 2: each point is one measurement; trials 3-8: each point is mean of 8
measurements collected as two sets of 4 measurements at 15-s intervals 5 min apart. Standard errors are omitted
for clarity.

S. K. MEIDEL AND P. O. YUND

60-

A) Mar 3-5

50-

n = 55

40-

30-

20-

n T

n 3° 16 8 Fl

10-

G 32 16 4
m m r^i i i

1 Ir— il: 1 : II :ln|::|

i i i i i i i i i i i i i i r

3/3 3/4 3/5

60-

64 B) Mar 22-24

50-

-.•

n = 352

40-

30-

20-

10-

nrl 4 ri

" i i i i t i i i i i i i i i i i i

3/22 3/23 3/24

G C) Mar 30 - Apr 1

50-

n = 200

40-

30-

20-

10-

^ n

\ i i i i i i r

OOOOOOOOOOOOOOOOO

ooooooooooooooooo

rl </-, co — ^t ^ '-o O* r) i/-i oo — -t rr> ^ C1 <N

— — — CN rj

3/30 3/3 1 4/ 1

Time and date (1999)

Figure 4. Temporal distribution of fertilizations at Christmas Cove
station 3 (surface) during 1999 sample intervals. (A) March 3-5. (B) March
22-24. (C) March 30-April 1. n, total number of embryos counted per
station (pooled from two to three baskets). Developmental stage corre-
sponding to each inferred fertilization time is indicated above the bar (2. 4,
8. 16. 32. and 64. number of cells; B, unhatched blastula; G. gastrulal.
When sperm availability extended through two sample intervals (A. C).
fertilization times are inferred from both samples.

Fig. 4C). During the most widespread sperm release event
in 2000 (March 17-19). fertilizations at 4 of the 5 ChC
station/depth combinations were fairly evenly spread over
about 24 h (Fig. 5 A, B, D, E). but at the fifth station/depth
combination virtually all sperm arrived during a much
shorter time interval (Fig. 5C). Although the distribution of
fertilizations in three apparent pulses at the four stations and
depths (Fig. 5 A. B. D. E) is an artifact of lower temporal
resolution later in development, resolution was nevertheless

sufficient to distinguish fairly consistent sperm availability
from a single, shorter pulse (Fig. 5C). During the corre-
sponding fertilization event at C1C (March 15-17. 2000).
fertilizations at stations 1 to 3 were also distributed over
about 24 h (Fig. 6B-D). At CIC. fertilizations in the March
8-10. 2000 event at station 1 occurred in two major and one
minor pulse spread over about 27 h (Fig. 6A).

Sea urchins collected at the end of the field survey at ChC
and CIC had intermediate to high gonad indices relative to
levels previously recorded for 5. droebachiensis off the
Maine coast (Cocanour and Allen, 1967). Mean gonad
indices (± SE) were as follows: ChC females, 19.9% ±
4.7% (n ---- 3), males, 12.3%' ± 2.5% (n == 7); CIC
females, 14.2% ± 8.8% (n = 5), males, 9.1% ± 7.0%
in -- 6). Consequently, additional spawning is likely to
have taken place later in the season, after sampling ceased.

Discussion

More than 75% of eggs of the temperate sea urchin
Strongylocentrotus droebachiensis were generally viable
for 48 h when kept in the laboratory in aged (but otherwise
untreated) seawater. and viability through 72 h ranged from
50% to 75% in most females (Fig. 1 ). Subsamples from later
time intervals that were isolated prior to formaldehyde
addition continued to develop normally through gastrulation
(unpub. data). Consequently, fertilization appears to be a
reasonable assay of true egg longevity, and does not merely
indicate a prolonged ability to elevate a fertilization enve-
lope. Overall, our egg longevity values are greater than
earlier estimates of 8 h in sterilized seawater (Wahle and
Peckham. 1999) and 24 h in filtered seawater (Pennington.
1985), but shorter than the 1-2 weeks for sea urchin eggs
kept under axenic conditions (Epel ct <//., 1998). Variation
both within and among studies, coupled with observations
of egg damage in our field surveys, suggests that egg
viability is not static but is instead affected by a combina-
tion of endogenous and exogenous factors. Variation in egg
longevity among the different females in our laboratory
experiment (Fig. 1 ) illustrates the presence of endogenous
individual variation. Epel el al. (1998) attribute the extreme
egg longevity in their study to the removal of bacterial
contaminants that can cause the lysis of eggs under labora-
tory conditions. We observed another form of exogenous
damage to eggs in some field samples subject to rough
weather (pers. obs.). especially when sediment particles
became trapped in the egg baskets. Although damage from
sediment abrasion may simply represent a basket artifact, it
may also be indicative of a type of egg damage that occurs
in nature. Factors controlling egg longevity may ultimately
prove to play a significant role in determining fertilization
levels in natural spawns.

When we exposed eggs to sperm slowly diffusing from a
spawning male's spines and tests (which functionally pro-

SEA URCHIN FERTILIZATION DYNAMICS

oo -

A) Station 1

it = 561

80-

60-

40-

20-

n -

3/17

3/18

3/19

100 -

B) Station 2 (surface)

C) Station 2

-r:

H = 229

(bottom)

80-

n= 102

40-

20-

1 i

i i i i i i i i

1 1 1 1 1 1 1 1

1 1 I I I I I I

1 1 1 1 1 1 1 I

3/17 3/18

3/19

3/17 3/18

3/19

100 -

D) Station 3 (surface)

E) Station 3 (bottom)

n = 259

n = 29

80-

60-

40-

H B
G :•: n

20-

1 1 1

0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 I I I I I I I I I I I 1 1 1 1 I I I I

OOOOOOOOOOOOOOOOO OOOOOOOOOOOOOOOOO
OOOOOOOOOOOOOOOOO OOOOOOOOOOOOOOOOO
cl "~. oo — ^* <*"' <o o* r] u~. oo — *t f~: O C1' (^l n ^~. oo — ^f fi o CT rj Vi oo — ^ *^"< ^o w^ f~J

3/17

3/18

3/19

3/17

3/18

3/19

Time and date (2000)

Figure 5. Temporal distribution of fertilizations at Christmas Cove during sample interval March 17-19.
2000. (A) Station 1. (B) Station 2 (surface). (C) Station 2 (bottom). (D) Station 3 (surface). (E) Station 3
(bottom). ;i, total number of embryos counted per station (pooled from three baskets). Developmental stage
corresponding to each inferred fertilization time is indicated above the bar (B, unhatched blastula: H. hatched
blastula; G, gastrula).

longs male spawning duration, even though sperm release
per se may have been short in duration), we recorded higher
fertilization levels with time at most downstream locations
(Fig. 2). Even though current velocities at our experimental
site were often considerable and always exceeded the 0.13
m • s~' sperm diffusion threshold suggested by Thomas
( 1994) during at least some portion of each trial (Fig. 3),
62% of our males still had substantial sperm clinging to
their spines and tests when retrieved at the end of the 3-h
period (similar observations are reported in Pearse et al..

1988. for a female S. droebachiensis). Hence even our 3-h
experiment probably underestimates the total time-inte-
grated fertilization levels of fixed-position eggs downstream
of a spawning male. Short-term fertilization experiments
that use sperm-filled syringes to mimic males (Pennington.
1985; Levitan, 1991: Levitan et al.. 1992; Wahle and Peck-
ham. 1999) completely bypass this effect.

The relatively long period of egg viability in 5. droe-
bachiensis makes it feasible to use sperm-permeable baskets
of eggs to assess spatial and temporal patterns of sperm

S. K. MEIDEL AND P. O. YUND

60-

A) Mar 8-10

50-1

Station 1

—

11= 172

40-

30-

20-

10-

1
3/

i i
8

i i i i

3/9

i i i i i i i i
3/10

60-

B)Mar 15-17

50-

Station 1

n = 217

40-

. i

30-

20-

10-

3/17

60-

OMar 15-17

50-

Station 2

n = 92

40-

30-

20-

10-

1 1 1

1 1 1 I

i i i i i i i i

3/15

3/16

3/17

60-

H D) Mar 15-17

50-

—

Station 3

40-

,, = 42

30-

20-

10-

0 1 1 1 1 1 1 I i i I i i I I I I I

— — — D (^

3/15 3/16 3/17

Time and date (2000)

Figure 6. Temporal distribution of fertilizations at Clarks Cove during
sample intervals March 8-10 (A) and March 15-17, 2000 (B-D). (A) Station
1. (B) Station 1. (C) Station 2. (D) Station 3. n, total number of embryos
counted per station (pooled from three baskets). Developmental stage corre-
sponding to each inferred fertilization time is indicated above the bar (32 and
64, number of cells; B, unhatched blastula; H, hatched blastula; G, gastnila).

availability in nature. Our preliminary application of this
method detected several sperm-release events in two small
populations, one occurring naturally (ChC) and one estab-

lished experimentally (C1C). Several features of the de-
tected sperm-release events are noteworthy. First, total
time-integrated fertilization levels were highly variable,
ranging from 3.3% to 62% fertilization (when sperm were
detected). We emphasize that our experimental design is a
hybrid between an experiment and a true survey, because
egg position was under experimental control but sperm
release occurred naturally. Furthermore, we do not know the
actual location of the male or males that spawned, though
repeated observations of the distribution of sea urchins
during the survey suggest that spawners were likely to be
near stations 1 or 2 at both sites (animals were never present
at station 3 at ChC, or stations 3 or 4 at C1C). Given these
considerations, great care should be exercised when inter-
preting the absolute fertilization levels reported here. Vari-
ation in fertilization levels among sample dates probably
reflects the number and proximity of spawning males, but it
may be erroneous to conclude that either the higher or lower
levels assayed truly represent fertilization levels in natural
spawns.

Second, the spatial sampling scheme adopted during
2000 permits some inferences about the spatial scale of
sperm availability. Some sperm-release events appear to be
highly localized (e.g.. ChC, February 19, 2000, station 1
only; C1C. March 10, 2000, station 1 only), with eggs at one
station fertilized while eggs a few meters away were not.
These sperm distributions are consistent with a pattern of
localized sperm availability as indicated by field fertiliza-
tion experiments conducted with sea urchins (Pennington,
1985; Levitan, 1991; Levitan el al., 1992; Wahle and Peck-
ham, 1999). At other times sperm were present throughout
much larger areas (e.g., March 19, 2000, at ChC, March 17,
2000. at C1C). During these widespread sperm-availability
events, fertilization levels were often appreciable in much of
the site (along a 12-m linear transect at ChC, and within an
~72-m2 triangle at C1C). Regardless of the actual location
or number of males spawning, the spatial distribution of
fertilizations is more extensive than predicted by simple
field fertilization experiments (though more consistent with
predictions from a whole-population spawning model; Levi-
tan and Young, 1995).

Third, the single sample that provides fertilization levels
at different depths at multiple stations (ChC stations 2 and
3, March 19, 2000) indicated higher fertilization levels near
the surface than near the bottom. At least close to shore in
shallow water, spawned sperm may tend to be concentrated
near the surface rather than near the bottom. The distribu-
tion is particularly interesting because eggs are negatively
buoyant and hence generally assumed to remain near the
bottom.

The distribution of developmental stages in retrieved
samples during our 1999 field survey indicated that fertili-
zations during natural sperm release events may accrue over
as much as 48 h (Fig. 4). The decreased temporal resolution

SEA URCHIN FERTILIZATION DYNAMICS

in our 2000 survey nevertheless produced temporal patterns
that were consistent with fertilization over about 24 h in
most samples (Figs. 5, 6). Fertilizations that occur over
extended time periods could be the result of continuous
extrusion of sperm from a single male's gonopores, the
diffusion of sperm from a gamete reservoir that has accu-
mulated on a male's test, or spawning by multiple males at
different times.

Our approach of using egg baskets in field experiments
and surveys can potentially be criticized because eggs were
held stationary at relatively high concentrations instead of
being allowed to move and disperse with the currents. If
eggs are rapidly transported away from the female during
natural spawns, and thus quickly diluted, egg longevity
would be less important in determining cumulative fertili-
zation levels than suggested by our results. But because
eggs are spawned in a viscous mass that tends to remain on
the test (Thomas. 1994). restraining eggs in baskets may
adequately approximate natural spawns and provide a good
estimator of fertilization levels under a range of flow con-
ditions. Future work should address the outstanding ques-
tion of where eggs are actually fertilized: in the egg mass on
a female's test (i.e., a fixed location), as they transition from
that mass into the water column (still essentially a fixed
location), or in the mainstream of flow. The answer to this
question is likely to vary with habitat and flow regime.

Our study suggests that details of the reproductive biol-
ogy of sea urchins can potentially have considerable effects
on fertilization levels in the field and that caution should be
used when extrapolating fertilization levels in natural
spawns from experiments that circumvent these apparent
adaptations. We suggest that successful fertilization in sea
urchins may result not only from short-term exposure to
highly concentrated sperm from a nearby male, but also
from long-term exposure to more dilute sperm from a num-
ber of more distant males. This proposed scenario is similar
to our understanding of fertilization in brooding inverte-
brates with mechanisms to capture dilute sperm (e.g., Yund.
1998; Bishop, 1998), and has also been suggested for tube-
dwelling broadcasters that can move sperm-containing wa-
ter past spawned eggs (M. E. Williams and M. G. Bentley.
University of St. Andrews, Scotland, unpub. obs.). We
suggest that fertilization in other broadcast-spawning inver-
tebrates may not be fundamentally all that different.

Acknowledgments

We thank Tim Miller and the staff of the Darling Marine
Center for their assistance. Many thanks also go to Matt
Babineau and Amy Gilbert whose help in the lab and field
was invaluable. We are also grateful to Rick Wahle for
many useful discussions throughout this study, to Ed Myers

for access to his dock at the C1C site, and to two anonymous
reviewers for helpful comments on an earlier version of this
manuscript. Funding was provided by the National Science
Foundation (OCE-97-30354). This is contribution no. 364
from the Darling Marine Center.

Literature Cited

Babcock, R. C., C. N. Mundy, and D. Whitehead. 1994. Sperm
diffusion models and in situ confirmation of long-distance fertilization
in the free-spawning asteroid Acanthaster planci. Biol. Bull. 186:
17-28.

Bishop, J. D. D. 1998. Fertilization in the sea: Are the hazards of
broadcast spawning avoided when free-spawned sperm fertilize re-
tamed eggs'1 Prm: R. Sue. Lund. B 265: 725-73 1 .

Chia, F.-S., and L. R. Bickell. 1983. Echinodermata. Pp. 545-620 in
Reproductive Biology of Invertebrates, Vol. II: Spermatogenesis and
Sperm Function. K. G. Adiyodi and R. G. Adiyodi, eds. John Wiley,
New York,

Cocanour, B., and K. Allen. 1967. The breeding cycles of a sand dollar
and a sea urchin. Comp. Biochem. Phvsiol. 20: 327-331.

Denny. VI. W., and M. F. Shibata. 1989. Consequences of surf-zone
turbulence for settlement and external fertilization. Am. Nat. 134:
859-889.

Epel. D., M. Kaufman. L. Xiao, H. Kibak. and C. Patton. 1998.
Enhancing use of sea urchin eggs and embryos for cell and develop-
mental studies: method for storing spawned eggs for extended periods.
Mol. Biol. Cell. 9: 182a.

Himmelman, J. H. 1975. Phytoplankton as a stimulus for spawning in
three marine invertebrates. J. Exp. Mar. Biol. Ecol. 20: 199-214.

Lasker, H. R., D. A. Brazeau. J. Calderon, M. A. Coffroth, R. Coma,
and K. Kim. 1996. In situ rates of fertilization among broadcast
spawning gorgonian corals. Biol. Bull. 190: 45-55.

Levitan, D. R. 1991. Influence of body size and population density on
fertilization success and reproducthe output in a free-spawning inver-
tebrate. Biol. Bull. 181: 261-268.

Levitan, D. R. 1995. The ecology of fertilization in free-spawning in-
vertebrates. Pp. 123-156 in Ecology of Marine Invertebrate Larvae.
L. R. McEdward. ed. CRC Press. Boca Raton, FL.

Levitan, D. R., and C. Petersen. 1995. Sperm limitation in the sea.
Trends Ecol Evol. 10: 22S-231.

Levitan, D. R., and C. M. Young. 1995. Reproductive success in large
populations: empirical measures and theoretical predictions of fertili-
zation in the sea biscuit Clypeaster rosaceus. J. Exp. Mar. Biol. Ecol.
190: 221-241.

Levitan, D. R., M. A. Sewell. and F.-S. Chia. 1991. Kinetics of
fertilization in the sea urchin Strongylocentrotus franciscanus: inter-
action of gamete dilution, age. and contact time. Biol. Bull. 181:
371-378.

Levitan, D. R., M. A. Sewell. and F.-S. Chia. 1992. How distribution
and abundance influence fertilization success in the sea urchin Strongy-
locentrotus franciscanus. Ecology 73: 248-254.

Pearse. J. S., D. J. McClary, M. A. Sewell, W. C. Austin, A. Perez-
Ruzafa. and VI. Byrne. 1988. Simultaneous spawning of six species
of echinoderms in Barkley Sound. BC. Invenebr. Reprod. Dev. 14:
279-288.

Pearson, G. A., and S. H. Brawley. 1996. Reproductive ecology of
Fucus distichus (Phaeophyceae): an intertidal alga with successful
external fertilization. Mar. Ecol. Prog. Ser. 143: 211-223.

Pennington. J. T. 1985. The ecology of fertilization of echinoid eggs:

S. K. MEIDEL AND P. O. YUND

the consequences of sperm dilution, adult aggregation, and synchro-
nous spawning. Bio/. Bull. 169: 417-430.

Petersen, C. W., R. R. Warner, S. Cohen, H. C. Hess, and A. T. Sewell.
1992. Variable pelagic fertilization success: implications for mate
choice and spatial patterns of mating. Ecology 73: 391-401.

Starr, M., J. H. Himmelman, and J.-C. Therriault. 1990. Direct
coupling of marine invertebrate spawning with phytoplankton blooms.
Science 247: 1071-1074.

Starr, M., J. H. Himmelman, and J.-C. Therriault. 1992. Isolation and
properties of a substance from the diatom Phaeodactylum tricornutum
which induces spawning in the sea urchin Strongylocentrotus droe-
bachiensis. Mar. Ecol. Pmg. Ser. 79: 275-287.

Starr, M., J. H. Himmelman, and J.-C. Therriault. 1993. Environ-
mental control of green sea urchin, Strongylocentrotus droebachiensis,

spawning in the St. Lawrence estuary. Con. J. Fish. At/iiat. Sci. 50:

894-901.
Stephens, R. E. 1972. Studies on the development of the sea urchin

Strongylocentrotus droebachiensis. 1. Ecology and normal develop-
ment. Bio/. Bull. 142: 132-144.
Thomas, F. I. M. 1994. Physical properties of gametes in three sea

urchin species. J. Exp. Bin/. 194: 263-284.
Wahle, R. A., and H. Peckham. 1999. Density-related reproductive

trade-offs in the green sea urchin Strongylocentrotus droebachiensis.

Mar. Bio/. 134: 127-137.
Vund, P. O. 1998. The effect of sperm competition on male gain curves

in a colonial marine invertebrate. Ecology 79: 328-339.
Yund. P. O. 20110. How severe is sperm limitation in natural populations

of marine free-spawners? Trends Ecol. Evol. 15: 10-13.

Reference: Bio/. Bull. 201: 95-103. (August 2001)

Biogeography of Asterias: North Atlantic
Climate Change and Speciation

JOHN P. WARES
Duke University Zoology, Box 90325, Durham. North Carolina 27708

Abstract. Fossil evidence suggests that the seastar genus
Asterias arrived in the North Atlantic during the trans-
Arctic interchange around 3.5 Ma. Previous genetic and
morphological studies of the two species found in the At-
lantic today suggested two possible scenarios for the spe-
ciation of A. ntbens and .4. forbesi. Through phylogenetic
and population genetic analysis of data from a portion of the
cytochrome oxidase I mitochondrial gene and a fragment of
the ribosomal internal transcribed spacer region. I show that
the formation of the Labrador Current 3.0 Ma was probably
responsible for the initial vicariance of North Atlantic As-
terias populations. Subsequent adaptive evolution in A.
forbesi was then possible in isolation from the European
species A. rubens. The contact zone between these two
species formed recently, possibly due to a Holocene found-
ing event of A. rubens in New England and the Canadian
Maritimes.

Introduction

The North Atlantic Ocean is populated by hundreds of
taxa which invaded from the North Pacific following the
opening of the Bering Strait about 3.5 million years ago
(Ma; Durham and MacNeil, 1967; Vermeij. 1991 ). Some of
these species have maintained genetic contact with source
populations in the Pacific until recently (Palumbi and Kess-
ing, 1991; van Oppen et ai. 1995), but many of them have
subsequently differentiated from the source populations and
are now recognized as distinct species (e.g.. Gosling, 1992;
Reid et ai. 1996; Collins et al.. 1996). Circumstantial
evidence suggests strongly that the seastar genus Asterias
(Echinodermata: Asteroidea: Asteriidae: Asteriinae) partic-
ipated in the trans-Arctic interchange (Worley and Franz,

Received 28 September 2000; accepted 10 May 2001.
Current address: Dept. of Biology, University of New Mexico, Castetter
Hall, Albuquerque, NM 87131. E-mail: jpwares@unm.edu

1983; Vermeij. 1991 ). Today, two species are recognized in
the North Atlantic; A. forbesi on the North American coast,
primarily from Cape Hatteras to Cape Cod (Franz et al..
1981 ). and A. rubens on the American coast primarily from
Cape Cod northward (Franz et al., 1981), and on the Euro-
pean coast from Iceland to western France (Clark and
Downey. 1992; Hay ward and Ryland. 1995). American
populations of A. rubens have been previously described as
A. vulgaris, a junior synonymy (Clark and Downey. 1992).
These species co-occur over a broad range of the North
American continental shelf centered on Cape Cod (Gosner.
1978; Menge, 1979; Franz et al.. 1981).

Two current hypotheses attempt to explain the recent
speciation between A. forbesi and A. rubens. Schopf and
Murphy (1973) suggested that they were a germinate spe-
cies pair formed by a late Pleistocene (0.02-2.5 Ma) vicari-
ance event (i.e., a separation of populations) at Cape Cod.
possibly due to lower sea levels during glacial maxima.
There is some evidence for hybridization between these
seastars. but the separation could be maintained by localized
adaptation to the different thermal regimes north and south
of Cape Cod (Franz et al., 1981). However, this thermal
boundary was latitudinally unstable throughout the Pleisto-
cene (Cronin. 1988) and only in the past 20,000 years
(Holocene) has it returned to its current state. If the geo-
graphical isolation between these taxa was recent, as pro-
posed in Schopf and Murphy (1973), then strong natural
selection within each region has prevented widespread hy-
bridization.

The second hypothesis, based on morphological and pa-
leoceanographic evidence, suggested a late Pliocene (ap-
proximately 2.5-5 Ma) separation of Asterias into distinct
North American and European species, followed by a Ho-
locene recolonization of North America by the European
species A. niben* (Worley and Franz, 1983) This hypoth-
esis would therefore suggest that the differentiation between

J. P. WARES

Table 1

Collection sites for individuals of each species in this study

Species [Population]

Location

Sample size

A. rnhens [North America] Maine (44:N. 69°W) 12

Nova Scotia (46°N. 62° W) 6

Newfoundland (50°N, 55°W) 5

.4. mbens [Europe] Iceland (64°N. 22°W) 2

Norway (63°N, 10°E) 8

Ireland (53°N, 10:E) 10

France (48°N. 3°E) 5

A.forbesi North Carolina (34°N, 76°W) 5

Cape Cod (41°N, 70° W) 3

A. tinnirensis Sea of Japan (43°N, 131 °E) 4

Leprasrerius s/>. Iceland (64°N. 22°W) 1

Voucher specimens are being maintained in the marine invertebrate
collections of C. W. Cunningham at Duke University.

the two Atlantic species is entirely due to long-term isola-
tion. Thus, subsequent physiological adaptations to warmer
water in A. forbesi (Franz et ai, 1981 ) are independent of
the speciation event. Essentially, the distinction between
these species reflects either primary divergence due to se-
lection or secondary contact following vicariance (Endler.
1977).

In this study, mitochondrial and nuclear sequence data
were collected from populations of A.forbesi and A mbens
throughout North America and Europe, as well as from
populations of the Pacific sister taxon A. tinnirensis (Clark
and Downey. 1992). Phylogenetic and population genetic
assays were used to test the hypotheses described above. It
appears that Worley and Franz (1983) were remarkably
accurate in suggesting a Pliocene speciation followed by a
recent invasion of A. mbens from Europe, even in their
prediction of details of timing, mechanisms, and effects.
Although selection may have driven some of the diver-
gence, it now seems clear that the initial separation of A.
mbens and A. forbesi is due to late Pliocene changes in
climate and ocean current flow, whereas North American
populations of A. mbens are very recent arrivals.

Materials and Methods

Asterias specimens were collected from intertidal sites
listed in Table 1 . Tube feet were immediately placed in 95%
ethanol or DMSO buffer (0.25 M EDTA pH 8.0. 20%
DMSO, saturated NaCl; Seutin et cil., 1991). Species were
identified on the basis of key morphological characters
described in Clark and Downey (1992) and Hay ward and
Ryland (1995).

DNA extraction and amplification

DNA was phenol-extracted from each specimen follow-
ing the protocol in Hillis et ul. (1996). These extractions

were stored at — 80°C. PCR amplification of an approxi-
mately 700-bp portion of the mitochondrial cytochrome c
oxidase I (COI) protein-encoding gene was performed using
the primers LCO1490 and HCO 2198 from Folmer et al.
(1994). Amplification was performed in 50-;u,l reactions
containing 10-100 ng DNA, 0.02 mM each primer, 5 jul
Promega 10X polymerase buffer, 0.8 mM dNTPs (Pharma-
cia Biotech), and 1 unit Taq polymerase (Promega). Reac-
tions took place in a Perkin-Elmer 480 thermal cycler with
a cycling profile of 94: (60 s) -40° (90 s) -72° ( 150 s) for
40 cycles. The internal transcribed spacer (ITS) region was
amplified under similar conditions, with an annealing tem-
perature of 50°C and with primers ITS4 and ITS5 (White et
ul.. 1990). For each individual, sequences were obtained for
three to four clones, and the consensus sequence was ob-
tained to eliminate Taq error.

PCR products were prepared for sequencing and were
cycle-sequenced as in Wares (2001) using both PCR prim-
ers. COI sequences representing each individual in this
study have been deposited with GenBank (AF240022-
240081 ); ITS sequences were only obtained for 10 individ-
uals, representing each species and region, and are also
accessible in GenBank (AF346608-AF346617). Sequences
were aligned and edited for ambiguities using complemen-
tary fragments in Sequencher 3.0 (Genecodes Corp., Cam-
bridge, MA). No gaps or poorly aligned regions occurred in
the COI alignment, but missing characters were trimmed
from the ends of the alignment to produce equal sequence
lengths for all individuals. In the ITS alignment, all missing
or ambiguous characters, including gaps, were removed.
Consensus sequences were exported as a NEXUS file for
subsequent analysis in PAUP*4.0b4a (Swofford. 1998).

Phylogenetic analysis

A heuristic search for the set of most-parsimonious trees
based on the COI data was performed using PAUP*4.0b4a
(Swofford, 1998). Trees were rooted using Leptasterias
polaris (Asteriinae) and individuals of A. tinnirensis. Start-
ing trees were obtained via stepwise addition, with simple
addition sequence. Tree-bisection-reconnection was used
for branch swapping, and branches were collapsed if the
maximum branch length was zero.

Maximum-likelihood (ML) phylogenies were also gener-
ated in PAUP*. The best-fit model for all likelihood anal-
yses (HKY with F-distributed rate variation; Hasegawa et
ai, 1985; Yang, 1994) was determined by adding parame-
ters until the likelihood description of the neighbor-joining
tree did not significantly improve (Goldman. 1993; Cun-
ningham et <//., 1998), using the likelihood-ratio test of
ModelTest (Posada and Crandall, 1998). A series of boot-
strap replicates (100 ML replicates, heuristic search) using
PAUP* were performed to determine support for interspe-
cific relationships in the clade. Estimates of the transition-

BIOGEOGRAPHY OF ASTER/AS

transversion ratio for the HKY model, along with the
gamma-distributed parameter for among-site rate heteroge-
neity, were held constant for bootstrap replicates. A maxi-
mum likelihood phylogeny of the ITS sequence data was
also generated using the appropriate best-fit model (F81:
equal rates among sites, unequal base frequencies).

Estimates of speciation time within the North Atlantic
require an estimate of the mutation rate (/n). Because pale-
ontological evidence suggests that Asterias arrived in the
North Atlantic during the trans-Arctic interchange about 3.5
Ma (Worley and Franz, 1983; Vermeij, 1991). and because
climatic changes shortly thereafter would have prevented
additional trans-Arctic migration, this date was used to
calibrate the divergence between the Pacific species A.
iiinurensis and the North Atlantic taxa. Other species, in-
cluding the echinoderm Strongylocentrotus pullidus. have
clearly maintained more recent connections across the Arc-
tic (Palumbi and Kessing. 1991). However. S. pallidus
appears to be more tolerant of Arctic conditions than Aste-
rias (Worley and Franz. 1983: Palumbi and Kessing. 1991 ).

The ML estimate of the internal branch length separating
the sister taxa (representing net nucleotide divergence d.
Nei and Li, 1979) was used to estimate the appropriate
mutation rate /u, (Edwards and Beerli, 2000), where ju. = 0.5
<//(3.5 Ma). Estimates were obtained for the full COI data
set (first, second, and third codon positions), as well as third
position only. Use of the third-position estimate circum-
vents problems with branch length estimation when there is
strong rate variation (Wares and Cunningham, in press), as
well as problems with the potential influences of non-
neutral evolution.

Haplotype networks may be more appropriate represen-
tations of genealogical relationships within species than are
outgroup-rooted phylogenetic trees, because ancestral hap-
lotypes are still present in the population (Crandall and
Templeton. 1996). Methods associated with haplotype net-
works were used to determine the root haplotype for A.
nihens. Determination of the root haplotype prevents spu-
rious conclusions about ancestry among populations. Net-
works were created using a parsimony criterion in the
program TCS (alpha version 1.01, Clement et al., 2000); at
the same time, a Bayesian analysis of the likelihood that
parsimony is violated (Templeton el ui, 1992) was per-
formed to ensure that the data set was unlikely to be com-
plicated by homoplasy.

The ML root was determined using GeneTree (Griffiths
and Tavare, 1994); the likelihood of each possible rooted
gene tree was determined under an infinite-alleles model.
This model assumes that there are no multiply substituted
nucleotide sites. The method allows for receding of char-
acters so that independent substitutions are analyzed sepa-
rately, but this was not an issue with the A. rubens COI data.
The relative likelihood of each tree in comparison with all

other possible rooted trees was calculated using 107 simu-
lations in GeneTree.

Tests of rate constancy

Likelihood-ratio tests (Felsenstein. 1988; Goldman,
1993) were used to test the hypothesis that the data collected
were consistent with a constant-rate Poisson-distributed
process of substitution (molecular clock). This procedure
ensures that the data can be used to estimate the time of
divergence between A. rubens and A. forbesi. The ML
phylogeny was estimated using the best-fit model, and then
the likelihood of this phylogeny was recalculated while
constraining the estimate to fit the molecular clock model.
These likelihood (L) estimates were used to calculate the
^-distributed test statistic 8 = 2[ln(L0) - ln(L,)]. with
(n - 2 ) degrees of freedom where n is the number of taxa
in the tree.

Neutrality tests

Because adaptive selection may have played a role in the
divergence between A. rubens and A. forbesi (given a short
divergence time; Schopf and Murphy. 1973), polymorphism
data for each species were input to DNAsp v.3.5 (Rozas and
Rozas, 1999) to test for patterns of non-neutral evolution.
Within each species. Tajima's ( 1989) test generates a beta-
distributed parameter indicating the difference in two esti-
mates (polymorphic sites and number of alleles) of diver-
sity. Significantly low statistics can indicate non-neutral
evolution (Tajima, 1989). Additionally, a McDonald-Kreit-
man test (McDonald and Kreitman. 1991) was performed on
each pairwise set of species polymorphism data to deter-
mine whether selection has played a role in the divergence
between A. rubens and A. forbesi. Also, DNAsp was used to
calculate haplotype diversity (//, see eqn. 8.4 in Nei. 1987)
and sampling variance for each species or population.

Results

The COI data set (60 individuals) includes 627 charac-
ters, of which 484 are constant, 49 are parsimony-uninfor-
mative, and 94 are parsimony-informative. Base frequencies
are 33.7% A. 19.6% C. 21.6% G, and 25.1% T for this
fragment. Most of the substitutions (92.3%) are at third-
position sites; overall, 63% of all third-position characters
are polymorphic. These third-position sites are heavily AT-
biased (39.0% A. 15.1%- C, 1 1.8% G, and 33.9% T).

The best-fit model (HKY + F) was used to estimate
distances among individuals to determine whether there is
any evidence for saturation at third-position characters in
the COI coding region. A plot of pairwise genetic distances
versus number of third-position substitutions does not indi-
cate any pattern of saturation (data not shown); in fact, all of
the information within each species is based on third-posi-

J. P. WARES

tion substitutions. Additionally, the best-fit model was re-
estimated for this character partition; likelihood-ratio tests
indicate that the HKY model with invariant sites (/ =
0.213) and no rate variation describes the third-position
data effectively. The Asterias data sets do not reject the
molecular clock model, whether all positions are considered
(P = 0.163). or only third positions (P = 0.231).

Maximum-likelihood analysis was used to determine the
interspecific gene tree, using all codon positions and the
HKY + T model (Tr:Tv 8.256, a = 0.0608. four rate
classes). The ML tree (L = 1472.87) is presented in Figure
1A, including all individuals sampled within A. annirensis,
A. nibens, and A. forbesi. Bootstrap support is indicated on
the tree, with each species being fully resolved in 100% of
replicates. The Pacific species A. amurensis is basal to a
strongly supported clade of Atlantic species in this phylog-
eny.

Following exclusion of missing and ambiguous charac-
ters in the ITS data set (length of fragment varies from 413
to 482 bases when gaps included), these data include 368
characters of which 343 are constant. 1 is parsimony-unin-
formative. and 24 are parsimony-informative. Indels did not
vary within species and were removed (analysis with
gapped characters included produced nearly identical re-
sults). Parsimony analysis produced a single most-parsimo-
nious tree of 25 steps, and the ML phylogeny (best-fit model
F81, no rate variation) is shown in Figure IB. Under a
variety of mutational models, this phylogeny is statistically
indistinct from the COI phylogeny in Figure 1A. Likeli-
hood-ratio tests indicate that, in addition to a similar inter-
specific topology, branch lengths on the COI and ITS phy-
logenies are proportional (P > 0.10), though the substitution
rate is significantly different (P < 0.05). Bootstrap replicates of
the ITS data also indicate strong support for differentiation
among these species. The ITS data do not reject a molecular
clock model.

Divergence among these species is indicated in Table 2.
HKY -I- F distances in the COI fragment indicate that A.
amurensis, A. forbesi, and A. nibens have been isolated
from each other for a similar amount of time; assuming
trans-Arctic isolation around 3.5 Ma. A. nibens and A.
forbesi have been separated for at least 3.0 Ma. Although
the estimated divergence date is higher when all codon
positions are included (Table 2), and these data do not reject
a molecular clock, neutrality tests (see below) suggest that
some second-position substitutions may be under selection.
Therefore, third-position sites may be more appropriate for
the divergence estimate. The estimated divergence time is
also higher when the ITS data are used; however, there is no
reason to believe that speciation predated the appearance of
Asterias in the North Atlantic, and the long branch leading
to A. forbesi is not easily explained since it appears in both
phylogenies (one using a protein-coding gene, one using
untranslated spacer region data). This longer branch appears

to influence the age estimates of the COI (all positions) and
ITS data sets strongly.

A McDonald-Kreitman test (McDonald and Kreitman,
1991) rejects a pattern of neutral substitution between A.
nibens and A. forbesi (P < 0.01 , Table 3). Despite branch
lengths that do not reject the molecular clock model, there
is an excess of amino acid replacement substitutions be-
tween the Atlantic species. The replacement substitutions
between A. nibens and A. forbesi do not include any first-
position substitutions. Half (8/16) of the amino acid substi-
tutions do not involve a change in charge or polarity,
whereas almost half (7/16) of the changes substitute a basic
residue for an uncharged or nonpolar residue. However,
there does not seem to be an obvious pattern to these
changes between A. nibens and A. forbesi. Other species
comparisons do not reject the neutral model of substitution
(Table 3). Within each species, Tajima's (1989) test is
nonsignificant (A. amurensis, D = 0.837, P > 0.10; A.
forbesi, D = -0.705, P > 0.10; A. nibens, D = -1.482,
P > 0.10), indicating that there is no reason to suspect
non-neutral evolution in the intraspecific comparisons.

Additionally, Bayesian analysis (Templeton el at., 1992;
Clement ct ai, 2000) of the COI data within A. rubens
indicates greater than 95% confidence that the intraspecific
gene tree is parsimonious. The ML root haplotype is found
on both coasts of the Atlantic (Fig. 1A, Haplotype B), and
this haplotype is at least an order of magnitude more likely
to be the ancestral haplotype than any other haplotype of A.
rubens (likelihood index = 0.857). All North American
haplotypes are also found in Europe; the unique haplotypes
found in Europe contribute to a significantly higher allelic
diversity (P < 0.0 1 . Table 4). The ITS data are consistent
with the COI data in that there is no allelic diversity among
North American and European individuals of A. nibens
(n = 6).

Discussion

Understanding the mechanisms that are responsible for
the divergence of Asterias nibens and A. forbesi first re-
quires that the timing of their divergence be estimated.
Estimates based on the molecular calibrations reported here
suggest that these species last shared a common ancestor at
least 3.0 Ma (Table 2), not long after the genus first arrived
in the North Atlantic (around 3.5 Ma; Worley and Franz,
1983; Vermeij, 1991 ). Note, however, that asterozoan skel-
etons are rarely preserved in the fossil record, because they
lack rigidly articulated skeletons and rapidly disintegrate
(Barker and Zullo, 1980); indeed, fossils of A. forbesi have
been reported only twice, each time in Pleistocene intergla-
cial sediments. Thus, little direct evidence points to the first
appearance of Asterias in the North Atlantic (Durham and
MacNeil, 1967; Worley and Franz, 1983), and the biogeo-
graphic data used in this paper is therefore based on con-

BIOGEOGRAPHY OF ASTERIAS

A. forbesi

10C>

-L1

H Norway
Norway
Haplotype A
(n=16)
1— Ireland

rubens

100

Haplotype B

(«=14)

- Ireland
1— Ireland A
Haplotype C (n=l)
- France
- France
Norway
Haplotype D («=2)
e and

100

_jT

100 *— — 1~

— 0.01 substitutions/site

100

Ireland
Iceland

Iceland

Newfoundland

A. ruu€tis

100

Maine

-1

A. forbesi

0.005 substitutions/site

Figure 1. Phylogenetic trees for Asierias generated using the best-tit maximum likelihood model in each
data set (COI: HKY + T; ITS: F81 ). (A) Cytochrome c oxidase I phylogeny of inter- and intraspecific Asterias
relationships. Here all characters (first, second, and third position) are included; an identical topology is found
using parsimony or distance methods, or looking at third-position characters alone. Bootstrap support for each
species is indicated by the numbers below each branch. These data do not reject a molecular clock model. The
divergence across the Arctic (between A. amurensis and the Atlantic speciesl is considered to be 3.5 Ma; this
generates an estimate of about 3.0 Ma for the divergence between A. rubens and A. amurensis (see Table 2 and
Discussion). Haplotypes A— D of A. rubens are found on both the North American and European coasts (A:
Maine (n = 8), Nova Scotia (n = 2), Newfoundland (n = 2), Iceland in = 1 I. Norway (H = 2), Ireland (H =
I ); B: Maine (n = 2). Nova Scotia (n = 4). Newfoundland (n = 3). Iceland (H = 1 ). Ireland (H = 2). France
(n = 2); C: Maine (n = 1 ). Norway (n = 3), Ireland (n = 2). and France (n = 1 ); D: Ireland (n = 1 ), and
Maine (n = 1 )). Amphi-Atlantic haplotype B is the maximum likelihood root (index = 0.857). (B) Internal
transcribed spacer ( ITS ) phylogeny of inter- and intraspecific Asterias relationships. Likelihood ratio tests do not
reject a hypothesis of proportional branch lengths (P > 0. 10) suggesting that, aside from substantial differences
in substitution rate, the two phylogenies are equivalent representations of interspecific differentiation. A nearly
identical phylogeny is reconstructed when indels are included in the ITS data.

sistent fossil evidence from other cold temperate species
that participated in the trans-Arctic exchange. Nevertheless,
there is reason to believe that Asterias also spread from the
Pacific to the Atlantic at about 3.5 Ma (Worley and Franz.
1983). Miocene and early Pliocene temperatures were

around 5°-6°C warmer in the North Atlantic and Arctic,
permitting the initial trans-Arctic passage of temperate spe-
cies (Berggren and Hollister. 1974; Vermeij, 1991 ), but then
two dramatic changes were initiated around 3.0 Ma that
appear to play a role in speciation within the North Atlantic.

100 J. P. WARES

Table 2

Internal branch lengths (based on best-fit likelihood model} separating Asterias species (lower triangle*, all 3 matrices)

All characters

A. atnurensis

A. rubens

A. forbesi

A. anmrensis

tL = 1.954 x 10~8 ± 8.63 x 10"9

IJL = 2.665 X 10~8 ± 9.59 X 10~"

A. nibens

0.13678 ± 0.06044

3.59 Ma

A. forbesi

0.18658 ± 0.06715

0.16576 ± 0.04595

3rd position only

A. anmrensis

A. rubens

A. forbesi

A. amurensis

ju, = 6.689 x 10"" ± 3.36 x 10~8

M, = 9.751 x 10~8 ± 3.74 x lO""

A. rubens

0.48084 ± 0.2352

2.96 Ma

A. forbesi

0.68254 ± 0.26168

0.49270 ± 0.15661

ITS-1

A. anmrensis

A. rubens

A. forbesi

A. amurensis

p. = 5.142 x 10~9 ± 2.04 x 10~9

/M = 7.188 x 10~9 ± 2.40 X 10~"

A. nibens

0.0361 ± 0.0143

3.84 Ma

A. forbesi

0.0500 ± 0.0168

0.0470 ± 0.0163

The calibration date of 3.5 Ma is used to obtain the mutation rate ^. for comparisons between A. amurensis and the Atlantic species. The estimated
divergence time between A. rubens and A. forbesi is based on the mean of this calibrated mutation rate (cytochrome c oxidase I [COI] all positions, top;
COI 3rd position only, middle; internal transcribed spacer (ITS) 1. bottom).

* In each matrix, the lower triangle containing the internal branch lengths is made up of the matrix cells below the diagonal line of empty cells
representing comparisons within the same value. The upper triangle contains the estimated mutation rates and estimated divergence data.

At that time, warm North Atlantic currents were dis-
placed by the formation of the cold-water Labrador Current.
This event created a significant thermal gradient in the
North Atlantic, and tropical-temperate faunas were abruptly
replaced with polar and subpolar faunas on the continental

Table 3

McDonald-Krehman tests on each Asterias species pair using
cytochrome c oxidase I (COI) translated data

Species pair

Fixed differences

Polymorphisms

A. rubens-A. forbesi

Synonymous

Nonsynonymous

;

P < 0.001

Synonymous

Nonsynonymous

P > 0.05

A. forbesi-A. amurensis

Synonymous

Nonsynonymous

/' > 0.15

Only the comparison between A. rubens and A. forbesi indicates a
significant departure from neutral evolution. A two-tailed Fisher's exact
test was used for each set of comparisons.

shelf off Nova Scotia and the rest of New England (Berg-
gren and Hollister, 1974; Worley and Franz, 1983; Cronin,
1988). As Northern Hemisphere glaciation began, (lie
present-day latitudinally controlled faunal provincialization
was established as well (Berggren and Hollister, 1974). This
dramatic cooling of the northwestern North Atlantic prob-
ably initiated the separation of North Atlantic Asterias into
European and North American populations with very little
genetic contact (Worley and Franz, 1983). Subsequent
Pleistocene glaciation would have prevented the long-term

Table 4

Comparisons of haplotype diversity (\\, see eqn. 8.4 in Nei 1987.
calculated in DNAsp 3.50, ROMS and Rozas 1999) for the cytochrome c
oxidase I fragment in each species and population of A. rubens

Species/Population Haplotype diversity (H) cr

Asterias rubens 0.793 0.00138

North America 0.597 0.00395

Europe 0.893 0.00143

Asterias forbesi 0.964 0.00596

Asterias amurensis 0.999 0.03125

European populations of A. rubens have significantly higher allelic
diversity than North American populations (P < 0.01); this finding is
supported by nonparametric haplotype sampling in Wares (2000).

BIOGEOGRAPHY OF ASTERIAS

101

establishment of populations in New England, as most of
the North American coast from Long Island Sound north-
ward was covered by a kilometer of ice during glacial
maxima (Kelley et til., 1995).

Pacific and Atlantic populations of other species appear
to have had more recent trans-Arctic genetic contact than
the estimates above would suggest for Asterias (Palumbi
and Kessing. 1991; van Oppen el al., 1995). Moreover,
rapid climatic fluctuations (Cronin, 1988: Roy et al.. 1996)
during the Pleistocene could have permitted large-scale
changes in the geographic range of cold temperate species.
However, both the sea urchin Strongylocentrotus pal/idus
(Palumbi and Kessing, 1991) and the red alga Phycodrys
nihens (van Oppen et al., 1995) appear to have greater
tolerance for Arctic waters than Asterias does. Worley and
Franz (1983) report that expansion of Asterias populations
into habitats as far north as Greenland only occurs period-
ically, and that these populations cannot tolerate colder
waters (Franz et al.. 1981 ). However, the indirect morpho-
logical and paleontological evidence is bolstered by the
molecular evidence, which strongly suggests that A. rubens
and A. forbesi diverged shortly after their ancestral lineage
separated from the Pacific A. amurensis. The estimates of
mutation rate presented here are very similar to other esti-
mates for both the COI fragment (Knowlton and Weigt,
1998; Schubart et al.. 1998; Wares, 2001; Wares and Cun-
ningham, in press) and the ITS fragment (Schlotterer et al..
1994; van Oppen et al.. 1995). Thus these data strongly
support earlier inferences of a late Pliocene trans-Arctic
passage and subsequent speciation within the Atlantic.

An analysis of genealogical patterns within A. rubens
confirms that the North American populations of this spe-
cies are descendants of a recent colonization from Europe
that probably followed the most recent glacial maximum
(about 20.000 BP, Holder et al.. 1999). The genealogical
data presented here fit several important patterns that sug-
gest a recent range expansion (Wares. 2000). All North
American haplotypes are identical to the most-common
European haplotypes (Fig. 1A). Generally, invading haplo-
types are the most deeply nested haplotype in the European
(putative source) population. This is to be expected, because
deeply nested ancestral haplotypes are often the most com-
mon (Castelloe and Templeton. 1994), and therefore have a
higher probability of participating in long-distance dispersal
events. Haplotype B (Fig. 1 A) is a good illustration of this
expectation — it is closely related to each other haplotype
and has a high copy number in both European and American
populations. These observations contribute to the high like-
lihood (85.7<7r, more than an order of magnitude greater
likelihood than any other haplotype) that this is the ancestral
allele in A. rubens.

Additionally, allelic diversity is significantly lower in
North American A. rubens than in Europe (Table 4), a signal
of recent range expansion (Hewitt, 1996; Austerlitz et al..

1997). However, the North American colonization is diffi-
cult to date because there are no unique haplotypes in North
America; ancestral allelic polymorphism tends to inflate
indirect estimates of population size and age (Kuhner el al..
1998: Edwards and Beerli, 2000). The lack of unique di-
versity in North America also prevents the meaningful use
of other phylogeographic methods; for instance, statistics of
the geographic dispersion of haplotypes (for review see
Templeton, 1998) are uninformative (Wares, unpubl. data).
This is primarily because even closely related individuals
(identical haplotypes) are distributed across the entire geo-
graphic range of A. rubens. It is possible that the multiple
shared alleles between Europe and North America represent
a multiple-invasion history; Asterias larvae are planktotro-
phic and disperse in the water column for 6 or more weeks
(Clark and Downey, 1992).

There is evidence that natural selection has played some
role in the overall divergence between these species. A
significant number of amino acid replacement substitutions
distinguish A. rubens from A. forbesi (Table 3), all of them
reflecting second- or third-position nucleotide substitutions.
There is no obvious pattern to the amino acid replacements,
as most of them involve substitutions among uncharged or
nonpolar amino acids. Two of the three species in the genus
Asterias are found in cold-temperate waters, while A.
forbesi is found in the warmer mid- Atlantic region (Schopf
and Murphy, 1973; Franz et al., 1981 ). Many of the phys-
iological differences between A. rubens and A. forbesi
(Franz et al., 1981 ) reflect this latitudinal distribution. How-
ever, the possibility that these amino acid substitutions are
related to physiological differences in the warm-temperate
A. forbesi has never been tested. The difference in temper-
ature between the habitats of A. rubens and A. forbesi is
unlikely to contribute to differences in metabolic rate that
could accelerate the mutation rate (for review see Rand,
1994). Nevertheless, this hypothesis is worth examination,
because A. forbesi is supported by relatively long branches
in both the COI and the non-coding ITS region (Table 2,
Fig. IB). If natural selection is playing a role in the amino
acid divergences of the mitochondrial COI gene between A.
rubens and A. forbesi, there is no reason why a noncoding
nuclear sequence should reflect the same increase in diver-
gence rate.

In conclusion, the biogeographic response of Asterias to
late Pliocene climatic and oceanographic change fits a pat-
tern predicted by Worley and Franz (1983). Following the
arrival of Asterias in the North Atlantic around 3.5 Ma
(Worley and Franz, 1983; Venneij, 1991). populations were
established on both the European and North American
coasts during a period when the North Atlantic was as much
as 5-6°C wanner (Berggren and Hollister. 1974). The for-
mation of the Labrador Current 3.0 Ma rapidly changed the
faunal composition of the intertidal Canadian Maritimes and
New England coast, and Asterias populations in this region

J, P. WARES

probably went extinct. An American population survived
under the conditions of the mid-Atlantic coast and Gulf
Stream waters (A.forbesi), and the European population (A.
rubens) has recently recolonized the cold-temperate shores
of New England and the Canadian Maritimes. Thus, the
zone of sympatry between these two species appears to be a
zone of secondary contact. Hybridization is considered rare
between these species (Schopf and Murphy, 1973; Worley
and Franz, 1983), but whether behavioral mechanisms
(Franz et ai, 1981) or gametic recognition mechanisms
(Hellbergand Vacquier, 1999; Fernet, 1999) are responsible
is unclear.

The genetic data presented here illustrate a strong con-
cordance between paleoceanographic changes and indirect
estimates of speciation between the North Atlantic Asterias
species. The species boundaries are phylogenetically quite
distinct, and the divergence estimates based on these genetic
data appear to support a late Pliocene, rather than late
Pleistocene or Holocene, separation. A better understanding
of the balance between oceanographic and climatic changes
in the late Pliocene and Pleistocene, and of the response of
species based on varying life-history characters to these
changes, will enable us to predict the responses of other taxa
(Cunningham and Collins, 1998: Wares and Cunningham,
in press).

Acknowledgments

1 thank G. Manchenko, A. Ingolfsson, J. Maunder, B.
O'Connor, D. Garbary, D. M. Rand, and C. Damiani for
assistance in the field collecting seastars. The manuscript
was greatly improved thanks to discussions with T. Turner
and the suggestions of two anonymous reviewers. These
analyses were done in the laboratory of C. W. Cunningham,
whose aid during this and other projects was invaluable. A
National Science Foundation Dissertation Improvement
Grant (NSF DEB-99-72707) to J. P. W. funded this study.

Literature Cited

Austerlitz, F., B. Jung-Muller, B. Godelle, and P.-H. Gouyon. 1997.

Evolution of coalescence times, genetic diversity and structure during

colonization. Theor. Popul. Biol. 51: 148-164.
Barker, L., and V. A. /ullo. 1980. Asterias forbesi (Desor) (Asterozoa.

Asteroidea) from the Pleistocene "Coquina" at Fort Fisher, New

Hanover County, North Carolina. J. Elislui Mitchell Sci. Soc. 96:

39-44.
Berggren, W. A., and C. D. Hollister. 1974. Paleogeography, paleobio-

geography, and the history of circulation in the Atlantic Ocean. Soc.

Econ. Paleontol. Mineral. 20: 126-186.
Castelloe, J., and A. R. Templeton. 1994. Root probabilities for in-

traspecific gene trees under neutral coalescent theory. Mo!. Phylogenet.

E\-ol. 3: 102-113.
Clark, A. M., and M. E. Downey. 1992. Starfishes of the Atlantic.

Chapman & Hall. London.
Clement, M., D. Posada, and K. A. Crandall. 2000. TCS: a computer

program to estimate gene genealogies. Mol. Ecol 9: 1657-1660.
Collins, T. M., K. Frazer, A. R. Palmer, G. J. Vermeij, and W. M.

Brown. 1996. Evolutionary history of northern hemisphere Nucella
(Gastropoda. Muricidael: molecular, morphological, ecological, and
paleontological evidence. Evolution 50: 2287-2304.

Crandall, K. A., and A. R. Templeton. 1996. Applications of intraspe-
citic phylogenetics. Pp. 81-99 in New Uses for New Phytogenies, P. H.
Harvey. A. J. Brown. J. Maynard Smith, and S. Nee, eds. Oxford
University Press, Oxford.

Cronin, T. M. 1988. Evolution of marine climates of the U. S. Atlantic
coast during the past four million years. Philos. Trans. R. Soc. Land. B
318: 661-678.

Cunningham, C. W., and T. M. Collins. 1998. Beyond area relation-
ships: extinction and recolonization in molecular marine biogeography.
Pp. 297-321 in Molecular Ecology and Evolution: Approaches and
Applications. B. Schierwater, B. Streit, G. Wagner, and R. DeSalle,
eds. Birkhauser Verlag. Basel.

Cunningham, C. W., H. Zhu, and D. M. Hillis. 1998. Best-fit maxi-
mum-likelihood models for phylogenetic inference: Empirical tests
with known phylogenies. Evolution 52: 978-987.

Durham, J. W., and F. S. MacNeil. 1967. Cenozoic migrations of
marine invertebrates through the Bering Strait region. Pp. 326-349 in
The Bering Land Bridge. D. M. Hopkins, ed. Stanford University Press,
Stanford.

Edwards, S. V., and P. Beerli. 2000. Perspective: Gene divergence,
population divergence, and the variance in coalescence time in phylo-
geographic studies. Evolution 54: 1839-1854.

Endler, J. A. 1977. Geographic Variation. Speciation. and Clines.
Princeton University Press, Princeton, NJ.

Felsenstein, J. 1988. Phylogenies from molecular sequences: inference
and reliability. Annu. Rev. Genet. 22: 521-565.

Folmer, O., M. Black, W. Hoeh, R. Lutz, and R. Vrijenhoek. 1994.
DNA primers for amplification of mitochondria! cytochrome c oxidase
subunit I from diverse metazoan invertebrates. Mol. Mar. Biol. Bio-
icclmol. 3: 294-299.

Franz, D. R., E. K. Worley, and A. S. Merrill. 1981. Distribution
patterns of common seastars of the middle Atlantic continental shelf of
the northwest Atlantic (Gulf of Maine to Cape Hatteras). Biol. Bull.
160: 394-418.

Goldman. N. 1993. Statistical tests of models of DNA substitution. J.
Mol. Evol. 36: 182-198.

Gosling. E. M. 1992. Systematics and geographical distribution of Myti-
lus. Pp. 1-20 in The Mussel Mytilus: Ecology. Physiology. Genetics,
and Culture. E. M. Gosling, ed. Elsevier, Amsterdam.

Gosner, K. L. 1978. A Field Guide to the Atlantic Seashore From the
Ba\ of Fundv to Cape Hatteras. Houghton Mifflin. Boston.

Griffiths, R. C., and S. Tavare. 1994. Sampling theory for neutral
alleles in a varying environment. Philos. Trans. R. Soc. Land. B 344:
403-410

Hasegawa, M., H. Kishino, and T. Yano. 1985. Dating of the human-
ape splitting by a molecular clock of mitochondria! DNA. J. Mol, Evol.
21: 160-174.

Hayward, P. J., and J. S. Ryland. 1995. Handbook of the Marine
Fauna of North-West Europe. Oxford University Press, Oxford.

Hellberg, M. E., and V. D. Vacquier. 1999. Rapid evolution of fertil-
ization selectivity and lysin cDNA sequences in teguline gastropods.
Mol. Biol. Evol. 16: 839-848.

Hewitt, G. M. 1996. Some genetic consequences of ice ages, and their
role in divergence and speciation. Biol. J. Linn. Soc. 58: 247-276.

Hillis. D. M., B. K. Mable, A. Larson, S. K. Davis, and E. A. Zimmer.
1996. Nucleic acids IV: Sequencing and cloning. Pp. 321-384 in
Molecular Systematics. D. M. Hillis. C. Moritz. and B. K. Mable, eds.
Sinauer, Sunderland. MA.

Holder, K., R. Montgomerie, and V. L. Friesen. 1999. A test of the
glacial refugium hypothesis using patterns of mitochondrial and nu-

BIOGEOGRAPHY OF ASTERIAS

103

clear DNA sequence variation in rock ptarmigan (Lagoons inutus}.
Evolution S3: 1936-1951).

Kelley, J. T., A. R. Kelley, and S. Appollonio. 1995. Landtorms of the
Gulf of Maine. Pp. 19-38 in From Cap? Cod to the Buy of Fiuu/v: an
Environmental Atlas of the Gulf of Maine. P. W. Conkling. ed. MIT
Press, Cambridge, MA.

Knimlton. N.. and L. A. \Veigt. 1998. New dates and new rates for
divergence across the Isthmus of Panama. Proc. R. Soc. Loud. B 265:
2257-2263.

Kuhner, M. K., J. Vamato, and J. Felsenstein. 1998. Maximum like-
lihood estimation of population growth rates based on the coalescent.
Genetics 149: 429-43-4.

McDonald, ,). H., and M. Kreitman. 1991. Adaptive protein evolution
at the Adli locus in Drosophila. Nature 351: 652-654.

Menge, B. A. 1979. Coexistence between the seastars Asterias vulgaris
and A. forbesi in a heterogeneous environment: a non-equilibrium
explanation. Oecologia 41: 245-272.

Nei, M. 1987. Molecular Evolutionary Genetics. Columbia University-
Press. New York.

Nei, M., and \V.-H. I,i. 1979. Mathematical model for studying genetic
\ anation in terms of restriction endonucleases. Proc. Natl. Acait. Sci.
USA 76: 5269-5273.

Palumbi. S. R.. and B. D. Kessing. 1991. Population biology of the
trans-Arctic exchange: mtDNA sequence similarity between Pacific
and Atlantic sea urchins. Evolution 45: 1790-1805.

Fernet, B. 1999. Gamete interactions and genetic differentiation among
three sympatric polychaetes. Evolution 53: 435-446.

Posada. D.. and K. A. Crandall. 1998. Modeltest: testing the model of
DNA substitution. Bioinfonnatics 14: 817-818.

Rand, D. 1994. Thermal habitat, metabolic rate, and the evolution of
mtDNA. Trends Ecol. Evol. 9: 125-131.

Reid. D. G., E. Rumbak, and R. H. Thomas. 1996. DNA. morphology
and fossils: phylogeny and evolutionary rates of the gastropod genus
Littorina. Philcs. Trans. R. Soc. Land. B 351: 877-895.

Roy, K., J. W. Valentine, D. Jablonski, and S. M. Kidwell. 1996.
Scales of climatic variability and time averaging in Pleistocene biotas:
implications for ecology and evolution. Trends Ecol. Evol. 11: 458-463.

Rozas, J.. and R. Rozas. 1999. DnaSP version 3: an integrated program
for molecular population genetics and molecular evolution analysis.
Bioinfonnatics 15: 174-175.

Schlotterer, C., M. Hauser, A. von Haeseler, and D. Tautz. 1994.
Comparative evolutionary analysis of rDNA ITS regions in Drosoph-
ila. Mol. Biol. Evol. 11: 513-522.

Schopf, T. J. M., and L. S. Murphy. 1973. Protein polymorphism of the

hybridizing seastars Asterias forbesi and Asterias vu/garis and impli-
cations for their evolution. Biol. Bull. 145: 589-597.
Schubart, C. I)., R. Diesel, and S. B. Hedges. 1998. Rapid evolution to

terrestrial life in Jamaican crabs. Nature 393: 363-365.
Seutin. G., B. N. White, and P. T. Boag. 1991. Preservation of avian

blood and tissue samples for DNA analyses. Can. J. Zool. 69: 82-90.
Swofford, D. L. 1998. PAUP*. Phylogenetic Analysis Using Parsimony

(* and Other Methods). Version 4. Sinauer Associates, Sunderland.

MA.
Tajima, F. 1989. Statistical method for testing the neutral mutation

hypothesis by DNA polymorphism. Genetics 123: 585-595.
Templeton, A. R. 1998. Nested clade analyses of phylogeographic data:

testing hypotheses about gene flow and population history. Mol. Ecol.

7: 381-397.
Templeton. A. R., K. A. Crandall. and C. F. Sing. 1992. A cladistic

analysis of phenotypic associations with haplotypes inferred from

restriction endonuclease mapping and DNA sequence data. III. Cla-

dogram estimation. Genetics 132: 619-633.
van Oppen, M. J. H., S. G. A. Draisma, J. L. Olsen, and W. T. Stam.

1995. Multiple trans-Arctic passages in the red alga Phycoctrys

rubens: evidence from nuclear rDNA ITS sequences. Mar. Biol. 123:

179-188.
Vermeij, G. 1991. Anatomy of an invasion: the trans-Arctic interchange.

Paleobiology 17: 281-307.
Wares, J. P. 2000. Abiotic influences on the population dynamics of

marine invertebrates. Ph.D. dissertation. Duke University. Durham.

NC.
Wares, J. P. 2001. Patterns of speciation inferred from mitochondria!

DNA in North American Chthamaliis (Cirripedia: Balanomorpha:

Chthamaloidea). Mol. Phylogenet. Evol. 18: 104-116.
Wares, J. P., and C. W. Cunningham. Phylogeography and historical

ecology of the North Atlantic intertidal. Evolution (in press).
White, T. J., T. Bruns, S. Lee, and J. Taylor. 1990. Amplification and

direct sequencing of fungal ribosomal RNA genes for phylogenetics.

Pp. 315-322 in PCR Protocols: A Guide to Methods and Applications.

M. A. Innis. D. H. Gelfand. J. J. Snmsky. and T. J. White, eds.

Academic Press. San Diego.
Worley, E. K., and I). R. Franz. 1983. A comparative study of selected

skeletal structures in the seastars Asterias forbesi. A. vulgaris, and A.

rubens with a discussion of possible relationships. Proc. Biol. Soc.

Wash. 96: 524-547.
Yang, Z. 1994. Maximum likelihood phylogenetic estimation from DNA

sequences with variable rates over sites: Approximate methods. J. Mol.

Evol. 39: 306-314.

Reference: Biol. Bull. 201: 104-119. (August 2001)

Systematics and Biogeography of the Jellyfish
Aurelia labiata (Cnidaria: Scyphozoa)

LISA-ANN GERSHWIN
Cabrillo Marine Aquarium, San Pedro, California 90731 and CSU Northrid?>e, California 91330

Abstract. The hypothesis that the common eastern North
Pacific Aurelia is A. aurita is falsified with morphological
analysis. The name Aurelia lahiata is resurrected, and the
species is redescribed, to refer to medusae differing from A.
aurita by a suite of characters related to a broad and elon-
gated manubrium. Specifically, the oral arms are short,
separated by and arising from the base of the fleshy manu-
brium. and the planulae are brooded upon the manubrium
itself, rather than on the oral arms. Aurelia aurita possesses
no corresponding enlarged structure. Furthermore, the num-
ber of radial canals is typically much greater in A. lahiata,
and thus the canals often appear more anastomosed than in
A. aurita. Finally, most A. labiaia medusae possess a 16-
scalloped bell margin, whereas the margin is 8-scalloped in
most A. aurita. Separation of the two forms has previously
been noted on the basis of allozyme and isozyme analyses
and on the histology of the neuromuscular system. Partial
18S rDNA sequencing corroborates these findings. Three
distinct moiphotypes of A. lahiata, corresponding to sepa-
rate marine bioprovinces, have been identified among 17
populations from San Diego. California, to Prince William
Sound. Alaska. The long-undisputed species A. limhata may
be simply a color morph of A. labiata, or a species within a
yet-unelaborated A. lahiata species complex. The first
known introduction of Aurelia cf. aurita into southern Cal-
ifornia waters is documented. Although traditional jellyfish
taxonomy tends to recognize many species as cosmopolitan
or nearly so, these results indicate that coastal species, such
as A. labiata, may experience rapid divergence among iso-
lated populations, and that the taxonomy of such species
should therefore be scrutinized with special care.

Received 16 December 1998; accepted 5 April 2001.
Current address: Dept. of Imegrative Biology, University of California.
Berkeley. CA 94720. E-mail: gershwin@socrates.berkeley.edu

Introduction

Perhaps had Darwin not been afflicted with seasickness,
he might have noticed the bewildering array of geographi-
cally varying jellyfish morphologies. Some of his contem-
poraries documented species separated by only short dis-
tances but differing greatly in appearance (Eschscholtz.
1829; Brandt. 1835, 1838; Agassiz, 1862; Haeckel, 1879,
1880). Morphological distinctions have since been reported
for populations of Cassiopea from separate islands of the
Caribbean (Hummelinck, 1968), Mastigias in different
lakes of Palau (Hamner and Hauri, 1981), and Aurelia
scyphistomae from various parts of the Thames estuary
(Lambert, 1935). In his studies of the genus Cyanea, Brewer
( 1 99 1 ) reported distinct morphotypes that could be corre-
lated with isolated locations in Long Island Sound, USA;
these observations resurrected a long-standing argument
about species distribution and recognition criteria of North
Atlantic Cyanea. Nineteenth-century taxonomists recog-
nized different species, corresponding to a latitudinal gra-
dation, on both sides of the Atlantic. Cyanea arctica Peron
and Lesueur, 1810, was known as the boreal species from
Europe to North America. In the western Atlantic. C. fulva
L. Agassiz. 1 862, was found along the mid-Atlantic states,
while the form south of the Carolinas was recognized as C.
versicolor L. Agassiz. 1862. In the eastern Atlantic, C.
capillata (Linnaeus. 1746) was established as the northern
European species, while C. lamarckii Haeckel, 1880. was
identified in warmer southern European waters. This pattern
of biodiversity was largely overlooked by twentieth-century
taxonomists. who often lumped the forms and recognized
only C. capillata (Mayer. 1910; Bigelow. 1914; Stiasny and
van der Maaden. 1943; Kramp, 1961; Calder, 1971; Larson,
1976).

The scarcity of biogeographic studies of jellyfishes may
be, in part, attributable to the unclear systematics of these

104

SYSTEMATICS AND BIOGEOGRAPHY OF AURELIA LAB/ATA

105

Figure 1. Original illustrations of Aurelia labiata, showing greatly enlarged munubrium: (A) lateral view of
medusa; (B) oblique view of subumbrella. (Reprinted from Chamisso and Eysenhardt. 1821).

animals. Color differences, patterns of pigmentation, and
anatomical variation led to the description of many nominal
species during the expeditions of the nineteenth century (see
Mayer, 1910; Kramp, 1961). The range of variation in
jellyfishes is not well understood, and species definitions are
often vague, focusing only on the few most obvious char-
acters. For example, if one sees a flat, whitish medusa with
four horseshoe-shaped gonads. most tend to think it must be
Aurelia aiirita. The details of anatomy have not been scru-
tinized closely. Therefore, significant morphological differ-
ences have not been detected, and inappropriate identifica-
tions and erroneous conclusions regarding biogeography
have been made. The systematic tangle and biogeographic
mistakes are common throughout the medusan taxa, though
I focus herein on Aurelia.

Mayer (1910) recognized 13 unique forms of Aurellia
(the spelling was later formally changed back to Aurelia by
Rees, 1957), and sorted these forms into three morpholog-
ical groups:

1. A. aiirita (Linnaeus, 1746) sensu Lamarck, 1816, and
its seven varieties, described as A. cniciata Haeckel.
1880, A. colpota Brandt, 1835 [sensu Gotte, 1886] (as
=A. coerulea von Lendenfeld, 1884), A. flavidula
Peron and Lesueur. 1810 [incorrectly listed as 1809)
(as =A. habanensis Mayer, 1900). A. hyalina Brandt.
1835. A. dubia Vanhfiffen, 1888. A. vitiana Agassiz
and Mayer. 1899. and A. imirginalis L. Agassiz. 1862

2. A. labiata Chamisso and Eysenhardt, 1821 [incor-
rectly listed as 1820[. with three varieties, described as

A. clausa Lesson, 1829, A. limbata (Brandt. 1835)
[incorrectly listed as 1838], and A. inaldivensis Big-
elow. 1904
3. A. solida Browne, 1905

Mayer distinguished A. labiata and its varieties from the
other two groups based primarily on the degree of scallop-
ing of the bell margin, being 16-notched in the former and
8-notched in the latter. He subsequently found a specimen
of A. iinritti at Tortugas. Florida, closely resembling A.
labiata, leading him to conclude that A. labiata was prob-
ably derived as a mutation from A. aiirita (Mayer, 1917).
Kramp also wavered on the validity of A. labiata, first
recognizing the species in his 1961 synopsis, then later
regarding it as doubtful (1965, 1968). Most recently, au-
thors such as Russell (1970). Larson (1990), and Arai
( 1997) have recognized two valid species: A. limbata, which
is primarily arctic and has a conspicuous brown bell margin,
and A. auritci. whose name has been treated as the senior
synonym of all others. Russell (1970) followed Kramp
(1965, 1968) in regarding all other species as varieties,
whereas Larson (1990) and Arai (1997) simply did not
mention any other species.

The source of this confusion is unclear, as the original
description of A. labiatu was quite specific. Translated from
Latin, "It differs from A. aiirita by its very long oral lips.
Marginal tentacles were not observed, but are without a
doubt present. Arms appressed to the bell. Diameter of the
bell nearly a foot" (Chamisso and Eysenhardt. 1821). The
focus of the description and its accompanying illustrations
is the strikingly unique elongated manubrium (Figs. 1,2).

106

L. GERSHWIN

Figure 2. Aurelia labiata. adult medusa, from Monterey Bay. Califor-
nia.

although this character is rarely mentioned in later revisions.
Furthermore, the characteristically short oral arms arising
from the base of the manubrium were mentioned as being
held close to the bell, a trait that is readily apparent in live
specimens. Ironically, the commonly accepted character of
16 marginal scallops is not mentioned, although it is subtly
illustrated. It is unclear why certain key characters of the
original description have been ignored by later workers.

Disorder in the nomenclature of Aurelia worldwide has
caused confusion about the identity of the species in the
eastern North Pacific. Depending on the author, one to three
species have been recognized. Most authors have applied
the name A. aurita to all forms. Some distinguish ,4. lim-
bata, although this appears to have been occasionally con-
fused with A. labiata (Zubkoff and Lin, 1975; Greenberg et
al., 1996). When A. labiata has been recognized, it has been
separated from A. aurita only by the doubling of marginal
scallops (Hand. 1975; Kozloff, 1974). Although A. labiata
was originally described from California, most reports of
the species (apparently incorrectly) are from regions outside
the eastern North Pacific.

Throughout all the confusion, several studies have re-
ported differences between the eastern North Pacific Aurelia
and those of other regions, yet failed to elaborate the sys-
tematics. Chia et al. ( 1984) found that the muscle system in
Puget Sound polyps is distinct from that of polyps from
Plymouth. England. Zubkoff and Lin (1975) observed pe-
culiar banding in the isozyme patterns of Aurelia scyphis-
tomae from Puget Sound, Washington, that caused them to
wonder whether this population may belong to a species
other than A. aurita. Similarly. Greenberg et al. (1996)
could distinguish two groups on their allozyme patterns: one
group consisted of two populations of A. "aurita" from
Japan (one from Tokyo Bay, and one aquarium-raised) plus

a population that was apparently introduced to San Fran-
cisco Bay; and the second group consisted of wild medusae
from Monterey Bay, California, and Vancouver, British
Columbia. They further distinguished the two groups on the
basis of morphology, using manubrium length and the
highly anastomosed condition of the radial canals.

To test the hypothesis that the common eastern North
Pacific Aurelia is A. aurita, I compared the morphology of
17 populations of Aurelia from San Diego, California, to
Prince William Sound, Alaska, to the morphology of A.
aurita from Europe, and A. flavidula from the eastern
United States, as described and figured by Agassiz (1862),
Mayer (1910), Kramp (1961). Russell (1970), and many of
the references therein. The conclusions that I have drawn on
morphological characters are consistent with those emerg-
ing from the enzyme analyses of Zubkoff and Lin (1975)
and Greenberg et al. (1996), the neuromuscular study of
Chia et al. (1984), and the DNA sequencing results of
J. Lowrie of the Cnidarian Research Institute (pers. comm.,
June 2000) — that is, that the common eastern North Pacific
Aurelia is not A. aurita. However, it does match the de-
scription of the species previously described as Aurelia
labiata Chamisso and Eysenhardt, 1821. Thus, I propose a
revalidation of A. labiata, and herein offer a redescription
and designate a neotype. In scrutinizing the morphology of
A. labiata. I further found that each population possesses
unique characters that cluster into three morphotypes cor-
responding to well-demarcated biogeographic provinces.
The purposes of this paper are to describe the morphological
and geographical variation in A. labiata and to stabilize the
nomenclature for the species. This is necessary as a basis for
further systematic investigation, for ongoing biodiversity
studies, and for proper management of species introduc-
tions.

Materials and Methods

Aurelia aurita and other fonns

Literature-based comparisons were made using the Euro-
pean form, Aurelia nuriia, and are denoted traditionally
(e.g., Aurelia aurita). The full breadth of literature used for
comparison is too massive to list here, but can be found in
the synonymies of Mayer (1910), Kramp (1961), and Rus-
sell (1970).

Literature-based comparisons were made with A. flav-
idula from the eastern United States, primarily following
Agassiz (1862) and the references in the synonymy of
Mayer (1910).

Literature-based comparisons were made to the boreal A.
limbata using Brandt (1835. 1838). Vanhoffen (1906),
Kishinouye (1910), Bigelow (1913, 1920), Uchida (1934),
Bigelow (1938), Kramp (1942), Stiasny and van der Maa-
den (1943), Naumov ( 1961 ), Uchida and Nagao (1963), and
Faulkner (1974).

SYSTEMATICS AND BIOGEOGRAPHY OF AURELIA LABIATA

107

Comparisons were made using live, captive medusae
descended from a Japanese population (cultured at Cabrillo
Marine Aquarium); although the phylogenetic relationship
between the European and Japanese forms is still in ques-
tion, they are structurally similar — that is. they both lack the
enlarged manubrium characteristic of A. labiata.

Comparisons were also made on some live, wild medusae
from Spinnaker Bay, Long Beach, California, which pos-
sessed the A. aurita body form, and on the descriptions of
Greenberg et al. (1996) for the introduced San Francisco
Bay form. Live representatives of Greenberg's population at
Foster City could not be found. References made to forms
that possess the A. aurita body type but are of uncertain
taxonomic affiliation are denoted non-traditionally (e.g.,
Aurelia "aurita" or Aurelia cf. aurita). This includes the
captive Japanese form, as well as introduced forms.

Systematics of Aurelia labiata

Attempts were made to locate the holotype at the follow-
ing institutions: The California Academy of Sciences (San
Francisco) (CAS), Institut Royal des Sciences Naturelles de
Belgique (Brussels), Museum fur Naturkunde (Berlin), Mu-
seum National D'Histoire Naturelle (Paris), Museum of
Comparative Zoology (Harvard). Nationaal Natuurhisto-
risch Museum (Leiden), National Museum of Natural His-
tory (Washington), Natural History Museum (London),
Zoological Institute (St. Petersburg). Zoological Museum
(Copenhagen), and the Zoological Museum (Moscow Uni-
versity). All would have been reasonable depositories or
recipients of a transfer of a holotype of a California species
found by European explorers on a Russian expedition of
that time. However, none had A. labiata type material nor
knew where it might be kept; indeed, it appears doubtful
that specimens were originally collected and deposited.
Thus, my observations were made on animals from near the
type locality and from many other regions along the Pacific
Coast of North America.

A neotype was designated in order to stabilize the taxon-
omy of the species, and is deposited in the California
Academy of Sciences in San Francisco. The original type
locality could not be identified. Chamisso and Eysenhardt
( 1 82 1 ) recorded the species from "New California," and a
map in Schweizer (1973) indicates only somewhere near
San Francisco Bay. However, specimens that I collected
near San Francisco Bay were in poor shape, so the most
intact representative specimen from the available material
was selected from Monterey Bay (ca. 100 miles to the
south). Morphological differences were not apparent be-
tween specimens from San Francisco and Monterey, except-
ing those attributable to collection.

I preferentially examined live medusae in the wild to
avoid artifacts of captivity and preservation; however, cul-
tured and captive medusae were observed supplementally.

In the wild, mature and immature medusae were collected
from July 1995 to March 2000 by hand and by dip nets from
nine locations in California (Coronado Island. San Diego:
Newport Beach; Spinnaker Bay, Long Beach; Catalina Is-
land; Marina del Rey; Santa Barbara; Monterey Bay; Sau-
salito, San Francisco Bay; Tomales Bay), and from New-
port, Oregon; Poulsbo. Washington; Friday Harbor, San
Juan Island, Washington; and Brentwood Bay, Saanich In-
let. British Columbia. Cultured and captive medusae were
examined at the Birch Aquarium at Scripps. San Diego,
California (San Diego A. labiata): Cabrillo Marine Aquar-
ium, San Pedro, California (both Japanese Aurelia "aurita"
and Long Beach A. labiata): Monterey Bay Aquarium.
Monterey, California (Japanese A. "aurita" and Monterey
A. labiata): Oregon Coast Aquarium. Newport. Oregon
(Japanese A. "aurita" and Newport A. labiata): Point Defi-
ance Zoo and Aquarium. Tacoma, Washington (A. labiata
from Poulsbo, Washington); and the Seattle Aquarium, Se-
attle. Washington (A. labiata from Poulsbo, Washington).
In addition to the above observations, characters were as-
sessed as much as possible from a videotape taken in July
1996 of medusae from Prince William Sound, Alaska; from
photographs of A. labiata from Steamer Bay, Alaska (Barr
and Barr, 1983) and A. liiubata from Amchitka Island.
Alaska (Faulkner. 1974); and from preserved specimens
from the Farallon Islands, California.

Measurements were taken on 7-20 live medusae from
each of the following locations: Coronado Island. Newport
Beach. Spinnaker Bay. Marina del Rey. Monterey Bay,
Tomales Bay. Newport (OR). Poulsbo. and Brentwood Bay.
Each medusa was individually dipped out of the water with
a bucket and measured immediately with a vernier caliper or
ruler to the nearest millimeter. Bell diameter (BD) was
typically measured with the specimen lying flat on its ex-
umbrellar surface. Manubrium length (ML) was usually
measured with the animal in the water with the manubrium
projecting upward, but captive medusae from Newport
(OR) were measured with the manubrium hanging down-
ward in the water. Since the manubrium is stiff and carti-
laginous, its position did not appear to bias the measure-
ments. To account for the difference in size at maturity of
medusae from different populations, manubrium lengths
were normalized as a percentage of bell diameter.

In addition to the measurements described above, about
200 medusae from each population were cursorily examined
for the following characters, then released: manubrium
shape, number of marginal scallops, oral arm length, num-
ber of radial canals emanating from each gastro-genital
sinus, bell shape and color, and if female, the location and
pattern of larval brood.

German papers were translated with Power Translator
6.02 for Windows (Globalink).

108

L. GERSHWIN

FIG. 3. Comparative diagram of three morphotypes ofAurelia labiata
with A. aurila, subumbrellar and lateral views. (A) Aurelia aitrila. (B)
Southern morph, from Southern California Bight. (C) Central niorph. from
Santa Barbara, California, to Oregon. (D) Northern morph, from Puget
Sound, Washington, to Alaska. In A. aurita, manubrium is inconspicuous,
oral arms meet in the middle, the radial canals are few. and the margin has
8 scallops. In A. labiata. the manubrium protrudes below the bell margin,
which has 16 scallops, there are many radial canals, and the oral arms do
not meet. Darkened areas along oral arms (A. aurita) and manubrium (A.
labiata) indicate position of larval brood.

Results

Comparison with European Aurelia aurita (Fig. 3)

Medusae from every population that I studied in the
eastern North Pacific differed from published descriptions
of the European A. aurita but closely matched the original
description of A. labiata. Specifically, the A. labiata body
form is characterized by an enlarged, fleshy manubrium;
oral arms arising from the base of the manubrium; planulae
brooding upon the manubrium; up to 15 radial canals arising
from each gastro-genital sinus, and typically anastomosing
in older individuals; and secondary scalloping of the bell
margin between rhopalia (Fig. 3B-D). In contrast, the A.

aurita body type possesses no such enlarged manubrium
structure; the oral arms meet in the middle of the animal;
planulae are brooded upon the oral arms; typically only 3-5,
sometimes 7. radial canals arise from each gastro-genital
sinus; and secondary scalloping is rarely observed (Fig.
3 A).

Comparison with western Atlantic Aurelia "flavidula"

The nominal species Aurelia flavidula is another taxo-
nomic tangle that was somewhat resolved by Kramp ( 1942).
Kramp concluded that the yellow Greenlandic form seen by
Fabricius (1780) and named by Peron and Lesueur (1810)
was identical to A. limbata, later named by Brandt (1835),
and that calling the northern Atlantic American form A.
flavidula was a mistake by Agassiz (1862). Agassiz had
differentiated the western Atlantic A. "flavidula " from the
European A. aurita on the former having a marginal net-
work of anastomoses, the gonadal pouches closer together
and occupying fully 1/3 of the bell diameter, and differences
in the mouth fringes. Kramp further cautioned that using the
name A. flavidula would be confusing, so he gave the
common American Atlantic form the name A. occidentalis,
distinguishing it from A. aurita on the heavier anastomosing
of the radial canals; he later lumped it into A. aurita without
comment (Kramp, 1961).

Proper phylogenetic placement of both the Greenlandic
form and the common American Atlantic form must await a
revision of the genus Aurelia based on live material. For the
Greenlandic form, being yellow and having anastomosed
canals seem insufficient for concluding conspecificity with
the Alaskan A. limbata. Ideally, conspecificity should be
based on numerous characters inherited by common de-
scent, not by shared color. The importance of anastomosed
canals is discussed below. The American Atlantic form,
regardless of its identity, does not possess the enlarged
manubrium and related characters of A. labiata: whether it
is present along the Pacific coast of North America has not
yet been determined.

Systematics of Aurelia labiata

The common moon jellyfish found in 17 populations
from San Diego. California, to Prince William Sound,
Alaska, is characterized by the body form described by
Chamisso and Eysendardt (1821) for A. labiata. Many of
the references to Aurelia of the eastern North Pacific do not
contain illustrations or photographs; those that do are most
often based on the European morphology. In at least one
example, the same photograph is used in both West coast
and East coast American field guides (Audubon Society,
1981 ). A large body of literature has thus been responsible
for perpetuating the misidentification. The synonymy below
contains only the references that have figures or descriptions
positively referable to A. labiata sensu Chamisso and

SYSTEMATICS AND BIOGEOGRAPHY OF AURELIA HHIATA

Eysenhardt, 1 82 1 ; thus, even references to A. labiata are not
included below if they do not include the enlarged nianu-
brium. The remainder of references to eastern North Pacific
Aurelia are dealt with below in appropriate sections.

Aurelia labiata Chamisso and Eysenhardt, 1821
(Figs. 2; 3B-D)

Aurcllia labiata Chamisso and Eysenhardt, 1821: 358. pi. 28. fig, 1A,
B. — Mayer. 1910: 622. 628, in part, eastern North Pacific records only.

Medusa labiata. — Eschscholtz, 1829: 64.

Aurelia labiata.— de Blainville. 1834: 294. pi. 42. figs. 1. 2 (Cham. &
Eysen. illustrations). — Lesson, 1843: 377. — L. Agassiz, 1862: 160. —
A. Agassiz. 1865: 43.— Haeckel, 1880: 557 (monograph).— Fewkes,
1889a: 593 (Point Conception. Monterey; manubrium).— Torrey, 1909: 1 1
(coll. by Cham. & Eysen.).— Barr and Barr. 1983: 80. text fig. 28 (Field
Guide (= FG): AK).— Wrobel and Mills, 1998: 55 (FG: Pacific coast).—
Gershwin. 1999: 993-1000. in part (symmetry variation).

Aurelia aurita non Linnaeus 1758. — Hauser and Evans, 1978: 21 text
photo. 81 (commensal crab).— Snively, 1978: 152 text fig., pi. 77 (FG: BC,
WA, OR).— Gotshall, 1994: 24, fig. 40 (FG).

Aurelia sp. — Campbell, 1992: 12. 13. Back cover (photographs). —
Greenberg et ai. 1996: 401-409, in part, text fig 3, 4 (allozymes).

Moon jellyfish.— Malnig. 1985: 40 (photograph).— Stefoff, 1997: 9
(photograph).

Holotype. Apparently not extant.

Neotype. CASIZ 111024, Monterey Bay. CA, coll. 19
April 1997 by D. Wrobel; gravid female, preserved 25-cm
bell diameter (BD), 12-cm manubrium length (ML).

Additional preserved material. CAS 20, Farallon Islands.
East Landing, coll. 14 Sep 1975 by D.R. Lindberg. CAS
95506, same data as CAS 20. CAS 95507, same data as
CAS 20. CAS 81306, Monterey Bay, Pacific Grove, coll. 13
Nov 1990 by N. Greenberg, ca. 15-cm BD, manubr. 6.5 cm.
CAS 81307, Monterey Bay, Pacific Grove, coll. 13 Nov
1990 by N. Greenberg, BD ca. 15 cm, ML ca. 6 cm. CAS
86767, 2 specimens, Vancouver Island, Sooke Basin, Roche
Cove. coll. 11 Sep. 1990 by N. Greenberg. 14.5-cm BD, 6
cm ML. CAS 81304, Monterey Bay, Pacific Grove, coll. 13
Nov 1990 by N. Greenberg, ca. 13-cm BD, ca. 4-cm ML.
CAS 81306, Monterey Bay, Pacific Grove, coll. 13 Nov
1990 by N. Greenberg. CAS 107800, 2 specimens,
Monterey (CA), coll. 30 July 1966 by Rofen. CAS 111016
and 1 1 1020, Brentwood Bay, Saanich Inlet, coll. 24 June
1996 by LG. CAS 111017. Point Defiance, Puget Sound,
coll. 5 April 1996 by LG. CAS 1 1 1021-1 1 1022, numerous
specs, Santa Barbara, coll. 30 Nov 1996 by S. Anderson.
CAS 1 1 1023, numerous specs, Marshall dock, Tomales Bay
(CA), coll. 30 June 1996 by LG. CAS 111227, Spinnaker
Bay, Long Beach (CA). coll. Sep 1995-Jan 1997 by L.
Gershwin. In addition, preserved, unregistered specimens
were examined from collections at Bodega Marine Labora-
tory, Cabrillo Marine Aquarium, Friday Harbor Laboratory,
and Santa Barbara Museum of Natural History.
Diagnosis. Aurelia with manubrium elongated, wide, pro-
truding below the bell margin when viewed laterally. Oral
arms shorter than bell radius, attached to base of manu-

brium. extending outward to bell margin or bent at 90°
angle typically counterclockwise. Bell margin 16-scalloped.
with a primary indentation at each of 8 rhopalia and a
secondary indentation midway between rhopalia. Older in-
dividuals typically with many radial canals arising from
each gastro-genital sinus; in some, the outer branches are
greatly anastomosed. Embryos and larvae brooded on the
manubrium or on stiff, shelf-like manubrial extensions,
rarely on the oral arms.
Redescription.

Medusa. (Based on mature tetramerous individuals.) Bell
typically quite flat at rest, in some subhemispherical; older
individuals may have raised hump over gonadal region.
Diameter at maturity ranging from 100 mm to 450 mm,
depending on population. Manubrium fleshy, rigid; rectan-
gular, pyramidal, or rounded in side view; variably ruffled at
4 corners; width approximately 1/3 of bell diameter; with
stiffened, whorled, perradial mesogleal extensions. Index of
manubrium length to bell diameter varying geographically,
longest in Oregon (.v = 37.2% ± 3.6%; n = 10. Newport),
shortest in southern California (x = 16.7% ± 2.6%; n = 7,
Spinnaker Bay, Long Beach). Oral arms 4, perradial,
straight or curved at 90° angles typically counterclockwise
(but occasionally variable), arising from base of manu-
brium; length short, reaching approximately to bell margin
(thus only ± 1/3 bell diameter); extending laterally outward
against subumbrellar surface of bell. In older cultured indi-
viduals, oral arms may hang downward. Size of subgenital
ostia varying, encircled by raised mesoglea in some indi-
viduals. Interradial and adradial canals typically un-
branched; perradial canals branched once, or in large indi-
viduals the gastro-genital sinus may overgrow the
trifurcation causing the perradial canal to appear un-
branched. Eradial canals branched. 4-12 arising from each
gastro-genital sinus. Some large specimens have conspicu-
ous anastomoses of canals on outer third of bell. Gastro-
genital sinuses interradial, 4. but varying from 1 to 8 (per-
haps more), in rounded to flattened horseshoe-shaped or
heart-shaped rings, with adaxially-pointing free ends. Bell
with 16 marginal scallops produced by 8 primary indenta-
tions at rhopalia located along the perradial and interradial
axes, with secondary indentations between adjacent rhopa-
lia. Bell transparent and colorless in juveniles and young
adults, becoming milky white, or tinted pinkish, purple,
peach, or bluish in older medusae. Color of gonad pale
pinkish or brownish in mature females, dark purple in
mature males, but often appearing white in males ready to
spawn.

Plumtlu. Elliptical to elongated; ciliated. Color most of-
ten white, but other colors found in certain populations:
lavender (Monterey), peach (Saanich Inlet), or yellow-ochre
(Spinnaker Bay). Planktonic or benthic locomotion by cili-
ary movement. Brooded on manubrium or its whorls.

Scvphistoma. Polyps 2-3 mm in height, with oral disk 1 -2

110

L. GERSHWIN

mm diameter. Manubrium short, cruciform. Septal funnels
conspicuous. Typically with 16 tentacles, alternating shorter
and longer: number of tentacles highly varied, often corre-
sponding to symmetry of parent medusae, parent polyp, or
offspring ephyrae. At Friday Harbor, Washington, and
Santa Cruz Island. California, scyphistomae typically with
20 tentacles. Color whitish to pale pinkish-orange. Habit
benthic. usually hanging downwards from underside of
docks, mussel shells, or rocks. Asexual proliferation by side
budding, stolon budding, or podocyst formation. See Chia et
al. (1984) for a histological study of the neuromuscular
system.

Strobila. Ranging from monodisk to polydisk with more
than 20 developing ephyrae. Color varying with locality:
cinnamon in southern California, buff in Monterey. Polyp
remaining flesh-colored or whitish. Strobilation time about
7 days; easily induced with periods of chilling.

Ephyra. Diameter 2-3 mm at release. With 8 marginal
arms, each with a terminal rhopalium flanked by 2 lappets.
Nematocysts scattered over the exumbrellar surface. Num-
ber of arms and rhopalia highly varied, not always in
correspondence with each other or within a clone. Color
same as the strobila: cinnamon or pale butt.
Type locality. Monterey Bay, California.
Distribution. I have collected A. luhinui from Saanich Inlet,
British Columbia, to San Diego, California. To the north, I
was able to confirm its presence in Prince William Sound,
Alaska, from a videotape; the species has also been photo-
graphed at Steamer Bay, in southeast Alaska (Barr and Barr,
1983). Its range may extend southward into the waters off
Baja California. Mexico. The species generally occurs in
bays and harbors where it is easily collected from jetties and
boat slips, but medusae have been observed drifting in open
waters off Santa Barbara, California (S. Anderson, Univ.
California Santa Barbara, pers. comm., Nov. 1996), near
Monterey Bay. California (D. Wrobel. Monterey Bay
Aquarium, pers. comm.. Oct. 1996; D. Powell, Monterey
Bay Aquarium, pers. comm.. May 1997). off Newport.
Oregon (D. Compton, Oregon Coast Aquarium, pers.
comm., June 1996). and in Puget Sound (LG. pers. obs.,
June 1996). The polyps generally strobilate in early spring,
and the medusae quickly mature, spawn, and die by mid-
summer or early fall. In some years and in some localities,
the population of medusae is present throughout the year
(Spinnaker Bay, LG. pers. obs.; Monterey. D. Wrobel. pers.
comm.).

Biogeography

Observations of 1 7 populations from San Diego. Califor-
nia to Prince William Sound. Alaska have shown that the
species can be reliably subdivided into three easily distin-

1
•g- 0.35

! 03

j! 025

— |

1 0.2

E 015

•c
•§ 01

5 005
n

Populations

Figure 4. Average manubrium lengths of Japanese Aurelia cf. uuriia
and nine populations of A. labiata. Japanese = Aurelia cf. auritu. cultured
at Cabrillo Marine Aquarium. Northern morph: Saanich = Saanich Inlet.
British Columbia; Pt. Def. = Poulsbo, Washington (cultured at Pt. Defi-
ance Aquarium); Seattle = Poulsbo. Washington (cultured at Seattle
Aquarium). Central morph: Newport = Newport. Oregon (captive at
Oregon Coast Aquarium); Tomales : Tomales Bay, California;
Monterey = Monterey. California. Southern morph: Marina = Marina del
Rey. California; Spinnaker = Spinnaker Bay. California; Coronado =
Coronado Island, California. Between morphotype comparison. ANOVA:
F = 42.595. df = 3.5, P = 0.001.

guishable geographical morphotypes. Though bell diameter
is highly variable with environmental conditions, even
among nearby populations (Lucas and Lawes, 1998), ma-
nubrium length, expressed as a percentage of bell diameter,
differs significantly among the three forms (Fig. 4,
ANOVA; F == 42.595, df = 3,5. P = 0.001). These three
forms are easily distinguished as follows (summary in Table
1 ). Following the synopsis of each form is a list of literature
that pertains to Aurelia from the region, but contains insuf-
ficient information for positive determination.

Southernmost form (Fig. 3B). Manubrium a wide,
rounded frilly mound, not distinctly pyramidal. Radial ca-
nals few to many, possibly dependent on age; adradials
particularly wide in San Diego medusae. Oral arms typically
straight, not curved. Planulae ranging in color from white to
ochre to bright orange, brooded in a reticulating pattern on
frills of manubrium. Bell colorless to milky whitish; some
individuals with dark purple tentacles. Male gonads dark
purple, female gonads pale pink. Typical maximum size, 35
cm. Marina del Rey medusae with pronounced rhopalial
hoods set up off the margin.

Known range. California, from San Diego to Marina del
Rey. possibly extending north to Ventura and south into
Baja California. Populations are apparently isolated and
discontinuous; not observed at Oceanside. Dana Point. Los
Angeles Harbor, or Malibu. Reported at Catalina Island.
Local residents at Ventura Harbor and Channel Islands
Harbor tell of seeing an occasional medusa or two; it is
currently unclear if they are this form. Typically occurring
until late spring, occasionally into autumn.

SYSTEMATICS AND BIOGEOGRAPHY OF AURELIA LABIATA

Table I

Comparison ofmorphotype characters, Aurelia labiata

Character

Northern morph

Central morph

Southern morph

Manuhrium length (x) (as % hell diam)

22.98%

37.15%

16.73%

±0.04%

±0.03%

site

Poulsbo/Saanich

Newport

Spinnaker Bay

Manuhrium shape

pyramidal

long and tapered

rounded

Oral arm length

1/3 bell diam

1/3+ hell diam

1/3 hell diam

Oral arm shape

± straight

counterclockwise

straight

# canals per sinus

7-9

7-15

5-7

Anastomosing

heavy

very heavy

moderate

Bell size

to 12 cm

to 45 cm

to 25 cm

Bell color

whitish or peach

purple, pink, or white

whitish

Planula color

white or peach

white or purple

white, ochre, or orange

Literature.

Aurelia aurila, — MacGinitie and MacGinitie, 1949: 131, text fig. 32
(growth, strobilation, Newport Bay). — MacGinitie and MacGinitie, 1968:
131. text fig. 32 (growth, strobilation, Newport Bay). — Reish, 1972: 25.
text fig. 26 (FG: Southern CA).— Allen, 1976: 22. 75 (FG: Southern
CA).— Reish. 1995: 38, fig. 31 (FG: Southern CA).

Central fonn (Fig. 3C>. Manubrium extremely elongated,
rectangular and tapering. Canals numerous, typically
heavily anastomosed in largest individuals. Oral arms
straight or bent counterclockwise. Planulae distinctly laven-
der, brooded in teardrop-shaped clumps on the base of
manubrium or on shelves. Scyphistomae pale buff colored.
Medusae from Monterey, California tending to be distinctly
purple; Santa Barbara, California, medusae often pale pink.
Gonads dark purple in males, pale brown in females. Di-
ameter of captive medusae from Newport, Oregon, recorded
to 45 cm, with longest manubrium being 17 cm!

Known range. Santa Barbara (including Channel Islands),
California to Newport, Oregon. Likely occurring, but un-
confirmed, along the outer coast of southern Washington
state. Abundant in late summer.

Literature.

Aurelia labiata. — Fewkes, 1889b: 122 (Santa Barbara Channel; pink). —
Boyd, 1972 (fouling organism; Bodega Harbor, CA).— Pearcy, 1972: 354
(Oregon).— Hand, 1975: 95 (FG: Central CA).

Aurelia aitrita. — ?Galigher, 1925: 94 (scyphistomae; Monterey. CA). —
Hamner and Jenssen, 1974:833-848. text fig. 1 (growth and degrowth.
Tomales Bay, CA).— Shenker. 1984: 619-630 (abundance; OR).— Abbott.
1987: 28 (morphology; Monterey).— Keen and Gong, 1989: 735-744 (scy-
phistoma clonal growth; Tomales Bay, CA). — Niesen. 1997: 43 (FG:
Northern CA). — Rigsby. 1997: 207 (Monterey Bay).

Aurellia labiata.— Light ct ai. 1954: 41 (FG: central CA).

Aure/lia aurira.— Hedgpeth, 1962: 52, text fig. B (FG: Northern CA).

Aurelia sp. — Gottshall et a/.. 1965: 149 (prey of blue rockfish; Bodega.
Monterey, Morro Bay). — Pereyra and Alton. 1972: 448 (near Columbia
River. OR).

Northernmost form (Fig. 3D). Manubrium low. pyrami-
dal. Many parallel radial canals in mature individuals, giv-
ing a lacy appearance to the bell. Oral arms more or less
straight, but may be variable in the same individual in

Departure Bay specimens (M. Arai, Pacific Biological Sta-
tion. Nanaimo, BC. pers. comm. 2000). Planulae variably
colored; brooded at the base of the manubrium and on
manubrial shelves. Overall coloration peach or whitish, with
gonads dark purple in males, pale brown in females. At
Poulsbo, Washington, maximum diameter approximately 12
cm; brooded planulae white, appearing as a wash or haze
rather than in discrete bundles. At Saanich Inlet. British
Columbia, medusae larger, to approximately 15-cm diame-
ter during my study, but reported to range from 14-29-cm
(Hamner ct <//.. 1994); brooded planulae peach-colored.

Known range. Puget Sound. Washington, to Prince Wil-
liam Sound, Alaska; mainly occurring in late spring.

Literature.

Aurelia labiata.— Carl, 1963: 101 (FG: BC).— Kozloff. 1974: 22. in part
(FG: WA).

Aurelia limhuta.— ?Stiasny. 1922: 522 (Vancouver).— ?van der Maaden,
1939: 33 (rhopalial folds; Vancouver).

Aurelia aitrita.— Bigelow. 1913: 98 (marginal scallops; Puget Sound). —
Clemens. 1933: 16 (Canada).— Kozloff, 1973: 62, text photo 10 (FG:
WA).— Arai and Jacobs. 1980: 120 (medusivory: BC).— Mills. 1981: 22
(seasonality; Puget Sound).— Kozloff, 1983: 56. text photo 13 (FG;
WA). — Chia et «/., 1984: 69-79 (scyphistoma structure; Puget Sound). —
Larson, 1986: 107-120 (chemical composition; Saanich Inlet). — Kozloff,
1987: 65 (FG: Pacific Northwest). — Larson. 1987: 93-100 (carbon cycling;
Saanich Inlet). — Strathmann, 1987: 76 (development; Puget Sound). —
Strand and Hamner. 1988: 409-414 (prey of Phucellup/wra, Saanich In-
let).— Norris. 1989: 381-393 (fossilization).— Arai. 1991: 363 (chemical
predation cues; BC).— Keen. 1991: 1-176 (scyphistoma biology; Tomales
Bay). — Fautin and Lowenstein. 1992: 13 (polyp and medusa proteins). —
Hamner ft al., 1994: 347-356 (sun migration; Saanich Inlet).

Aurelia sp.— MacGinitie, 1955: 120 (color range; Pt. Barrow, AK). —
Zubkoff and Lin. 1975: 915 (isozymes).

In addition to the literature apparently attributable to each
form above, a large body of literature exists which pertains
to Aurelia of the eastern North Pacific but cannot be attrib-
uted to a single region as described above. Many of these
references do not illustrate the species, or in some cases, use
general drawings or photographs from other locations.

112

L. GERSHWIN

Aureliti uiirihi. — Johnson and Snook. 1927: 82, text fig 62 (FG). —
Guberlet, 1936: 45. text photo (FG: Northwest).— Guberlet, 1949: 45. text
photo (FG: Northwest). — Hartman and Emery. 1956: 307 (CA). — Guber-
let. 1962: 45. text photo (FG: Northwest).— Flora and Fairbanks. 1966: 50,
Fig. 42: (FG: BC. WA. OR).— Johnson and Snook. 1967: 82. text fig 62
(FG).— Brusca and Brusca, 1978: 52, text fig. 22 (FG: CA).— McLachlan
and Ayres. 1979: 47, text photo (FG: Pacific Northwest). — Gotshall and
Laurent, 1980: 40. text photo 40 (FG, Pac. coast). — Haderlie ci ai, 19X0:
52. pi. 3.22 (FG: CA).-Audubon Society. 1981: 363, in part. pi. 502 (photo
is of A. aurita, possibly outside NE Pacific). — Austin, 1985: 71 (Alaska to
southern California). — McConnaughey and McConnaughey. 1985: 466, pi.
384 (photo is of A. aurita, but may have been taken elsewhere). — Ricketts
et al.. 1985: 303, text fig. 316 (FG).— Farmer, 1986: 111 (FG; AK to so.
CA). — Parsons. 1986: 18 (sting treatment). — Connor and Baxter. 1989: 53
(in kelp forest). — Amos. 1990: 36. in part, Alaska to southern California
(photo is of Aequorea sp. (Cnidaria: Hydrozoa) but attributed to A. au-
rita).— Larson, 1990: 546-556 (distribution). - Larson and Arneson. 1990:
130-136 (California).— Niesen. 1994: 48, text fig. 4-33 (FG: CA).— Thuesen
and Childress, 1994: 84-96 (enzyme activity; southern and central CA).

Aurelia (and Aurelia sp. ). — Ricketts and Calvin. 1939: 244. text fig. 109
(FG).— Wells, 1942: 146, text fig. (FG).— Ricketts and Calvin, 1948: 144,
244. text fig. 109 (FG).— Ricketts and Calvin. 1952: 328, text fig. 109
(FG).— Smith. 1962: 13, text fig. 10 (FG: Pac. Northwest).

Aurellia aurira. — Light. 1941: 19 (invert, manual). — Ricketts and Cal-
vin. 1968: 264, text fig. 266 (FG).

Aurellia (and Aurellia sp.). — Tierney ft ai. 1967: 26. text fig. (FG).

Jellyfish.— Ulmer. 1968 (children's book).

A urelia labiata.— North. 1976: 153 (FG: CA).— Austin. 1985:71 (Alas-
ka to central CA).

A second Aurelia introduction

A second population of Aurelia "aurita." apparently in-
troduced, has recently been found at Spinnaker Bay. Long
Beach, California (the first was found at South San Fran-
cisco Bay, California, by Greenberg et ai, 1996). It is
impossible to know exactly when it first appeared; however,
I have been working closely with the Spinnaker Bay pop-
ulation since 1995, and have only observed this other form
since 1997. Morphologically, it is allied to the European and
Japanese forms. However, preliminary 18S rDNA partial
sequence analyses indicate that it is similar to a population
from Fort Lauderdale. Florida (J. Lowrie, Cnidarian Re-
search Institute, pers. comm., June 2000). Lowrie has fur-
ther found that the Spinnaker Bay population clusters into at
least four genetic subpopulations, one closely related to
island populations, one as described above, and two appar-
ent hybrid forms. This pattern is evident in the morphology
as well. Since 1997, both A. labiata and A. "aurita" medu-
sae have been observed side by side, as well as some that
possess characters of both.

Comparison with Japanese Aurelia "aurita"

The Japanese form of A. "aurita" closely matches the
descriptions of the European form (e.g., Russell, 1970), and
thus differs morphologically from A. lahiata in a similar
manner. Kishinouye (1891) described a form from Tokyo
Bay, Japan, named Aurelia japonica; it was said to differ

from A. aurita in having prominent subgenital cavities and
in having broad and folded lobes on the proximal halves of
the oral arms (Kirkpatrick, 1903). Whether this form is
identical to the European form or to A. flavidula, or to the
Japanese material presently raised in American public
aquariums, has not yet been determined and is beyond the
scope of this paper.

Notes on Aurelia limbata

Upon casual inspection. A. limbata appears to be unmis-
takable because of its chocolate-brown marginal pigment
band (see Audubon Magazine, Jan. 1974 cover, for an
excellent photograph). It also appears to be distinctive in
having relatively few tentacles and in the extreme anasto-
mosing of the radial canals in all growth stages. However,
closer examination may show A. limbata of the Arctic to be
a fourth morph of A. labiata. or possibly even a color
variant of the northern form. Mayer (1910) regarded A.
limbata as a variety of A. labiata, apparently based on its
having 16 marginal scallops. I have not had the opportunity
to examine specimens of A. limbata, but written descrip-
tions, drawings, and photographs reveal additional similar-
ities. Like the northern form of A. labiata. A. limbata has a
triangular protruding manubrium and many radial canals
emanating from each gastro-genital sinus (Kishinouye,
1910; Faulkner, 1974; but the former character is not ap-
parent in Mertens's illustrations published by Brandt in
1838). In addition. A. limbata shares with the Marina del
Rey, California, population of A. labiata the peculiar char-
acter of large and conspicuous rhopalial hoods that are well
above the bell margin. There has been some debate about
the phylogenetic meaning of wrinkles in the rhopalial pits
(see Uchida, 1934; van der Maaden. 1939); this character
has not been checked in A. lahiata. Furthermore, the anas-
tomosing of the radial canals is far more developed in A.
limbata. If the two nominal species are eventually regarded
as conspecih'c, the name A. labiata would have chronolog-
ical priority. More logically, A. limbata may be a separate
species in an undefined species complex currently known as
A. labiata.

Discussion

Biogeographical and svstematic implications

Most twentieth century authors regard Aurelia aurita as
cosmopolitan, occurring abundantly the world over, and
some recognize Aurelia limbata of the Arctic Ocean as the
only other species in the genus. These notions are dispelled
by the present results. Not only is A. aurita replaced along
the American Pacific coastline by A. labiata, but the latter is
also divided into three morphologically distinctive forms
coincident with established bioprovinces. Furthermore,
there is some evidence that A. limbata may be a color morph

SYSTEMATICS AND BIOGEOGRAPHY OF AURELIA LABIATA

or possibly even a separate species within the clade cur-
rently known as A. labiata. Thus, the Aurelici group may
actually consist of numerous local species, as was indicated
by Lambert (1935 ). Hummelinck ( 1968), Hamner and Hauri
( 1981 ). and Brewer ( 1991 ) for other taxa, or possibly even
more than one genus. Future molecular analysis of the
morphotypes may elucidate the degree of differentiation.

One of the predictions of this hypothesis is that additional
populations of Anrelia found along the Pacific coast of
North America may be assignable among the three morpho-
types, according to morphology and latitude. The eastern
North Pacific flow patterns are consistent with the morpho-
logical differences of the jellyfishes, with both currents and
morphologies diverging in the vicinity of Point Conception,
California, and Puget Sound. Washington. The three regions
corresponding to the morphotype ranges are coincident with
the Calitbrnian. Oregonian. and Aleutian bioprovinces of
molluscs (Hall. 1964: fig. 5; see also Valentine. 1966: fig.
1 ). Although molluscan provinces appear to be determined
by sustained reproductive water temperatures (Hall. 1964).
the cause of similar distribution in Anrelia is currently
without explanation. Logically, temperature could play a
role, but Anrelia is able to grow and reproduce continually
in the laboratory in a wide range of temperatures, both
cooler and warmer than the ambient ocean temperature
(unpubl. notes). It is well documented that the distributions
of benthic groups such as molluscs (Campbell and Valen-
tine. 1977; Roy et al. 1998) and algae (Abbott and Hollen-
berg. 1976) conform to biogeographical provinces. In con-
trast, the ranges of pelagic taxa are typically thought to be
ill-defined at the fine scale, being confined primarily by the
great gyres, if not cosmopolitan (Lalli and Parsons, 1993;
Nybakken. 1993). For a nearshore pelagic invertebrate such
as A. labiata. this generalization does not hold true. Further
studies should examine Anrelia and other widespread
coastal medusae in regions with similar latitudinal gradi-
ents, that is. eastern and western continental shores in both
hemispheres.

Several recent studies may become important in our
understanding of nearshore medusa distribution. First, Hell-
berg (1996) examined differential gene flow between one
coral species that brooded its larvae and another with pe-
lagic, feeding larvae: he found greater genetic subdivision in
the brooding species. Likewise, Anrelia spp. and Cyanea
spp. are planula brooders, and thus may have less gene flow
among populations than previously assumed. Second, Co-
wen et al. (2000) found that simulated larvae do not disperse
as readily as generally thought. Indeed, it appears that
dispersal in some cases may be overestimated by nine
orders of magnitude. Medusae, like larvae, are not passive
particles. Rather, their dispersal ability is subject to their
own behaviors as well as to diffusion and mortality. Many
medusae swim actively against a gentle current, or drop
lower in the water column to avoid currents (pers. obs.);

these behaviors may serve as anti-dispersal mechanisms.
Finally. Barber et al. (2000) found a sharp genetic break in
nearby populations of the mantis shrimp Haptosquilla />///-
chella in Indonesia, and suggested the presence of a sort of
"marine Wallace's line." Even though the stomatopod lar-
vae are planktonic, and thus have the means to disperse over
great distances, it appears that they do not. Whether the
same explanation can be applied to Anrelia remains to be
shown.

Because so much of the coastline is hospitable to A.
labiata, it is helpful to ask whether other similar species
may be present as well. Currently there is no evidence of
endemic species other than A. labiata, excepting the unre-
solved nomenclatural questions relating to A. limbata. How-
ever, it is easy to imagine that other forms may have been
overlooked in a similar way as A. labiata. or that within the
species I herein recognize as A. labiata. numerous cryptic
species exist. The recent scientific literature abounds with
discoveries of cryptic species, such as one recent startling
example, wherein the fungal Gibberella fiijiknroi species
complex was found to comprise 45 phylogenetic species
(O'Donnell et al.. 1998)! Given that many of the popula-
tions of A. labiata along the eastern North Pacific coast are
uniquely diagnosable. and that these diagnosable forms
partition into the three latitudinal morphotypes, the possi-
bility of cryptic species seems high. Indeed, Greenberg et al.
(1996) hypothesized restricted gene flow between eastern
Pacific populations, based on significant allele frequency
differences. Thus, the biogeographic pattern in A. labiata
may represent cladogenesis in action, or possibly even a
splitting event of the recent past. I hesitate at this time to
recognize the three forms as distinct species, or for that
matter to assign the eastern North Pacific forms to a new
genus, although it is clear that the three forms are quite
different from one another and from A. anrita. Although
scyphozoan population genetics have not yet been studied in
depth, some cnidarians have surprisingly low rates of ge-
netic divergence (see Knowlton. 2000), so species conclu-
sions should be made cautiously. Thus, until the clade
currently known as A. anrita is resolved, it is difficult to
comment with confidence on the internal and external rela-
tionships of the morphotypes of A. labiata. However, this
does beg the questions of species concept and species rec-
ognition criteria.

Taxonomic characters

Throughout most of the twentieth century, it was custom-
ary to recognize medusan taxa based on certain key char-
acters, reeardless of distribution and discrete forms of vari-
ation; that is, all populations possessing a small number of
aiven characters were thought to be one species. For exam-
ple, in the Pelagiidae. the character of tentacle number has
been so highly regarded that a large and conspicuous spe-

114

L. GERSHWIN

cies was incorrectly classified, favoring a tentacle number
over all other characters combined (Gershwin and Collins,
2001 ). The same reasoning seems to have applied to Aiire-
lia, favoring the "essence" of A. aurita over all other char-
acters. This appears to have resulted in excessive lumping
for many taxa. In contrast, I have employed a phylogenetic
perspective, bringing together data from morphology, ge-
ography, and genetics to evaluate a lineage's history. How-
ever, some characters are still worthy of further comment.
as they have led to confusion in the past.

Perhaps the most ignored character is the best key in
separating A. labiata from A. aurita. Greenberg et al. ( 1996)
used manubrium length in distinguishing the American
form from the Japanese form, but failed to notice the asso-
ciated changes in the relationship of the oral arms to each
other and the altered brooding habits (Figs. 1, 2. 3B-D). To
summarize, in A. labiata the oral arms are relatively short,
about one-third the bell diameter, and project outward from
the base of the fleshy manubrium. In addition, the larvae are
brooded on the manubrium or on the rigid manubrial
shelves. In contrast. A. aurita lacks the fleshy manubrium;
consequently, the oral arms meet at the mouth and are about
one-half the bell diameter. Furthermore, the brood pouches
for the larvae line the upper portions of the oral arms. Thus,
the large manubrium of A. labiata relates to a suite of
morphological and functional differences from A. aurita.

Kramp (1913) considered the anastomosed canals to be a
distinctive character in separating the Greenlandic form of
A. (inritu (as A. flavidula) from the typical form, and most
descriptions of A. limbata include this character. However,
the canals of some captive medusae of both A. labiata and
A. "aurita" eventually become heavily anastomosed (F.
Sommer, Monterey Bay Aquarium, pers. comm., and my
own unpublished observations), possibly attributable to the
phenomenon of growth and degrowth (Hamner and Jenssen,
1974). This was not taken into consideration by Greenberg
et al. ( 1996). in claiming that the anastomoses could be used
as a reliable character for distinguishing eastern Pacific
Aurelia from western Pacific Aurelia. Indeed, their North
American medusae were held captive nearly a year, whereas
their Asian medusae were held only for 2 months. Although
this character does seem more conspicuous in large speci-
mens of A. lahiata than in A. "aurita, " this may be due to
the increased number of canals in A. labiata: that is, many
canals anastomosing may give the appearance of a finer
mesh than one would expect in an individual with fewer
canals. This too (extra canals) was not taken into account by
Greenberg et al. (1996). A closer study of anastomosis of
canals might be helpful in future taxonomic studies.

Some authors have reported that the number of canals
arising from the gastro-gonadal sinuses is taxonomically
unreliable because it is associated with size and rate of
growth (Stiasny, 1922; Bigelow, 1938; Kramp, 1942, 1965;
Russell. 1970). Indeed. I have observed that older, larger

individuals do tend to have more canals than smaller,
younger individuals. However, old. large A. aurita typically
have 1 or 2 eradial canals arising in each space between
interradial and adradial canals (for a total of 5-7 canals
arising from each gonad). whereas old, large A. labiata
typically have 3-6 eradials per side (for a total of 9-15 total
per gonad). However, in the closely related A. limbata,
Stiasny (1922) and Bigelow (1938) argued that the number
of radial canals and the degree of branching are both useful
characters. Curiously, medusae of the northern and central
forms tend to possess greater numbers of radial canals than
do medusae of the southern form.

The taxonomic significance of the 16-scalloped bell mar-
gin is currently unclear. Medusae from all endemic eastern
North Pacific populations that I have observed possess this
scalloping, in some cases quite conspicuously so. However,
use of this character to distinguish species has been criti-
cized by Kramp (1965). citing that in A. limbata the sec-
ondary scalloping is lost in preservation, and agreeing with
Bigelow (1913) that the degree of scalloping is merely due
to contraction of the bell. Because of its occasional occur-
rence in A. aurita. the secondary scalloping should not be
used as the distinguishing taxonomic character of A. labiata
as has been done in the past. However, it remains one of
several useful field characters for A. labiata and may prove
useful in similarly distinguishing other species worldwide.

Confusion has arisen regarding certain specimens from
Nanaimo. British Columbia. Stiasny (1922) and van der
Maaden (1939) assigned them to A. limbata: whereas
Kramp ( 1942) identified them as a variety of A. aurita based
on the width of their radial canals. I have not yet examined
these specimens. However, Stiasny's (1922) description is
consistent with A. labiata, namely, the 16-scalloped margin
and the 5-9 radial canals issuing from each gastrovascular
sinus.

At present. A. labiata appears to be a temperate endemic
restricted to the eastern North Pacific. However, this leaves
a series of references to medusae with 16 marginal scallops
as A. labiata, although their morphological characteristics
and geographic locations suggest that they are not. Avail-
able drawings and a photograph all clearly show 16 scallops
of the margin, but do not show a protruding manubrium or
numerous radial canals (Mayer. 1910. 1917; Uchida, 1928).
Since the illustrations of Chamisso and Eysenhardt (1821)
indicate a large manubrium. I exclude medusae that lack this
character from this classification. However, I have not ex-
amined specimens from the following sources for complete
diagnostic characteristics.

Aiiivlliu /<;/>/<»<;.— Mayer 1910: 628, fig. 398 (A. limbata as var. of A.
Uihiutii; Philippines). — Light, 1914a: 294 (harmless); Philippines). — Lite,
I914b: 200 (Philippines).— Mayer. 1915: 160, 1S2 (A. labiata derived from
A. aurita). — Mayer. 1917: 205. text fig. 11 (Philippines and Tortugas. Flor-
ida).— Light, 1921: 31 (Philippines). — Bigelow. 1938: 167 (synonymous with
A. aurita).

Aurelia labiata. — Stiasny, 1919: 93 (Malay Archipelago). — Stiasny.

SYSTEMATICS AND BIOGEOGRAPHY OF AURELIA IARIATA

115

1926: 244 (Philippines; ,4. labiata is a variety of A. aiirita). — Uchida 1928:
373-376 (pentamerous. Palau). — Stiasny. 1931: 140 (-specimen a( British
Museum). — Stiasny. 1935: 34 (Aroe Islands). — Stiasny. 1937: 207 (East
Indies). — Ranson, 1945: 60. 61 (review of genus).— Kramp 1961: 340
(taxonomy). — Kramp. 1965: 262-263. plate 1 rig. 1 (A. labiata same as A.
auriun.— Kramp 1968: 68 (discusses A. labiata). — Russell 1970: 140
(discussion of synonymy). Powell, 1975: 6 (New Zealand I.

Two reports of A. labiata in Hawaii (Chu and Cutress.
1954: 9; Devaney and Eldredge. 1977: 1 1 1) are worthy of
attention. Drawings I made in 1993 from live animals in the
Waikiki Aquarium appear to be of A. luhiutii. However,
preserved specimens from the same location examined in
1997 lacked the enlarged manubrium. At this time, I pro-
visionally include Hawaiian Aurelia with A. labiata. but
firm determination must wait until additional live and pre-
served material can be examined. The Oahuan form appears
to be introduced, as it was not reported until 1954, but the
origin of the introduction is not yet known (J. T. Carlton,
Mystic Seaport. Mystic. CT, and L. G. Eldredge. B. P.
Bishop Museum. Honolulu. HI, pers. cornm.).

Thus far, little consensus exists over what characters are
taxonomically reliable for jellyfishes over a wide range of
populations. To further confound the problem, immature
specimens of closely related species often bear a striking
resemblance. However, recent rearing of Japanese Aurelia
"aiirita" and Monterey A. labiata in the same aquarium
yielded distinctive morphs consistent with the two species
(M. Schaadt, Cabrillo Marine Aquarium. San Pedro, CA,
pers. comm., Oct. 1999). Although I have herein distin-
guished only the northern, central, and southern morphs,
medusae from each of the 1 1 locations were easily identi-
fiable. The ability to distinguish morphological characteris-
tics associated with particular populations of Aurelia spp.
will not only help to resolve the phylogeny of the group, but
may also help in identifying the origins of introductions
such as those in Spinnaker Bay, California; San Francisco
Bay. California (Greenberg el al, 1996); and Oahu, Hawaii
(J.T. Carlton and L.G. Eldredge, pers. comm., 1998).

Field key to the eastern North Pacific forms of Aurelia

1. Bell lacking secondary notches between adjacent rho-
palia. margin 8-scalloped. Lacking broad and/or elongated
manubrium. Currently known only from South San Fran-
cisco Bay and Spinnaker Bay cf. A. aitrila

1 '. Bell with secondary notches between adjacent rhopa-
lia, appearing 16-scalloped. Possessing conspicuously broad
and/or elongated manubrium 2

2. Bell with conspicuous chocolate-brown margin. Pri-
marily Arctic A. liiiibuta

2'. Bell lacking brown margin 3

3. Manubrium greatly elongated, tapering rectangular in
shape. Generally found Pt. Conception. CA, to northern
Oregon. Color variable from white to purple to pink. Often
very large, to 45 -cm or more . . . A. labiata. central morph

3'. With manubrium protruding in lateral view, but much
less than one-third bell diameter 4

4. Manubrium pyramidal. Generally found in and north
of Puget Sound. Color variable from white to peach. Typ-
ically small. 12-15 cm A. labiata. northern morph

4'. Manubrium rounded. Generally found south of Pt.
Conception. Color typically milky white, occasionally with
dark tentacles A. labiata, southern morph

Acknowledgments

I thank the staff and volunteers of the Cabrillo Marine
Aquarium for unwavering encouragement, Susan Gershwin
and Norma Kobzina for tracking down obscure references,
Richard Harbison for translation of Chamisso and Eysen-
hardt (1821). Eric Hochberg for valuable museum and
manuscript assistance, Claudia Mills and Allen Collins for
stimulating discussions and help in a multitude of ways,
Freya Sommer for sharing her knowledge and passion for
jellyfishes, Gary Williams for his artwork and taxonomic
guidance, Dave Wrobel for the beautiful photograph repro-
duced in Figure 2, the countless friends and colleagues who
provided valuable suggestions on previous versions of the
manuscript. Sincerest thanks to Mary Arai for providing
assistance beyond the normal standard for review, and to an
anonymous reviewer for additional helpful comments. In
addition. I am indebted to the following people and institu-
tions for help in obtaining specimens and information (in
alphabetical order): Leslee Yasukochi and Eric Johnson at
Birch Aquarium at Scripps; Jim Ulcickas at the Bluewater
Grill. Newport Beach. California; Cadet Hand and staff at
Bodega Marine Lab; Chris Mah at California Academy of
Sciences; researchers and students at Friday Harbor Labs;
Freya Sommer. Dave Wrobel, Dave Powell, and Ed Seidel
at Monterey Bay Aquarium; Dave Compton and Polly Delle
at Oregon Coast Aquarium; researchers and staff at Oregon
Institute of Marine Biology; John Carlyle at Point Defiance
Zoo and Aquarium; Yogi and Kathy Carolsfeld at Saanich
Inlet; Erin Johnston and Shaun Larson at Seattle Aquarium;
Spinnaker Bay and Spinnaker Cove homeowners; Thomas
Shirley and Jennifer Boldt at University of Alaska; Shane
Anderson at UC Santa Barbara; Rossi Marx at University of
Victoria; and Joyce and Stuart Welch at Tomales Bay. I am
thankful for financial support from the Friends of Cabrillo
Marine Aquarium, the Howard Hughes Medical Institute
Undergraduate Research in Biological Sciences Program,
and the University of California. Berkeley. UCMP Contri-
bution #1727.

Literature Cited

Abbott. D. P. 1987. Ohsi'iving Marine Invertebrates. Stanford Univer-
sity Press. Stanford. CA.

Abbott. I. A., and G. J. Hollenberg. 1976. Marine Algae of California.
Stanford University Press, Stanford. California.

L. GERSHWIN

Agassiz, L. 1862. Contributions to the Natural History of the United

States of America. Little, Brown, Boston.
Agassiz, A. 1865. North American Acalephae. Mem. Mus. Comp. Zool.

Harvard College 1(2): 1-234.
Agassiz, A. and A. G. Mayer. 1899. Acalephs from the Fiji Islands.

Bull. Mus. Comp. Zool. Harvard 32(4): 157-189.

Allen, R. K. 1976. Common Intertiilal Invertebrates of Southern Cali-
fornia, Rev. ed. Peek Publications. Palo Alto. CA.
Amos, S. H. 1990. Familiar Seashore Creatures. The Audubon Society

Pocketguide. Alfred A. Knopf, New York.
Arai, M. N. 1991. Attraction of Aurelia and Aequorea to prey. Hvdro-

biolot-ia 216/217: 363-366.
Arai, M. N. 1997. ,4 Functional Biology of Scyphozoa. Chapman and

Hall. London.
Arai, M. N., and J. R. Jacobs. 1980. Interspecific predation of common

Strait of Georgia planktonic coelenterates: laboratory evidence. Can. J.

Fish. Aauat. Sci. 37: 120-123.
Audubon Society. 1981. The Auduhon Societv Field Guide to North

American Seashore Creatures. Alfred A. Knopf. New York.
Austin, \V. C. 1985. An Annotated Checklist of Marine Invertebrates in

the Cold Temperate Northeast Pacific. Khoyatan Marine Laboratory,

Cowichan Bay, BC, Canada.
Barber, P. H., S. R. Palumbi, M. V. Erdmann, and M. K. Moosa. 2000.

A marine Wallace's line'' Nature 406: 692-693.
Barr, N., and L. Barr. 1983. Under Alaskan Seas. Alaska Northwest

Publishing Company, Anchorage.
Bigelow, H. B. 1904. Medusae from the Maldive Islands. Bull. Mus.

Comp. Zool. Harvard 39: 245-269.
Bigelow. H. B. 1913. Medusae and Siphonophorae collected by the U.S.

Fisheries Steamer 'Albatross' in the north-western Pacific. 1906. U.S.

Nut. Mus. Proc. Wash.. 191.1. 44: 1-119.
Bigelow, H. B. 1914. Fauna of New England. 12. List of the medusae

craspedotae. siphonophorae, scyphomedusae. ctenophorae. Boston Soc.

Nat. Hist. Occas. Pap. 7(part 12): 1-37.
Bigelow, H. B. 1920. Medusae and Ctenophores from the Canadian

Arctic Expedition. 1913-1918. Part H. Medusae and Ctenophora. Rep.

Canadian Arctic Exped. 8: 19 pp.

Bigelow, H. B. 1938. Plankton of the Bermuda Oceanographic Expedi-
tions. VIII. Medusae taken during the years 1929 and 1930. Zooloifica.

N.Y. 23 (part 2): 99-189.
Boyd, M. J. 1972. Fouling community structure and development in

Bodega Harbor. California. Ph.D. dissertation. University of California,

Davis.
Brandt,,]. F. 1835. ProJroinus descriptionis animalium ah H. Mertensio

in nrbis terrantm circumnavigations obsen'atonnn. Fascic. I. /V/v/'m.

Acalephas Discophoras et Siphonophoras, nee non Echinoderrnata

conlinens. Sumptibus Academiae. Petropoli.
Brandt, J. K. 1838. Austiihrliche Beschreibung der von C.H. Mertens

auf seiner Weltumsegelung beobachteten Schimiquallen. Mem. Acad.

Sci. St.-Petersh. Sci. Nat. Series 6. 2: 237-41 1.

Brewer, R. H. 1991. Morphological differences between, and reproduc-
tive isolation of. two populations of the jellyfish Cyanea in Long Island

Sound. USA. Hydrobiologia 216/217: 471-477.
Browne, K. T. 1905. Scyphomedusae. Fauna Geogr. Maldive & Lac-

cadirc Archcpclaiios 2(suppl. I): 958-971.
Brusca. G. J., and R. C. Brusca. 1978. A Naturalist's Seashore Guide.

Mad River Press, Eureka, CA.
Calder, I). R. 1971. Nematocysts of polyps of Aurelia. Chrysaora and

Cyanea and their utility in identification. Trans. Am. Microsc. Soc. 90:

269-274.
Campbell, C. A., and J. W. Valentine. 1977. Comparability of modern

and ancient marine faunal provinces. Paleobiology 3: 49-57.
Campbell, E. 1992. A Guide to the World of the Jellyfish. Monterey Bay

Aquarium Foundation, Monterey, CA.

Carl, G. C. 1963. Guide to Marine Life of British Columbia. British
Columbia Provincial Museum, Victoria, BC, Canada.

Chamisso, A., and C. G. Eysenhardt. 1821. De animalibus quibusdam
e classe Vermium Linneana. in circumnavigatione terrae, auspicante
Comite N. Romunzoff duce Ottone de Kotzebue, annis 1815-1818 per
acta, observatis. Nova Ada Acad. Caesar. Leap. Carol. 10: 345-374,
pi. 24-33.

Chia, F.-S., H. M. Amerongen, and D. G. Petaja. 1984. Ultrastructure
of the neuromuscular system of the polyp of Aurelia aurita L.. 1758
(Cnidaria, Scyphozoa). J. Morphol. 180: 69-79.

Chu, G. W. T. C., and C. E. Cutress. 1954. Human dermatitis caused
by marine organisms in Hawaii. Proc. Hawaii. Acad. Sci. 29th Annual
Meeting: 9.

Clemens, W. A. 1933. A Checklist of the Marine Fauna and Flora of the
Canadian Pacific Coast. National Research Council of Canada, Ot-
tawa.

Connor, J., and C. Baxter. 1989. Kelp Forests. Monterey Bay Aquar-
ium. Monterey, CA.

Cowen, R. K., M. M. L. Kamazima, S. Sponaugle, C. B. Paris, and
D. B. Olson. 2000. Connectivity of marine populations: Open or
closed? Science 287: 857-859.

de Blainville, H. M. I). 1834. Manuel d'Actinologie, ou de Zoophytolo-
gie. F. G. Levrault, Paris.

Devaney, D. M. and L. G. Eldredge. 1977. Class Scyphozoa. Pp.
108-118 in Reef and Shore Fauna of Hawaii, Section I: Protozoa
through Ctenophora, D. M. Devaney and L. G. Eldredge, eds. Bishop
Museum Press, Honolulu. HI.

Eschscholtz, F. 1829. System der Acalephen. Einc aiisfiilirliche Beschreib-
ung aller medusenartigen Strahhiere. F. Dummlert Berlin.

Fabricius, O. 1780. Fauna Groen/andica. Impensis loannis Gottlob
Rothe, Hafniae et Lipsiae.

Farmer, W. M. 1986. Seashore Discoveries. Farmer, Santee, CA.

Faulkner. D. 1974. (Cover Photograph). Audubon January 1974.

I .iiiiin. D. G., and J. M. Lowenstein. 1992. Scyphomedusae and their
polyps are the same immunologically: implications for systematics.
Comp. Biochem. Physio/. B 102(1): 13-14.

Fewkes, J. W. 1889a. On a few Califomian medusae. Am. Nat. 23:
591-602.

Fewkes, J. W. 1889b. New invertebrata from the coast of California.
Bull. Essex lust. 21: 99-146.

Flora, C. J., and E. Fairbanks. 1966. The Sound and the Sea. 2nd ed.
Pioneer Printing. Bellingham. WA.

Galigher, A. E. 1925. Occurrence of larval stages of Scyphozoa in the
Elkhorn Slough. Monterey Bay. Calif. Am. Nat. 59: 94-96.

Gershwin, L. 1999. Clonal and population variation in jellyfish symme-
try. ./. Mar. Biol. Assoc. UK 79: 993-1000.

Gershwin, L., and A. G. Collins. 2001. A preliminary phylogeny of
Pelagiidae (Cnidaria. Scyphozoa), with new observations of Chrysaora
co/orata comb. nov. J. Nat. Hist, (in press).

Gotshall, D. W. 1994. Guide to Marine Invertebrates Alaska to Baja
California. Sea Challengers, Monterey, CA.

Gotshall, D. W., and L. L. Laurent. 1980. Pacific Coast Subtidal
Marine Invertebrates. Sea Challengers. Los Osos. CA.

Gotshall, D. W., J. G. Smith, and A. Holbert. 1965. Food of the blue
rockfish, Sebastodes mystinus. Calif. Fish Game 51(3): 147-162.

Greenberg, N., R. L. Garthwaite, and D. C. Potts. 1996. Allozyme and
morphological evidence for a newly introduced species of Aurelia in
San Francisco Bay. California. Mar. Biol. 125: 401-410.

Gubtrlet, M. L. 1936. Animals of the Seashore. Binfords & Mort,
Portland. OR.

Guberlet, M. L. 1949. Animals of the Seashore, rev. ed. Binfords &
Mort, Portland, OR.

Guberlet, M. L. 1962. Animals of the Seashore. 3rd ed. Binfords &
Mort, Portland, OR.

SYSTEMATICS AND BIOGEOGRAPHY OF AURELIA LABIATA

117

Haderlie, E. C., C. Hand, and VV. B. Gladfelter. 19X0. Cnidaria (Co-
elenterata): The Sea Anemones and Their Allies. Pp. 40-75 in Inter-
ndal Invertebrates of California. R. H. Morris. D. P. Abbott, and E. C.
Haderlie, eds. Stanford University Press, Stanford. CA.

Haeckel. E. 1879. Das System tier Medusen: Erster Tlieil finer Monog-
raphic tier Medusen. G. Fischer, Jena.

Haeckel, E. 1880. System tier Acraspeden: Zweite Ha/fie des System der
Medusen. Denkschriften. Jena.

Hall, C. A. Jr. 1964. Shallow-water marine climates and molluscan
provinces. Ecology 45: 226-234.

Hamner, W. M., and I. R. Hauri. 1981. Long-distance horizontal
migrations of zooplankton (Scyphomedusae: Mastigias), Limnol.
Oceanogr. 26: 414-423.

Hamner, \V. M., and R. M. Jenssen. 1974. Growth, degrowth, and
irreversible cell differentiation in Aurelia aurita. Am. Zoo/. 14: 833-
849.

Hamner, W. M., P. P. Hamner, and S. W. Strand. 1994. Sun-compass
migration by Aurelia aurita (Scyphozoa): population retention and
reproduction in Saanich Inlet, British Columbia. Mar. Biol. 119: 347-
356.

Hand, C. 1975. Scyphozoa. Pp. 94-96 in Light's Manual: Intertidal
Invertebrates of the Central California Coast. 3rd ed. R. I. Smith and
J. T. Carlton. eds. University of California Press. Berkeley.

Hartman. O., and K. O. Emery. 1956. Bathypelagic coelenterates.
Limnol. Oceanogr. 1: 304-312.

Hauser, H., and B. Evans. 1978. The Living World of the Reef. Walker.
New York.

Hedgpeth. J. \V. 1962. Introduction to Seashore Life of the San Fran-
cisco Bay Region and the Coast of Northern California. University of
California Press, Berkeley.

Hellberg, M. E. 1996. Dependence of gene flow on geographic distance
in two solitary corals with different larval dispersal capabilities. Evo-
lution 50(3): 1167-1175.

Hummelinck, P. VV. 1968. Caribbean scyphomedusae of the genus Cas-
siopea. Studies of the Fauna of Curacao and other Caribbean Islands.
No. 97. 25: 1-57.

Johnson. M. E., and H. J. Snook. 1927. Seashore Animals of the Pacific
Coast. Macmillan, New York.

Johnson. M. E., and H. J. Snook. 1967. Seashore Animals of the Pacific
Coast. Dover Publications. New York.

Keen, S. L. 1991. Clonal dynamics and life history evolution in the
jellyfish Aurelia aurita. Ph.D. dissertation. University of California.
Davis.

Keen, S. L., and A. J. Gong. 1989. Genotype and feeding frequency
affect clone formation in a marine cnidarian. Fund. Ecol. 3: 735-745.

Kirkpatrick, F. Z. S. 1903. Notes on some medusae from Japan. Annu.
Mag. Nat. Hist.. Ser. 7 12: 615-621.

Kishinouye, K. 1891. Aurelia japonica. Zoo/. Mag. (Tokyo) 3(33): 289,
pi. 7. '

Kishinouye, K. 1910. Some medusae of Japanese waters. J. Coll. Sci.
Tokyo. 27: 1-35.

Knowlton. N. 2(MIO. Molecular genetic analyses of species boundaries in
the sea. Hydrobiologia 420: 73-90.

Kozloff. E. N. 1973. Seashore Life ofPuget Sound, the Strait of Georgia,
and the San Juan Archipelago. University of Washington Press. Seat-
tle.

Kozloff. E. N. 1974. Keys to the Marine Invertebrates of Puget Sound,
the San Juan Archipelago, and Adjacent Regions. University of Wash-
ington Press, Seattle.

Kozloff, E. N. 1983. Seashore Life of the Northern Pacific Coast. An
Illustrated Guide to Northern California. Oregon. Washington, anil
British Columbia. University of Washington Press, Seattle.

Kozloff, E. N. 1987. Marine Invertebrates of the Pacific Northwest.
University of Washington Press. Seattle.

Kramp, P. L. 1913. Medusae collected by the "Tjalfe" Expedition.

Vidensk. Medd. Dan. Naturhist. Foren. 65: 257-286.
Kramp. P. L. 1942. Medusae: The 'Godthaab' Expedition 1928. Medd.

Gronl. Bd. 81(1): 1-168.
Kramp, P. L. 1961. Synopsis of the medusae of the world. J. Mar. Biol.

Assoc. UK 40: 1-469.
Kramp, P. L. 1965. Some medusae (mainly Scyphomedusae) from

Australian coastal waters. Trans. R. Soc. S. Aust. 89: 257-278.
Kramp. P. L. 1968. The Scyphomedusae collected by the Galathea

Expedition 1950-52. Vidensk. Medd. Dan. Naturhist. Foren. 131:

67-98.
Lalli. C. M., and T. R. Parsons. 1993. Biological Oceanography: An

Introduction. Butterworth-Hcinemann. Oxford.
Lamarck. 1816. Histoirc natiirelle des Animaux sans Vertebres. Ver-

diere, Paris.
Lambert, F. J. 1935. Observations on the scyphomedusae of the Thames

Estuary and their metamorphoses. Trav. Stat. Zool. Wimereux. 12:

281-307.
Larson, R. J. 1976. Marine flora and fauna of the Northeastern United

States. Cnidaria: Scyphozoa. NOAA Tech. Rep. NMFS Circ. 397: 1-18.
Larson, R. J. 1986. Water content, organic content, and carbon and

nitrogen composition of medusae from the northeast Pacific. J. Exp.

Mar. Biol. Ecol. 99: 107-120.
Larson, R. J. 1987. Respiration and carbon turnover rates of medusae

from the NE Pacific. Camp. Biochem. Physwl. 87A: 93-100.
Larson, R. J. 1990. Scyphomedusae and cubomedusae from the Eastern

Pacific. Bull. Mar. Sci. 47: 546-556.
Larson, R. J., and A. C. Arneson. 1990. Two medusae new to the coast

of California: Carybdea marsupialis (Linnaeus, 1758), a cubomedusa

and Phyllorhi-a punctuta von Lendenfeld, 1884. a rhizostome scy-

phomedusa. Bull. South. Calif. Acad. Sci. 89(3): 130-136.
Lesson, R. P. 1829. Voyage medical autour du monde execute stir la

con-ette La Coquille pendant les annees 1822-25. Zoologie. Paris.
Lesson, R. P. 1843. Histoire Natiirelle des Zoophytes. Acalephes. Li-

brairie encyclopedique de Roret. Paris.
Light, S. F. 1914a. Another dangerous jellyfish in Philippine waters.

Philipp. J. Sci. B9(3): 291-295.
Light, S. F. 1914b. Some Philippine Scyphomedusae. including two new

genera, five new species, and one new variety. Philipp. J. Sci. 9:

195-231.

Light. S. F. 1921. Further notes on Philippine scyphomedusan jelly-
fishes. Philipp. J. Sci. 18: 25-32.
Light, S. F. 1941. Laboratory and Field Text in Invertebrate Zoology.

Associated Students Store, University of California. Berkeley.
Light, S. F., R. I. Smith, F. A. Pitelka, D. P. Abbott, and F. M.

Weesner. 1954. Intertidal Invertebrates of the Central California

Coast. University of California Press, Berkeley.
Linnaeus, C. 1746. Fauna svecica, sistens animalia Sveciae regni:

Quadrupedia. Ares. Amphibia. Pisces. Insecta. Venues, distributa per

classes & ordines. genera & species, cum differentiis specierum. syn-

uimtus autoruin, nominibus incolarum, locis habitationum. descrip-

tionibus insectorum. Sumtu & literis L. Salvii, Stockholm.
Linnaeus, C. 1758. Systema Naturae. 10th ed. Impensis L. Salvii. Hol-

miae.
Lucas, C. H., and S. Lawes. 1998. Sexual reproduction of the scy-

phomedusa Aurelia aurita in relation to temperature and variable tood

supply. Mar. Biol. 131(4): 629-63X.
MacGinitie, G. E. 1955. Distribution and ecology of the marine

invertebrates of Point Barrow, Alaska. Smithson. Misc. Collect. 128:

1-201.

MacGinitie, G. E., and N. MacGinitie. 1949. Natural History of Ma-
rine Animals. McGraw-Hill. New York.

118

L. GERSHWIN

MacGinitie, G. E., and N. MacGinitie. 1968. Natural History of Ma-
rine Animals, 2nd ed. McGraw-Hill, New York.

Malnig, A. 1985. Where the Waves Break: Life at the Edge of the Sea.
Carolrhoda Books, Minneapolis.

Mayer, A. G. 1900. Some medusae from Tortugas. Florida. Bull. Mus.
Comp. Zool. Harvard College 37(2): 13-82.

Mayer, A. G. 1910. Medusae of the World. Vols. I ami II. the Hydrome-
dusae. Vol. 3: The Scyphomedusae. Carnegie Institution. Washington,
DC.

Mayer, A. G. 1915. Medusae of the Philippines and of Torres Straits.
Being a report on the Scyphomedusae collected by the U.S. Fisheries
Bureau steamer 'Albatross' in the Philippine Islands and Malay Archi-
pelago, 1907-1910, and upon the medusae collected by the expedition
of the Carnegie Institution of Washington to Torres Straits, Australia,
in 1913. Pap. Tortugas Lab. 8: 157-202.

Mayer, A. G. 1917. Report upon the Scyphomedusae collected by the
United States Bureau of Fisheries steamer "Albatross" in the Philippine
Islands and Malay Archipelago. Bull. U. S. Nat. Mus. 100( 1 ): 175-233.

McConnaughey, B., and E. McConnaughey. 1985. The Audubon So-
ciety Guides: Pacific Coast. Alfred A. Knopf, New York.

McLachlan, D. H., and J. Ayres. 1979. Fieldbook of Pacific Northwest
Sea Creatures. Naturegraph, Happy Camp, CA.

Mills, C. E. 1981. Seasonal occurrence of planktonic medusae and
ctenophores in the San Juan Archipelago (NE Pacific). Wasmann
J. Biol. 39: 6-29.

Naumov. D. V. 1961. Stsifoidnye meduzy morei S.S.S.R. [Scyphome-
dusae of the seas of the USSR.] (In Russian). Oprede/ite/i po Faune
SSSR 75: 1-98.

Niesen, T. M. 1994. Beachcomber's Guide to California Marine Life.
Gulf Publishing. Houston. TX.

Niesen, T. M. 1997. Beachcomber's Guide to Marine Life of the Pacific
Northwest. Gulf Publishing, Houston. TX.

Norris, R. D. 1989. Cnidarian taphonomy and affinities of the Ediacara
biota. Lethaia. 22: 381-393.

North, W. J. 1976. Undenvater California. University of California
Press. Berkeley.

Nybakken, J. W. 1993. Marine Biology: an Ecological Approach. 3rd
ed. HarperCollins, New York.

O'Donnell, K., E. Cigelnik, and H. I. Nirenburg. 1998. Molecular
systematics and phylogeography of the Gibbere/la fiijikuroi species
complex. Myco/ogia 90(3): 465-493.

Parsons, C. 1986. Dangerous Marine Animals of the Pacific Coast.
Helm Publishing, San Luis Obispo, CA.

Pearcy, W. G. 1972. Distribution and ecology of oceanic animals off
Oregon. Pp. 351-377 in The Columbia River Estuary and Adjacent
Ocean Waters. A. T. Pruter and D. L. Alverson, eds. University of
Washington Press, Seattle.

Pereyra, W. T., and M. S. Alton. 1972. Distribution and relative abun-
dance of invertebrates off the Northern Oregon Coast. Pp. 444-474 in
The Columbia River Ettuan' and Adjacent Ocean Waters. A. T. Pruter
and D. L. Alverson, eds. University of Washington Press, Seattle.

Peron, F., and C. A. Lesueur. 1810. Tableau des caracteres generiques
et specifiques de toutes les especes de meduses connuesjusqu'ace jour.
Ann. Mus. Hist. Nat. Paris 14: 325-366.

Powell, A. W. B. 1975. Native Animals of New Zealand. Auckland
Institute and Museum, Auckland.

Ranson, G. 1945. Scyphomeduses provenant des Campagnes du Prince
Albert ler de Monaco. Series Resiiltats des campaniles scientific/lies,
Monaco 106: 1-92.

Rees, W. J. 1957. Proposed validation under the plenary powers of the
generic name 'Aure/ia' Lamarck 1816 (class Scyphozoa). Bull. Zool.
Nomencl. 13(1957-1958): 26-28.

Reish, D. J. 1972. Marine Life of Southern California. Forty-Niner
Shops, Long Beach. CA.

Reish, D. J. 1995. Marine Life of Southern California. 2nd ed. Kendall/

Hunt Publishing, Dubuque, IA.
Ricketts. E. F., and J. Calvin. 1939. Between Pacific Tides. Stanford

University Press, Stanford, CA.
Ricketts, E. F., and J. Calvin. 1948. Between Pacific Tides, rev. ed.

Stanford University Press, Stanford. CA.
Ricketts, E. F., and J. Calvin. 1952. Benveen Pacific Tides. 3rd ed.,

revised by Joel W. Hedgpeth. Stanford University Press, Stanford, CA.
Ricketts, E. F. and J. Calvin. 1968. Between Pacific Tides. 4th ed.,

revised by Joel W. Hedgpeth. Stanford University Press, Stanford, CA.
Ricketts, E. F., J. Calvin, and J. Hedgpeth. 1985. Bertveen Pacific

Tides. 5th ed., revised by D.W. Phillips. Stanford University Press,

Stanford, CA.
Rigsby, M. 1997. Open waters. Pp. 178-207 in Natural History of the

Monterey Bay National Marine Sanctuary. Monterey Bay Aquarium

and NOAA, Monterey. CA.
Roy, K., D. Jablonski, J. W. Valentine, and G. Rosenberg. 1998.

Marine latitudinal diversity gradients: Tests of causal hypotheses. Proc.

Nat. Acud. Sci. USA. 95: 3699-3702.
Russell, F. S. 1970. Medusae of the British Isles. II. Pelagic Scyphozoa

with a Supplement to the First Volume on Hvdromedusae. Cambridge

University Press, Cambridge.
Schweizer, N. R. 1973. A Poet among Explorers: Chamisso in the South

Seas. Verlag Herbert Lang, Bern.
Shenker, J. M. 1984. Scyphomedusae in surface waters near the Oregon

coast, May-August, 1981. Estuarine Coastal Shelf Sci. 19: 619-632.
Smith, L. 1962. Common Seashore Life of the Pacific Northwest. Nat-
uregraph, Healdsburg. CA.

Snively, G. 1978. Exploring the Seashore in British Columbia, Wash-
ington and Oregon. Gordon Soules Book Publishers. Vancouver, BC,

Canada.

StefofT. R. 1997. Jellyfish. Benchmark Books. Tarrytown, NY.
Stiasny, G. 1919. Die Scyphomedusen-Sammlung des Naturhistorischen

Reichsmuseums in Leiden. II. Stauromedusen. Coronatae. Semaeo-

stomae. Zool. Meded. 5: 66-98.
Stiasny, G. 1922. Papers from Dr. Th. Mortensen's Pacific Expedition

1914-1916. XII. Die Scyphomedusen-sammlung von Dr. Th.

Mortensen nebst anderen Medusen aus dem Zoologischen Museum der

Universitat in Kobenhagen. Vidensk. Medd. Naturhist. Foren. 73: 513-

558.
Stiasny, G. 1926. Uber Einige Scyphomedusen von Puerto Galera, Min-

doro (Philippinen). Zool. Meded. 9: 239-248.
Stiasny, G. 1931. Die Rhi/ostomeen-Sammlung des British Museum

(Natural History) in London. Zool. Meded. 14: 137-78.
Stiasny, G. 1935. Die Scyphomedusen der Snellius expedition. Verh. K.

Akad. Wet. Amst.. Sect. 2. 34(6): 1-44.
Stiasny, G. 1937. Biological results of the Snellus Expedition. III. Die

f undone der Scyphomedusen und Tornarien. Temminckia 2: 203-210.
Stiasny, G., and H. van der Maaden. 1943. Uber Scyphomedusen aus

dem Ochotskischen und Kamtschatka Meer nebst einer Kritik der

Genera Cyanea und Desmonema. Zool. Jahrb. iSyst.) 76(3): 227-266.
Strand, S. W., and W. M. Hamner. 1988. Predatory behavior of Pha-

cellophora camtschatica and size-selective predation upon Aure/ia

aiinta (Scyphozoa: Cnidaria) in Saanich Inlet. British Columbia. Mar.

Biol. 99: 409-414.
Strathmann. M. F. 1987. Reproduction and Development of Marine

Invertebrates of the Northern Pacific Coast. University of Washington

Press, Seattle.
Thuesen, E. V., and J. J. Childress. 1994. Oxygen consumption rates

and metabolic enzyme activities of oceanic California medusae in

relation to body size and habitat depth. Biol. Bull. 187: 84-98.
Tierney, R. J., J. W. timer, L. J. Waxdeck, H. N. Foster, and J. R.

Eckenroad. 1967. Exploring Tidal Life Along the Pacific Coast.

Tidepool Associates. Oakland, CA.

SYSTEMATICS AND BIOGEOGRAPHY OF AURELIA LABI ATA

119

Torrey. H. B. 1909. The Leptomedusae of the San Diego region. Univ.

Calif. Puhl. Zoo/. 6: 11-31.
I'ohidu. T. 1928. Short notes on medusae. 1. Medusae with abnormal

\\mnietry. Annot. Zool. Jpn. 2: 373-376.
I'chida. T. 1934. A saemostome medusa with some characters of rhiz-

ostomae. Proc. Imp. Acad. 10: 698-700.
I'chida, T., and Z. Nagao. 1963. The metamorphosis of the Scyphomc-

dusa. Aurelia limbata (Brandt). Annot. Zool. Jpn. 36: 83-91.
Ulmer, J. \V. 1968. Exploring Our Coast (Seashore Discovery Book I).

Oecologica. Tomales. CA.
\'alentine, J. \V. 1966. Numerical analysis of marine molluscan ranges

of the extratropical northeastern Pacific shelf. Limnol. Oceanogr. 11:

19S-21 1.
van der Maaden, H. 1939. Uber das Sinnesgriibchen von Aitrelia auritci

Linne. Zool. An:. 125: 29-35.

Vanhoffen, E. 1888. Untersuchungen ueber Semaeostome und Rhizo-
stome Medusen. Bihl. Zool. 1(3): 1-52.

Vanhoffen, E. 1906. Acraspedae. Nordisches Plankton 6< 1 1 ): 40-64.

von Lendenfeld, R. 1884. The scyphomedusae of the southern hemi-
sphere. Part III. — Conclusion. IV. Ordo — Discomedusae. Proc. Linn.
Soc. N.S.W. 9: 259-306.

Wells, H. 1942. Scuxhuri' Life. California State Department of Educa-
tion. Sacramento.

\\ robel. D., and C. Mills. 1998. Pacific Coast Pelagic Invertebrates: A
Guide to the Common Gelatinous Animals. Sea Challengers and
Monterey Bay Aquarium. Monterey, CA.

Zubkoff, P. L., and A. L. Lin. 1975. Isozymes of Aurelia aurita scy-
phistomae obtained from different geographical locations. Pp. 915-930
in Isozymes. IV. Genetics und Evolution. C. L. Markert, ed. Academic
Press, New York.

Marine

Biological

Laboratory

Woods Hole

Massachusetts

One Hundred and Third Report

for the Year 2000

One Hundred and Twelfth Year

Officers of the Corporation

Sheldon J. Segal, Chairman of the Board of Trustees

Frederick Bay. Co-Vice Chair

Mary J. Greer. Co-Vice Chair

John E. Dowling. President of the Corporation

John E. Burris, Director and Chief Executive Officer

William T. Speck, Interim Director and Chief Executive Officer

Mary B. Conrad, Treasurer

Robert E. Mainer, Clerk of the Corporation

Contents

Report of the Director and CEO Rl

Report of the Treasurer R6

Financial Statements R7

Report of the Library Director R18

Educational Programs

Summer Courses R20

Special Topics Courses R24

Other Programs R32

Summer Research Programs

Principal Investigators R35

Other Research Personnel R36

Library Readers R37

Institutions Represented R38

Year-Round Research Programs R43

Honors R57

Board of Trustees and Committees R64

Administrative Support Staff R68

Members of the Corporation

Life Members R71

Members R72

Associate Members R83

Certificate of Organization R86

Articles of Amendment R86

Bylaws R86

Publications . . R91

Photo credits:

E. Armstrong— R3 (bottom), R4 (top), R20, R21,

R24, R27, R35, R47, R55
K. Begos— R38
D. Buffam— R2 (bottom)
M. Dornblaser— R68
J. Dowling— R30
L. Colder— R64
Gray Museum of the Marine Biological

Laboratory — R57
R. Hanlon— R43
R. Howard— R4 (bottom), R18

A. Kuzirian — R6

B. Liles— R71

H. Luther— R23. R46
J. Montgomery — R2 (top)
P. Presley— Rl
A. Rader— R86

Report of the Director
and Chief Executive Officer

It is with great pleasure that I write this report as the
Marine Biological Laboratory's newest Director and Chief
Executive Officer. My relationship with the MBL has
grown and expanded in rewarding and exciting ways
during the past twenty-five years. I am now pleased to
have the opportunity to serve as Director of this esteemed
Laboratory. I first came to the MBL as a student and then
returned as an investigator for several summers. My role
expanded when I was elected to the Laboratory's Board
of Trustees in 1994. and again when I joined the
Discovery Campaign Steering Committee. In 1999. I
succeeded Mel Cunningham as Chair of the Development
Committee. Since being appointed Interim Director upon
John Burris's departure in the summer of 2000, I've had
a wonderful opportunity to view the inner workings of
this remarkable institution.

I think it's fair to say that the Marine Biological
Laboratory is stronger and healthier both financially and
programmatically than it has ever been in its history. In
this report, I'll review what has led us to this point, share
with you some highlights from the year 2000, and discuss
where the Trustees and I see the Laboratory going in the
next few years.

The Discovery Campaign

The Marine Biological Laboratory concluded its first
comprehensive fundraising campaign — Discovery: The
Campaign for Science at the Marine Biological
Laboratory — in December 2000. Our goal was to raise
$25 million for a variety of initiatives at the MBL. When
we began planning for the campaign, some felt that this
goal was a stretch for the institution. Thanks to the
generosity of thousands of Trustees. Corporation
Members, Associates, Alumni, Staff Members,
Foundations, and Friends of the Laboratory, the MBL far
surpassed that goal, raising more than $41 million by the
end of the year 2000 in support of research, education,
the library and physical plant, and the annual fund.

Funds raised through the Discovery Campaign have

already had a major impact on the Laboratory's
educational and research programs. One of the most
obvious achievements of the Campaign is the construction
of the C. V. Starr Environmental Sciences Building,
which will become the new home of The Ecosystems
Center in 2001. Thanks to the Campaign we also
established the Josephine Bay Paul Center for
Comparative Molecular Biology and Evolution and hired
two new assistant scientists there (Michael Cummings and
Jennifer Wernegreen); added five new summer courses
and the Semester in Environmental Sciences Program for
undergraduates to our education roster; created more than
a dozen endowed scholarships for students and endowed
fellowships for young researchers; established a program
in scientific aquaculture in the Marine Resources Center;
endowed the director's chair of the Marine Resources
Center; and expanded our public outreach efforts through
the creation of the Robert W. Pierce Visitors Center.

In addition, we raised funds to support endowed
lectureships for the summer courses and an annual lecture
in Bioethics starting in the summer of 2001, and to help
shore up the Laboratory's aging physical plant. Moreover,
we received gifts to permanently endow the maintenance
of the Waterfront Park and the Pierce Visitors Center.
Finally, thanks to gifts to the Discovery Campaign, the
Library has been air-conditioned and the Crane House on
Millfield Street has been refurbished and added to our
year-round housing inventory.

Physical Plant

We've also been able to tackle some other long-
overdue maintenance projects on campus. For example,
the crumbling section of seawall near the Lillie Building
has been reconstructed. By the summer of 2001, the Brick
Dormitory will have been renovated and furnished for
year-round use. Cottages at Memorial Circle have been
updated and de-leaded, and we have begun renovations at
Devils Lane. The research laboratories in the Lillie
Building are being renovated to accommodate expanding

R2 Annual Report

year-round research programs in the Bay Paul Center,
BioCurrents Research Center, and Architectural Dynamics
in Living Cells Program. We've also added fresh paint
and carpeting to the Meigs Room, and have begun
painting and replacing lighting and other fixtures
throughout the Swope Building.

Our plans also include renovating summer research
laboratories in the Whitman Building. We expect to begin
modestly renovating the Homestead building, which, once
vacated by the staff of The Ecosystems Center, will
eventually become home to the administrative offices of
Financial Services. Education, Human Resources, and The
Biological Bulletin.

The Biological Bulletin

The Marine Biological Laboratory's journal, The
Biological Bulletin, celebrated a major milestone in 2000.
Edited by Michael J. Greenberg of the University of
Florida's Whitney Laboratory, the journal has been
publishing peer-reviewed articles of general biological
interest for more than 100 years. During the summer of
2001 the journal will launch a new initiative by
publishing articles electronically with HighWire Press of
Stanford University.

Education

During the summer of 2000, the MBL's Educational
Program offered a record 22 summer and special topics
courses. Three hundred and thirty-five course directors
and faculty members taught 490 advanced graduate and
postdoctoral students in the courses last summer. An
additional 315 guest lecturers and instructors participated
in the courses as well. From all accounts, the quality of
our students improves every year.

We offered a symposium on the history of biology and
a workshop in microbial diversity designed for middle

and high school teachers. Last summer brought quite a
few undergraduates to the MBL as well, through a variety
of Research Experience for Undergraduate Programs. One
program focused on Marine Models, another was
coordinated by the Boston University Marine Program.
and others were offered by the Marine Resources and
Ecosystems Centers. I'm pleased to report that funding
has been allocated for two additional research programs
for undergraduates beginning in summer 2001.

The MBL's own semester-long undergraduate program.
The Semester in Environmental Sciences, offered by the
staff of The Ecosystems Center, completed its 3rd year in
2000 with 1 5 students participating. The consortium of
colleges whose students come for the fall semester
continues to grow, currently numbering more than 40
members.

Research

The summer research program ran at full capacity
during the summer of 2000. One hundred and thirty-two
investigators used all of our available lab space. In fact,
one applicant had to set up his research in a dark room.
The majority of the investigators (60%) were professors/
chief scientists, followed by associate professors (20%)
and postdoctoral fellows (10%). The balance was
comprised of assistant scientists and graduate students.

I'm proud to report that for the second year in a row
an MBL Summer Scientist — Avram Hershko of the
Technion in Israel — has won the prestigious Lasker
Award (Clay Armstrong won this award in 1999). This
award is second only to the Nobel Prize in significance in
science. Dr. Hershko will deliver a Friday Evening
Lecture during the summer of 2001. I'm also pleased to
be able to count two of the year 2000' s Nobel Prize
winners as members of the MBL family: Paul Greengard
of Rockefeller University, an alumnus of the Embryology
Course and a former faculty member of the Neurobiology
Course, and Eric Kandel of Columbia, a past MBL

Report of the Director and CEO R3

investigator and Corporation Member. These awards
validate the tremendous significance and impact the
MBL's research and educational programs have on the
biology community at large.

The MBL's research fellowship program hosted 21
investigators during the summer of 2000. The range of
the research being undertaken by these scientists was
remarkable, and the caliber of their backgrounds scored
high by the Fellowship Committee and our external
advisors. The Science Writing Fellowship Program also
continued to figure prominently among print and
broadcast journalists for the outstanding opportunity it
affords them to work alongside scientists to learn about
the process of doing science.

The Ecosystems Center

Research is and will always be a key mission of the
MBL. We have seen a continued growth in our resident
research programs. The Ecosystems Center, directed by
Jerry Melillo and John Hobbie, now numbers more than
60 staff, and its funding base has more than doubled
during the past 5 years. It is now is in excess of $7
million. Thirty research projects are underway around the
globe, from Siberia to Martha's Vineyard. In 2000 The
Ecosystems Center celebrated its 25th anniversary with a
weekend-long celebration. The festivities included an
open house, a one-day symposium complete with a visit
by Rep. William Delahunt of the Massachusetts 10lh
District, and a reunion clambake at the Swope Center.
More than 50 Ecosystems Center alumni from all over the
world traveled to Woods Hole to celebrate the success of
the Center's first 25 years and to discuss the future of
ecosystems science.

The Josephine Bay Paul Center

The Bay Paul Center for Comparative Molecular
Biology and Evolution, under Mitch Sogin's direction.

currently has 33 scientists and support staff. The Center's
project to sequence the genome of the parasite Giardia is
nearly complete.

For the first time, the MBL has received a prestigious
gift from the Keck Foundation. This $1 million award
will establish the W. M. Keck Ecological and
Evolutionary Genetics Facility at the Bay Paul Center.
Microbial ecologists, molecular evolutionists, and genome
scientists from the Bay Paul Center, The Ecosystems
Center, and other scientific groups within the Woods Hole
community will form a coalition to study how the genes
of millions of microbes work together to influence
biogeochemical processes within ecosystems.

The BioCurrents Research Center

The NIH BioCurrents Research Center, directed by
Peter Smith, has increased in size and now numbers 1 1
scientists, thanks to the recent addition of Drs. Orian
Shirihai and Stefan McDonough to the scientific staff.
Among their many research projects. Smith and his
colleagues continue to collaborate with Dr. Barbara
Corkey of Boston Medical Center on the study of how
cells process insulin. They are currently fine tuning
instruments that will enable them to monitor the
movement and release of glucose, insulin, and calcium
within pancreatic beta cells, the goal being to learn more
about how diabetes type II works at a cellular level.
Another exciting collaboration is underway between the
BioCurrents Research Center and the Bay Paul Center to
study the evolution, diversity, and physiology of
organisms living in extreme environments — like the hot
vents of the deep oceans and extremely acidic (battery
acid-like) ecosystems.

The Marine Resources Center

Research using DNA fingerprinting to assess paternity
and reproductive patterns and population structure in the

R4 Annual Report

local squid fishery — valued at $33 million annually —
continues in the Marine Resources Center (MRC). under
the direction of Roger Hanlon. Work on how polarized
vision is used by the squid to help detect prey is also a
focal point. During the Campaign, a landmark gift from
Honorary Trustee Ellen Grass established the first
endowed Directorship at the MBL. This gift, the grant
from the Schooner Foundation to establish the Program in
Scientific Aquaculture, and a recent anonymous grant of
$500,000 ensures future vitality for the MRC. The MRC
is also currently in the process of hiring three faculty-
level scientists and a scientific aquaculturist.

I've only touched on a few of the MBL's resident
research initiatives. In addition to these research centers,
the MBL is home to a score of investigators' research
programs that focus on a range of topics including
infertility, microscopy, learning and memory, and the
effects of lead poisoning on children.

Trustees will start developing a 5- to 10-year strategic
plan — a map charting the direction that the Laboratory
will take in both research and education in the coming
years. This plan will further strengthen and position the
Laboratory to serve science and society.

As we continue to build the year-round research
programs, plans have been developed to add a new year-
round research program in Global Infectious Diseases and
Parasitism. Parasites cause debilitating and often lethal
diseases in billions of people around the world. The
World Health Organization estimates that one in ten are
infected by one or more of the five major parasitic
diseases: schistosomiasis, filariasis, malaria,
trypanosomiasis, and leishmaniasis. The MBL is already a
leader in the field of parasitology and infectious disease,
hosting two major international parasitology meetings and
offering a world-renowned course in the Biology of
Parasitism each summer. This new program will build on
the Laboratory's existing strengths in this field and take
advantage of the high throughput technologies and
scientific expertise available in the Bay Paul Center,
creating a one-of-a-kind research environment that fosters
interactions between parasitologists and experts in
molecular biology, phylogenetics, and environmental
microbiology. The Trustees agree that this is a strong and
important addition to the MBL's year-round research
portfolio.

On the education side, Mitch Sogin and Clare Eraser,
one of our newest Trustees, are planning to offer an
exciting and novel course in genomics. This course will
premiere in Fall 2002. We hope to offer more and more
cutting-edge courses throughout the year in the future.

The Library

The MBLAVHOI Library continues to expand both its
print and electronic serial collections. More than 2000
full-text electronic journals are now available on our
scientists' desktops through the Library's web site. The
entire collection has grown to more than 200,000
volumes, occupying all the space the Library has
available in Woods Hole. Storage issues are currently
being addressed by providing more electronic access to
journals and by sending some volumes off campus to the
Harvard Depository.

Looking Ahead

It's an exciting time for the Marine Biological
Laboratory. Now more than ever, the Trustees are
committed to building and strengthening the MBL's year-
round research program. Within the next year, the

Trustees

The Trustees elected four new Board members and
reappointed one Trustee to the Class of 2005 at their
November 4, 2000 meeting. Dr. Porter W. Anderson, who
completed his first term on the Board this year, was

Report of the Director and CEO R5

appointed to a second term. He is joined by Dr. Claire M.
Fraser, President and Director of The Institute for Genomic
Research in Maryland; Mr. George Logan, Chairman of the
Board and Organizer of the Valley Financial Corporation as
well as Principal of the Wood Park Capital Corporation in
Roanoke, VA; Robert A. Prendergast, Professor of
Ophthalmology and Associate Professor of Pathology at The
Wilmer Institute at The Johns Hopkins University School of
Medicine, Baltimore, MD; and John W. Rowe, M.D.,
President and CEO of Aetna Inc. Thomas S. Crane, Co-
ordinator of Mintz Levin Cohn Ferris Glovsky and Popeo's
Health Care Fraud and Abuse and Corporate Compliance
practice group serving the firm's Boston and Washington,
DC, offices, was elected Clerk of the Corporation.

Sheldon Segal, John Dowling, and Mary B. Conrad were
reelected to serve as Chairman of the Board, President of the
Corporation, and Treasurer, respectively. Trustee Al Zeien
was elected Vice Chair of the Board. The Board also
thanked retiring members Fred Bay, Marty Cox, Mary
Greer, William Steere. and Gerald Weissmann for their
tireless efforts on behalf of the Laboratory.

In Memoriam

As this report was going to press, we were saddened to
learn of the tragic deaths of Jim and Alma Ebert, who
were killed on May 22, 2001, in a car accident while
traveling from Baltimore to Woods Hole for the summer.
Jim was President of the MBL Corporation from 1970 to
1978 and again from 1990 to 1998. He was Director of
the Laboratory from 1970 to 1978, a Trustee from 1964
to 1968, and was named Director Emeritus in 2000. Alma
was active in the MBL Associates, volunteering her time
and energy on behalf of the Laboratory, and supporting
Jim during his tenure as Director.

For five decades the MBL has benefited from Jim's
considerable knowledge and experience. He was
instrumental in bringing significant funding to the
Laboratory, and his guidance and insight were key to the
MBL's success. The loss of these dear friends will be
deeply felt by the MBL family for many years.

—William T. Speck

•• : .,

Report of the Treasurer

The Marine Biological Laboratory had another
impressive operating year in 2000 that was partially offset
by weak near-term investment portfolio returns.
Auspicious growth in Operating Support and the decline
in the Equity Markets were the major contributors to the
mixed results.

Three areas of Operating Support showed double-digit
increases. The growth in Government Grants accelerated
to 14.7% over 1999 results and represented an all-time
high of 45.2% of Total Operating Support. Fees for
Conferences and Services grew even faster, up 17.1%.
Short-term Investment Income also grew by 13.1% as a
result of stronger interest rate returns on a larger portfolio
of Cash & Cash Equivalents, Short-Term Investments,
and the Assets Held by the Bond Trustee. This had a very
favorable impact, particularly on the Change in
Unrestricted Net Assets from Operations. It increased
from only $138 thousand in 1999 to $1.3 million in 2000.
This represented a very strong 9.5% Operating Margin.

Reviewing our Non-Operating Activities, we expanded
our Investment in Plant to $4.64 million, more than
doubling what was done in 1999. Total Contributions,
again, exceeded $10 million in the final year of our
Discovery Campaign with almost 45% going toward Plant
improvements. On the other hand, MBL experienced $2.1
million, or 3.9%, in realized and unrealized investment
losses. We also utilized $1.4 million from our standard
spending rate draw. This impacted our Long-term
Investment portfolio, which fell slightly in value for the
first time since 1994.

Even with this, MBL reported a $3.2 million Total
Change in Net Assets. This represented the sixth year of
positive change, but represented only a 4.3% Return on
Average Net Assets.

MBL's 2000 Balance Sheet experienced some
significant changes from 1999. Assets grew by over $1 1
million due to double-digit growth of 16.4% in Net Plant
Assets, increased liquidity, and added Assets held by the
Bond Trustee, which was a result of the $10.2 million
Variable Rate Revenue Bonds issued March 8, 2000. The
Bond refinanced $2.3 million of higher cost debt, with the
balance of the proceeds being used to make capital
improvements to MBL's educational, research, and
housing facilities. Even with this increased debt, MBL
has a sound Leverage Ratio (Unrestricted and
Temporarily Restricted Net Assets-to-Debt) of 5.26X at
year-end 2000. Also note our strong operational returns
resulted in an improved Debt Coverage Ratio of 11. 6X
over previous years. One last positive sign to note is a $3
million increase in the Laboratory's Unrestricted Net
Assets.

In summary, the Laboratory completed an effective
leverage of its financial strength, closed a very successful
fundraising campaign, and demonstrated strong
operational returns. This more than offset the marginal
decline in portfolio performance, and we remain well
poised to continue our capital improvement efforts.

— Mary B. Conrad

Financial Statements

PricewaterhouseCoopers LLP
One International Place
Boston MA 021 10
Telephone (f>17) 478 5000
C.\. simile ((,17) 478 3900

REPORT OF INDEPENDENT ACCOUNTANTS

To the Board of Trustees of
Marine Biological Laboratory:

In our opinion, the accompanying balance sheet of Marine Biological Laboratory (the "Laboratory") at
December 3 1 , 2000 and the related statements of activities and of cash flows for the year then ended present
fairly, in all material respects, the financial position of the Laboratory as of December 31, 2000, and the
changes in its net assets and its cash flows for the year then ended in conformity with accounting principles
generally accepted in the United States of America. These financial statements are the responsibility of the
Laboratory's management; our responsibility is to express an opinion on these financial statements based on
our audit. The prior year summarized comparative information has been derived from the Laboratory's 1999
financial statements, and in our report dated April 7, 2000, we expressed an unqualified opinion on those
financial statements. We conducted our audit in accordance with auditing standards generally accepted in the
United States of America. Those standards require that we plan and perform the audit to obtain reasonable
assurance about whether the financial statements are free of material misstatement. An audit includes
examining, on a test basis, evidence supporting the amounts and disclosures in the financial statements. An
audit also includes assessing the accounting principles used and significant estimates made by management,
as well evaluating the overall financial statement presentation. We believe that our audit provides a
reasonable basis for our opinion.

Our audit was conducted for the purpose of forming an opinion on the basic financial statements taken as a
whole. The supplemental schedule of functional expenses as of December 3 1 , 2000 is presented for the
purpose of additional analysis and is not a required part of the basic financial statements. Such information
has been subjected to the auditing procedures applied in the audit of the basic financial statements and, in
our opinion, is fairly stated, in all material respects, in relation to the basic financial statements taken as a
whole.

April 6. 2001

MARINE BIOLOGICAL LABORATORY

BALANCE SHEET

As of December 31, 2000

(With Comparative Totals as of December 31. 1999)

ASSETS 2000 1999

Cash and cash equivalents $ 3,583,033 $ 1 ,942,285

Short-term investments, at market 3,599,833 3,182,537

Accounts receivable, net of allowance for doubtful accounts of $47,222 in 2000 and $59,978 in 1999 1.109,706 1,158,073

Current portion of pledges receivable 5,026,750 3,974,385

Receivables due for costs incurred on grants and contracts 2,036,734 1,380,766

Other current assets 352.983 306,5 1 8

Total current assets 15.709.039 11,944,564

Assets held by bond trustee 5,423,615

Long-term investments, at market 44,494.649 45,001,493

Pledges receivable, net of current portion 2,433,292 3,498,787

Plantassets.net 23,423,156 20,118,725

Other assets 206.280 —

Total long-term assets 76,180,922 68,619,005

Total assets $91,690,031 $80,563,569

LIABILITIES AND NET ASSETS

Current portion of long-term debt
Accounts payable and accrued expenses
Deferred income and advances on contracts

Total current liabilities

Annuities and unitrusts payable
Long-term debt, net of current portion
Advances on contracts

Total long-term liabilities

Total liabilities
Commitments and contingencies

2.073,375
1.016.060

3.089.435

1,393,735

10.200.000

1.230.743

12.824.478
15,913,913

$ 267,404

1,957,508

656.745

2.881.657

1,460,948
2,056,692
1.574,758

5.092.398
7.974.055

Net assets:
Unrestricted
Temporarily restricted
Permanently restricted

Total net assets

Total liabilities and net assets

22,903,287
30,752,413
22.120.418

75.776.118
$ 91,690,031

19.887.437
33.349,244
19.352.833

72.589.514

The accompanying notes are an integral part of the financial statements.

MARINE BIOLOGICAL LABORATORY

STATEMENT OF ACTIVITIES

For the Year Ended December 3 1 , 2000

(With Comparative Totals for the Year Ended December 31, 1999)

Operating support and revenues:
Government grants
Private contracts
Laboratory rental income
Tuition, net

Fees for conferences and services
Contributions
Investment income
Miscellaneous revenue
Present value adjustment to annuities
Net assets released from restrictions

Total operating support and revenues

Expenses:
Research
Instruction

Conferences and services
Other programs (Note 2)

Total expenses
Change in net assets before nonoperating activity

Nonoperating revenue:
Contribution to Plant:
Private
Government
Release from restriction

Invested in Plant

Total investment income and gains/losses
Less: investment earnings used for operations

Reinvested (utilized) investment income and gains/losses

Total change in net assets
Net assets, beginning of year

Net assets, end of year

Temporarily Permanently

Unrestricted

Restricted Restricted

2000

1999

$14,048,464

$ $

$14,048,464

$12,248,442

1.697.062

— —

1.697.062

1,819,240

1.598,373

— —

1,598,373

1,548,168

543,305

— —

543.305

537,835

4.407,311

— —

4,407,3 1 1

3,765,039

1.693.185

2.347.731 1.908,528

5,949.444

8,620.519

1,736.186

594.530

2.330.716

2.060,478

468.482

— —

468.482

466.903

—

55.176

55,176

(30,533)

4,144.547

(4.249.547) 105,000

—

30,336.915

(1,252.110) 2,013.528

31,098,333

31,036,091

17.799.627

17,799,627

14,147,645

5.626.223

— —

5,626,223

4,742.287

1.307.458

— —

1,307,458

2,252.842

4,261.327

— —

4,261,327

5,297,773

28,994.635

28,994,635

26.440,547

1.342,280

(1,252.110) 2.013.528

2,103,698

4.595,544

404.018

4.109.597 125.000

4.638,615

1,757.319

—

— —

—

198.443

1,615.142

(1.615,142) —

2,019,160

2,494.455 125.000

4.638.615

1,955,762

(284,514)

(2.484.3811 629.057

(2.139.838)

5,938.476

(61,076)

( 1 .354.795 ) —

(1.415.871)

(1.262.020)

(345,590)

(3.839,176) 629.057

(3.555,709)

4.676,456

3.015.850

(2,596,831) 2.767,585

3,186,604

11,227.762

19,887,437

33,349,244 19,352,833

72,589.514

61,361.752

$22.903.287

$30,752.413 $22.120.418

$75.776.118

$72,589,514

The accompanying notes are an integral fart of the financial statements.

MARINE BIOLOGICAL LABORATORY

STATEMENT OF CASH FLOWS

For the Year Ended December 3 1 , 2000

(With Comparative Totals for the Year Ended December 31, 1999)

Cash flows from operating activities:
Change in net assets

Adjustments to reconcile change in net assets to net cash provided by (used in) operating activities:
Depreciation and amortization

Unrealized (appreciation) depreciation on investments
Realized gain on investments

Present value adjustment to annuities and unitrusts payable
Contributions restricted for long-term investment and annuities
Provision for bad debt
Provision for uncollectible pledges
Change in certain balance sheet accounts:

Accounts receivable

Pledges receivable

Grants and contracts receivable

Other current assets and other assets

Accounts payable and accrued expenses

Deferred income

Annuities, and unitrusts payable

Advances on contracts

Net cash provided by operating activities

Cash flows from investing activities:
Purchase of property and equipment
Proceeds from sale of investments
Purchase of investments

Net cash used in investing activities

Cash flows from financing activities:

Payments on annuities and unitrusts payable
Receipt of permanently restricted gifts
Annuity and unitrusts donations received
Bond issuance
Payments on long-term debt

Net cash provided by financing activities

Net increase in cash and cash equivalents
Cash and cash equivalents at beginning of year

Cash and cash equivalents at end of year

2000

$3,186,604

1999

$11,227.762

1.791,975

1,562.487

6.700,396

(3.544.380)

(3.886.669)

(1.639,795)

(55,176)

30,533

(2.033,528)

(2,485,624)

—

36,968

423.982

—

48.367

47,489

(410,852)

(3,010,156)

(655,968)

150,317

(252,745)

251,390

1 15,867

(100,233)

359,315

193,872

(73,167)

68,112

(344.015)

302,368

4,914,386

3,091,110

(5.096,406)

(2,145.041)

68.837.634

63,101,047

(76,930.252)

(65.485.238)

(13,189.024)

(4,529,232)

(96,316)

(49,897)

2.033,528

2,438,148

102.270

47,476

10.200.000

—

(2,324.096)

(243,274)

9,915,386

2,192,453

1 ,640,748

754.331

1,942,285

1,187,954

S3.583.033

$1,942.285

The accompanying notes are an integral pan of the financial statements.

RIO

Financial Statements Rll

Marine Biological Laboratory
Notes to Financial Statements

1. Background

The Marine Biological Laboratory dhe "Laboratory") is a private, independent not-for-profit research and educational institution dedicated to
establishing and maintaining a laboratory and station for scientific study and investigation, and a school for instruction in biology and natural history.
The Laboratory was founded in 1888 and is located in Woods Hole, Massachusetts.

2. Significant Accounting Policies
Basis of Presentation

The accompanying financial statements have been prepared on the accrual basis of accounting and in accordance with the principles outlined in the
American Institute of Certified Public Accountants" Audit Guide, "Not-For-Profit Organizations." The financial statements include certain prior-year
summarized comparative information in total but not by net asset class. Such information does not include sufficient detail to constitute a presentation
in conformity with generally accepted accounting principles. Accordingly, such information should be read in conjunction with the Laboratory's
financial statements for the year ended December 31, 1999. from which the summarized information was derived.

The Laboratory classifies net assets, revenues, and realized and unrealized gains and losses based on the existence or absence of donor-imposed
restrictions and legal restrictions imposed under Massachusetts State law. Accordingly, net assets and changes therein are classified as follows:

Unrestricted

Unrestricted net assets are not subject to donor-imposed restrictions of a more specific nature than the furtherance of the Laboratory's mission.
Revenues from sources other than contributions are generally reported as increases in unrestricted net assets. Expenses are reported as decreases in
unrestricted net assets. Gains and losses on investments and other assets or liabilities are reported as increases or decreases in unrestricted net assets
unless their use is restricted by explicit donor stipulations or law. Expirations of temporary restrictions on net assets, that is, the donor-imposed
stipulated purpose has been accomplished and/or the stipulated time period has elapsed, are reported as reclassifications between the applicable classes
of net assets and titled "Net assets released from restrictions."

Temporarily Restricted

Temporarily restricted net assets are subject to legal or donor-imposed stipulations that will be satisfied either by the actions of the Laboratory, the
passage of time, or both. These assets include contributions for which the specific, donor-imposed restrictions have not been met and pledges,
annuities, and unitrusts for which the ultimate purpose of the proceeds is not permanently restricted. As the restrictions are met. the assets are released
to unrestricted net assets. Also, realized/unrealized gains/losses associated with permanently restricted gifts which are not required to be added to
principal by the donor are classified as temporarily restricted and maintain the donor requirements for expenditure.

Permanently Restricted

Permanently restricted net assets are subject to donor-imposed stipulations that they be invested to provide a permanent source of income to the
Laboratory. These assets include contributions, pledges and trusts which require that the corpus be invested in perpetuity and only the income be made
available for program operations in accordance with donor restrictions.

Performance Indicator

Nonoperating revenues include realized and unrealized gains on investments during the year as well as investment income on the master pooled
investments and revenues that are specifically for the acquisition or construction of plant assets. Investment income from short-term investments and
investments held in trust by others are included in operating support and revenues. To the extent that nonoperating investment income and gains are
used for operations as determined by the Laboratory's total return utilization policy (see below), they are reclassified from nonoperating as
"Investment earnings used for operations" to operating as "Investment income" on the statement of activities. All other activity is classified as
operating revenue.

Cash and Cash Equivalents

Cash equivalents consist of resources invested in overnight repurchase agreements and other highly liquid investments with original maturities of
three months or less.

Financial instruments which potentially subject the Laboratory to concentrations of risk consist primarily of cash and investments. The Laboratory
maintains cash accounts with one banking institution.

Investments purchased by the Laboratory are carried at market value. Donated investments are recorded at fair market value at the date of the gift.
For closely held non-publicly traded investments, management determines the fair value based upon the most recent information available from the
limited partnership. For determi nation of gain or loss upon disposal of investments, cost is determined based on the first-in. first-out method.

R12 Annual Report

Investments with an original maturity of three months to one year, or those that are available for operations within the next fiscal year, are classified
as short-term. All other investments are considered long-term. Investments are maintained primarily with three institutions.

In 1924, the Laboratory became the beneficiary of certain investments, included in permanently restricted net assets, which are held in trust by others.
The Laboratory has the continuing rights to the income produced by these funds in perpetuity, subject to the contractual restrictions on the use of
such funds. Accordingly, the trust has established a process to conduct a review every ten years by an independent committee to ensure the Laboratory
continues to perform valuable services in biological research in accordance with the restrictions placed on the funds by the agreement. The committee
met in 1994 and determined that the Laboratory has continued to meet the contractual requirements. The market values of such investments are
$7,904,545 and $7,275,488 at December 31, 2000 and 1999. respectively. The dividend and interest income on these investments, included in
unrestricted support and revenues, totaled $201.407 and $221.882 in 2000 and 1999, respectively.

Investment Income and Distribution

For the master pooled investments, the Laboratory employs a total return utilization policy that establishes the amount of the investment return made
available for spending each year. The Finance Committee of the Board of Trustees has approved a spending policy that the withdrawal will be based
on a percentage of the 12 quarter average ending market values of the funds. The market value includes the principal plus reinvested income, realized
and unrealized gains and losses. Spending rates in excess of 5%, but not exceeding 7%, can be utilized if approved in advance by the Finance
Committee of the Board of Trustees. For fiscal 2000 and 1 999. the Laboratory obtained approval to expend 6% of the latest 1 2 quarter average ending
market values of the investments.

The net appreciation on permanently and temporarily restricted net assets is reported together with temporarily restricted net assets until such time
as all or a portion of the appreciation is distributed tor spending in accordance with the total return utilization policy and applicable state law.

Investment income on the pooled investment account is allocated to the participating funds using the market value unit method (Note 4).
Held by Bond Trustee

Assets held by bond trustee relate to assets held by an outside trustee under the March 1, 2000 loan and trust agreement. Per the prospectus, these
funds may be used solely for capital projects as determined by the Laboratory's Board of Directors. At December 31, 2000, these assets were invested
in a qualified QIC under a funding agreement with an insurance company.

Plant Assets

Buildings and equipment are recorded at cost. Donated facility assets are recorded at fair market value at the date of the gift. Depreciation is computed
using the straight-line method over the asset's estimated useful life. Estimated useful lives are generally three to five years for equipment and 20 to
40 years for buildings and improvements. Depreciation is not recorded for those assets classified as construction-in-process as they have not yet been
placed into service. Depreciation expense for the years ended December 31, 2000 and 1999 amounted to $1.791.975 and $1,562.487. respectively,
and has been recorded in the statement of activities in the appropriate functionalized categories. When assets are sold or retired, the cost and
accumulated depreciation are removed from the accounts and any resulting gain or loss is included in unrestricted income for the period.

Annuities and Unitrusts Payable

Amounts due to donors in connection with gilt annuities and unitrusts are determined based on remainder value calculations, with varied assumptions
of rates of return and payout terms.

Deferred Income and Advances on Contracts

Deferred income includes prepayments received on Laboratory publications and advances on contracts to be spent within the next year. Advances
on contracts includes funding received for grants and contracts before it is earned. Long-term advances are invested in the master pooled account
until they are expended.

Revenue Recognition

Sources of revenue include grant payments from governmental agencies, contracts from private organizations, and income from the rental of
laboratories and classrooms for research and educational programs. The Laboratory recognizes revenue associated with grants and contracts at the
time the related direct costs are incurred or expended. Recovery of related indirect costs is recorded at predetermined fixed rates negotiated with the
government. Revenue related to conferences and services is recognized at the time the service is provided, while tuition revenue is recognized as
classes are offered. The tuition income is net of student financial aid of $579,790 and $527.258 in 2000 and 1999, respectively. Fees for conferences
and other services include the following activities: housing, dining, library, scientific journals, aquatic resources and research services.

Contributions

Contribution revenue includes gifts and pledges. Gifts are recognized as revenue upon receipt. Pledges are recognized as temporarily or permanently
restricted revenue in the year pledged and are recorded at the present value of expected future cash flows, net of allowance for unfulfilled pledges.
Gifts and pledges, other than cash, are recorded at fair market value at the date of contribution.

Expenses

Expenses are recognized when incurred and charged to the functions to which they are directly related. Expenses that relate to more than one function
are allocated among functions based upon either modified total direct cost or square footage allocations.

Financial Statements R13

Other programs expense consists primarily of fundraising, year-round labs and library room rentals, costs associated with aquatic resource sales and
scientific journals. Total fundraising expense for 2000 and 1999 is $1,156,656 and $1.008,920, respectively.

Use of Estimates

The preparation of financial statements in conformity with generally accepted accounting principles requires management to make estimates and
assumptions that affect the reported amounts of assets and liabilities and disclosure of contingent assets and liabilities at the date of financial
statements and the reported amounts of revenues and expenses during the reporting period. Actual results could differ from those estimates.

Tax-Exempt Status

The Laboratory is exempt from federal income tax under Section 501(c)(3) of the Internal Revenue Code.

Reclassification

Certain prior year balances have been reclassified to conform with the current year presentation.

3. Investments

The following is a summary of the cost and market value of investments at December 31, 2000 and 1999:

Market

Certificates of deposit
Money market securities
U.S. Government securities
Corporate fixed income
Common stocks
Mutual funds
Limited partnerships

Total investments

Cost

2000

1999

2000

1999

40.000

$ 40,000

764.969

1,781.128

764,969

1,781,128

2,300.738

69,125

2,165,197

69,951

2,412,548

2,364,068

2,537,913

2.536.808

16,144,089

15,665,205

16,318,538

10,608,588

19.909,549

26,664,204

19,306,250

23,851,004

6,522,589

1,600,300

5.324,442

958.982

48,094,482

$ 48.184,030

$ 46,457,309

$ 39.846,461

Investment portfolios for the years ended December 31, 2000 and 1999 are as follows:

Market

Short-Term Investments

Certificates of deposit
Money market
Mutual funds
Common stocks in transit

Total short term
Long-Term Investments

Pooled investments:

Master pooled investments
Separately invested:

General Chase Trust

Library Chase Trust

Annuity and unitrusts investments

Total long term
Total investments

Cost

2000

1999

2000

1999

$ 40,000

$ 40.000

377,654

233,938

377,654

233.938

3.102,515

2,875,480

3,085.445

2.965.273

79,664

33,119

79,664

33,119

$3.599,833

$34.116.704

6,204.107
1,700,438
2.473.400

44.494,649
$48.094,482

$3,182,537

$35.354.938

5,717,108
1,558,380
2.371.06?

45.001.493
$48.184,030

$3,648,491

$33,153,390

5,654,623
1,543,691

2.522,842

42.874,546
$46.457,309

$3,272,330

$27,514,505

5.335.721
1.448.569
2.275.336

36.574,131
$39.846.461

R14 Annual Report

For the years ended December 31, 2000 and 1999. the Laboratory recorded net realized gains of $3.886,669 and $1,639.795; net unrealized losses
(gains) of $6,700,396 and $(3,544.380); and dividend and interest income of $1.588,734 and $1,533,579, respectively.

4. Accounting for Pooled Investments

Certain net assets are pooled for investment purposes. Investment income from the pooled investment account is allocated on the market value unit basis,
and each fund subscribes to or disposes of units on the basis of the market value per unit at the beginning of the calendar quarter within which the
transaction takes place. The unit participation of the funds at December 31, 2000 and 1999 is as follows:

2000

7999

Unrestricted
Temporarily restricted
Permanently restricted
Advances on contracts

Pooled investment activity on a per-unit basis was as follows:

11,290
40,042

73.724
5,396

1 30.452

8,573
42,351
65,789

5,557

122.270

Unit value at beginning of year
Unit value at end of year

Total return on pooled investments

2000

$ 283.37

261.53

$ (21.84)

7999

$ 225.51
283.37

$ 57.86

5. Long-Tenn Debt

Long-term debt consisted of the following at December 31:

2000

Variable rate (63% at December 31, 1999) Massachusetts Industrial
Finance Authority Series 1992 A Bonds payable in annual install-
ments through 2012

6.63% Massachusetts Industrial Finance Authority Series 1992B
Bonds, payable in annual installments through 2012

5.8% The University Financing Foundation, Inc. payable in monthly
installments through 2000

5.8% The University Financing Foundation. Inc. payable in monthly
installments through 2002

Variable rate (4.75%) Massachusetts Development Finance Agency
Bonds payable in annual installments from 2006 through 2030

10.200.000
$ 10,200.000

7999

$ 890.000

1.175.000

120,929

138,167

$ 2,324.096

In March 2000, the Massachusetts Development Finance Agency issued on behalf of the Laboratory a series of Variable Rate Revenue Bonds (the
"Bonds") in the amount of $10.200.000. The initial interest rate on the issue was 3.65% and is reset weekly. At December 31, 2000, the rate was 4.75%.
The bonds are scheduled to mature on February 1 , 2030. The Laboratory is required to make interest payments only for the first five years. The first
principal payment is due February 1. 2006 with incremental increases through maturity. The proceeds of these bonds were used to finance the capital
improvements of the Laboratory's educational, research, and administrative facilities, specifically the construction and equipping of the Environmental
Sciences building. A portion of the proceeds was also used to extinguish all of the Laboratory's prior debt obligations.

As collateral for the bonds, the Laboratory has entered into a Letter of Credit Reimbursement Agreement which is set to expire on March 15. 2007.
The Letter of Credit is in an amount sufficient to pay the aggregate principal amount of the bonds and up to 46 days' interest.

The agreements related to these bonds subject the Laboratory to certain covenants and restrictions. Under the most restrictive covenant of this debt, the
Laboratory is required to maintain a debt service coverage ratio.

In 1992, the Laboratory issued $1,100,000 Massachusetts Industrial Finance Authority (MIFA) Series 1992A Bonds with a variable interest rate and
$ 1 .500,000 MIFA Series 1 992B with an interest rate of 6.63%. Interest expense totaled $33.20 1 for the year ended December 3 1 . 2000. The Series 1 992
A and B Bonds were scheduled to mature in December 2012, but were retired on March 8, 2000 with the new bond proceeds.

On March 17, 1998, the Laboratory entered into a ten-year interest rate swap contract in connection with the Series 1992A Bonds. This contract was
canceled as part of the extinguishment of old debt and new debt issuance on March 8. 2000.

Financial Statements R15

In 1 996, the Laboratory borrowed $500.000 with an interest rate of 5.8% per annum from the University Financing Foundation, Inc. The interest expense
for the year ended December 31. 2000 was $1.950. The loan was paid off in March 2000 with the new bond proceeds.

In 1997, the MBL borrowed $250.000 with an interest rate of 5.8% per annum from the University Financing Foundation. Inc. The interest expense
for the year ended December 31. 2000 was $2,140. This loan was scheduled to mature in 2002 but was paid off in connection with the new debt issued
in March of 2000.

The Laboratory has a line of credit agreement with a commercial bank from which it may draw up to $1.000.000. The line of credit has an interest rate
of prime plus 1/2 percent. The line expires May 29, 2001. No amounts were outstanding under this agreement at December 31, 2000 and 1999.

6. Plant Assets

Plant assets consist of the following at December 3 1 :

Land
Buildings
Equipment
Construction in process

2000

$ 702,908

35,236,087

5,059.022

4,681,629

7999

$ 702.908

33.702.485

4,667.026

1.510.821

Total
Less: Accumulated depreciation

Plant assets, net

45,679,646
(22.256.490)

$23.423.156

40.583,240
(20,464.515)

$20,118,725

7. Retirement Plan

The Laboratory participates in the defined contribution pension plan of TIAA-CREF (the "Plan"). The Plan is available to permanent employees who
have completed two years of service. Under the Plan, the Laboratory contributes 10% of total compensation for each participant. Contributions
amounted to $862,850 and $785,509 for the years ended December 31. 2000 and 1999, respectively.

8. Pledges

Unconditional promises to give are included in the financial statements as pledges receivable and the related revenue is recorded in the appropriate net
asset category. Unconditional promises to give are expected to be realized in the following periods:

In one year or less

Between one year and five years

After five years

2000

$ 5,026,750
3,021,752

7999

$ 3,974,385

3,632,683

202,948

Total

8,048.502

7.810.016

Less: discount of $168,460 in 2000 and $236.844 in 1999 and
allowance of $420,000 in 2000 and $100,000 in 1999

(588.460)

(336.X44)

$ 7,460.042

$ 7,473.172

9. Postretiremen! Benefits

The Laboratory accounts for its postretirement benefits under Statement No. 106, "Employers' Accounting for Postretiremen! Benefits Other than
Pensions." which requires employers to accrue, during the years that the employee renders the necessary service, the expected cost of benefits to be
provided during retirement. As permitted, the Laboratory has elected to amortize the transition obligation over 20 years commencing on January 1, 1994.

The Laboratory's policy is that all current retirees and certain eligible employees who retired prior to June 1 , 1 994 will continue to receive postretirement
health benefits. The remaining current employees will receive benefits; however, those benefits will be limited as defined by the Plan. Employees hired
on or after January I. 1995 will not be eligible to participate in the postretirement medical benefit plan.

R16 Annual Report

The following tables set forth the Plan's funded status as of December 31:

Change in benefit obligation

Postretiremen! benefit obligation at beginning of year
Service cost
Interest cost
Actuarial gain
Benefits paid

Postretiremen! benefit obligation at end of year

Change in plan assets

Fair value of plan assets at beginning of year
Employer contribution
Actual return on plan assets
Benefits paid

Fair value of plan assets at end of year

Funded status

Unrecognized actuarial gain
Unrecognized net obligation at transaction

Accrued postretiremen! benefit cost

Less estimated amount payable within one year and classified
as a current liability

Accrued postretiremen! benefit cost, net of current portion

Weighted-average assumptions as of December 3 1
Discount rate

2000

2.043,659

23,020

149.574

(87,740)

(136.844)

1,991,669

936,149

182.776

56,465

(136.844)

1 .038,546

(953,123)
(185,377)
1.128.691

(9,809)

$ (9,809)

7.50%

1999

2,171,119

28,231

134,533

(174.966)

(115.258)

2,043.659

820,645

192,082

38.680

(115.258)

936.149

(1.107,510)

(125,351)
1.215.513

(17,348)

$ (17.348)

8.00%

For purposes of measuring the benefit obligation, an 8.0% annual rate of increase in the per capita cost of covered health benefits was assumed for 2000.
The rate was assumed to decrease gradually to 5% in 2005 and remain at that level thereafter.

Components of net periodic benefit cost
Service cost
Interest cost

Expected return on assets
Amortization of net obligation at transition
Recognized net actuarial loss

Net periodic benefit cost

Impacl of 1% increase in healih care cost trend:
on interest cost plus service cost during past year
on accumulated postretiremen! benefit obligation

Impact of 1 % decrease in health care cost trend:
on interest cost plus service cost during past year
on accumulated postretiremen! benefit obligation

2000

$ 23,020

149,574

(69,524)

86,822

(14.655)

$ 175,237

14,271
41,263

(11,946)
(233,324)

1999

28,231

134,533

(61,425)

86.822

(5,385)

$ 182,776

(71,626)
(456,863)

(10,559)
(235,728)

Pension plan assets consist of investments in a money market fund.

1C'

£ i 1 i§ 2

rl rl 2 £ o

ON ~~~"

rt-

sD <"*"* sO r — "

— ' OC t*~i r] r*-(

S! 1

S — m o r-
r- ON m ^t

r- rj ON ^c >n
oo <n ON >n ON

— s—

ON'

m* ri rl

— • r I

r, so so 0 fN

— cs r*i m ri

r- sc oo m, oc

sC ^ x: ON •-!

i 1

V~i

m O rt DO r-
r) "rf O DO so

— ON sO ~~ DC

o ^0 o r~" o

<N h.

<*•* rn rj

-J — — r*-.

-f n

rn r-1 ON

{, -C tr,

in so

SO sD OO

OO m,

DO r-

a 5 •?

= ^ C

ON'

rn 1 1 1 r-"

r- 1 m

— 1 t' ri
1 r-

•5" ^ ^

So ^o o

Tt O oo rj o

u £

O ri m
oc o -3- •<•»•

ri oo ~f ON in

— in o O

Q •;..

oc o ^' o

r-' 1 ON' r-'

O — — r-

| 3

ri r-

C/3

•<

00 V~| O w~.

cn t*- t-

w
(J

§; S °° ?.

ID in f^

ri r-' so'
in m m

i S ^

•-i

m, — so

V* "*"

— f-1

— -t C rt r

O O O sC

1' s

O fi (N r- ^C

— r*-, ON sC sC

-t — DC in,

O m, m,

^S ^

m" so* sc 1 1 sc DC
r- o 1 r-i

OC \O DO

m r- (N

^c — «•*"'

c/:
Z

o m r— DC o ^

sC C -~ —

\C ON ^t r- oc C

^, — ?i J^;

02
O

"•

ON o* in" i in o c

ri Tf i in TT

rj | — ^t

•^

ri ON -n! 00 0 -

i i/-) — -i \o \c r i

- O DC sD — t

Z =

— <•*") m, rj ^c C

•o

>~ ^

rn — ' ^t ""^ O r

j m n ON — oo

•a

—

•"- Si

O 1 • ~~ '^l

r~~ oc — m.
\O r- ON CN

o
g

Q
UJ

L? ' =
* 1

ac ae ^ m ON •-

3 00 m 0 —

ri — r- sC1 sC s£
r- sC O ri rj r

; — m! rr, -t
i in oc ri oc

.£

sC oc V, <-*-, rj C
ri r- ri -rt s£

: -t — oc s£

02
^

'•§

^c — —

-C

J! -S
3 o

C .1

o „

'•<=,

ON ON

oo m ON

O ON ON

ON ON

oo r~-
r- o

n rj

O rn
rn in

-t
-t

(-*-,

r- 1
ri

t a
§ a

oc — —

i O ri

so'

^ o

ri r*-,

— -t

^ 5 ^

1 g S

O
sO

ri
O

c "

2 u

—

— —

Is w

3- -a

-T ri r-' oo oc r*- r-

O -t oo -t — -1- ri
•I- ri — ON r- — u~.

If „

= 1 i s

-3 ,

E S
SO

R17

Report of the Library Director

During the past several years a major paradigm shift
has clearly taken place in the MBLAVHOI Library. We
now have more than 2000 full-text electronic journals
available on the network. The library web site is the
starting point for content rich information that is being
delivered to the much heralded "scientists' desktop."

The simple act of checking in a journal and placing it
on the shelf or requesting an Interlibrary Loan now
requires the use of various pieces of software like
Prospero, CLIO, OCLC Microenhancer. OCLC Passport,
Ariel, Microsoft Office. EDI, ABLE, URSA, and various
modules of Mariner, as well as online delivery service
software for statewide courier services: FedEx, UPS.
CISTI, NTIS. and ISI. The inauguration of information
delivery via our web site also employs the use of SQUID,
Geobrowser. LUCID. MySQL, Ultra Edit, Adobe GoLive,
Adobe Premiere. Adobe Acrobat, Omni Page, Web Star.
Fetch, Quid Pro Quo, Microsoft Office, Portfolio.
Graphics Converter. Home Page, SSH, FileMaker. BB
Edit, Illustrator. Photoshop, PageMaker, and Apache, and
languages such as PHP3, Perl, and SQL. Obviously,
"instant" delivery of information requires many hours of
staff time implementing major software and hardware
infrastructure installations to support this effort.

This instant information drive is powerful, but intellectual
ownership and archival requirements are elusive in the world
of ePublishers and libraries. Print subscriptions still arrive
daily, and electronic journals seem to disappear from a web
site at whim. We are making choices in an age of disruptive
technologies and value -changing economies. Still, much was

accomplished in 2000 in the library. The emphasis this year
was on expanding the serial collection, both print and
electronic. The collection has grown to more than 200,000
volumes — occupying all the physical space we have
available in Woods Hole.

Space anil Renovations

Providing space for library resources is a constant concern
for library patrons and staff. Some of the storage problems
have been addressed by providing more electronic access to
journals and sending some volumes off campus to the
Harvard Depository. A Feasibility Study performed by Jay
Lucker. Library Consultant, and Stephen Hale, Architect,
presented several design ideas to the Trustees. Along with
the major recommendation for additional space, the study
resulted in a redesign of the equipment and furniture in the
catalog room, which allowed more computer terminals for
patrons, and the installation of a "window" to the reference
desk for easy access to "live" reference information. In
addition, the WHOI Archives fini hed a compact shelving
project that encompasses 2130 square feet and resulted in
1 1 ,200 linear feet; it will allow for more aggressive record
management and 15 years' added growth in archival space
projections.

The major improvement to the current library space in
the Lillie building was the installation of a new HVAC
system that supplies heat in the winter and cooling in the
summer to the stack area, the library office, and the
reading rooms. This joint venture, financially supported

R1S

Library Director's Report R19

by both MBL and WHOI. is preventing the wild
temperature swings that can be so damaging to the
collections. This is a key improvement and the basis for
any conservation and preservation program.

Special Collection and Rare Books

Dr. Garland Allen and Carol Winn identified nearly
2500 volumes in the open stacks, dating from the early
19th century, that require preservation and increased
security to protect their plates and illustrations. Our Rare
Books Room is filled to capacity, so we must find
additional space for these materials in the coming years.

The Mary Sears collection, which included individual
pieces of the Challenger and Siboga expeditions, was
cataloged and indexed this year. Dr. Arthur Humes'
collection was also processed; it included a collection of
exotic shells. Also acquired and added to the Florence
Gould Collection in the Rare Books was Guillaume
Rondelet's Libri de Piscibus Marinis (1554). This volume
is now the oldest book in the collection and one of the
first books to describe marine organisms and fishes.

Electronic Access

As access to information becomes more interactive and
information retrieval moves at breakneck speed, the
importance of web design and accessibility heightens. The
library's web page will continue to be in "re-design" mode
with the addition of new resources and services. A new staff
member, Amy Stout. Digital Systems and Services
Coordinator, is in charge of posting and monitoring the use
of this integral part of the library's services. Major upgrades
to the library's software took place this year, which resulted
in a new look and feel to the web interface, allowing more
flexibility in customizing displays for patrons.

Electronic access to the Oxford English Dictionary and
web versions of Zoo Record and PsycINFO were new
additions to the library holdings this year. The library
joined JSTOR, a project that provides digital archives of
classic serials in the general sciences, ecology, and
botany. JSTOR gives us access, for example, to the entire
run of The Philosophical Transactions of the Royal
Society of London from Volume 1, Number 1 in 1665.

Cooperating Libraries

The Boston Library Consortium (BLC) received grant
funding from the Massachusetts Board of Library
Commissioners for the implementation of a virtual catalog and
interlibrary loan (ILL) direct distance borrowing project
(VirCat). This grant has made it possible for a growing number
of libraries in the consortium to allow patron initiated
borrowing from each other's collections without going through
the ILL librarians. A group of BLC libraries, including MBL/

WHOI. purchased ScienceDirect from Elsevier. This increased
our full-text electronic coverage of Elsevier titles from 1 1 1 to
850, which represents the combined holdings of Elsevier titles
by the BLC members along with an additional 400 Springer-
Verlag full-text eJournals.

Volunteers and Staff

Judy Ashmore, the Assistant Director for MBL Library
Operations, Marguerite (Peg) Costa, Cataloger, and
Margot Garritt, WHOI Archivist, together representing
more than 50 years of experience in the Library, retired
this year. Their work has been greatly appreciated by the
entire Woods Hole scientific community.

Eleanor Uhlinger, former Director of the Pell Marine
Science Library, joined the library as Assistant Director in
January 2001. Sha Li (Lisa), Director of Information
Services Center and Library for the South China Sea
Institute of Oceanology, Chinese Academy of Sciences in
Guangzhou, China, spent two months in the library on a
study visit learning new technology.

The volunteers in the Rare Books Room and Archives
in the Main library, as well as the volunteers in the Data
library, have provided invaluable assistance in helping to
organize and make these collections available for future
scientists. The oral history project at WHOI has been a
great success and will be of inestimable value as the 75th
anniversary of that institution approaches. Peg Costa
joined the ranks of volunteers and helps Carol Winn with
the original cataloging project in the rare books.

It is with extreme sorrow that I report that Dr. Robert
Huettner died in March 2001. He will be remembered as
someone who had a very real element of the spirit of
discovery and learning, a teacher who exuded enthusiasm as
well as knowledge. Bob and his wife, Millie, have been
volunteers in the Rare Books room for more than 10 years.

The MBLAVHOI Library hosted the Information Futures
Institute at the Jonsson Center in May and welcomed leaders
in the field of library science. Participation in these meetings
is important not only for the national recognition it affords,
but for the leadership these groups exercise in shaping the
future of research libraries.

The library has embraced the era of informatics.
Funded by the Jewett Foundation, extensive research is
underway creating an electronic Key system in taxonomy
and a taxonomic name server that will serve the academic
enterprise over the web. The library committee has
finished its strategic plan, which continues to support the
library's mission, and looks forward to a future providing
a collaborative and collegia! environment, with access to
information essential to scientific research, preservation of
materials for future generations, and teaching in the
Woods Hole scientific institutions.

— Catherine Norton

Educational Programs

Summer Courses

Biology of Parasitism: Modern Approaches
(June 8-August 11, 2000)

Directors

Pearce. Edward, Cornell University

Tschudi. Christian, Yale University School of Medicine

Faculty

Phillips, Meg, University of Texas Southwestern Medical Center
Russell, David, Washington University School of Medicine
Scott, Phillip, University of Pennsylvania
Selkirk, Murray. Imperial College of Science. Technology &

Medicine, United Kingdom

Sibley, David, Washington University School of Medicine
Ullu, Elisabetta, Yale University School of Medicine
Waters, Andrew P., Leiden University Medical Centre

Lecturers

Allen, Judith. University of Edinburgh

Artis. David, University of Pennsylvania

Bangs. Jay. University of Wisconsin-Madison

Beckers, Cornelis, University of Alabama, Birmingham

Beverley, Stephen. Washington University School of Medicine

Borst. Piet, Netherlands Cancer Institute

Burleigh. Barbara. Harvard School of Public Health

Cully, Doris, Merck & Co.

Dunne, David, Cambridge University

Fidock, David. Albert Einstein College of Medicine

Goldberg, Daniel. Washington University School of Medicine

Grencis. Richard, University of Manchester, United Kingdom

Guiliano. David

Gull, Keith. University of Manchester. United Kingdom

Hajduk. Steve, University of Alabama, Birmingham

Hoffman, Steve

Hunter, Christopher, University of Pennsylvania

Komuniecki, Richard, University of Toledo

Kopf, Manfred, Basel Institute for Immunology, Switzerland

Landfear. Scott. Oregon Health Sciences University

Langhorne. Jean. Medical Research Council

McKerrow, James

Mottram, Jeremy, University of Glasgow

O'Neill, Scott, Yale University School of Medicine

Parsons, Marilyn, Seattle Biomedical Research Institute

Preiser. Peter. Medical Research Council

Rathod. Pradip. Catholic University of America

Sacks. David. National Institutes of Health

Scherf, Artur, Institut Pasteur, France

Sher. Alan, National Institutes of Health

Sinnis. Photini. New York University School of Medicine

Tarleton, Rick, University of Georgia

Turco. Sam, University of Kentucky Medical Center

Wang. Ching Chung, University of California. San Francisco

Wirth. Dyann, Harvard School of Public Health

Teaching Assistants

Beatty, Wandy, Washington University School of Medicine
Djikeng, Appolinaire. Yale University School of Medicine
Hussein, Ayman, Imperial College of Science. Technology &

Medicine. United Kingdom

Jackson, Laurie, University of Texas Southwestern
Kissinger, Jessica. University of Pennsylvania
Lovett. Jennie, Washington University School of Medicine
MacDonald, Andrew. Cornell University

Morrissette. Naomi. Washington University School of Medicine
Reiner. Steven
van der Wei, Annemarie, Biomedical Primate Research Centre, The

Netherlands

Course Assistants

Chipperfield, Caitlin Nadine, Cornell University
Johnson, Ben, Cornell University

Students

Andersson, John, Karolinska Institut

D'Angelo, Maxinuliano. University of Buenos Aires

Dolezal, Pavel, Charles University. Prague

Ferreira, Ludmila, Universidade Federal de Minas Gerais

Figueiredo, Luisa, Institut Pasteur

Gilk. Stacey, University of Vermont

Lamb. Tracey. University of Edinburgh

Lowell, Joanna, Rockefeller University

Martins. Gislaine, University of Sao Paulo

Murta, Silvane, Centro de Pesquisas "Rene Rachou," Brazil

O'Donnell. Rebecca, University of Melbourne

Ralph, Stuart, University of Melbourne

Sehgai, Alftca, Tata Institute of Fundamental Research India

Tangley, Laura, U.S. News & World Report. Science Writer

Triggs. Veronica. University of Wisconsin. Madison

Ulbert, Sebastian, Netherlands Cancer Institute

Villarino, Alejandro, University of Pennsylvania

Embryology: Concepts and Techniques in
Modern Developmental Biology
(June 18-Jnly 29, 2000)

Directors

Bronner-Fraser, Marianne, California Institute of Technology
Fraser, Scott, California Institute of Technology

R20

Educational Programs R21

Faculty

Adoutte. Andre, University of Paris-Sud. France
Blair, Seth S., University of Wisconsin, Madison
Carroll, Sean, University of Wisconsin, Madison
Collazo, Andres, House Ear Institute
Ettensohn, Charles, Carnegie Mellon University
Harland. Richard, University of California. Berkeley
Henry, Jonathan, University of Illinois, Urbana
Krumlauf. Robb. National Institute for Medical Research
Levine, Michael. University of California, Berkeley
Martindale, Mark. Kewalo Marine Laboratory
Niswander, Lee, Memorial Sloan-Kettering Cancer Center
Rothman, Joel, University of California, Santa Barbara
Saunders, John, Jr., Marine Biological Laboratory
Schupbach. Trudi, Princeton University
Shankland, Martin. University of Texas, Austin
Soriano, Philippe, Fred Hutchinson Cancer Research Center
Wieschaus, Eric, Princeton University
Wray. Gregory, Duke University
Zeller. Robert, University of California. San Diego

Lecturers

Davidson, Eric, California Institute of Technology
Holland, Linda, University of California, San Diego
Hopkins, Nancy, Massachusetts Institute of Technology
Joyner. Alexandra, New York University School of Medicine
Rosenthal, Nadia. Massachusetts General Hospital, East
Smith, William. University of California, Santa Barbara
Stern, Claudio, Columbia University

Teaching Assistants

Allison. Toby, Howard Hughes Medical Institute
Atit, Radhika. Memorial Sloan-Kettering Cancer Center
Baker, Clare, California Institute of Technology
Garcia-Castro, Martin, California Institute of Technology
Gendreau, Steve, Exelixis, Inc.
Kuhlman. Julie, University of Oregon

Lane. Mary Ellen. University of Massachusetts Medical Center
Lartillot, Nicolas, University of Paris-Sud. France
Liu, Karen, University of California, Berkeley
Maduro. Morris, University of California, Santa Barbara
Mariani. Francesca, University of California, Berkeley
Micchelli, Craig, University of Wisconsin, Madison
Ober, Elke, University of California, San Francisco
Seaver, Elaine, University of Hawaii
Tabin, Clifford, Harvard Medical School
Tobey, Allison. Memorial Sloan-Kettering Cancer Center
Trainor, Paul. Medical Research Council, United Kingdom
Wallingford, John, University of California, Berkeley
Walsh, Emily. University of California. San Francisco
Williams. Terri A.. Yale University
Wilson. Valerie, University of Edinburgh

Course Assistants

Hurwitz, Mark, Marine Biological Laboratory
Stringer. Kristen. Marine Biological Laboratory
Wylie. Matthew. Marine Biological Laboratory

Students

Aspock. Gudrun, University of Basel

Ballard. Victoria, University of Surrey. United Kingdom

Bates, Damien. Murdoch Childrens Research Institute

Beckhelling, Clare. Marine Biology Station, France

Bellipanni, Gianfranco. University of Pennsylvania

Cheeks, Rebecca. University of North Carolina. Chapel Hill

Dichmann, Darwin, Hagedorn Research Institute

Dorman, Jennie, University of Washington

Ellertsdottir, Elin, University of Freiburg

Espinoza, Nora, Louisiana State University

Ezin, Max. University of Virginia

Field, Holly, University of California. San Francisco

Gong, Ying, California Institute of Technology

Gross, Jeffrey. Duke University

Huber. Jennifer, University of Hawaii

Imai. Kazushi. Columbia University

Javaherian, Ashkan, Cold Spring Harbor Lab

Jiang, Di, National Institutes of Health

Khokha, Mustafa, University of California. Berkeley

Kyrkjebo, Vibeke, Sars Centre

Lee, Vivian, Oregon Health Sciences University

Mansfield, Jennifer, Columbia University

Marx, Vivien, Freelance Science Journalist

Nasevicius, Aidas, University of Minnesota

Prud'homme, Benjamin, CNRS

Skromne, Isaac, Princeton University

Warkman, Andrew, University of Western Ontario

Microbial Diversity (June 11 -July 27, 2000)

Directors

Harwood, Caroline, University of Iowa
Spormann, Alfred, Stanford University

Faculty

Overmann, Jorg, University of Oldenburg
Schmidt, Thomas, Michigan State University

Lecturers

Delong, Edward. Monterey Bay Aquarium Research Institute

Gaasterland, Terry. Rockefeller University

Greenberg. E. Peter, University of Iowa

Groisman. Eduardo A.. Washington University School of Medicine

McFall-Ngai. Margaret, University of Hawaii

Omston. Nicholas, Yale University

Parsek, Matthew, Northwestern University

Rainey, Paul. Oxford University

Schoolmk, Gary. NIH/NIAID

R22 Annual Report

Stemmer, Pirn. Maxygen, Inc.

Visscher. Pieter, University of Connecticut

Walker, Graham. Massachusetts Institute of Technology

Weinstock. George, University of Texas. Houston

Teaching Assistants
Johnson. Hope. Stanford University
Leadbetter. Jared. University of Iowa
Lepp, Paul, Stanford University
Schaefer. Amy. University of Iowa

Course Coordinator

Hawkins, Andrew, University of Iowa

Course Assistant

Ament, Nell, Marine Biological Laboratory

Students

Barak. Yoram, Hebrew University

Begos. Kevin, Winston-Salem Journal. Science Writer

Blake. Ruth. Yale University

Buckley, Daniel. Michigan State University

Callaghan. Amy. Rutgers University

Goldman. Robert. University of Houston

Hansel. Colleen. Stanford University

Kadavy, Dana. University of Nebraska. Lincoln

Kirisits. Mary Jo, University of Illinois, Urbana-Champaign

Lester, Kristin, Stanford University

Lin, Li-hung, Princeton University

MacRae, Jean. University of Maine

McCance, James, Leicester University. England

McMullin. Erin, Penn State University

Neretin, Lev, Shirshov Institute of Oceanography

Powell, Sabrina, University of North Carolina, Chapel Hill

Scott, Bari. SoundVision Productions Science Writer

Simpson. Joyce. University of Illinois, Urbana

Singh, Brajesh. Imperial College

Stevenson, Bradley. Michigan State University

Ward. Dawn. University of Delaware

Zaar. Annette. Universitat Freiburg

Neural Systems and Behavior

(June ll-August 4, 2000)

Directors

Carr, Catherine, University of Maryland
Levine. Richard. University of Arizona, Tucson

Faculty

Brodtuehrer, Peter. Bryn Mawr College

Dudchenko, Paul. University of Stirling

Ferrari, Michael. University of Arkansas

French. Kathleen. University of California, San Diego

Glanzman. David. University of California. Los Angeles

Kelley, Darcy. Columbia University

Knierim. James. University of Texas Medical School

Kristan, William, University of California, San Diego

Nadim. Farzan. Rutgers University

Nusbaum, Michael, University of Pennsylvania School of Medicine

Prusky, Glen. The University of Lethbridge. Canada

Roberts. William, University of Oregon

Szczupak. Lidia, Universidad de Buenos Aires
Weeks. Janis, University of Oregon
Wood, Emma, University of Edinburgh
Zakon, Harold, University of Texas, Austin

Lecturers

Augustine. George. Duke University

Korn, Henri, Pasteur Institut

Maler. Leonard, University of Ottawa

Pflueger, Hans-Joachim, Freie Universitat Berlin

Ribera, Angela, University of Colorado Health Science Center

Schwartz, Andrew, The Neuroscience Institute

Walters, Edgar T., University of Texas Medical School

Teaching Assistants
Armstrong, Cecilia, University of Washington
Beenhakker. Mark. University of Pennsylvania
Blitz. Dawn Marie. University of Chicago
Bower, Mark, University of Arizona, Tucson
Chen, Shanping, House Ear Institute
Chitwood, Raymond. Baylor College of Medicine
Coleman. Melissa. St. Joseph's Hospital
Duch. Carsten, University of Arizona, Tucson
Gamkrelidze. Georgi, Lucent Technology
Gerrard, Jason, University of Arizona. Tucson
Goodman. Miriam B., Columbia University
Hill, Andrew, Emory University
Masino, Mark. Emory University
McAnelly, Lynne, University of Texas, Austin
Otis, Thomas. University of California. Los Angeles
Parameshwaran, Suchitra, University of Maryland
Philpot, Benjamin, Brown University
Scares, Daphne, University of Maryhnd
Stell, Brandon. University of California, Los Angeles
Villareal. Greg. University of California, Los Angeles
Yong. Rocio. University of California. Los Angeles
Zee. M. Jade. University of Oregon

Course Assistants

Aimers. Lucy. Marine Biological Laboratory
Psujek. Sean. Marine Biological Laboratory

Students

Akay, Turgay, University of Cologne

Archie. Kevin, University of Southern California

Billimoria, Cyrus, Brandeis University

Black, Michael, Arizona State University

Boyden, Edward, Stanford University

Bradford, Yvonne. University of Oregon

Cardin, Jessica. University of Pennsylvania

Dasika, Vasant. Boston University

Ding, Long. University of Pennsylvania

Froemke, Robert. University of California, Berkeley

Grammer. Michael, University of Southern California

Hubbard, Aida, University of Texas, San Antonio

Hunt. Barbekka. University of Colorado, Boulder

Karmarkar, Uma. University of California, Los Angeles

Konur, Sila, Columbia University

Oestreich, Joerg. University of Texas, Austin

Rela. Lorena, University of Buenos Aires

Sinha. Shiva. University of Maryland

Siuda. Edward, Michigan State University

Tobin, Anne-Elise. Emory University

Educational Programs R23

Neurobiology (June ll-Aitgmt 12, 2000)

Directors

Faher. Donald. Albert Einstein College of Medicine
Lichtman. Jeff W.. Washington University School of Medicine

Section Director

Greenberg, Michael. Children's Hospital

Faculty

Denk, Winfried, Max-Planck Institute for Medical Research

Can. Wenbiao, New York University School of Medicine

Griffith. Leslie. Brandeis University

Harris. Kristen, Boston University

Hart. Anne. Massachusetts General Hospital

Heuser, John E., Washington University School of Medicine

Howell. Brian. National Institutes of Health

Khodakhah. Kamran. University of Colorado School of Medicine

Lambert. Nevin. Medical College of Georgia

Lin, Jen-Wei. Boston University

Nedivi. Elly. Massachusetts Institute of Technology

Nowak. Linda. Cornell University

Reese. Thomas. National Institutes of Health

Sanes. Joshua, Washington University Medical School

Schweizer. Felix. University of California. Los Angeles

Shaman. Steven. Children's Hospital

Smith, Carolyn. National Institutes of Health

Terasaki. Mark, University of Connecticut Health Center

Thompson. Wesley J., University of Texas

Van Vactor. David, Harvard Medical School

Wong. Rachel, Washington University School of Medicine

Lecturers

Barres. Ben A.. Stanford University School of Medicine

Bean, Bruce. Harvard University

Conchello. Jose-Angel, Washington University

Ghosh, Anirvan, Johns Hopkins University School of Medicine

Linden. David, Johns Hopkins University

McCleskey. Edwin, Oregon Health Sciences University

McMahan, Uel, Stanford University School of Medicine

Miller, Chris. Brandeis University

Sigworth, Fred. Yale University

Smith, Stephen, Stanford University School of Medicine

Tsien, Roger, University of California, San Diego

Turrigiano, Gina

Teaching Assistants

Pereda. Alberto. Albert Einstein College of Medicine
Petersen. Jennifer. National Institutes of Health
Tumey. Stephen. Washington University
Walsh. Mark. Washington University School of Medicine

Course Assistants

Chiu. Delia. Marine Biological Laboratory
Nover. Harris. Marine Biological Laboratory

Students

Ang. Eugenius. Yale University

Kettunen. Petronella. Karolinska Institute!

Khabbaz, Anton. Princeton University/Lucent Technologies

Livet. Jean, IBDM, Marseille

Long, Michael, Brown University

McKellar. Claire. Harvard University

Misgeld, Thomas. Max-Planck-Institute of Neurobiology. Martinsried,

Germany
Nelson. Laura. National Institute for Medical Research. United

Kingdom

Ruta, Vanessa, The Rockefeller University
Weissman. Tamily, Columbia University
Yasuda, Ryohei, Teiko University Biotech Research Center
Zhong. Haining, Johns Hopkins University

Physiology: The Biochemical and Molecular
Basis of Cell Signaling (June ll-July 22, 2000)

Directors

Garbers. David. University of Texas Southwestern Medical Center
Reed. Randall. Johns Hopkins University School of Medicine

Faculty

Furlow. John. University of California, Davis
Lockless, Steve. University of Texas Southwestern Medical Center
Noel. Joseph. Salk Institute

Prasad. Brinda. Johns Hopkins School of Medicine
Quill, Timothy, University of Texas Southwestern Medical Center
Ranganathan. Rama. University of Texas Southwestern Medical
Center

R24 Annual Report

Verdecia, Mark, Salk Institute

Wedel. Barbara, University of Texas Southwestern Medical Center

Zhao. Haiqing. Johns Hopkins School of Medicine

Zielinski. Raymond, University of Illinois, Urbana

Isenberg Lecturer

Hudspeth. A., James, Rockefeller University

Lecturers

Armstrong, Clay, University of Pennsylvania

Buck, Linda, Harvard Medical School

Clapham, David. Harvard Medical School

Devreotes. Peter. Johns Hopkins University School of Medicine

Dixon, Jack, University of Michigan Medical School

Ehrlich. Barbara, Yale University

Eraser, Scott, California Institute of Technology

Freedman, Leonard. Memorial Sloan-Kettering Cancer Center

Hilgemann, Donald W., University of Texas Southwestern

Medical Center

Huganir, Richard, Johns Hopkins University School of Medicine
Jaffe. Lionel, Marine Biological Laboratory
MacKinnon. Roderick, Rockefeller University
Mangelsdorf, David, University of Texas Southwestern

Medical Center

Oprian, Daniel. Brandeis University
Stamler, Jonathan S., Duke University Medical Center
Wilkie. Thomas, University of Texas Southwestern Medical Center

Course Coordinator

Lemme, Scott, University of Texas Southwestern Medical Center
Rossi. Kristen. University of Texas Southwestern Medical Center

Students

Brclid/.e. Tinatin. University of Miami School of Medicine
Carroll, Michael. University of Newcastle upon Tyne, United

Kingdom

Colon-Ramos. Daniel. Duke University

Cordeiro. Maria, Sofia Instituto Gulbenkian de Ciencia. Portugal
Costa, Patricia, University of Rio de Janeiro
Cotrufo, Tiziana, Scuola Normale Superiore
Crespo-Barreto, Juan, University of Puerto Rico
Cruz, Georgina, University of South Florida
Dayel. Mark, University of California, San Francisco
Fleegal. Melissa. University of Florida
Fleischer, Jorg, University of Hohenheim
Glater. Elizabeth. Brown University

Jhaveri, Dhanisha, Tata Institute of Fundamental Research

Johansson, Viktoria, Goteborg University

Mah, Silvia, Scripps Institution of Oceanography

Marrari, Yannick, Villefranche Sur Mer

Meister, Jean-Jacques, Swiss Federal Institute of Technology

Menna, Elisabetta, Institute of Neurophysiology, Pisa

Nguyen, Anh, University of Kansas

Petrie, Ryan, University of Calgary

Rankin, Kathleen, Oberlin College

Rodeheffer, Carey, Emory University

Rodgers, Erin, University of Alabama, Birmingham

Seipel, Susan, Rutgers University

Sen, Subhojit. Tata Institute of Fundamental Research

Shatkin-Margolis, Seth, Duke University

Shilkrut, Mark, Technion-Israel Institute of Technology

Takai, Erica, Columbia University

Zeidner, Gil, Weizmann Institute of Science

Special Topics Courses

Analytical and Quantitative Light Microscopy
{May 4-May 12, 2000)

Directors

Sluder, Greenfield, University of Massachusetts Medical School
Wolf, David, BioHybrid Technologies Inc.

Faculty'

Amos, William B., Medical Research Council. United Kingdom

Cardullo, Richard, University of California, Riverside

Gelles. Jeff. Brandeis University

Inoue, Shinya, Marine Biological Laboratory

Oldenbourg. Rudolf, Marine Biological Laboratory

Salmon, Edward, University of North Carolina, Chapel Hill

Silver, Randi, Cornell University Medical College

Spring, Kenneth, National Institutes of Health

Straight, Aaron, Harvard Medical School

Swedlow, Jason. University of Dundee

Lecturer

McCrone. Walter, McCrone Research Institute

Teaching Assistants

Grego, Sonia, University of North Carolina, Chapel Hill
Hinchcliffe, Edward, University of Massachusetts Medical School
Pollard, Angela, BioHybrid Technologies

Course Coordinator

Miller. Rick. University of Massachusetts Medical School

Students

Abraham, Clara, University of Chicago

Alvarez, Xavier, N.E. Regional Primate Research Center, Harvard

Medical School

Andrews. Paul, University of Dundee
Bonnet, Gregoire, Rockefeller University
Bravo-Zanoguera, Miguel. University of California, San Diego
Cohen, David, Cornell University Medical College
Connett, Marie, Westvaco Forest Sciences Lab
Crittenden, Sarah, University of Wisconsin, Madison

Educational Programs R25

D'Onofrio, Terrence. Pennsylvania State University

Faruki. Shamsa. Wadsworth Center

Gasser. Susan. Swiss Cancer Institute

Handwerger, Korie. Carnegie Institution of Washington

Hunter, Edward. Q3DM

Jansma. Patricia. University of Arizona

Kraft. Catherine. University of Pittsburgh

Lee, Michelle. Harvard Medical School

Lowe. Christopher, University of California. Berkeley

Lu. Bai. National Institutes of Health/NICHD

Maldonado. Hector, Universidad Central del Carihe

Matsumoto. Vutaka. University of Colorado

McKinney. Leslie, Uniformed Services University

Morelock. Maurice. Boehringer Ingelheim Pharmaceuticals

Mundigl. Olaf. Roche Diagnostics

Mycek. Mary-Ann. Dartmouth College

Provencal. Bob. Los Alamos National Laboratory

Sanabria. Priscila. Universidad Central del Caribe

Sedwick. Caitlin. University of Chicago

Tang. Jay. Indiana University

Tirnauer, Jennifer. Harvard Medical School

Xu. Fang. The Hospital for Special Surgery

Frontiers in Reproduction: Molecular and
Cellular Concepts and Applications
(May 21-July I, 2000)

Directors

Hunt. Joan. University of Kansas Medical Center

Mayo. Kelly. Northwestern University

Schatten. Gerald. Oregon Health Sciences University

Faculty

Ascoli, Mario, University of Iowa College of Medicine

Campbell, Keith, PPL Therapeutics

Camper. Sally, University of Michigan Medical School

Chan. Anthony. Oregon Health Sciences University

Croy. Barbara Anne. University of Guelph. Canada

Dominko. Tanja. Oregon Regional Primate Research Center

Gibori. Geula. University of Illinois

Hunt. Patricia A.. Case Western Reserve University

Jaffe. Launnda. University of Connecticut Health Center

Moore. Karen. University of Florida

Morris. Patricia. The Rockefeller University

Mukherjee, Abir, Northwestern University

Nilson, John. Case Western Reserve Medical School

Page. Ray. PPL Therapeutics Inc.

Pedersen. Roger. University of California. San Francisco

Shupnik. Margaret. University of Virginia Medical Center

Smith, Lawrence, University of Montreal

Terasaki. Mark. University of Connecticut Health Center

Wakayama. Teruhiko, Rockefeller University

Weigel. Nancy. Baylor College of Medicine

Lecturers

Balczon, Ronald. University of South Alabama
Behringer. Richard, University of Texas
Charo. Alta. University of Wisconsin, Madison
Compton. Duane. Dartmouth Medical School
Crowley, William, Massachusetts General Hospital
De Sousa. Paul, Alexandre Roslin Institute
Fazleabas, Asgerally. University of Illinois

Hennighausen. Lothar, National Institutes of Health, NIDDK

Mitchison. Timothy, Harvard Medical School

Myles. Diana, University of California

Ober. Carole. University of Chicago

Orth, Joanne. Temple University School of Medicine

Palazzo. Robert, University of Kansas

Piedrahita. Jorge, Texas A&M University

Reijo Pera. Renee, University of California

Richards, Jo-Anne, Baylor College of Medicine

Ruderman, Joan. Harvard Medical School

Shenker, Andrew. Children's Memorial Hospital. CMIER

Sluder. Greenfield. University of Massachusetts Medical School

Stearns, Tim

Strauss, Jerome, University of Pennsylvania Medical Center

Tilly. Jonathan L., Massachusetts General Hospital

Wall. Robert, U.S. Department of Agriculture

Wessel, Gary. Brown University

Woodruff. Teresa. Northwestern University

Teaching Assistants

Berard. Mark. University of Michigan

Carroll. David. Florida Institute of Technology

Giusti. Andrew, University of Connecticut Health Center

Gray. Heather. University of Chicago

Greenwood. Janice, University of Guelph

Hmkle. Beth, University of Connecticut Health Center

Hodges. Craig. Case Western Reserve University

Jaquette, Julie, University of Iowa

Malik. Nusrat, Baylor College of Medicine

Miller, Michelle, Oregon Health Sciences University

Payne, Christopher, Oregon Regional Primate Research Center

Runft. Linda, University of Connecticut Health Center

Saunders. Thomas, University of Michigan

Takahashi, Diana, Oregon Regional Primate Research Center

Voronina. Ekaterina. Brown University

Week, Jennifer, Northwestern University

Course Coordinators

Burnett. Tim, University of Kansas Medical Center
Marin Bivens, Carrie. University of Massachusetts
McMullen, Michelle, Northwestern University
Petroff, Margaret, University of Kansas Medical Center
Simerly, Calvin, Oregon Regional Primate Research Center

Students

Alberio, Ramiro. Ludwig-Maximilian University, Germany

Allegrucci, Cinzia, Perugia University, Italy

Ashkar, Ali. University of Guelph

Berkowitz. Karen. University of Pennsylvania

Chong. Kowit-Yu, Oregon Regional Primate Research Center

Diaz, Lorenza. INNSZ

Graham, Kathryn. Oregon Health Sciences University

Greenlee. Anne. Marshneld Medical Research Foundation

Heifetz, Yael, Cornell University

Keller. Dominique. Texas A&M University

Lavoie. Holly, University of South Carolina

Majumdar, Subeer, National Institute of Immunology

Powell, Jacqueline, Morehouse School of Medicine

Richard. Craig, Magee-Wornen's Research Institute

Sahgal. Namita, Kansas University Medical Center

Zhang, Gongqiao, University of Virginia

R26 Annual Report

Fundamental Issues in Vision Research

(August 13-25, 2000)

Directors

Masur, Sandra K.. Mount Sinai School of Medicine
Papermaster, David, University of Connecticut Health Center

Faculty

Barlow. Robert, Syracuse University

Barres, Ben A., Stanford University School of Medicine

Beebe, David C.. Washington University School of Medicine

Berson, Eliot L., Harvard Medical School

Bok, Dean, University of California, Los Angeles

Dickersin, Kay, Brown University

Dowling, John E., Harvard University

Fisher, Richard, National Institutes of Health

Gordon, Marion. Rutgers College of Pharmacy

Hamm, Heidi E., Northwestern University Medical School

Horton. Jonathan. University of California

Horwitz. Joseph. University of California, Los Angeles

Lang, Richard A.. New York University School of Medicine

LaVail, Jennifer, University of California. San Francisco

Lavker. Robert. University of Pennsylvania

Lehrer. Robert, University of California, Los Angeles

Leske, M. Cristina, State University of New York. Stony Brook

Liberman. Ellen. National Institutes of Health

Malchow, Robert. University of Illinois. Chicago

Masland. Richard, Massachusetts General Hospital

Nathans, Jeremy. Johns Hopkins University School of Medicine

Niederkom, Jerry Y., University of Texas Southwestern Medical Center

Overbeek, Paul A., Baylor College of Medicine

Piatigorsky, Joram, National Institutes of Health

Raviola, Elio. Harvard Medical School

Shatz. Carla, Harvard Medical School

Stambolian, Dwight. University of Pennsylvania

Sugrue. Stephen P., University of Florida College of Medicine

Wasson, Paul. Harvard Medical School

Lecturers

Assad. John, Harvard Medical School

Hernandez, M. Rosario. Washington University School of Medicine

Moses, Marsha, Children's Hospital, Boston

Russell, Paul, National Institutes of Health

Students

Al-Khatib, Khaldun, University of Illinois, Chicago

Bernstein, Audrey. Mount Sinai Medical School

Birnbaum. Andrea, University of Illinois, Chicago

Camelo. Serge, Institut Pasteur

Cronin. Carolyn, University of Virginia

Gaudio. Paul, Yale University

Goh, Meilan Stephanie, University of Illinois, Chicago

Hartford, April, University of Louisville

Jessani. Nadim, Scripps Research Institute

Jiang. Shunai. Emory University

Kenyon, Kristy. Massachusetts Eye and Ear Infirmary

Libby, Richard, Medical Research Council, United Kingdom

Liu, Xiaorong. University of Virginia

Mahajan, Vinit, University of California. Irvine

Pennesi, Mark, Baylor College of Medicine

Pittman, Kristi, North Carolina State University

Rose, Linda, University of Maryland

Ruttan, Gregory. University of Miami, Florida

Sagdullaev. Botir, University of Louisville

Shestopalov, Valery, Washington University

Medical Informatics (May 28-June 3, 2000)

Director

Masys, Daniel, University of California, San Diego

Faculty

Canese, Kathi. National Library of Medicine

Cimino, James, Columbia University

Friedman, Charles, University of Pittsburgh

Giuse. Nunzia, Vanderbilt University Medical Center

Hightower, Allen, Centers for Disease Control and Prevention

Kingsland, Lawrence, National Library of Medicine

Lindberg, Donald, National Library of Medicine

McDonald, Clement. Regenstrief Institute

Miller. Randolph. Vanderbilt University Medical Center

Nahin. Annette. National Library of Medicine

Ozbolt, Judy, Vanderbilt University Medical Center

Stead. William. Vanderbilt University Medical Center

Wheeler. David. National Library of Medicine

Students

Athreya, Balu, DuPont Hospital for Children
Barnes, Judith, Ingham Regional Medical Center

Educational Programs R27

Belts. Eugene. Medical College of Georgia

Blalt. Jod\. Health Care Financing Admiimiiation

Brill, Peter. Trover Foundation

Brown. Janis. University of Southern California

Clintworth. William. University of Southern California

Cohn. Wendy, University of Virginia

Cowper, Diane. Hines VA Hospital

Cooper. Natasha. Penn State College of Medicine

Desai. Sundeep. Northwestern Medical Faculty Foundation

Ebbeling. Kelly. University of Wisconsin. Madison

Fulda. Pauline. Louisiana State University

Halsted, Deborah. Houston Academy of Medicine

Harris. Anthony. University of Maryland

Levine. Alan. University of Texas. Houston

Jenson, James, University of New Mexico

Klingen, Donald, Riverside Regional Medical Center

Kubal. Joseph. VA Information Resource Center

Mcknight. Michelynn. Norman Regional Hospital

Obijiofor, Chioma, Bioresources Development and Conservation

Program

Schwartz. Marilyn, Naval Medical Center, San Diego
Smith. John. University of Alabama. Birmingham
Sooho. Alan. Battle Creek Veterans Administration
Stocking. John, University of Louisville
Strachan. Dina. King/Drew Medical Center
Thibodeau. Patricia, Duke University
Vaidya. Vinay, University of Maryland
Woeltje, Keith. Medical College of Georgia
Yamamoto, David. University of California, Los Angeles
Zick. Laura, Clarian Health

Medical Informatics (October 1-7, 2000)

Director

Cimino. James. Columbia University

Faculty

Bakken. Suzanne. Columbia University
Cimino, Chris, Albert Einstein College of Medicine
Friedman. Charles. University of Pittsburgh
Jenders, Robert. Columbia University
Kingsland, Lawrence, National Library of Medicine
Lindberg. Donald. National Library of Medicine
Masys, Daniel. University of California. San Diego
McCray, Alexa, National Library of Medicine
Nahin, Annette. National Library of Medicine
Perednia, Douglas, Association of Telehealth Providers
Starren. Justin. Columbia University
Wheeler. David. National Library of Medicine

Students

Amend. Clifford. Care First Blue Cross Blue Shield

Babu. Ajit. St. Louis VA Medical Center

Baer. Michael. Lebanon Veterans Admin. Medical Center

Barclay. Allan, Indiana University School of Medicine

Burke, Cynthia. Hampton University

Byrd, Vetria, University of Alabama, Birmingham

Dam. Steven, University of Western Ontario

Davis. Wayne. University of Michigan Medical School

DiPiro. Joseph. University of Georgia

Fernandes, John, Chicago College Osteopathic Medicine

Frank. Christine. Rush-Presbyterian-St. Luke's Medical Center

Gallardo, Gladys, Universidad Central del Caribe
Gamble, James. Maniilaq Health Center
Gill, Jagjit. Mayo Clinic and Foundation
Goodwin, Cheryl. Swedish Medical Center
Guarcello. Catherine. St. Elizabeth's Medical Center
Jones, Dixie. LSU Health Science Center
Kelly, Catherine. Massachusetts General Hospital
Mackowiak. Leslie. Duke University Health System
McKoy. Karen. Lahey Clinic

Moser, Stephen. University of Alabama. Birmingham
Murray. Kathleen. University of Alaska Anchorage
Pepper, David, University Medical Center
Riesenberg, Lee, Ann Guthrie Healthcare System
Sathe. Nila. Vanderbilt University Medical Center
Sullivan, Eileen, University of New Mexico
Taylor. Vera, Morehouse School of Medicine
Wellik. Kay, Mayo Clinic Scottsdale

Wiedermann. Bernhard. Children's National Medical Center,
Washington

Methods in Computational Neuroscience
(July 30-August 26, 2000)

Directors

Bialek, William. NEC Research Institute
de Ruyter, Rob. NEC Research Institute

Faculty

Abbott, Lawrence, Brandeis University

Colby, Carol, University of Pittsburgh

Collett, Thomas. University of Sussex

Dan, Yang, University of California. Berkeley

Delaney. Kerry, Simon Fraser University. Canada

Doupe, Allison, University of California, San Francisco

Ermentrout. Bard. University of Pittsburgh

Ferster. David. Northwestern University

Gelperin. Alan. Bell Laboratories

Hopfield. John, Princeton University

Johnston. Daniel. Baylor College of Medicine

Kelley. Darcy. Columbia University

Kleinfeld. David, University of California. San Diego

Kopell. Nancy. Boston University

Marder, Eve. Brandeis University

Markram, H., University of California

Miller. K. D.. University of California. San Francisco

R28 Annual Report

Mitra. Partha, AT&T Bell Laboratories

Nemenman. Ilya, NEC Research Institute

Rieke. Fred, University of Washington

Seung. H. Sebastian, Massachusetts Institute of Technology

Sigvardt, Karen. University of California. Davis

Solla, Sara A., Northwestern University Medical School

Sompolinsky, Maim, The Hebrew University. Israel

Tank. David. AT&T Bell Laboratories

Tishby, Naftali. The Hebrew University, Israel

Tsodyks, Michail. Weizmann Institute of Science

Zucker, Steven, Yale University

Lab Instructor

Jensen, Roderick, Wesleyan University

Microinjection Techniques in Cell Biology

(May 16-23, 2000)

Director

Silver, Robert. Marine Biological Laboratory

Faculty

Klaessig. Suzanne, Cornell University
Kline, Douglas, Kent State University
Shelden. Eric. University of Michigan
Wilson, Susan, Cornell University

Teaching Assistant

Miller, Roy Andrew. Kent State University

Lecturers

Baylor, Denis, Stanford University Medical Center

Berry. Michael, Princeton University

Koberle. Roland. Universidade de Sao Paulo. Brasil

Laughlin. Simon Barry. Cambridge University. United Kingdom

Logothetis. Nikos, Max-Planck-Institute for Biological Cybernetics

Srinivasan, Mandyam V., Australian National University. Australia

Teaching Assistants

Aguera y Areas, B., Princeton University
Lewen, Geoffrey David. NEC Research Institute
White, John, Boston University

Course Assistants
Jensen, Kate. Marine Biological Laboratory
Purpura. Keith, Marine Biological Laboratory

Students

Cabot, Ryan. University of Missouri

Caswell. Wayne, Lahey Clinic

Combelles. Catherine. Tufts University

Davies. Daryl, University of Southern California

Dong. Lily. UT Health Science Center, San Antonio

Geraci. Fabiana, University of Palermo

Gilbert, Joanna, Harvard Medical School

Gundersen-Rindal. Dawn, U.S. Department of Agriculture

Harwood, Claire, University of Pennsylvania

Hawash. Ibrahim. Purdue University

Howe. Charles, Stanford University

Kay, EunDuck, Doheny Eye Institute

Kline-Smith, Susan. Indiana University

Macdonald, Jennifer, Medical University of South Carolina

Nguyen. Hong-Ngan, University of Louisiana of Lafayette

Okusu. Akiko, Harvard University

von Dassow, Peter, Scripps Institute of Oceanography

Webb. Bradley. Queen's University

Widelitz. Randall. University of Southern California

Yang. Jin, Duke University, HHM1

Students

Aharonov-Barki, Ranit, Hebrew University

Bartlett, Edward, University of Wisconsin. Madison

Bodekin. Clara, Boston University

Boudreau. Christen (Beth), Baylor College of Medicine

Feinerman, Ofer, Wiezmann Institute of Science

Felsen, Gidon. University of California. Berkeley

Globerson. Amir. Hebrew University

Giitig. Robert. University of Freiburg

Jin, Dezhe, University of California, San Diego

Kang. Kukjin, Hebrew University

Krishna, B. Suresh, New York University

Lauritzen, Thomas. University of California. San Francisco

Parthasarathy. Hemai. Nature America

Paz. Ron. Hebrew University

Petereit. Christian, Universitat Bielefeld

Pierce, John. Vibration & Sound Sol. Ltd.

Rokni. Uri, Hebrew University

Schreiber, Susanne. Humboldt Universitat Berlin

Shi, Songhai, Cold Spring Harbor Laboratory

Sirota. Anton, Rutgers University

Szalisznyo, Krisztina, Hungarian Academy of Science

Taylor. Dawn. Arizona State Lmiversity

Ulanovsky. Nachum, Hebrew University

Werfel. Justin. Massachusetts Institute of Technology

Modeling of Biological Systems
(March 25-May 4, 2000)

Director

Silver, Robert. Marine Biological Laboratory

Faculty

Boston. Raymond C.. University of Pennsylvania

Cheatham. Thomas E.. University of Utah

Herzfeld. Judith, Brandeis University

Hummel. John. Argonne National Laboratory

Kollman. Peter. University of California. San Francisco

Moate. Peter. University of Pennsylvania

Pearson. John, Los Alamos National Laboratory

Petsko, Greg A., Brandeis University

Ponce Dawson. Silvina. Ciudad Universitaria. Argentina

Students

Genick. Ulrich. The Salk Institute

Ginsberg. Tara, University of Texas, Houston

Hershberg. Uri. Hebrew University

Immerstrand. Charlotte. Linkoping University. Sweden

Jiang, Yi. Los Alamos National Laboratory

Educational Programs R29

Mosavi, Leila, University of Connecticut Health Center
Quinteiro, Guillermo. University of Buenos Aires
Teng. Ching-Ling. University of Virginia
Uppal. Hirdesh. Punjab Veterinary Vaccine Institute, India

Molecular Biology of Aging (August 12-18, 2000)

Directors

Guarente. Leonard P., Massachusetts Institute of Technology
Wallace. Douglas, Emory University School of Medicine

Faculty

Austad. Steven, University of Idaho

Beal, M. Flint, Cornell University

Bohr. Vilhelm A., National Institutes of Health

Campisi. Judith. Lawrence Berkeley National Laboratory

Culotta. Valeria L., Johns Hopkins University

de Lange. Titia, The Rockefeller University

Hanawalt, Philip. Stanford University

Johnson, Thomas. University of Colorado

Jones. Dean P., Emory University

Kenyon. Cynthia. University of California, San Francisco

Kim, Stuart. Stanford, University School of Medicine

Lithgow. Gordon J., University of Manchester

Martin, George, University of Washington School of Medicine

McChesney. Patricia, University of Texas Southwestern

Medical Center

Price, Donald L., Johns Hopkins University School of Medicine
Richardson, Arlan, University of Texas Health Science Center,

San Antonio

Ruvkun. Gary, Massachusetts General Hospital
Tanzi. Rudolph E , Harvard Medical School
Tower. John, University of Southern California
Van Voorhies, Wayne, University of Arizona, Tucson
Wright. Woodnng E., University of Texas Southwestern

Medical Center

Lecturers

Finch. Celeb. LIniversity of Southern California

Hekimi. Siegfried, McGill University

Wemdruch, Richard H., Veterans Administration Hospital

Teaching Assistants

Coskun. Elif Pinar, Emory University School of Medicine
Ford, Ethan, Massachusetts Institute of Technology
Kerstann, Keith, Emory University School of Medicine
Kokoszka, Jason, Emory University
Levy, Shawn, Vanderbilt-Ingram Cancer Center
Marcimak. Robert, Massachusetts Institute of Technology
McVey, Mitch, Massachusetts Institute of Technology
Murdock, Deborah, Emory University

Course Coordinator

Burke. Rhonda E., Emory University School of Medicine

Course Assistant
Ament. Nell, Marine Biological Laboratory

Students

Bailey, Adina, University of California, Berkeley
Baur. Joe. UT Southwestern Medical Center, Dallas

Bordone. Laura, University of Minnesota

Cui. Wei, Roslin Institute, Edinburgh

Cypser. James, University of Colorado

Filosa. Stefania, 1IGB-CNR

Furfaro, Joyce. Pennsylvania State University

Harper. James. University of Idaho

Huang. Xudong. Massachusetts General Hospital

Johnson. Kristen, Purdue University

Konigsberg, Mina, Universidad Autonoma Metropolitana

Kostrominova, Tatiana, University of Michigan

Luo, Yuan, University of Southern Mississippi

Munoz, Denise, University of Buenos Aires/UC Berkeley

Peel, Alyson, The Buck Center for Research in Aging

Podlutsky, Andrej, National Institute on Aging

Radulescu, Andreea. Albert Einstein College of Medicine

Srivivsan, Chandra, University of California, Los Angeles

Tong, Jiayuan (James), Cold Spring Harbor

Zaid, Ahmed, Stockholm University

Molecular Mycology: Current Approaches to
Fungal Pathogenesis (August 7-25. 2000)

Directors

Edwards, John, Jr., Harbor-UCLA Medical Center
Magee. Paul T., University of Minnesota
Mitchell, Aaron P.. Columbia University

Faculty

Filler, Scott, Harbor-UCLA Medical Center
Heitman, Joseph. Duke University Medical Center
Rhodes, Judith, University of Cincinnati Medical Center
White, Theodore. Seattle Biomedical Research Institute

Lecturers

Cushion. Melanie, University of Cincinnati

Doering, Tamara. Washington University School of Medicine

Fink, Gerald, Whitehead Institute

Kozel, Thomas, University of Nevada School of Medicine

Kwon-Chung, June. National Institutes of Health

Levitz, Stuart. Boston University

Magee. Beatrice. University of Minnesota

Puziss, John, Proteome, Inc.

Quinn, Cheryl, Pharmacia & Upjohn

Scherer, Stewart, Rosetta Inpharmatics

Teaching Assistants
Flenniken, Michelle, Montana State University
Johnston, Douglas, Harbor-UCLA Medical Center
Lengeler, Klaus B., Duke University Medical Center

Course Assistant
Martin, Sam, Marine Biological Laboratory

Students

Askew, David. University of Cincinnati
Austin, W. Lena, Howard University
Blankenship. Jill. Duke University
Burr, Ian, Pfizer Central Research
Francis, Susan. University of Washington
Hochstenbach. Frans, University of Amsterdam
Ibrahim, Ashraf, Harbor-UCLA Medical Center

R30 Annual Report

Lo, Hsiu-Jung, National Health Research Institutes
Mol, Pietemella, University of Amsterdam
Munro, Carol, University of Aberdeen
Perea, Sofia. University of Texas
Spellberg. Brad, Harbor-UCLA Medical Center
Spreghini, Elisabetta. Yale University
Toenjes, Kurt. University of Vermont
Wasylnka. Julie, Simon Fraser University

Neural Development and Genetics of Zebrafish
(August 13-26, 2000)

Directors

Dowling, John E., Harvard University

Hopkins, Nancy, Massachusetts Institute of Technology

Faculty

Chien, Chi-Bin. University of Utah Medical Center

Collazo. Andres, House Ear Institute

Eisen, Judith S., University of Oregon

Fetcho, Joseph, State University of New York, Stony Brook

Hanlon, Roger, Marine Biological Laboratory

Houart, Corrine. University College London, United Kingdom

Kimmel. Charles. University of Oregon

Lin. Shuo. Medical College of Georgia

Neuhauss, Stephan, Max-Planck-Institut fur Entwicklungsbiologie,

Germany

Talbot, William S., Stanford University School of Medicine
Wilson, Stephen. University College London, United Kingdom

Lecturers

Astrofsky, Keith, Massachusetts Institute of Technology
Fraser, Scott, California Institute of Technology

Teaching Assistants

Amacher, Sharon, University of California. Berkeley

Clarke. Jon, University College London. United Kingdom

Fadool, James, Florida State University

Granato. Michael. University of Pennsylvania

Lyons, David. University College London

Mazanec, April, University of Oregon

Mullins, Mary. University of Pennsylvania

Perkins, Brian. Harvard University

Pomrehn, Andrea, Stanford University

Wagner. Daniel, University of Pennsylvania Medical School

Walker-Durchanck, Charline. University of Oregon
Waterbury. Julie. University of Pennsylvania

Course Coordinator

Schmitt. Ellen. Harvard University

Facility Technician
Linnon. Beth. Marine Biological Laboratory

Course Assistant
Bradley, Margaret, Marine Biological Laboratory

Students

Challa, Anil Kumar, Ohio State University
Croall, Dorothy, University of Maine
Darimont, Beatrice, University of Oregon
Kaneko, Maki, University of Houston
Leung, Fung Ping, Hong Kong University
Levkowitz, Gil. Weizmann Institute of Science
Lupo, Giuseppe. University of Pisa

Maldonado. Ernesto. Massachusetts Institute of Technology
Mangoli. Maryam. University College London. United Kingdom
Meyer. Martin. Stanford University
Naco, Grace. Johns Hopkins School of Medicine
Nelson. Ralph. National Institutes of Health
Niell. Cristopher, Stanford University
Schneider, Valerie, Harvard Medical School
Starr, Catherine, The Rockefeller University
Yvon, Anne-Marie, University of Massachusetts, Amherst

Neurobiology & Development of the Leech
(August 13-September 1, 2000)

Directors

Calabrese. Ronald L.. Emory University
Sahley, Christine. Purdue University
Shankland, Martin, University of Texas, Austin

Faculty

Ali. Declan. Hospital for Sick Children

Baader. Andreas. Universitat Bern, Switzerland

Bissen, Shirley. University of Missouri

Blackshaw, Susanna. University of Oxford. United Kingdom

Brodfuehrer. Peter, Bryn Mawr College

Carbonetto, Salvatore, Montreal General Hospital, Canada

Drapeau, Pierre, McGill University, Canada

Fernandez de Miguel, Francisco, Universidad Nacional Autonoma

de Mexico

Masino, Mark. Emory University
Modney. Barbara, Cleveland State University
Muller. Kenneth. University of Miami School of Medicine
Nicholls. John. SISSA. Italy
Weisblat. David. University of California, Berkeley

Lecturer

Macagno, Eduardo. Columbia LIniversity

Course Assistant

Johnson, Ben, Marine Biological Laboratory

Educational Programs R31

Students

Carrasco. Rosa. Purdue University

Duan. Yuanli, University of Miami

Kuo. Dian-Hun. University of Texas. Austin

Kwon. Hyung-wook, University of Arizona

Rela. Lorena. University of Buenos Aires

Scimemi. Annalisa. SISSA. Italy

Song. Mi Hye. University of California. Berkeley

Trueta. Citlali. UNAM

Weber, Douglas. Arizona State University

West. Morris, University of Florida

Yashina, Irene. University of Illinois at Chicago

Zoccolan, Davide, SISSA, Italy

Optical Microscopy and Imaging in the
Biomedical Sciences (October 11-19, 2000)

Director

Izzard, Colin, State University of New York. Albany

Faculty

DePasquale, Joseph. New York State Department of Health
Hard. Robert. State University of New York. Buffalo
Inoue. Shinya. Marine Biological Laboratory
Maxfield. Frederick. Cornell University Medical College
Murray. John. University of Pennsylvania School of Medicine
Piston. David M., Vanderbilt University
Spring. Kenneth. National Institutes of Health
Swedlow, Jason. University of Dundee. UK

Lecturers

Hinsch. Jan, Leica, Inc.

Keller. H. Ernst. Zeiss Optical Systems

Oldenbourg. Rudolf. Marine Biological Laboratory

Teaching Assistant

Sigurdson. Wade. State University of New York. Buffalo

Course Associate
Snyder. Kenneth. State University of New York. Buffalo

Course Assistant

Pierini. Lynda. Weill Medical College of Cornell University

Students

Arudchandran, Ramachandran. National Institutes of Health

Christensen. Trace. Mayo Clinic

Diez. Stefan. Max-Planck-Institute

Dobrun/. Lynn. University of Alabama. Birmingham

Flett, Alexander, University of Dundee

Furie. Bruce. Harvard Medical School

Garcia-Mata. Rafael. University of Alabama. Birmingham

Caspar. Claudia. Montreal General Hospital

Goldsworthy. Michael. Memorial University of Newfoundland

Gross, Peter. Beth Israel Deaconess Medical Center

Hagting. Anja. Wellcome/CRC Institute

Holtom. Gary. Pacific Northwest National Laboratory

Hu, Ke, University of Pennsylvania

Islam. Mohammad. University of Pennsylvania

Karlsson, Christina. Karolinska Institute

Linser, Paul, Whitney Lab, University of Florida

Martinez, Angle, Harvard Medical School

Martins. Gabriel, State University of New York. Buffalo

North. Alison, Rockefeller University

Ono. Yasuko, University of Arizona

Praetorius, Jeppe, National Institutes of Health

Qiao. Jizeng. Massachusetts General Hospital

Rice. Marian. Mount Holyoke College

Schmidtke, David. University of Oklahoma

Rapid Electrochemical Measurements
(May 11-15, 2000)

Director

Gerhardt. Greg, University of Kentucky

Faculty

Cass. Wayne, University of Kentucky

Currier, Theresa, University of Kentucky

Gratton. Alain. McGill University

Hoffman. Alex, National Institutes of Health

Huettl, Peter, University of Kentucky

Palmer. Michael. University of Colorado Health Science Center

Porterfield, David, University of Missouri-Rolla

Purdom, Matt. University of Kentucky

Stanford. John. University of Kentucky

Sulzer, David, Columbia University

Surgener. Stewart, University of Kentucky

Teaching Assistants
Burmeister. Jason. University of Kentucky
Pomerleau. Francois. McGill University

Course Coordinator

Lindsay. Robin. University of Kentucky

Students

Ahmad. Laura. Eli Lilly & Company

Bruno. John, The Ohio State University

Byrd. Kenneth, Indiana University School of Medicine

Cho. Sunyoung, Kyunghee University, Korea

Espey, Michael. National Institutes of Health

Fadel. Jim. Vanderbilt University

Grinevich, Vladimir, University of Kentucky

Hull, Elaine, State University of New York, Buffalo

Jackson, Mark, Yale University

Jow, Brian, Wyeth-Ayerst Research

Judy, Jack, University of California, Los Angeles

Kusano. Kiyoshi. National Institutes of Health

Lan. Esther. Lmiversity of California. Los Angeles

Lee, Irwin, Harvard Medical School

Maidment. Nigel, L'niversity of California, Los Angeles

Montanez. Sylvia, University of Texas Health Science Center

Olazabal, Daniel. Rutgers University

Perry, Kenneth. Lilly Research Labs

Phillips. Janice. University of St. Andrews

Reid. Stephen, University of Saskatchewan

Salvatore, Michael. Louisiana State University Health Sciences Center

Siapas. Athanassios. Massachusetts Institute of Technology

Walker, Eric. University of California, Los Angeles

Wilbrecht, Linda, Rockefeller University

R32 Annual Report

Workshop on Molecular Evolution

July 30-Aiigust 11, 2000

Director

Cummings, Michael, Marine Biological Laboratory

Faculty

Beerli. Peter. University of Washington

Edwards. Scolt, University of Washington

Eisen, Jonathan. Institute for Genomic Research

Felsenstein. Joseph, University of Washington

Fraser, Claire M., Institute for Genomic Research

Huelsenbeck. John P., University of Rochester

Kuhner, Mary. University of Washington

Lewis, Paul O., University of Connecticut

Maddison. Wayne P.. University of Arizona

Meyer, Axel, University of Konstanz, Germany

Patel. Nipam, University of Chicago

Pearson. William. University of Virginia Health Sciences Center

Rand. David. Brown University

Rice, Ken, Bioinformatics

Riley, Margaret A.. Yale University

Swofford, David. Smithsonian Institution

Thompson. Steven. Biolnfo 4U

Voytas, Daniel F.. Iowa State University

Yokoyama, Shozo, Syracuse University

Lecturer

Yoder. Anne D.. Northwestern University Medical School

Teaching Assistants

Amaral-Zettler, Linda. Marine Biological Laboratory
Babin, Josephine. Louisiana State University
Church. Sheri A., University of Virginia
Dennis. Paige M.. University of Massachusetts
FrantzDale. Ben

McArthur, Andrew. Marine Biological Laboratory
Medina. Monica. Marine Biological Laboratory
Myers, Daniel. Marine Biological Laboratory
Pritham, Ellen, University of Massachusetts
Reed, David, Louisiana State University
Waring. Molly E., Marine Biological Laboratory

Students

Allender. Charlotte, Southampton University

Ardell, David. Uppsala University

Barbour. Jason, University of California. San Francisco

Baric. Sanja, University of Innsbruck

Bedard, Donna. Rensselaer Polytechnic Institute

Birungi. Josephine. Yale School of Medicine

Borenstein, Seth. Knight Ridder Newspapers

Boykin. Laura. University of New Mexico

Calcagnotto. Daniela. America Museum of Natural History

Cipriano, Frank. San Francisco State University

Drozdowicz. Yolanda, University of Pennsylvania

Eick. Brigitte. University of Cape Town

Erpenbeck. Dirk. University of Amsterdam

Ganter. Philip, Tennessee State University

Garcia Saez, Alberto, Alfred Wegener Institute

Garcia, Martin, UNAM

Gurgel, Carlos, University of Louisiana, Lafayette

Handley, Scott. Centers for Disease Control and Prevention

Hanel, Reinhold. University of Innsbruck

Held, Christoph, Unive.-sitat Bielefeld

Holland. Brenden. University of Hawaii

Johns, Susan, University of California. San Francisco

Joseph, Leo, Academy of Natural Sciences

Kalia, Awdhesh, Yale University

Kim, Hyigyung. Smithsonian Institution

Kulathinal, Rob, McMaster University

Liu, Ji, University of Georgia

Longnecker. Krista, Oregon State University

Lundholm, Nina, University of Copenhagen

Mark Welch. David. Harvard University

McLaughlin. Ian. PE Biosystems

McMahon. Kathenne. University of California. Berkeley

Moncayo. Abelardo, University of Texas

Munroe. Stephen. Marquette University

Nepokroeff, Molly, Smithsonian Institution

Newman, Lucy. University of Maryland

Nilsen, Frank, Institute of Marine Research

O'Connor. Daniel. L'niversity of California, San Diego

Olson, Julie. Harbor Branch Oceanographic Institution

Pannacciulli. Federica, University of Genoa

Pellegrino. Katia. Brigham Young University

Perez, Ernesto, Universite Libre de Bruxelles

Perez-Losada. Marcos. Brigham Young University

Phillips. Louise. University of Melbourne

Regnery, Russell, Centers for Disease Control & Prevention

Rhoads. Allen. Howard University

Richardson. Paul. Joint Genome Institute

Rokas. Antonis. University of Edinburgh

Salzburger, Walter. University of Innsbruck

Sankale, Jean-Louis. Harvard School of Public Health

Stone, Karen, University of Alaska, Fairbanks

Tiffin. Peter. University of California. Irvine

Utiger, Urs, Zoologisches Museum Zurich

Vasiliou, Vasilis. University of Colorado Health Sciences Center

Vincent. Martin. Centers for Disease Control and Prevention

Watson. Linda. Miami University

Westneat. Mark, Field Museum of Natural History

Wilgenbusch, James. Smithsonian Institution

Wilmotte. Annick, University of Liege

Won. Yong-Jin. Rutgers University

Xie. Gang (Gary). Los Alamos National Laboratory

Other Programs

Marine Models in Biological Research

Undergraduate Program

(June 13-Augmt 11, 2001)

Directors

Browne, Carole L., Wake Forest University

Tytell. Michael. Wake Forest University School of Medicine

Facility'

Allen. Nina S.. North Carolina State University

Browne. Carole. Wake Forest University

Furie. Barbara, Harvard School of Medicine

Furie, Bruce. Harvard School of Medicine

Gould. Robert, New York State Institute for Basic Research

Hanlon, Roger. Marine Biological Laboratory

Educational Programs R33

Malchow, R. Paul. University of Illinois. Chicago

Mensinger. Allen. University of Minnesota. Duluth

Palazzo. Robert, University of Kansas

Rome. Lawrence, University of Pennsylvania

Tytell. Michael, Wake Forest University School of Medicine

Wainwnght. Norman. Marine Biological Laboratory

Sciniinir Speakers

Augustine, George. Duke University Medical Center
Ehrlich. Barbara. Yale University School of Medicine
Gallant, Paul. National Institutes of Health
Hill. Susan, Michigan State University
Oldenbourg, Rudolf, Marine Biological Laboratory
Sloboda. Roger. Dartmouth College

Students

Fornwalt. Brandon. University of South Carolina
Gilles. Nicole. University of Minnesota
Gupton. Stephanie. North Carolina State University
Hembree. Chad. Wake Forest University
Kingston, Margaret. Wake Forest University
Lee. Tony. Duke University
Levin. Tracy. Smith College
Mangiamele. Lisa. Colgate University
Rosenkranz. Naomi. Yeshiva University
Szucsik, Amanda. Rutgers University
Zerbe. Jamie, University of Kansas

NASA Planetary Biology Internship
(June-September 2000)

Directors

Dolan. Michael F., University of Massachusetts
Margulis. Lynn. University of Massachusetts

Interns

Amponsah-Manager. Kwabena. University of Ghana

Clarkson. William. Oxford University

Delaye, Luis, National Autonomous University of Mexico

Finarelli. John, University of New Hampshire

Lamb. David. University of North Dakota

Lawson. Jennifer. University of Illinois. Chicago

Lloret y Sanchez. Lourdes. National Autonomous University of

Mexico

Mikuki. Jill A.. Portland State University
O'Donnell. Vicki. National University of Ireland. Maynooth
Richards. Thomas. Southampton University

Sponsors

Arrhemus. Gustaf. Scripps Institution of Oceanography

Cady. Sherry. Portland State University

Des Marais, David, NASA Ames Research Center

Gogarten. Peter. University of Connecticut

Hinkle. C. Ross, Kennedy Space Center

Nierzwicki-Bauer, Sandra. RPI

Pohorille, Andrew, NASA Ames Research Center

Priscu, John. Montana State University

Roberts, Michael S.. Kennedy Space Center

Rothschild. Lynn, NASA Ames Research Center

Semester in Environmental Science
(September 4-December 15, 2000)

Administration

Hobbie. John E.. Director

Foreman. Kenneth H.. Associate Director

Moniz, Polly C.. Administrative Assistant

Faculty

Deegan. Linda A.
Foreman, Kenneth H.
Giblin. Anne E.
Hobbie. John E.
Hopkinson, Charles S.. Jr.
Hughes. Jeffrey
Melillo. Jerry M.
Nadelhoffer. Knute J.
Neill. Christopher
Peterson, Brace J.
Rastetter. Edward B.
Shaver, Gaius R.
Vallino, Joseph J.
Williams. Mathew

Research and Teaching Assistants
Eldridge. Cynthia
Gay, Marcus
Kwiatkowski. Bonnie
Micks. Patricia
Morrisseau. Sarah
Tholke. Kns

SES Students

Angeloni. Catherine A.. Wheaton College
Bandstra. Leah M.. Beloit College
Businski. Tara N.. Bates College
Chiarelli. Robyn N.. Brandeis University
Creswell. Joel E., Macalester College
Dalsimer, Heather S.. Dickinson College
Hayes, Alison B.. Lawrence University
Johnson. Rebecca T, Oberlin College
Karasack. Rebecca D.. Dickinson College
Krumholz, Jason S.. Lawrence University

R34 Annual Report

Lawrence. Corey R., Clarkson University
Schwartz, Jessica C., Connecticut College
Shayler, Hannah A., Connecticut College
Taylor. Catherine A.. Brandeis University
Teeters, Kelsa E., Brandeis University

SPINES — Summer Program in Neuroscience,
Ethics and Survival (June I0-Juty 8, 2000)

Directors

Martinez, Joe, University of Texas, San Antonio
Townsel, James G.. Meharry Medical College

Faculty

Augustine. George, Duke University
Berger-Sweeney, Joanne E.. Wellesley College
Escalona de Motta. Gladys. University of Puerto Rico
Etgen, Anne, Albert Einstein College of Medicine
Fox, Thomas O., Harvard University Medical School
Gonzalez-Lima, Francisco, University of Texas
Maynard, Kenneth I., Massachusetts General Hospital
Zukin, R. Suzanne. Albert Einstein College

Villareal. Greg. University of California, Los Angeles
Whittle. Chris. University of Alaska. Fairbanks

Teachers' Workshop: Living in the Microbial
World (August 13-19, 2000)

Directors

Dugas, Jeff, University of Connecticut, Storrs
Olendzenski, Lorraine, University of Connecticut, Storrs

Faculty

Dorritie, Barbara. Cambridge Rindge and Latin School, Cambridge.

MA
Wier. Andrew. University of Massachusetts, Amherst

Presenters

Amils. Ricardo, Autonomous University of Madrid. Spain
Edgcomb. Virginia. Marine Biological Laboratory
Margulis. Lynn. University of Massachusetts. Amherst
Stolz. John. Duquesne University
Wainwright. Norm. Marine Biological Laboratory

Lecturers

Kravitz. Edward, Harvard Medical School
Wyche, James. Brown University

Teaching Assistant

Hohmann, Christine. Morgan State University

Course Coordinator

Garcia. Elizabeth. University of Texas, San Antonio

Students

Boomer, Akilah. Johns Hopkins University
Colon, Wanda, University of Puerto Rico
Davis. Kamisha, University of Utah
Kamendi. Harriet. Howard University
Lorge. Greta, University of Michigan
Mercado, Eduardo, Rutgers University
Reyes. Rosario, University of Oregon
Rodriguez. Gustavo. Purdue University
Vidal. Luis. University of Puerto Rico

Teacher Participants

Barker. Jean. Pleasant Lea Junior High School. Lee's Summit. MO
Brothers. Chris. Falmouth High School. MA
Campbell. LeeAnne. Mashpee High School. MA
Demetriou, Christina, Astor School. Dover, United Kingdom
Dugan. Maureen. Nashoba Regional High School. Bolton. MA
Ebberly. Stuart, Astor School, Dover, United Kingdom
Estabrooks, Gordon, Boston Latin School, MA
Fenske, Sue, Bernard J. Campbell Junior High School. Lee's Summit.

Jaye, Robert, Solomon Schecter Day School, MA
Johnson, Linda, Nauset Regional Middle School, Orleans. MA
Kamborian. Kimberly. Fenway High School. Boston. MA
Kuhn. Gale, Amherst Regional High School, MA
Pamco. Suzanne. Fenway High School. Boston. MA
Soracco, Marlene. Bourne High School. MA
Stupples, Eileen, Sir Roger Manwood School, Kent. United

Kingdom

Trask. Janet, Mashpee High School. MA
Trimarchi. Ruth. Amherst Regional High School. MA
Tuite. Deb. Nauset Regional Middle School. Orleans. MA
Veneman, Val, Amherst Regional High School, MA
Virchick. Garret. Fenway High School. Boston. MA
Watts. Ngaire, Sir Roaer Manwood School. Kent. L'nited Kingdom

Summer Research Programs

Principal Investigators

Antic. Srdjan. Yale University School of Medicine
Armstrong. Clay. University of Pennsylvania
Armstrong. Peter B.. University of California, Davis
Augustine. George J.. Duke University Medical Center

Baker. Robert. New York University Medical Center
Barlow, Robert B., Jr., State University of New York Health

Science Center
Beauge, Luis, Institute de Investigacion Medica "Mercedes y Martin

Ferreyra," Argentina

Belluscio, Leonardo, Duke University Medical Center
Ben-Jonathan. Nira. University of Cincinnati
Bennett, Michael V. L., Albert Einstein College of Medicine
Bodznick. David. Wesleyan University
Boron. Walter. Yale University Medical School
Boyer. Barbara. Union College
Boyle. Richard. Oregon Health Sciences University
Brady. Scott T.. The University of Texas Southwestern Medical

Center. Dallas

Brown. Joel. Albert Einstein College of Medicine
Browne. Carole. Wake Forest University School of Medicine
Bruzzone. Roberto. Institut Pasteur. France
Burger. Max M.. Friedrich Miescher Institut. Switzerland
Burgess. David. Boston College
Burgos. Mario. Universidad Nacional de Cuyo. Argentina

Changeux, Jean-Pierre. Institut Pasteur. France

Chappell. Richard L.. Hunter College, City University of New York

Chiao, Chuan-Chin. University of Maryland

Clay, John. National Institutes of Health

Cohen. Lawrence B.. Yale University School of Medicine

Cohen. William D.. Hunter College, City University of New York

De Weer. Paul. University of Pennsylvania School of Medicine

Devlin. C. Leah. Penn State University

DiPolo. Remaldo, Instituto Venezoiano Investigaciones Cientificas.

Venezula
Dodge. Frederick. State University of New York Upstate Medical

University

Edds-Walton. Peggy. University of California. Riverside
Ehrlich. Barbara, Yale University School of Medicine

Fadool. Debra Ann, Florida State University
Fay. Richard, Loyola University of Chicago

Field. Christine. Harvard University Medical School

Fishman. Harvey M., University of Texas Medical Branch, Galveston

Gadsby. David, Rockefeller University

Garcia-Blanco. Mariano. Duke University Medical Center

Gerhart. John. University of California. Berkeley

Giuditta. Antonio. University of Naples, Italy

Goldman, Robert D.. Northwestern University Medical School

Gould. Robert. New York State Institute for Basic Research

Groden. Joanna. University of Cincinnati

Haimo. Leah. University of California. Riverside

Hale, Melina, State University of New York. Stony Brook

Haydon, Philip, Iowa State University

Heck. Diane. Rutgers University

Hershko, Avram. Technion-Israel Institute of Technology, Israel

Highstein, Steven M., Washington University School of Medicine

Hill. Susan Douglas. Michigan State University

Hines, Michael. Yale University School of Medicine

Hotmann. Johann. Stanford University

Holmgren, Miguel. Harvard University Medical School

Holz. George, New York University School of Medicine

Johnston, Daniel. Baylor College of Medicine
Jones. Teresa. National Institutes of Health

Kaczmarek, Leonard, Yale University School of Medicine
Kaminer, Benjamin, Boston University School of Medicine
Kaplan, Barry. National Institutes of Mental Health
Kaplan, Ilene M., Union College
Kaupp, U.B., Institut fur Biologische Informationsverarbeitung.

Germany

Khan, Shahid. Albert Einstein College of Medicine
Kier, William. University of North Carolina. Chapel Hill
Kirschner. Marc. Harvard University Medical School
Koulen. Peter, Yale University School of Medicine
Kuhns. William, The Hospital for Sick Children, Canada
Kuner. Thomas. Duke University Medical Center

Lafer, Eileen M.. University of Texas Health Science Center

Landowne. David. University of Miami School of Medicine

Langford. George. Dartmouth College

Laskin. Jeffrey. University of Medicine and Dentistry of New Jersey

Laufer, Hans. University of Connecticut

LaVail, Jennifer. University of California, San Francisco

LeBaron. Richard. University of Texas. San Antonio

Lenzi, David. University of Virginia School of Medicine

Levitan. Irwin, University of Pennsylvania Medical Center

Link, Brian, Harvard University

R35

R36 Annual Report

Lipicky, Raymond J., Food and Drug Administration
Llinas, Rodolfo R., New York University Medical Center

Magee, Jeff. Louisiana State University Medical Center
Malchow. Robert Paul. University of Illinois, Chicago
Malgaroli. Antonio, University of Milan. Italy
Martinez, Joe, University of Texas, San Antonio
McFarlane. Matthew, New York University Medical Center
McNeil. Paul, Medical College of Georgia
Mensinger. Allen. University of Minnesota. Duluth
Messerli, Mark. Purdue University
Mitchison, Timothy. Harvard University Medical School
Moore, John W.. Duke University Medical Center
Mooseker, Mark, Yale University

Nasi, Enrico, Boston University School of Medicine

Ogden, David, National Institute for Medical Research
Ogunseitan, Oladele A., University of California. Irvine

Palazzo. Robert. University of Kansas

Pant. Harish. National Institutes of Health

Parysek. Linda. University of Cincinnati

Paydarfar. David. University of Massachusetts Medical School

Rakowski Robert F., Finch University of Health Sciences/The Chicago

Medical School

Ratner, Nancy, University of Cincinnati
Reese. Thomas S.. National Institutes of Health
Rieder, Conly, Wadsworth Center
Ripps. Harris, University of Illinois College of Medicine
Rome, Larry. University of Pennsylvania
Rosenbaum, Joel, Yale University
Russell, John M., Syracuse University

Saggau, Peter. Baylor College of Medicine

Salmon. Edward. University of North Carolina, Chapel Hill

Schmolesky, Matthew, University of Utah

Sloboda. Roger D.. Dartmouth College

Spiegel. Evelyn. Dartmouth College

Spiegel. Melvin. Dartmouth College

Srinivas, Miduturu. Albert Einstein College of Medicine

Steinacker, Antoinette, University of Puerto Rico

Sugimori. Mutsuyuki, New York University Medical Center

Telzer, Bruce, Pomona College

Tilney. Lewis. University of Pennsylvania

Trinkaus, John P., Yale University

Tytell, Michael, Wake Forest University School of Medicine

Udvadia, Ava, Duke University Medical Center

Wadsworth. Pat, University of Massachusetts

Wang. Jing. Lucent Technologies

Weidner. Earl. Louisiana State University

White. Thomas. Harvard University Medical School

Whittaker, J. Richard, University of New Brunswick, Canada

Wills. Zachary. Harvard University Medical School

Yamoah, Ebenezer. University of California, Davis
Young, Iain. University of Pennsylvania

Zecevic, Dejan P.. Yale University School of Medicine
Zimmerberg, Joshua. National Institutes of Health

Zottoli, Steven, Williams College

Zukin, R. Suzanne. Albert Einstein College of Medicine

Other Research Personnel

Abe, Teruo. Niigata University Brain Research Institute, Japan

Ahmed, Tanweer. University of Leeds. United Kingdom

Allen, Nina, North Carolina State University

Altamirano, Anibal. Instituto de Investigacion Medica "Mercedes y

Martin Ferreyra." Argentina
Angarita. Benjamin. Williams College
Artigas, Pablo. Rockefeller University
Asokan. Rengasamy, University of California, Davis
Atherton. Jill, Allegheny College

Basanei. Gorka, National Institutes of Health

Bauer, Sharon, Hunter College

Bendiksby. Michael. Duke University Medical Center

Berberian. Graciela. Instituto de Investigacion Medica "Mercedes y

Martin Ferreyra." Argentina
Bergamaschi. Andrea, University S. Raffaele, Italy
Bertetto. Lisa, Wesleyan University
Bingham. Eula. University of Cincinnati Medical School
Bonacci. Lisa, Hunter College

Bornstein. Gil. Technion — Israel Institute of Technology. Israel
Boudko. Dmitri, University of Florida
Boyle. Richard. Oregon Health Sciences University
Breitwieser. Gerda, Johns Hopkins University School of Medicine
Bucior. Iwona, Friedrich Miescher Institute, Switzerland

Callender. Delon, Hunter College

Chou, Ying-Hao, Northwestern University

Clarkson. Melissa, University of Kansas

Clegg. Janet, University of California. Riverside

Clegg. Michael. University of California, Riverside

Colvin. Robert. Ohio University

Desai, Arshad. European Molecular Biology Laboratory. Germany
Djunsic. Maja, Yale University School of Medicine
Doussau, Frederic. Duke University Medical Center
Dunham. Philip, Syracuse University

Easter. Joshua, Williams College

Eddleman, Christopher, Texas Tech Medical School

Escalada. Arthur, University of Barcelona Medical School, Spain

Eyman, Maria, University of Naples, Italy

Faas, Guido, Baylor College of Medicine
Forger. Daniel, Courant Institute

Gace. Arian. Louisiana State University

Gainer. Harold, National Institutes of Health

Galbraith. James A.. National Institutes of Health

Gallant. Paul E.. National Institutes of Health

George. Paul. Brown University

Gerosa-Erni. Daniela, Fnedrich Miescher Institute, Switzerland

Gilles, Nicole, University of Minnesota

Gioio, Anthony, National Institutes of Health

Goda. Makoto, Kyoto University, Japan

Goldman. Anne E., Northwestern University Medical School

Gomez. Maria del Pilar, Boston University School of Medicine

Grant. Philip. National Institutes of Health

Summer Research R37

Guplon. Stephanie. North Carolina State University
Gyoeva, Fatima K., Institute of Protein Research, Russia

Hardin. Robert, Brigham and Women's Hospital

Harper. Claudia. Massachusetts Institute of Technology

Harrington. John. University of South Alabama, Mobile

Harrow. Faith. Hunter College

Harwood. Claire. University of Pennsylvania

Hembree, Walter. Wake Forest University

Hernandez, Carlos. New York University School of Medicine

Hernandez. Ruben, University of Texas, San Antonio

Hitt, James, State University of New York Health Science Center

Hiza, Nicholas, Williams College

Hogan, Emilia, Yale University Medical School

Hussain, Mohammad, Albert Einstein College of Medicine

Hutchins. Heidi, National Institutes of Health

Innocenti, Barbara. Iowa State University

Janowitz. Tobias. Yale University
Johenning. Friedrich, Yale University
Jonas. Elizabeth, Yale University
Jones. Kendrick. Brown University

Kamino, Kohtaro, Tokyo University School of Medical and

Dental, Japan

Kang. Guoxin. New York University School of Medicine
Kapoor. Tarun, Harvard University Medical School
Karson. Miranda, Michigan State University
Kingston, Margaret, Wake Forest University
Klimov. Andrei. University of Pennsylvania

Kopacek. Petr, Institute of Parasitology ASCR, The Czech Republic
Koroleva. Zoya. Hunter College
Kreitzer. Matthew. University of Illinois. Chicago
Kumar. Mukesh, National Institutes of Health
Kuner. Thomas. Duke University Medical Center

Lambert, Justin, University of Arizona

Lee, Kyeng Gea, Hunter College

Lee, Licheng. Duke University

Levin, Tracy. Smith College

Levitan. Edwin. University of Pittsburgh School of Medicine

Liu. Vincent. New York University Medical Center

I.oboda. Andrey, University of Pennsylvania

Lovell, Peter. Whitney Laboratory

Lowe, Christopher, University of California, Berkeley

Maddox. Paul. University of North Carolina. Chapel Hill

Marder. Eve. Brandeis University

Marshall. Wallace, Yale University

Mclntyre. Charmian, Brandeis University

McQuiston. Rory. Baylor College of Medicine

Miller. Todd. Hunter College

Molina. Anthony. University of Illinois. Chicago

Morgan. Jennifer. Duke University Medical Center

Moroz, Leonid, University of Florida

Mutyambizi. Kudakwashe, Williams College

Noble, Peter. University of South Carolina

Oegema. Karen. European Molecular Biology Laboratory, Germany

Petersen. Jennifer. National Institutes of Health

Prasad. Kondury, University of Texas Health Science Center

Price, Nichole, Connecticut College

Qian, Haohua. University of Illinois. Chicago

Ramsey. David, Harvard University

Rapoport, Scott, University of California, San Diego

Rhodes, Paul, New York University Medical School

Ringel, Israel. Hebrew University. Israel

Rosenkranz. Naomi, Yeshiva University

Russell, James, National Institute of Health

Saidel, William, Rutgers University

Salzberg, Brian, University of Pennsylvania

Sandberg. Leslie, Dartmouth College

Schneider. Eric. Brown University

Schwartz, Lawrence. University of Massachusetts

Scotto. Lavina, National Institutes of Health

Shuster. Charles, Boston College

Simpson, Tracy, University of Hartford

Solzin, Johannes. Institut fur Biologische Informationsverarbeitung,

Germany

Stafford, Phillip, Dartmouth College
Stephens, Natalie, Williams College
Stockbridge, Norman, U.S. Department of Agriculture
Sul, Jai-Yoon, Iowa State University
Szucsik, Amanda, Rutgers University

Takahashi, Joseph, Northwestern University

Tani. Tomomi, Tokyo Metropolitan Institute of Medical Science, Japan

Taylor. Kevin, Wake Forest University

Thrower. Edwin, Yale University

Tokumaru, Hiroshi, Duke University Medical Center

Tokumaru. Keiko. Duke University Medical Center

Tran, Phong. Columbia University

Twersky, Laura. Saint Peter's College

Tyson. Cortni. Williams College

Viitanen. Liisa, Boston College

Wachowiak, Matt. Yale University School of Medicine
Wassersug, Richard, Dalhousie

Weyand, Ingo. Institut fur Biologische Informationsverarbeitung,
Germany

Yamaguchi. Ayako. Columbia University

Yoo, Soonmoon, University of Texas Medical Branch

Zakevicius. Jane M.. University of Illinois College of Medicine
Zerbe. Jamie. University of Kansas
Zhou, Yuehan, Yale University
Zochowski, Michal. Yale University

Library Readers

Naitoh, Yutaka, University of Hawaii

Nguyen. Michael P., University of Texas Medical Branch

Nierman. Jennifer, Williams CoMege

Abbott. Jayne. Marine Research
Ahmadjian, Vernon, Clark University
Allen, Garland, Washington University

R38 Annual Report

Alliegro, Mark, Louisiana State University Health Sciences Center
Alsup, Peggy, Tennessee Department of Health
Anderson, Everett, Harvard Medical School

Baccetti. Baccio. Institute of General Biology

Barry, Susan, Mount Holyoke College

Baylor, Martha, Marine Biological Laboratory

Benjamin, Thomas, Harvard Medical School

Bernhard, Jeffery, University of Massachusetts Medical Center

Bernheimer, Alan, New York University School of Medicine

Borgese, Thomas, Lehman College-CUNY

Boyer, John, Union College

Candelas, Graciela, University of Puerto Rico

Changeux, JeanPierre, Rand Fellowship

Child, Frank, Trinity College

Clarkson, Kenneth. Lucent Technologies

Cobb, Jewel P., California State University

Cohen, Seymour, American Cancer Society

Cooperstein, Sherwin J., University of Connecticut Health Center

Copeland, Donald, Marine Biological Laboratory

Corwin, Jeffrey. University of Virginia

Cowling, Vincent, Palm Beach, FL

De Toledo-Morrell, Leyla, Rush University
Epstein. Herman. Brandeis University

Fraenkel, Dan, Harvard Medical School

Frenkel. Krystyna, New York University School of Medicine

Galatzer-Levy, Robert, University of Chicago
German, James, Cornell University Medical College
Grossman, Albert. New York University Medical School
Gruner, John, Cephalon. Inc.

Harrington. John. University of South Alabama

Haubrich. Robert. Denison University

Haugaard, Niels, HUP Philadelphia

Herskovits, Zara, Belter Educational Center

Herskovits, Theodore. Fordham University

Hitchcock-DeGregorii. Sarah, Robert Wood Johnson Medical School

Hunter, Robert. Gartnaval Royal Hospital

Inoue. Sadayuki. McGill University
Issodorides. Marietta, Athens, Greece

Jacobson, Allan, University of Massachusetts Medical School
Jaye. Robert, Solomon Schechter Day School
Josephson. Robert K., University of California, Irvine

Kaltenbach, Jane, Mount Holyoke College
Karlin. Arthur, Columbia University
Kelly. Robert, University of Illinois
King. Kenneth. Falmouth. MA
Kornberg, Hans, Boston University
Krane. Stephen. Harvard Medical School

Laster. Leonard. University of Massachusetts Medical Center

Lee. John, City College of New York

Lesser, Carolyn, University of Wisconsin

Linck, Richard, University of Minnesota

Lorand, Laszlo, Northwestern University Medical School

Luckenbill, Louise. Ohio University

Mauzerall, David. Rockefeller University
Mitchell, Ralph. Harvard University
Mizell, Merle. Tulane University
Mizoguchi. Hazime. Johns Hopkins University

Nagel. Ronald, AECOM NYC

Naugle, John. National Aeronautics and Space Administration

Nickerson, Peter, SUNY Buffalo

Pappas, George D., University of Illinois. Chicago
Prendergast. Robert, John Hopkins University

Schippers, Jay, Resource Foundation
Shepro, David, Boston University
Siwicki, Kathleen, Swarthmore College
Spector, Abraham, Columbia University
Spotte, Stephen. University of Connecticut
Sundquist, Eric, USGS
Sweet, Frederick, Washington University

Trager, William. The Rockefeller University
Tweedell. Kenyon. University of Notre Dame
Tykocinski. Mark, University of Pennsylvania

Van Holde, Kensal. Oregon State University

Walton. Alan. Cavendish Lab

Warren, Leonard, University of Pennsylvania

Yevick, George, Stevens Institute of Technology

Domestic Institutions Represented

Academy of Natural Sciences
Alabama, University of, Birmingham
Alaska. University of. Anchorage
Alaska, University of. Fairbanks
Albert Einstein College of Medicine
Allegheny College
American Cancer Society
American Museum of Natural History
Argonne National Laboratory
Arizona State University
Arizona. University of, Tucson
Arkansas, University of

2000 Library Room Readers

Lucio Cariello

Stazione Zoologica A. Dohrn

Michael Clegg

Giuseppe D'Alessio
University of Naples

Robert Goldman

Northwestern University Medical School

Harlyn Halvorson

Marine Biological Laboratory

Michael Hines

Yale University School of Medicine

Alex Keynan

Israel Academy of Science

John Moore

Duke Medical Center

Michael Rabinowitz

Marine Biological Laboratory

George Reynolds
Princeton University

Ann Stuart
UNC Chapel Hill

Gerry Weissmann

NYU School of Medicine

Summer Research R39

Association of Telehealth Providers
AT&T Bell Laboratories

Battle Creek Veterans Administration

Baylor College of Medicine

Bell Laboratores

BioHybrid Technologies. Inc.

Biolnfo 4U

Bioinformatics

Bioresources Development and Conservation Programme

Blue Cross Blue Shield of Maryland

Boston College

Boston University

Boston University School of Medicine

Bowdoin College

Brandeis University

Brigham and Women's Hospital

Brigham Young University

Brown University

Bryn Mawr College

Buck Center for Research in Aging

California Institute of Technology

California. University of. Berkeley

California. University of. Davis

California, University of, Irvine

California, University of, Los Angeles

California. University of. Riverside

California. University of, San Diego

California, University of, San Francisco

California, University of, Santa Barbera

Care First Blue Cross Blue Shield

Carnegie Institution of Washington

Carnegie Mellon University

Case Western Reserve Medical School

Case Western Reserve University

Catholic University of America

Centers for Disease Control and Prevention

Cephalon. Inc.

Chicago College of Osteopathic Medicine

Chicago. University of

Children's Hospital. Boston

Children's Memorial Hospital — CMIER

Children's National Medical Center

Cincinnati University Medical Center

Cincinnati, University of

City College of New York

Clarian Health

Cleveland State University

Cold Spring Harbor Laboratory

Colorado University Health Science Center

Colorado. University of. Boulder

Colorado University School of Medicine

Columbia University

Connecticut College

Connecticut University Health Center

Connecticut, University of

Cornell University

Cornell University Medical College

Courant Institute

Delaware, University of

Denison University

Doheny Eye Institute

Duke University

Duke University Medical Center

DuPont Hospital for Children

Eli Lilly & Company

Emory University

Emory University School of Medicine

Exelixis, Inc.

Field Museum of Natural History

Finch University of Health Sciences

Florida Institute of Technology

Florida State University

Florida University Brain Institute

Florida University College of Medicine

Florida, University of

Food and Drug Administration

Fordham University

Fred Hutchinson Cancer Research Center

Georgia, University of
Guthrie Healthcare System

Hampton University

Harbor Branch Oceanographic Institution

Harbor-UCLA Medical Center

Hartford, University of

Harvard Medical School

Harvard School of Public Health

Harvard University

Hawaii. University of

Health Care Financing Administration

Mines VA Hospital

Hospital for Special Surgery

Hospital of the University of Pennsylvania

House Ear Institute

Houston Academy of Medicine

Houston. University of

Howard Hughes Medical Institute

Howard University

Howard University School of Medicine

Hunter College

Idaho, University of

Illinois, University of, Chicago

Illinois. University of, Urbana-Champaign

Indiana University

Indiana University School of Medicine

Ingham Regional Medical Center

Institute for Genomic Research

Iowa University College of Medicine

Iowa State University

Iowa, University of

Johns Hopkins University

Johns Hopkins University School of Medicine

Joint Genome Institute

Dartmouth College
Dartmouth Medical School
Deaconess Medical Center

Kansas University Medical Center
Kansas, University of
Kent State University

R40 Annual Report

Kentucky University Medical Center
Kentucky, University of
Kewalo Marine Laboratory
King/Drew Medical Center
Knight Ridder Newspapers

Lahey Clinic

Lawrence Berkeley National Laboratory

Lehman College, CUNY

Leica, Inc.

Lilly Research Labs

Los Alamos National Laboratory

Louisiana State University

Louisiana State University Health Sciences Center

Louisiana, University of, Lafayette

Louisville. University of

Loyola University of Chicago

Lucent Technologies

Magee-Women's Research Institute

Maine, University of

Maniilaq Health Center

Marine Biological Laboratory

Marquette University

Marshlield Medical Research Foundation

Maryland. University of, Baltimore County

Massachusetts Eye and Ear Infirmary

Massachusetts General Hospital

Massachusetts Institute of Technology

Massachusetts, University of, Amherst

Massachusetts. University of. Medical School

Maxygen. Inc.

Mayo Clinic and Foundation

McCrone Research Institute

Medical College of Georgia

Medical University of South Carolina

Meharry Medical College

Memorial Sloan-Kettering Cancer Center

Merck & Co.

Miami, University of

Miami University School of Medicine

Michigan State University

Michigan University Medical School

Michigan. University of

Midwestern University

Minnesota University Medical School

Minnesota. University of

Missouri, University of, Rolla

Montana State University

Monterey Bay Aquarium Research Institute

Morehouse School of Medicine

Morgan State University

Mount Holyoke College

Mount Sinai School of Medicine

Murdoch Institute

National Aeronautics and Space Administration

National Institute of Mental Health

National Institute on Aging. NIH

National Institutes of Health

National Library of Medicine

Nature America

Naval Medical Center. San Diego

Nebraska. University of. Lincoln

NEC Research Institute

Neuroscience Institute

Nevada University School of Medicine

New England Regional Primate Research Center

New Mexico, University of

New York Health Science Center, State University of

New York State Department of Health

New York State Institute for Basic Research

New York, State University of, Albany

New York, State University of, Buffalo

New York. State University of. Stony Brook

New York University

New York University Medical Center

New York University School of Medicine

Norman Regional Hospital

North Carolina State University

North Carolina, University of. Chapel Hill

Northwestern Medical Faculty Foundation

Northwestern University

Northwestern University Medical School

Notre Dame, University of

Oberlin College

Ohio State Llniversity

Ohio University

Oregon Health Sciences University

Oregon Regional Primate Research Center

Oregon State University

Oregon, University of

PE Biosystems

Penn State University

Pennsylvania State University College of Medicine

Pennsylvania University Medical Center

Pennsylvania. University of

Pennsylvania University School of Medicine

Pfizer Central Research

Pharmacia & Upjohn

Pittsburgh, University of

Pomona College

Princeton University

Proteome. Inc.

Puerto Rico. University of

Purdue University

Purdue University Cancer Center

Q3DM, Inc.

Regenstrief Institute

Rensselaer Polytechnic Institute

Riverside Regional Medical Center

Robert Wood Johnson Medical School

Roche Diagnostics

Rochester. University of

Rockefeller University, The

Rosetta Inpharmatics

Rush-Presbyterian-St. Luke's Medical Center

Rutgers College of Pharmacy

Rutgers University

Saint Peter's College

Salk Institute

San Francisco State University

Scripps Institution of Oceanography

Summer Research R41

Scripps Research Institute

Seattle Biomedical Research Institute

Smith College

Smithsonian Institution

Solomon Schechter Day School

SoundVision Productions

South Alabama. University of

South Carolina. University of

South Florida. University of

Southampton University

Southern California. University of

Southern Mississippi, University of

St. Elizabeth's Medical Center

St. Joseph's Hospital

St. Louis VA Medical Center

St. Mary's College of Maryland

Stanford University

Stanford University Medical Center

Stanford University School of Medicine

Stevens Institute of Technology

Swarthmore College

Swedish Medical Center

Syracuse University

Temple University School of Medicine

Tennessee Depanment of Health

Tennessee State University

Texas A&M University

Texas Tech Medical School

Texas University Health Science Center

Texas University Medical School

Texas. University of. Austin

Texas, University of. Houston

Texas. University of. San Antonio

Texas University Southwestern Medical Center

Toledo. University of

Trover Foundation

Tufts University

Tufts University School of Medicine

Tulane University

U.S. Department of Agriculture

U.S. News & World Report

Uniformed Services University

Union College

University of Medicine and Dentistry of New Jersey

Utah University Medical Center

Utah. University of

VA Information Research Center

VA Maryland Health Care System

Vanderbilt University

Vanderbilt University Medical Center

Vanderbilt-Ingram Cancer Center

Vermont. University of

Veterans Administration Hospital

Veterans Affairs Medical Center

Virginia University Health Sciences Center

Virginia University Medical Center

Virginia. University of

Wadsworth Center

Wake Forest University

Wake Forest University School of Medicine

Washington University

Washington, University of

Washington University School of Medicine

Weill Medical College of Cornell University

Wellesley College

Wesleyan University

Western Reserve Medical School

Westvaco Forest Sciences Lab

Whitehead Institute

Whitney Laboratory. University of Florida

Williams College

Winston-Salem Journal

Wisconsin, University of, Madison

Woods Hole Oceanographic Institution

Wyeth-Ayerst Research

Yale University

Yale University School of Medicine

Yeshiva University

Zeiss Optical Systems

Foreign Institutions Represented

Aberdeen, University of. Scotland
Albert-Ludwigs-Universitat Freiburg, Germany
Alfred Wegener Institute, Germany
Amsterdam. University of. The Netherlands
Australian National University, Australia

Basel Institute for Immunology, Switzerland

Basel. University of. Switzerland

Bern, University of. Switzerland

Bielefeld. University of. Germany

Biomedical Primate Research Centre. The Netherlands

Boehringer Ingelheim Pharmaceuticals, Inc.. Germany

Buenos Aires. University of. Argentina

Calgary. University of. Canada

Cambridge University, United Kingdom

Cape Town. University of. South Africa

Centre de Genetique Moleculaire, France

Centre National de la Recherche Scientifique — CNRS, France

Centro de Pesquisas "Rene Rachou." Brazil

Charles University. Prague, Czech Republic

Comision Nacional de Energia Atomica, Argentina

Copenhagen, University of. Denmark

Dalhousie University. Canada
Dundee. University of. Scotland

Edinburgh. University of. Scotland

European Molecular Biology Laboratory, Germany

Friedrich Miescher Institute. Switzerland
Freie Universitat. Berlin. Germany

Gartnaval Royal Hospital, Scotland
Genoa, University of, Italy
Glasgow, University of, Scotland
Goteborg University, Sweden
Guelph, University of. Canada

R42 Annual Report

Haaedorn Research Institute. Denmark

Hebrew University. Israel

Hebrew University Medical School, Israel

Hohenheim. University of. Germany

Hong Kong. University of

Hong Kong University of Science and Technology

Hospital for Sick Children. Canada

Humboldt Universitat Berlin. Germany

Hungarian Academy of Sciences, Hungary

IBDM. Marseille. France

Imperial College of Science, Technology and Medicine, United

Kingdom

Innsbruck. University of, Austria

Institut fur Biologische Informationsverarbeitung, Germany
Institut Pasteur, France

Institute of Cell. Animal, and Population Biology. Scotland
Institute of Neurophysiology, Pisa. Italy
Institute of Parasitology ASCR, The Czech Republic
Institute of Protein Research, Russia
Institute de Investigacion Medica "Mercedes y Martin Ferreyra,'

Argentina

Institute de Investigaciones Biomedicas, Spain
Institute Gulbenkian de Ciencia, Portugal
Institute Nacional de la Nutricion. Mexico
Institute Venezolano Investigaciones Cientificas, Venezuela
Istituto Intemazionale di Genetica e Biofisica. Italy

Karolinska Institute, Sweden
Koln, University of. Germany
Konstanz. University of, Germany
Kyoto University. Japan
Kyunghee University. Korea

Leeds, University of. United Kingdom

Leicester. University of. United Kingdom

Leiden University Medical Centre. The Netherlands

Lethbridge, University of. Canada

Liege. University of. Belgium

Linkoping University, Sweden

Ludwig-Maximilian University. Germany

Manchester. University of. United Kingdom

Marine Biology Station, France

Max-Planck-Institute for Biological Cybernetics. Germany

Max-PUmck-Institute for Medical Research. Germany

McGill University, Canada

McMaster University, Canada

Medical Research Council, United Kingdom

Melbourne, University of, Australia

Montreal General Hospital. Canada

Montreal. University of. Canada

Naples. University of. Italy

National Institute for Medical Research. United Kingdom

Netherlands Cancer Institute

New Brunswick, University of, Canada
Newcastle-upon-Tyne. University of. United Kingdom
Niigata University Brain Research Institute, Japan
Nobel Institute for Neurophysiology. Sweden

Oldenburg, University of, Germany
Ottawa. University of. Canada
Oxford University, United Kingdom

Palermo, University of, Italy
Perugia. University of. Italy
Pisa, University of. Italy
Porto, University of, Portugal
PPL Therapeutics Inc.. Scotland
Punjab Agricultural University, India

Rayne Institute, United Kingdom
Rio de Janeiro. University of. Brazil
Roslin Institute. Edinburgh, Scotland

Sao Paulo, University of. Brazil

Sars Centre, Norway

Saskatchewan. University of. Canada

Scuola Intemazionale Superiore di Studi Avanzati (SISSA). Italy

Scuola Normale Superiore, Italy

Shirshov Institute of Oceanology, Russia

Simon Fraser University. Canada

Sofia Institute Gulbenkian de Ciencia. Portugal

St. Andrews, University of, Scotland

Stirling, University of. Scotland

Stockholm University. Sweden

Surrey, University of. United Kingdom

Sussex, University of. United Kingdom

Swiss Federal Institute of Technology, Switzerland

Swiss Institute for Experimental Cancer Research, Switzerland

Sydney, University of. Australia

Tata Institute of Fundamental Research, India
Technion-Israel Institute of Technology. Israel
Teikyo University Biotechnology Research Center. Japan
Tokyo University School of Medical and Dental. Japan

Universidad Autonoma Metropolitana. Mexico
Universidad Nacional Autonoma de Mexico
Universidad Nacional de Cuyo. Argentina
Universidade Federal de Minas Gerais. Brazil
Universitat Freiburg, Germany
Universite Libre de Bruxelles, Belgium
Universite Paris-Sud. France
Uppsala University, Sweden

Veterinary Vaccine Institute. India

Weizmann Institute of Science. Israel
Western Ontario. University of, Canada

Zurich, University of, Switzerland

Year-Round Research Programs

Architectural Dynamics in Living
Cells Program

Established in 1992. this program focuses on architectural dynamics
in living cells — the timely and coordinated assembly and disassembly of
macromolecular structures essential for the proper functioning, division,
motility. and differentiation of cells; the spatial and temporal
organization of these structures: and their physiological and genetic
control. The program is also devoted to the development and application
of powerful new imaging and manipulation devices that permit such
studies directly in living cells and functional cell-free extracts. The
Architectural Dynamics in Living Cells Program promotes
interdisciplinary research carried out by resident core and visiting
investigators.

Ki'Milcin Cure Investigators

Inoue, Shinya. Distinguished Scientist
Oldenbourg. Rudolf, Associate Scientist
Shribak. Michael. Staff Scientist

Staff

Knudson. Robert, Instrumental Development Engineer
Baraby. Diane. Laboratory Assistant
MacNeil. Jane. Executive Assistant

Visiting Investigators

Desai, Arshad. EMBL. Heidelburg. Germany

Fukui. Yoshio. Northwestern University Medical School

Coda. Makoto. Kyoto University, Japan

Keefe. David. Rhode Island Women and Infants Hospital

Liu. Lin. Rhode Island Women and Infants Hospital

Maddox. Paul, University of North Carolina-Chapel Hill

Mitchison. Timothy J.. Harvard Medical School

Salmon. Edward D.. University of North Carolina-Chapel Hill

Tran. Phong. Columbia University

The Josephine Bay Paul Center for

Comparative Molecular Biology

and Evolution

This Center employs the latest advances in phylogenetic theory,
computational biology, and high-throughput genome sciences to study
evolutionary processes that trace back to the first life forms on earth.
Through the application of high-powered statistical techniques, scientists
in the Josephine Bay Paul Center investigate how the evolution of genes
and genomes has driven phenotypic change at all levels of biological

organization. This holistic approach provides tools to quantify and
assess biodiversity and to identify genes and genetic mechanisms of
biomedical and environmental importance. We study all kinds of
microbes, their evolutionary history, their interactions with each other
and macroscopic forms of life, and how members of diverse microbial
communities contribute and respond to environmental change. Examples
of current research include: 1 ) a project supported by the National
Science Foundation to study the co-evolution of genomes for symbiotic
bacteria and their hosts; 2) investigations supported by the National
Institutes of Health to study expression and the complete genome
sequence of Giardia lamblia. a water-borne human pathogen that attacks
the intestinal tract and exacts a terrible toll on public health worldwide;
3) a computational biology program funded by the NIH. NASA, and
private corporations to integrate evolutionary theory with the functional
annotation of protein coding regions in bacterial genomes; and 4) an
interdisciplinary study supported by NASA and NSF to study life in
extreme environments on the planet earth in search of general principles
that can guide the quest for living forms elsewhere in the universe. The
Center encourages studies of genotypic diversity across all phyla and
promotes the use of modem molecular genetics and phytogeny to gain
insights into the evolution of molecular structure and function.

Our research activities are complemented by an active education
program. In addition to training postdoctoral fellows, the Josephine Bay
Paul Center offers the internationally recognized Workshop in Molecular
Evolution at the Marine Biological Laboratory, a workshop for
secondary educators titled Living in the Molecular World, and several
comprehensive web sites: 1 ) a description of research and education
associated with our membership in the Astrobiology Institute at the
Marine Biological Laboratory; 2) the interactive EcoCyc Project (an
interactive program that describes the metabolism of E. coli as well as
the identity and location of functional genes in the E. coli genome); 3)
the Giardia lamblia genome page (which provides annotated analyses
and current progress summaries from the MBL's Giardia lamblia
genome project); and 4) the Workshop in Molecular Evolution site
(which offers descriptions, information, and advice about sophisticated
software packages for phylogenetic inferences and analyses of
population biology data).

A generous gift from the Bay Paul Foundation and continuing annual
support from the G. Linger Vettlesen Foundation provided initial funding
in 1997 to form The Josephine Bay Paul Center for Comparative
Molecular Biology and Evolution. The Center has excellent resources
for studies of molecular biology and evolution, including well-equipped
research laboratories and a powerful computational facility. With a grant
from the W.M. Keck Foundation in 2000, the center established a
technology-rich Ecological and Evolutionary Genetics Facility. This
advanced laboratory provides a full range of high-throughput. DNA-
sequencing equipment, a DNA microarray facility and high-performance
computers. Several adjunct appointments and collaborative projects
strengthen research activities in the center. These activities include
interdisciplinary investigations of microbial diversity with scientists at
the Woods Hole Oceanographic Institution, molecular ecology studies at
the MBL Ecosystem Center's Plum Island LTER site, physiology

R43

R44 Annual Report

studies of acidophilic protists with the MBL BioCurrents Research
Center, and collaborative efforts to study mechanisms and patterns of
evolution with faculty of Harvard University, the Harvard School of
Public Health, and the University of Sydney, Australia. Future
expansion in the Josephine Bay Paul Center will focus upon molecular
evolution of global infectious disease and genome sciences.

Resident Core Investigators

Sogin, Mitchell. Director and Senior Scientist
Cornell, Neal, Senior Scientist
Cummings, Michael, Assistant Scientist
McArthur, Andrew, Staff Scientist II
Morrison, Hilary. Staff Scientist II
Riley. Monica. Senior Scientist
Wernegreen, Jennifer, Assistant Scientist

Adjunct Scientists

Halanych. Ken, Woods Hole Oceanographic Institution

Meselson, Matthew, Harvard University

Patterson, David, University of Sydney

Teske, Andreas, Woods Hole Oceanographic Institution

Laboratory of Neal Cornell

Dr. Neal Cornell, a senior scientist at the Marine Biological
Laboratory, played a key role in designing and attracting new faculty to
the Josephine Bay Paul Center for Comparative Molecular Biology and
Evolution. Dr. Cornell passed away in 2000. but all of us who knew
him cherish fond memories and harbor a deep gratitude for his
contributions to science and the MBL community. Research in Dr.
Cornell's laboratory, which continued to pursue his research goals
through the end of 2000, was concerned with the comparative molecular
biology of genes that encode the enzymes for heme biosynthesis. These
efforts placed particular emphasis on 5-aminolevulinate synthase, the
first enzyme in the pathway. Because the ability to produce heme from
common metabolic materials is a near universal requirement for living
organisms, these genes provide useful indicators of molecular aspects of
evolution. For example. 5-aminolevulinate synthase in vertebrate animals
and simple eukaryotes such as yeast and Plasmodium falciparum have
high sequence similarity to the enzyme from the alpha-purple subgroup
of eubacteria. This supports the suggestion that alpha-purple bacteria are
the closest contemporary relatives of the ancestor of eukaryotic
mitochondria. The analysis also raises the possibility that plant and
animal mitochondria had different origins. Aminolevulinate synthase
genes in mitochondria-containing protists are currently being analyzed to
obtain additional insight into endosymbiotic events. Also, genes of
primitive chordates are being sequenced to gain information about the
large-scale gene duplication that played a very important role in the
evolution of higher vertebrates. Other studies in the laboratory have
been concerned with the effects of environmental pollutants on heme
biosynthesis in marine fish, and it has been shown that polychlorinated
hiphenyls (PCBs) enhance the expression of the gene for
aminolevulinate synthase.

Laboratory of Michael P. Cummings

Our research is in the area of molecular evolution and genetics and
includes examination of patterns and processes of sequence evolution.
We use methods from molecular biology, population genetics,
systematics, statistics, and computer science. The basis for much of the
research is comparative; it includes several levels of biological
organization, and involves both computer-based and empirical studies. A
major research focus is analysis genotype-phenotype relationships using
tree-based statistical models (decision trees) and extension of this
methodology. Current investigations in this area examine how gene
sequence data can be used to understand and predict drug resistance in
tuberculosis, variation in color vision, and basic immune system
functions at the molecular level. For example, using drug resistance in
M\cobacterium tuberculosis as a model system, we are investigating
how well phenotype (level of drug resistance) can be predicted with
genotype information (DNA sequence data). Understanding evolution of
drug resistance, and developing accurate methods for its prediction using
DNA sequence data, can help in assessing potential resistance in a more
timely fashion and circumvent the need for culturing bacteria. More
generally, the relationship of genotype to phenotype is a fundamental
problem in genetics, and through these investigations we hope to gain
insight. The primary empirical work in the laboratory involves
examination of opsins, proteins involved in color vision, from local '
species of Odonata (dragonflies and damselflies). Other projects include
a review of genetic diversity in plants using coalescence-based analyses
and the genetic consequences of reserve designs in conservation.

Suff

Cornell. Neal W.. Senior Scientist
Faggart. Maura A., Research Assistant

Staff

Cummings, Michael P.. Assistant Scientist
Mclnemey, Laura A., Research Assistant II

Year-Round Research R45

Visiting Investigators

Clegg. Michael T., University of California, Riverside
Clegg. Janet, University of California. Riverside
Neel, Maile C., University of California, Riverside

Visiting Graduate Students

Church, Sheri A., University of Virginia

Garcia- Verela, Martin, Universidad Nacional Autonoma de Mexico

Undergraduates

Myers. Daniel S., Pomona College
Waring. Molly E.. Harvey Mudd College

Laboratory of Monica Riley

The genome of the bacterium Escherichia coli contains all of the
information required for a free-living chemoautotrophic organism to
live, adapt, and multiply. The information content of the genome can be
dissected from the point of view of understanding the role of each gene
and gene product in achieving these ends. The many functions of E. coli
have been organized in a hierarchical system representing the complex
physiology and structure of the cell. In collaboration with Dr. Peter
Karp of SRI International, an electronic encyclopedia of information has
been constructed on the genes, enzymes, metabolism, transport
processes, regulation, and cell structure of E. coli. The interactive
EcoCyc program has graphical hypertext displays, including literature
citations, on nearly all of E. call metabolism, all genes and their
locations, a hierarchical system of cell functions and some regulation
and transport processes.

In addition, the E. coli genome contains valuable information on
molecular evolution. We are analyzing the sequences of proteins of E.
coli in terms of their evolutionary origins. By grouping like sequences
and tracing back to their common ancestors, we learn not only about the
paths of evolution for all contemporary E. coli proteins, but we extend
even further back before E. coli, traversing millennia to the earliest
evolutionary times when a relatively few ancestral proteins served as
ancestors to all contemporary proteins of all living organisms. The
complete genome sequence of E. coli and sophisticated sequence
analysis programs permit us to identify evolutionarily related protein
families, determining ultimately what kinds of unique ancestral
sequences generated all of present-day proteins. The data developed in
the work has proven to be valuable to the community of scientists
sequencing other genomes. E. coli data serve as needed reference points.

Staff

Riley. Monica. Senior Scientist
Liang. Ping. Staff Scientist I
McCormack, Tom, Postdoctoral Scientist
Nahum, Laila, Postdoctoral Scientist
Pelegrini-Toole. Alida, Research Assistant II
Serres. Margerethe. Postdoctoral Scientist

Laboratory of Mitchell L. Sogin

This laboratory employs comparative phylogenetic studies of genes
and genomes to define patterns of evolution that gave rise to
contemporary biodiversity on the planet earth. The laboratory is
especially interested in discerning how the eukaryotic cell was invented
as well as the identity of microbial groups that were ancestral to
animals, plants, and fungi. The laboratory takes advantage of the
extraordinary1 conservation of ribosomal RNAs to define phylogenetic

relationships that span the largest of evolutionary distances. These
studies have overhauled traditional eukaryotic microbial classifications
systems. The laboratory has discovered new evolutionary assemblages
that are as genetically diverse and complex as plants, fungi, and
animals. The nearly simultaneous separation of these eukaryotic groups
(described as the eukaryotic "Crown") occurred approximately one
billion years ago and was preceded by a succession of earlier diverging
protist lineages, some as ancient as the separation of the prokaryotic
domains. Many of these early branching life forms are represented by
parasitic protists including Giardia lamklia, which is a significant
human parasite. Because of its medical importance and relevance to
understanding the evolutionary history of eukaryotes. we have initiated a
project to determine the entire genome sequence of Giardia lainMia. In
addition to identifying other genes that will be of value for unraveling
sudden evolutionary radiations that cannot be resolved by rRNA
comparisons, this project will provide insights into the presence or
absence of important biochemical properties in the earliest ancestors
common to all eukaryotic species. Finally, this project has revealed
important features of genome architecture that require a reconsideration
of available mechanisms for controlling gene expression in eukaryotes.

A second research theme is the study of microbial life in extreme
environments and molecular-based investigations of diversity and gene
expression in microbial communities. Using the ribosomal RNA
database and nucleic acid-based probe technology, it is possible to
detect and monitor microorganisms, including those that cannot be
cultivated in the laboratory. This strategy has uncovered new habitats
and major revelations about geographical distribution of
microorganisms. We are particularly interested in protists that thrive in
acid mine drainages and the characterization of physiological
mechanisms that allow their growth at extraordinarily low (<2.0) pH
levels. Our investigations of gene expression in microbial communities
is based upon the premise that microorganisms are the primary engines
of our biosphere. They orchestrate all key processes in geochemical
cycling, biodegradation. and in the protection of entire ecosystems from
environmental insults. They are responsible for most of the primary
production in the oceans. Microbial creatures of untold diversity have
complex chemistries, physiologies, developmental cycles, and behaviors.
With the powerful tools of high-throughput DNA sequencing and DNA
microarrays for massive parallel expression studies, we can directly
measure how microbial gene expression patterns in an entire ecosystem
respond to changing chemical and physical parameters. We will employ
an experimental paradigm that links biogeochemical processes with
ever-changing temporal and spatial distributions of microbial populations
and their metabolic properties. The concurrent measurement of
biogeochemical parameters, community-wide gene expression patterns,
and spatial descriptions of microbial populations offers a means to
understand the structure and function of biogeochemical machinery at
different levels of biological organization. We seek to discover the links
between biological diversity and the resilience and stability of
biogeochemical transformations.

Staff

Sogin. Mitchell L.. Director and Senior Scientist
Amaral Zettler. Linda. Postdoctoral Scientist
Beaudoin, David. Research Assistant
Bressoud. Scott. Laboratory Technician
Eakin. Nora, Research Assistant
Edgcomb. Virginia. Postdoctoral Scientist
Fair, Rebecca, Research Assistant
Gao. Lingqiu, Research Assistant II
Kim. Ulandt. Research Assistant
Kysela, David. Research Assistant
Laan. Maris. Research Assistant II

R46 Annual Report

Lim, Pauline, Executive Assistant
Luders. Bruce. Research Assistant
McArthur. Andrew. Postdoctoral Scientist
Medina, Monica, Postdoctoral Scientist
Morrison. Hilary G., Postdoctoral Scientist
Nixon. Julie. Postdoctoral Scientist
Sansone. Rebecca. Executive Assistant
Schlichter. Mimi, Executive Assistant
Shulman, Laura, Senior Research Assistant
Shakir, Muhammed Afaq. Postdoctoral Scientist

Visiting Investigators

Bahr. Michele. The Ecosystems Center
Campbell, Robert, Serono Laboratories. Inc.
Crump, Byron. The Ecosystems Center

Laboratory of Jennifer Wernegreen

The work in this lab focuses on the evolution of bacteria that
complete their life cycles within or closely related with eukaryotic host
cells. These symbiotic prokaryotes include well-known parasites as well
as obligately mutualistic bacteria that provide nutritional or other
benefits to their hosts. By virtue of their host associations,
endosymbionts may have smaller population sizes and experience
different selective forces than their free-living bacterial relatives. These
changes in population size and selection may each contribute to the
features shared by many endosymbiont genomes, such as low genomic

G + C (guanine + cytosinel contents, small genome sizes, and elevated
rates of DNA sequence evolution. Our research explores the molecular
and evolutionary mechanisms that shape these characteristics of
endosymbiont genomes, with a focus on mutualistic endosymbionts of
insects and obligate pathogens of animals.

One aim of this lab is to differentiate the effects of genetic drift,
directional mutation pressure, and natural selection on molecular
evolution of symbiotic and free-living bacteria. Our primary approach
has been to compare patterns of DNA sequence divergence at
homologous loci across symbiotic and related free-living bacterial
species. These comparisons have revealed a strong effect of genetic drift
and directional mutational pressure on sequence evolution in Buclmera
aphidicola. the vertically transmitted endosymbiont of aphids, compared
to its close free-living relative, Escherichia cn/i. Recently, our molecular
phylogenetic analyses have shown that Buclmera lacks horizontal gene
transfer that is typical of many free-living bacterial groups. On-going
and future work will explore the molecular evolution of other insect
endosymbionts in the gamma-3 Proteobacteria. including the obligate
bacterial associates of carpenter ants (Camponotus). We also employ full
genome comparisons to identify genes that have been lost in small
endosymbiont genomes, and to compare patterns of genome reduction in
mutualistic and pathogenic lineages. Of particular interest is the
substantial loss of proof-reading and DNA repair loci from several
symbiont genomes, which may elevate mutational rates and biases in
these species.

Staff

Wernegreen. Jennifer. Assistant Scientist

BioCurrents Research Center

The BioCurrents Research Center (BRC) is a national resource of the
National Institutes of Health, part of the Biomedical Technology
Resource Program of the NCRR. As with all such resources it has two
main goals: 1 1 to research and develop new biomedical technologies,
and 2) to make specialized technologies available to visiting biomedical
investigators. The emphasis of the BRC is on the physiology of cellular
transport mechanisms, particularly as they influence the boundary
conditions in the media adjacent to the plasma membrane. To this end
we develop new microsensor technologies that operate in a self-
referencing mode. We offer access to ion-selective, electrochemical, and
biosensor devices, coupled to advanced imaging techniques and
electrophysiological approaches — combinations unique to the BRC.

The BRC has seen a marked expansion in year 2000 after a
successful competitive renewal in December of 1999. This resulted in an
increase in staff, which included the appointment of two Assistant
Scientists: Stefan McDonough and Orian Shirihai. Two new postdoctoral
researchers also joined the group in 2000: Sung-Kwon Jung and
Andreas Hengstenberg. as did Laurel Moore and Robert Lewis in
support roles. Towards the end of 2000 we added Mark Messerli, who
works with both the BRC and Bay Paul Center.

The current structure of the resource comprises the core support
facility and three independent laboratories, as well as a number of
affiliate endeavors where the members of the Center work closely with
colleagues in the MBL and the regional medical schools. In particular,
we have strong links with the MBL program in Architectural Dynamics
in Living Cells, the Laboratory for Reproductive Medicine, and the Bay
Paul Center. Our involvement with regional hospitals includes Boston
Medical Center (diabetes). Massachusetts General Hospital (protein
trafficking), and Women and Infants (reproductive biology). In
summary, the core in-house research emphasis is on biophysics of
calcium transport and regulation (S. McDonough), the molecular biology

Year-Round Research R47

of transport processes (O. Shirihail. and sensor development and the
biology of transport mechanisms (P.J.S. Smith).

In addition, the BRC is developing an online database of
pharmacological compounds. The database has made considerable
progress over the past year and should be openly available by the
summer of 2001. It will be accessible through our web page at
<www.mbl.edu/BioCuirents>.

The Center supports an extensive outreach program to regional and
national universities, medical schools, and hospitals, and publishes
extensively in the field of cellular transport. Over the past year we have
continued to host a diverse group of visiting investigators whose studies
have ranged from ion transport and metabolic studies at the single cell
level to mapping ion flux associated with the olfactory sensilla of the
intact blue crab. Overall our emphasis remains on biomedical studies
using the specialized microsensors available, particularly those designed
to measure flux of calcium, potassium, hydrogen, oxygen, nitric oxide,
and ascorbate. Under development are the newer biosensors and electro-
optical probes.

The Center also maintains other core support facilities, such as a fully
equipped cell culture facility, electrode manufacture, and microinjection
systems which, as available, we also open to the general scientific
community.

Staff

Smith, Peter J.S.. Director and Senior Scientist
Hammar. Kasia. Research Assistant III
Hengstenberg, Andreas, Visiting Postdoctoral Fellow
Jung. Sung-Kwon, Postdoctoral Researcher
Lewis, Robert, Electronic Support
McDonough, Stefan. MBL Assistant Scientist
McLaughlin. Jane A.. Research Assistant HI
Messerli. Mark. NASA Research Fellow
Moore. Laurel. Database Development
Sanger. Richard H.. Research Assistant III
Shirihai, Orian, MBL Assistant Scientist

Laboratory of Stefan McDonough

Calcium ions trigger many cellular functions including
neurotransmission, muscle contraction, regulation of cell membrane
excitability, and the activation of enzymatic cascades. A major route of
calcium entry into a cell is through voltage-gated calcium ion channels,
proteins found in the plasma membrane of every excitable cell and
many nonexcitable cells. These proteins form channels that open to
allow a selective influx of calcium ions into the cell when the cell fires
an electrical spike. Calcium channels are current or potential targets for
clinical drugs treating cardiac arrhythmia, epilepsy, hypertension, pain,
diabetes, and brain damage after stroke.

Research in this laboratory focuses on the channels that conduct
calcium entry, the mechanisms controlling calcium levels within the cell,
and the tools to distinguish among different types of calcium channels.
Experiments are carried out using patch-clamp electrophysiology on
mammalian neurons, mammalian cardiac myocytes. or cloned calcium
channels expressed in nonexcitable cells. One effort, in collaboration
with the laboratories of Bruce and Barbara Furie and of Alan Rigby. is
to identify and characterize conotoxins that target voltage-gated ion
channels. Other experiments use the self-referencing ion-selective and
oxygen sensors of the BioCurrents Center, in collaboration with the
Laboratory of Peter Smith. Current areas of research include the effects
of zinc ions on calcium channels, a possible cause of ischemic neuronal
damage; calcium channel biophysics during the cardiac ventricular
action potential; the metabolic cost to the heart of maintaining calcium
homeostasis during resting and excited states; and the mechanisms of

activation of alternatively spliced forms of neuronal N-type calcium
channels.

Laboratory of Orian Shirihai

Erythroid differentiation involves expression of a set of unique
transport and enzymatic systems to support a robust induction of
hemoglobin synthesis. Active communication between the mitochondria!
matrix and cytosol is essential for heme biosynthesis. The first step,
production of aminolevulinic acid (ALA), occurs in the inner matrix.
ALA is transported to the cytosol and eventually converted to
coproporphynnogen III. which reenters the mitochondrion and. upon
further modifications, is joined with iron to form heme. This product is
then transported out of the inner matrix for assembly of cytochromes or
hemoglobin. Thus, at least four mitochondria! transport steps are
required. Although the enzymatic steps in heme synthesis are well
characterized, little is known about how biosynthetic intermediates are
shuttled across mitochondria] membranes. While malfunctioning of these
transporters most probably underlie hematologic and neurologic
diseases, their substrates are photoactivated toxic molecules used in
photo-dynamic therapy for cancer; the mechanism of transport into the
target organelle is of major interest.

A novel mitochondria! transporter, discovered by Dr. Shirihai. has
been the focus of research in the lah. This protein, named ABC-me (for
ATP-binding cassette-mitochondrial erythroid). localizes to the
mitochondria! inner membrane and is expressed at particularly high
levels in erythroid tissues of embryos and adults. ABC-me is induced
during erythroid maturation in cell lines and primary hematopoietic
cells, and its over-expression enhances hemoglobin synthesis in
erythroleukemia cells. Members of the ABC transporter superfamily
have been implicated in numerous human diseases, including cystic
fibrosis (CFTR), adrenoleukodystrophy (ALDP). Zellweger's syndrome

R48 Annual Report

(PMP70), progressive familial intrahepatic cholestasis (SPGP). and
Stargardt macular dystrophy (ABCR). To explore the functional role of
this transporter, the lab is generating a knockout mouse and cell line,
which would serve as a tool to study the biophysics and biochemistry of
this transporter as well as the phenotype appearing in the absence of this
gene. ABC-me represents a novel member of the ABC superfamily with
a potentially important role in erythroid development. In collaboration
with Dr. Weiss from the University of Pennsylvania and Dr. Orkin from
Harvard, we have recently cloned and sequenced the human homologue
of ABC-me and started screening multiple samples from candidate
patients send to us by physicians from the United States, Italy, and
England.

Laboratory of Peter J.S. Smith

The activities of this laboratory center on instrument development,
providing new insights into cellular transport mechanisms, and applying
these devices to biomedical problems. Much of the biological work is
done in collaboration with visiting investigators to the BRC. Over the
past year an increasing body of work has been undertaken using the new
amperometric microsensors capable of measuring single cell movement
of gases such as oxygen and nitric oxide. We continue to investigate the
metabolic cost of ion regulation in single cultured neurons.

In collaboration with Mitch Sogin of the Bay Paul Center, a new
research area was launched, investigating the transport physiology of
extremophilic organisms. The emphasis is to understand how membrane-
borne transport proteins continue to regulate a near neutral cytosol while
being exposed to acidic conditions of pH 1 or 2. This project is funded
through the NSF LEXEN program, attracting Mark Messerli to the
group, first as an MBL summer fellow but now on a full-time basis
funded by a NASA Fellowship.

In sensor design, we have made great progress with the new
generation of amperometric sensors, incorporating an immobilized
enzyme. Our attempts have focused on glucose and, thanks to the efforts
of Sung-Kwon Jung, our first single cell glucose flux measurements
have been achieved. Hybrid, electro-optical sensors have also been a
focus over the past year, where, working with visiting fellow Andreas
Hengstenberg, we have successfully built a micro-oxygen sensor on the
surface of a single mode fiber optic capable of stimulating a preloaded
cellular reporter molecule. In collaboration with Stefan McDonough, this
technology has been successful in imaging calcium activity while
recording oxygen uptake from a single cardiac myocyte.

Boston University Marine Program

Faculty

Atema. Jelle. Professor of Biology, Director
Dionne, Vincent, Professor of Biology
Golubic. Stjepko, Professor of Biology
Kaufman, Les. Associate Professor of Biology
Lobel. Phillip. Associate Professor of Biology
Voigt, Rainer. Research Associate Professor
Ward, Nathalie. Lecturer

Staff

Decarie. Linette. Senior Staff Coordinator
DiNunno. Paul, Research Assistant, Dionne Lab
Hall, Sheri. Program Manager
McCafferty. Michelle. Administrative Assistant
Gilbert, Niki. Program Assistant

Postdoctoral In vestigators

Grasso. Frank. Atema Laboratory
Kaatz. Ingrid, Lobel Laboratory
Trott. Thomas, Atema Laboratory

Visiting Fucitltv and Investigators

Hanlon, Roger. Marine Biological Laboratory
Hecker, Barbara, Hecker Consulting
Moore. Michael. Woods Hole Oceanographic Institution
Simmons, Bill. Sandia National Laboratory
Wamwright. Norman, Marine Biological Laboratory

Graduate Students
PhD Students

Existing
Cole. Marci
Dale, Jonathon
Dooley. Brad
Hauxwell, Jennifer
Herrold. Ruth
Kroeger, Kevin
Ma. Diana
Miller, Carolyn
Oliver. Steven
Stieve. Erica
Tomasky. Gabrielle
York, Joanna
Zettler, Erik

New

Frenz. Christopher

Skomal. Gregory

Masters Students

Existing

Allen, Christel
Atkinson, Abby
Bentis. Christopher
Bowen, Jennifer
Casper. Brandon
Cavanaugh. Joseph
Chichester. Heather
D'Ambrosio. Alison
Errigo, Michael
Evgenidou. Angeliki
Fredland. Inga
Frenz. Christopher
Grable. Melissa
Grebner. Dawn
Kollaros. Maria
Konkle, Anne
Lamb. Amy
Lawrence, David
Lever. Mark
Levine, Michael
Malley. Vanessa
Martel. David
Neviackas, Justin
Oweke. Ojwang William
Perez. Edmundo
Pugh. Tracy

Year-Round Research R49

Ramon, Marina
Ripley, Jennifer
Roycrot't. Karen
Smith. Spence
Stueckle, Todd
Sweeny. Melissa
Tuohy-Sheen, Elizabeth
Watson, Elise
Weiss. Erica
Wright, Dana

New

Bogomolni. Andrea
Bonacci. Lisa
deHart. Pieter
Estrada. James
Rice. Aaron
Rutecki. Deborah
Shriver, Andrea
Summers. Erin
Wittenmyer. Robert

Undergraduate Students

Spring 00
Kwong. Grace
Loewensteiner. David

Fall 00

Batson. Melissa
Bergan. Michael
Boynton. Seth
Burke, Alexandra
Buynevitch, Artem
Christie. Mark
Combellick. Lindsay
Darrell. Andrea
De Falco, Tomaso
Dewey. Hollis
Faloon, Kristine
Feeney, Brett
Hendricks. Amber
Hunt. Tamah
Kavountzis, Erol
Kim, Joanne
Kowalchuk. Lynn
Linehan, Candace
Lynch. Michael
Mattei, Bethany
McKay. Breda
McOwen. Micah
Miller, Jessica
Morano. Janelle
Newville. Melinda
Nichols, Dominica
Nguyen, Jean
O'Neil. Diane
Rohrbaugh, Lynne
Sorocco, Debra
Tubbs. Mollie
Wezensky, Eryn
Yopak, Kara
Zink. Rachel

Summer 2000 Interns
O'Connell. Timmy

Laboratory of Jelle Ate in a

Many organisms and cellular processes use chemical signals as their
main channel of information about the environment. All environments
are noisy and require some form of filtering to detect important signals.
Chemical signals are transported by turbulent currents, viscous flow, and
molecular diffusion. Receptor cells extract chemical signals from the
environment through various filtering processes. In our laboratory, fish,
marine snails, and Crustacea have been investigated for their ability to
use chemical signals under water. Currently, we use the lobster and its
exquisite senses of smell and taste as our major model to study the
signal-filtering capabilities of the whole animal and its narrowly tuned
chemoreceptor cells.

Research in our laboratory focuses on amino acids, which represent
important food signals for the lobster, and on the function and chemistry
of pheromones used in lobster courtship. We examine animal behavior
in the sea and in the lab. This includes social interactions and
chemotaxis. To understand the role of chemical signals in the sea we
use real lobsters and untethered small robots. Our research includes
measuring and computer modeling odor plumes and the water currents
lobsters generate to send and receive chemical signals. Other research
interests include neurophysiology of receptor cells and anatomical
studies of receptor organs and pheromone glands.

Laboratory of Vincent Dionne

Odors are powerful stimuli. They can focus the attention, elicit
behaviors (or misbehaviors), and even resurrect forgotten memories.
These actions are directed by the central nervous system, but they
depend upon the initial transduction of chemical signals by olfactory
receptor neurons in the nasal passages. More than just a single process
appears to underlie odor transduction, and the intracellular pathways that
are used are far more diverse than once thought. Hundreds of putative
odor receptor molecules have been identified that work through several
different second messengers to modulate the activity of various types of
membrane ion channels.

Our studies are being conducted with aquatic salamanders using
amino acids and other soluble chemical stimuli that these animals
perceive as odors. Using electrophysiological and molecular approaches,
the research examines how these cellular components produce odor
detection, and how odors are identified and discriminated.

R50 Annual Report

Laboratory of Les Kaufman

Current research projects in the laboratory deal with speciation and
extinction dynamics of haplochromine fishes in Lake Victoria. We are
studying the systematics, evolution, and conservation genetics of a
species flock encompassing approximately 700 very recently evolved
taxa in the dynamic and heavily impacted landscape of northern East
Africa. In the lab we are studying evolutionary morphology, behavior.
and systematics of these small, brightly colored cichlid fishes.

Another area of study is developmental and skeletal plasticity in
fishes. We are studying the diversity of bone tissue types in fishes,
differential response to mineral and mechanical challenge, and
matrophic versus environmental effects in the development of coral reef
fishes.

We also study the biological basis for marine reserves in the New
England fisheries. We are involved in collaborative research with
NURC. NMFS, and others studying the relative impact on groundfish
stocks of juvenile habitat destruction versus fishing pressure.

Laboratory of Phillip Label

Fishes are the most diverse vertebrate group and provide opportunities
to study many aspects of behavior, ecology, and evolution. We primarily
study 1) how fish are adapted to different habitats, and 2) behavioral
ecology of species interactions. Current research focuses on fish acoustic
communications.

We are also conducting a long-term study of the marine biology of
Johnston Atoll, Central Pacific Ocean. Johnston Atoll has been occupied
continuously by the military since the 1930s and has proven to be a
unique opportunity for assessing the biological impacts of island
industrialization and its effects on reefs. Johnston Atoll is the site of the
U.S. Army's chemical weapons demilitarization facility, JACADS.

Laboratory of Ivan Valiela

A focus of our work is the link between land use on watersheds and
consequences in the receiving estuarine ecosystems. The work examines
how landscape use and urbanization increase nutrient loading to
groundwater and streams. Nutrients in groundwater are transported to
the sea, and. after biogeochemical transformation, enter coastal waters.
There, increased nutrients bring about a series of changes on the
ecological components. To understand the coupling of land use and
consequences to receiving waters, we study the processes involved,
assess ecological consequences, and define opportunities for coastal
management.

A second long-term research topic is the structure and function of salt
marsh ecosystems, including the processes of predation. herbivory.
decomposition, and nutrient cycles.

Center for Advanced Studies in the
Space Life Sciences

In 1^45. the NASA Life Sciences Division and the Marine Biological
Laboratory established a cooperative agreement with the formation of
the Center for Advanced Studies in the Space Life Sciences (CASSLS at
MBL). The Center's overall goals are to increase awareness of the
NASA Life Sciences Program within the basic science community and
to examine and discuss potential uses of microgravity and other aspects
of spaceflight as probes to provide new insights to fundamental
processes important to basic biology and medicine.

Through symposia, workshops and seminars, CASSLS advises NASA
and the biological science community on a wide variety of topics.

Through fellowships. CASSLS supports summer research for
investigators in areas pertinent to the aims of NASA life sciences.

Since the Center began its operations in July 1995. more than 400
people have attended eight CASSLS workshops. Typically these
workshops last for two to four days and feature an international array of
scientists and NASA/International space agency staff. In many cases,
workshop chairs have a long-time association with the MBL. Workshop
schedules incorporate many opportunities for interaction and discussion.
A major outcome for workshops is the publication of proceedings in a
peer-reviewed journal. Moreover, our meetings introduce outstanding
biologists to research questions and prominent scientists involved in
gravitational biology and the NASA Life Sciences Program.

The Center sponsored one workshop in 2000: "Invertebrate Sensory
Information Processing: Implications for Biologically Inspired
Autonomous Systems," chaired by Dr. Frank Grasso. The Center
sponsored one Fellow during the summer of 2000: Dr. Mark Messerli,
Biology Department. Purdue University. He conducted research in
reaulation of cytoplasmic pH in eucaryotic acidophiles in collaboration
with Dr. Peter J.S. Smith and Dr. Mitchell Sogin of the Marine
Biological Laboratory. In addition, two scholars-in-residence worked
with the Center in 2000: Dr. Richard Wassersug of Dalhousie University
and Dr. Lawrence Schwartz of the University of Massachusetts,
Amherst. Finally, the Center worked with colleagues in Astrobiology
and the Josephine Bay Paul Center to offer a stimulating lecture series.

Staff

Blazis, Diana E.J., Director

Oldham. Pamela A., Administrative Assistant

Scholars-in-residence

Schwartz, Lawrence
Wassersug. Richard

The Ecosystems Center

The Ecosystems Center carries out research and education in
ecosystems ecology. Terrestrial and aquatic scientists work in a wide
variety of ecosystems ranging from the streams, lakes and tundra of the
Alaskan Arctic (limits on plant primary production) to sediments of
Massachusetts Bay (controls of nitrogen cycling), to forests in New
England (effects of soil warming on carbon and nitrogen cycling), and
South America (effects on greenhouse gas fluxes of conversion of rain
forest to pasture) and to large estuaries in the Gulf of Maine (effects on
plankton and benthos of nutrients and organic matter in stream runoff).
Many projects, such as those dealing with carbon and nitrogen cycling
in forests, streams, and estuaries, use the stable isotopes I3C and LN to
investigate natural processes. A mass spectrometer facility is available.
Data from field and laboratory research are used to construct
mathematical models of whole-system responses to change. Some ot
these models are combined with geographically referenced data to
produce estimates of how environmental changes affect key ecosystem
indexes, such as net primary productivity and carbon storage, throughout
the world's terrestrial biosphere.

The results of the Center's research are applied, wherever possible, to
the questions of the successful management of the natural resources ot
the earth. In addition, the ecological expertise of the staff is made
available to public affairs groups and governmental agencies who deal
with problems such as acid rain, coastal eutrophication, and possible
carbon dioxide-caused climate change.

The Semester in Environmental Science was offered again in Fall

Year-Round Research R51

2000. Fifteen students from seven colleges participated in the program.
The center also offers opportunities for postdoctoral fellows.

Administrative Staff

Hobble, John E., Co-Director

Melillo. Jerry M., Co-Director

Foreman, Kenneth H.. Associate Director. Semester in Environmental

Studies

Berthel, Dorothy J.. Administrative Assistant
Donovan. Suzanne J.. Executive Assistant
Moniz. Priscilla C.. Administrative Assistant, Semester in Environmental

Studies

Nunez, Guillermo. Research Administrator
Scanlon, Deborah G., Executive Assistant
Seifert. Mary Ann, Administrative Assistant

Scientific Staff

Hobbie. John E.. Senior Scientist
Melillo. Jerry M.. Senior Scientist
Deegan, Linda A.. Associate Scientist
Giblin. Anne E.. Associate Scientist
Herbert. Darrell A.. Staff Scientist
Holmes. Robert M., Staff Scientist
Hopkinson. Charles S.. Senior Scientist
Hughes. Jeffrey E.. Staff Scientist
Nadelhoffer. Knute J., Senior Scientist
Neill, Christopher, Assistant Scientist
Peterson. Bruce J., Senior Scientist
Rastetter. Edward B., Associate Scientist
Shaver, Gaius R., Senior Scientist
Steudler. Paul A., Senior Research Specialist
Tian, Hanqin. Staff Scientist
Vallino, Joseph J., Assistant Scientist
Williams. Mathew. Assistant Scientist

Educational Staff Appointments

Buzby. Karen. Postdoctoral Scientist
Cieri. Matthew D.. Postdoctoral Scientist
Crump, Byron. Postdoctoral Scientist
Garcia-Montiel. Diana C., Postdoctoral Scientist
LeDizes-Maurel, Severine, Postdoctoral Scientist
Kappel-Schmidt, Inger, Postdoctoral Scientist
Nordin. Annika, Postdoctoral Scientist
Raymond. Peter. Postdoctoral Scientist
Sommerkom. Martin. Postdoctoral Scientist
Tobias. Craig R., Postdoctoral Scientist
Williams, Michael R., Postdoctoral Scientist

Technical Staff

Ahrens, Toby. Research Assistant

Bahr. Michele P.. Research Assistant

Bettez. Neil D., Research Assistant

Burnette, Donald W.. Research Assistant

Claessens. Lodevicus H.J.M., Research Assistant

Colman. Benjamin P.. Research Assistant

Eldridge. Cynthia. Research Assistant

Fox, MaryKay. Research Assistant

Garritt. Robert H., Senior Research Assistant

Gay, Marcus O., Research Assistant

Goldstein. Joshua H.. Research Assistant

Jablonski. Sarah A., Research Assistant
Jillson. Tracy A., Research Assistant
Kelsey, Samuel. Research Assistant
Kicklighter, David W., Senior Research Assistant
Kwiatkowski. Bonnie L., Research Assistant
Laundre. James A.. Senior Research Assistant
Lezberg, Ann. Research Assistant
Lux, Heidi, Research Assistant
Merson. Rebekah. Research Assistant
Micks, Patricia, Research Assistant
Morriseau, Sara. Research Assistant
Nolin. Amy L.. Research Assistant
Nowicki, Genevieve, Research Assistant
O'Brien. Kathenne A.. Research Assistant
Otter, Marshall L., Research Assistant
Pan, Shufen, Research Assistant
Peterson, G. Gregory, Research Assistant
Regan. Kathleen M., Research Assistant
Ricca, Andrea. Research Assistant
Schwamb, Carol. Laboratory Assistant
Slavik. Karie A.. Research Assistant
Thieler, Kama K., Research Assistant
Tholke, Kristin S.. Research Assistant
Thomas. Suzanne M.. Research Assistant
Tucker, Jane, Senior Research Assistant
Vasiliou, David S., Research Assistant
Weston. Nathaniel B., Research Assistant
Wright. Amos, Research Assistant
Wyda. Jason C.. Research Assistant

Consultants

Bowles. Francis P., Research Systems Consultant
Bowles, Margaret C.. Administrative Consultant

Visiting Scientists and Scliolars

DeStasio. Bart, SES Faculty Fellow, Lawrence College

Koba. Keisuke, Graduate School of Informatics, Kyoto University. Japan

Laboratory of Aquatic Biomedicine

Our laboratory has developed the Spisula solidissima embryo model
to study mechanisms of neurotoxicology. We have shown that
polychlonnated biphenyls (PCBs) selectively target the nervous system
during development. We are now linking up and down regulation of the
p53 family of genes with neuronal cell development using new probes
developed by this laboratory.

In the second line of research, we are using the clam leukemia model
to investigate how environmental chemicals influence the progression of
leukemia. Further, we are studying whether mutations in p53 alter the
pathogenesis of leukemia in populations of Mya arenaria. Field work to
Nova Scotia showed that leukemia in Mya was also detected in Sydney.
N.S., which is heavily polluted with a variety of industrial chemicals.

Staff

Reinisch. Carol L., Senior Scientist
Cox, Rachel. Postdoctoral Scientist
Jessen-Eller. Kathryn, Postdoctoral Scientist
Kreiling. Jill. Postdoctoral Scientist
Stephens, Ray, Adjunct Scientist

R52 Annual Report

Visiting Scientists

Shohel, Stephen, University of California, San Francisco
Walker, Charles. University of New Hampshire

Student

Miller. Jessica. Boston University

Laboratory of Cell Communication

completed pollen and stratigraphic analyses, now being prepared for
publication, of the first transglacial lake core from a forested site
(Maicuru inselberg) in the eastern Amazon lowlands. Our collaborators
at the Florida Institute of Technology and the University of Cincinnati
identified chemical changes in the early sedimentary history of Lake
Pata in the western Amazon lowlands that show a strong synchroneity
with insolation changes associated with the precessional component of
astronomical climate forcing back to marine oxygen isotope stage 7, this
being the first such signal from the equatorial lowlands. In 2000 we also
concluded a paleoenvironmental reconnaissance of the Lake Nicaragua
region and are developing plans for raising a long core from the lake.

Established in 1994, this laboratory is devoted to the study of
intercellular communication. The research focuses on the cell-to-cell
channel, a membrane channel built into the junctions between cells. This
channel provides one of the most basic forms of intercellular
communication in organs and tissues. The work is aimed at the
molecular physiology of this channel, in particular, at the mechanisms
that regulate the communication. The channel is the conduit of growth-
regulating signals. It is instrumental in a basic feedback loop whereby
cells in organs and tissues control their number; in a variety of cancer
forms it is crippled.

This laboratory has shown that transformed cells lacking
communication channels lost the characteristics of cancer cells, such as
unregulated growth and tumorigenicity. when their communication was
restored by insertion of a gene that codes for the channel protein. Work
is now in progress to track the channel protein within the cells from its
point of synthesis, the endoplasmic reticulum. to its functional
destination in the plasma membrane, the cell-to-cell junction, by
expressing a fluorescent variant of the channel protein in the cells.
Knowledge about the cellular regulation of this process will aid our
understanding of what goes awry when a cell loses the ability to form
cell-to-cell channels and thus to communicate with its neighbors,
thereby taking the path towards becoming cancerous.

Another line of work is taking the first steps at applying information
theory to the biology of cell communication. Here, the intercellular
information spoor is tracked to its source: the macromolecular
intracellular information core. The outlines of a coherent information
network inside and between the cells are beginning to emerge.

Staff

Loewenstein. Werner, Senior Scientist
Rose, Birgit, Senior Scientist
Jillson, Tracy, Research Assistant

Laboratory of Paul Colinvaux

Staff

Colinvaux. Paul. Adjunct Scientist

Laboratory of Ayse Dosemeci

The laboratory investigates molecular processes that underlie synaptic
modification. The main project is to clarify how the frequency of
activation at a synapse can determine whether the synapse will he
potentiated (strengthened) or depressed (weakened) through the
participation of an enzyme called CaM kinase II. This enzyme is
regulated by autophosphorylation on distinct sites in the presence and
absence of calcium. Biochemical studies with isolated postsynaptic
density fractions are conducted to clarity functional consequences of
CaMKII autophosphorylation in response to sequential exposure to
calcium-containing and calcium-free media at different temporal
patterns.

In a related project, a new affinity-based method is being developed
for the preparation of postsynaptic density fractions of high purity. In
collaboration with Dr. Lucas Pozzo-Miller (University of Alabama.
Birmingham), we are studying changes in the activity of CaMKII in
hippocampal slices following high-frequency and low-frequency
electrical stimulation to generate long-term potentiation and long-term
depression, respectively. Related projects in collaboration with Dr.
Thomas Reese (NIH. NINDS) include studies on the redistribution of
CaMKII and structural changes in the post-synaptic density in response
to excitatory stimuli.

Staff

Dosemeci. Ayse. Adjunct Scientist

Visiting Invcstigatur

Pozzo-Miller. Lucas, University of Alabama

We have shown that accumulated pollen data now leave little doubt
that the Amazon lowlands remained forested without fragmentation
throughout glacial cycles. Changes in relative abundance of trees within
the highly diverse forests can be seen in the pollen record, however,
particularly in response to changing temperature. The pollen vocabulary
for the Amazon on which this conclusion is based has been codified in
our Amazon Pol/en Manual and Atlas with text in Portuguese as well as
English for the benefit of Brazilian researchers. We show that the
Amazon ecosystems yield two kinds of pollen signals: what might be
called the "classical" signal by wind-blown pollen, which allows
separation of biomes and many edaphically constrained facies of
Ama/on forests such as var-ea or igapo; and a species-rich signal from
animal-pollinated trees washed from the immediate watershed or
catchment of the sedimentary basin.

With our collaborators in Brazil and the University of Florida we

Laboratory of Barbara Furie and Bruce Furie

y-Carboxyglutamic acid is a calcium-binding amino acid that is found
in the conopeptides of the predatory marine cone snail, Conus. This
laboratory has been investigating the biosynthesis of this amino acid in
Conus and the structural role of •y-carboxyglutamic acid in the
conopeptides. This satellite laboratory relates closely to the main
laboratory, the Center for Hemostasis and Thrombosis Research, on the
Harvard Medical School campus in Boston: the main focus of the
primary laboratory is the synthesis and function of y-carboxyglutamic
acid in blood-clotting proteins and the role of vitamin K.

Cone snails are obtained from the South Pacific and maintained in the
Marine Resources Center. Until recently, the marine cone snail had been
the sole invertebrate known to synthesize y-carboxyglutamic acid (Gla).

Year-Round Research R53

The venomous cone snail produces neurotoxic conopeptides, some rich
in Gla, which it injects into its prey to immobilize it. To examine the
biosynthetic pathway for Gla. we have studied the Comix carboxylase
which converts glutamic acid to y-carboxyglutamic acid. This activity
has an absolute requirement for vitamin K. The Conux carboxylase
substrates contain a carboxylation recognition site on the conotoxin
precursor. Given the functional similarity of mammalian vitamin K-
dependent carboxylases and the vitamin K-dependent carboxylase from
Conns textile, we hypothesized that structurally conserved regions would
identity sequences critical to this common functionality.

Furthermore, we examined the diversity of animal species that maintain
vitamin K-dependent carboxylation to generate y-carboxyglutamic acid. We
have cloned full-length carboxylase homologs from the beluga whale
(Delphinaptenis leitcas) and toadfish (Opsanus tail}. In addition, we have
partially cloned the carboxylase gene from chicken (Gal/ns gallus), hagfish
(\l\\inc glutinosa), horseshoe crab (Limulus polyphemus), and cone snail
(Conns textile) in order to compare these structures to the known bovine,
human, rat. and mouse cDNA sequences. Comparison of the predicted
amino acid sequences identified a highly conserved 32-amino acid residue
region in all of these putative carboxylases. In addition, this amino acid
motif is also present in the Drosophila genome and identified a Drosophihi
homolog of the y-carboxylase. Assay of hagfish liver and Drosophila
demonstrated carboxylase activity in these non-vertebrates. More recently,
we hu\e cloned the entire vitamin K-dependent carboxylase gene from the
cone snail. Conns textile. The predicted amino acid sequence shows most
region-, are similar to the mammalian sequence, and that there is about 40%
sequence identity overall. These results demonstrate the broad distribution
of the vitamin K-dependent carboxylase gene, including a highly conserved
motif that is likely critical for enzyme function. The vitamin K-dependent
biosynthesis of •y-carboxyglutamic acid appears to be a highly conserved
function in the animal kingdom.

Novel y-carboxyglutamic acid-containing conopeptides have been
isolated from the venom of Conns textile. The amino acid sequence, amino
acid composition, and molecular weights of these peptides have been
determined. For several peptides. the cDNA encoding the precursor
conotoxin has been cloned. The three-dimensional structure of some of
these Gla-containing conopeptides are being determined by 2D NMR
spectroscopy. Complete resonance assignments of conotoxin P14.1 were
made from 2D 'H NMR spectra via identification of intraresidue spin
systems using 'H-'H through-bond connectivities. NOESY spectra provided
dN, dNN. and dN NOE connectivities and vicinal spin-spin coupling
constants 3JHNu were used to calculate <t> torsion angles. Structure
determination is nearing completion. The goal of this project is to determine
the structural role of y-carboxyglutamic acid in the Gla-containing
conotoxins and other y-carboxyglutamic acid-containing proteins.

Staff

Fune. Barbara C. Adjunct Scientist
Furie. Bruce. Adjunct Scientist
Begley. Gail, Scientist I
Czerwiec. Eva, Postdoctoral Fellow
Rigby, Alan. Adjunct Scientist
Stenflo. Johan. Visiting Scientist

Laboratory of Shiny a I none

Scientists in this laboratory study the molecular mechanism and
control of mitosis, cell division, cell motility, and cell morphogenesis,
with emphasis on biophysical studies made directly on single living
cells, especially developing eggs in marine invertebrates. Development
of biophysical instrumentation and methodology, such as the centrifuge
polarizing microscope, high-extinction polarization optical and video

microscopy, digital image processing techniques including dynamic
stereoscopic imaging, and exploration of their underlying optical theory
are an integral part of the laboratory's efforts.

Staff

Inoue, Shinya, Distinguished Scientist

Burgos. Mario, Visiting Scientist

Goda. Makoto. Visiting Scientist

Baraby. Diane. Laboratory Assistant

Knudson. Robert. Instrument Development Engineer

MacNeil, Jane. Executive Assistant

Laboratory of Rudolf Oldenbourg

The laboratory investigates the molecular architecture of living cells
and of biological model systems using optical methods for imaging and
manipulating these structures. For imaging cell architecture non-
invasively and non-destructively. dynamically and at high resolution, we
have developed a new polarized light microscope (Pol-Scope). The Pol-
Scope combines microscope optics with new electro-optical components,
video, and digital image processing for fast analysis of specimen
birefringence over the entire viewing field. Examples of biological
systems currently investigated with the Pol-Scope are microtubule-based
structures (asters, mitotic spindles, single microtubules); actin-based
structures (acrosomal process, stress fibers, nerve growth cones); zona
pellucida of vertebrate oocytes; and biopolymer liquid crystals.

Staff

Oldenbourg. Rudolf, Associate Scientist

Shribak. Michael. Staff Scientist

Knudson. Robert. Instrument Development Engineer

Baraby, Diane. Laboratory Assistant

Laboratory of Michael Rabinowitz

This laboratory investigates environmental geochemistry and
epidemiology. Areas of recent activity include modeling lead
bioavailability, writing a history of lead biokinetic models, performing a
case control survey of tea drinking and oral cancer in Taiwan,
quantifying the transport and fate of various sources of residential lead
exposure, and serving on several advisory boards of Superfund research
projects in Boston and New York.

Current activity focuses on characterizing lead paints and pigments.
Hundreds of lead poisoning lawsuits are filed every year against
landlords, but no compensation has ever been paid by the half dozen
companies that made lead pigments, because it has not been possible to
identify the specific manufacturer. This research has been funded by the
Eagle Picher Trust. Other activity, sponsored by HLID. involves using
stable isotopes of lead to determine the relative importance of various
household surfaces (doors, floors, windows, walls) as sources of indoor
dust lead levels. Dust lead is the major predictor of childhood lead
exposure and poisoning. This would allow for more focused deleading.

Another effort has been using historical fire insurance maps to locate
and identify unrecognized hazardous waste sites.

Staff

Rabinowitz. Michael. Associate Scientist

R54 Annual Report

Laboratory for Reproductive Medicine, Brown

University and Women and Infants Hospital,

Providence

Work in this laboratory centers on investigating cellular mechanisms
underlying female infertility. Particular emphasis is placed on the
physiology of the oocyte and early embryo, with the aim of assessing
developmental potential and mitochondria dysfunction arising from
mtDNA deletions. The studies taking place at the MBL branch of the
Brown Laboratory use some of the unique instrumentation available
through the resident programs directed by Rudolf Oldenbourg and Peter
J.S. Smith. Most particularly, non-invasive methods for oocyte and
embryo study are being sought. Of several specific aims, one is to use
the Pol-Scope to analyze the dynamic birefringence of meiotic spindles.
An additional aim is to study transmembrane ion transport using non-
invasive electro-physiological techniques available at the BioCurrents
Research Center. The newly developed oxygen probe offers the
possibility of looking directly at abnormalities in the mitochondria
arising from accumulated mtDNA damage. Our laboratory has also
focused on studying the mechanism underlying age-associated infertility
in terms of oocyte quality and has attempted to rescue developmentally
compromised oocytes or embryos through nuclear-cytoplasmic transfer
technology. We have characterized oxidative stress-induced
mitochondrial dysfunctions, developmental arrest, and cell death in early
embryos using animal models. Ultimately, this laboratory aims to
produce clinical methods for assessing preimplantation embryo viability.
an advance that will significantly contribute to the health of women and
children.

Staff

Keefe, David. Director
Liu. Lin, Adjunct Scientist
Trimarchi, James, Adjunct Scientist

Laboratory of Osainu Shimomitra

Aequorin, from the jellyfish Aequorea aequorea, was the first
calcium-sensitive photoprotein discovered by us in 1961. Because of its
high sensitivity to Ca2+ and biological harmlessness. the protein has
been widely used as a probe to monitor intracellular free calcium levels.
Aequorin is a unique protein that contains a high level of energy for
light emission in the molecule, and its structure has been the target of
many studies in the past. The complete 3-dimensiona] structure of
aequorin was finally obtained by X-ray crystallography 38 years after its
discovery, in collaboration with three other laboratories. Aequorin is
found to be a globular molecule having four helix-loop-helix "EF-hand"
domains, of which three can bind Ca2 + . The molecule contains
coelenterazine-2-hydroperoxide in its hydrophobic core cavity, as the
chromophoric ligand which decomposes into coelenteramide and carbon
dioxide accompanied by the emission of blue light.

Staff

Shimomura. Osamu. Senior Scientist, MBL, and Boston University

School of Medicine
Shimomura. Akemi. Research Assistant

Laboratory of Robert B. Silver

The members of this laboratory study how living cells make
decisions. The focus of the research, typically using marine models, is
on two main areas: the role of calcium in the regulation of mitotic cell
division (sea urchins, sand dollars, etc.) and structure and function
relationships of hair cell stereociliary movements in vestibular
physiology (oyster toadfish). Other related areas of study, i.e. synaptic
transmission (squid), are also, at times, pursued. Tools include video
light microscopy, multispectral, subwavelength, and very high-speed
(sub-millisecond frame rate) photon counting video light microscopy,
telemanipulation of living cells and tissues, and modeling of decision
processes. A cornerstone of the laboratory's analytical efforts is high
performance computational processing and analysis of video light
microscopy images and modeling. With luminescent, fluorescent, and
absorptive probes, both empirical observation and computational
modeling of cellular, biochemical, and biophysical processes permit
interpretation and mapping of space-time patterns of intracellular
chemical reactions and calcium signaling in living cells. A variety of in
vitro biochemical, biophysical, and immunological methods are used. In
addition to fundamental biological studies, the staff designs and
fabricates optical hardware, and designs software for large video image
data processing, analysis, and modeling.

Staff

Silver. Robert, Associate Scientist

Visiting Investigators

Hummel, John, Argonne National Laboratory
Jiang, Yi, Los Alamos National Laboratory
Keller, Bruce. SUNY Upstate Medical University
Kriebel, Mahlon, SUNY Upstate Medical University
Pappas, George, University of Illinois School of Medicine
Pearson. John. Los Alamos National Laboratory

Laboratory of Norman Wainwright

The mission of the laboratory is to understand the molecular defense
mechanisms exhibited by marine invertebrates in response to invasion
by bacteria, fungi, and viruses. The primitive immune systems
demonstrate unique and powerful strategies for survival in diverse
marine environments. The key model has been the horseshoe crab
Limitliis polyphemus. tinnitus hemocytes exhibit a very sensitive LPS-
triggered protease cascade which results in blood coagulation. Several
proteins found in the hemocyte and hemolymph display microbial
binding proteins that contribute to antimicrobial defense. Commensal or
symbiotic microorganisms may also augment the antimicrobial
mechanisms of macroscopic marine species. Secondary metabolites are
being isolated from diverse marine microbial strains in an attempt to
understand their role. Microbial participation in oxidation of the toxic
gas hydrogen sulfide is also being studied.

Staff

Wainwright. Norman. Senior Scientist
Child, Alice. Research Assistant
Williams. Kendra, Research Assistant

Visiting Investigator

Anderson, Porter. University of Rochester

Year-Round Research R55

Laboratory of Seymour Ziginan

This laboratory is investigating basic mechanisms of photooxidative
stress to the ocular lens due to environmentally compatible UVA radiation.
This type of oxidative stress contributes to human cataract formation. Other
studies are the search for and use of chemical antioxidants to retard the
damage that occurs. Cultured mammalian lens epithelial cells and whole
lenses in vitro are exposed to environmentally compatible UVA radiation
with or without previous antioxidant feeding. The following parameters of
lens damage are examined: molecular excitation to singlet states via
NADPH (the absorber): cell growth inhibition and cell death; catalase
maclivauon: cytoskeletal description (of actin. tubulin. integrins); and cell
membrane damage (lipid oxidation, loss of gap junction integrity and
intercellular chemical communications). Thus far. the most successful
antioxidant to reduce these deficiencies is alpha-tocopherol ( 10 ^ig/ml) and
tea polvphenols (especially from green tea). The preliminary phases of the
research are usually carried out using marine animal eyes (i.e., smooth
dogfish) as models. Our goal is to provide information that will suggest
means to retard human cataract formation.

Staff

Zigman. Seymour. Laboratory Director, Professor of Ophthalmology.

Boston University Medical School

Rafferty. Keen, Research Associate, Boston University Medical School
Rafferty. Nancy S.. Research Associate. Boston University Medical School
Zigman. Bunnie R., Laboratory Manager, Boston University Medical

School

The Marine Resources Center

The Marine Resources Center (MRC) — a modern, 32.000-square-foot
structure — features advanced facilities for maintaining and culturing
aquatic organisms essential to advanced biological, biomedical. and
ecological research. In addition to research, the MRC provides a variety
of important, complementary services to the MBL community through
its Aquatic Resources Division, its Aquaculture and Engineering
Division, and its administrative division.

The MRC and its life support systems have increased the ability of
MBL scientists to conduct research and have inspired new concepts in
scientific experiments. Vigorous research programs focusing on basic
biological and biomedical aquatic models are currently being developed
at the Center, including the Program in Scientific Aquaculture and the
Program in Sensory Biology and Neuroethology.

Research and educational opportunities for established investigators,
postdoctoral fellows, and graduate and undergraduate students are

available at the MRC. Investigators and students find that the MRC's
unique life support and seawater engineering systems make this a
favorable environment in which to conduct research using a variety of
aquatic organisms and flexible tank space for customized
experimentation on live animals.

Staff

Hanlon. Roger. Director and Senior Scientist

Carroll. James. Life Support Technical Assistant

Enos, Edward. Aquatic Resources Division Superintendent

Gilland, Edwin. Research Associate

Grossman, William, Marine Specimen Collector/Diving Safety Officer

Hanley. Janice. Water Quality and Animal Health Technician

Klimm. William. Licensed Boat Captain — R/V Gemma

Kuzirian, Alan. Associate Scientist

Linnon. Beth. Special Projects Coordinator

Mebane. William. Aquaculture and Engineering Division Superintendent

Santore. Gabrielle. Executive Assistant

Sexton, Andrew. Marine Organism Shipper

Smolowitz. Roxanna, MBL Veterinarian

Sullivan. Daniel, Boat Captain

Tassinari, Eugene. Senior Biological Collector

Whelan. Sean. Diver/Marine Specimen Collector

Summer and Full Employees and Volunteers

Buynevich. Artem, Work-study Student. Boston University

Carroll. Amanda, Volunteer

Dimond, Jay, Diver/Collector

Douton, Kate, AmeriCorps Assistant

Faloon. Kristine. Work-study Student. Boston University

Gudas. Chris. Diver/Collector

Kavountzis. Erol. Work-study Student, Boston University

Miraglia, Valentina. Volunteer. Universita di Napoli "Federico II." Italy

Potter. Chris, Diver/Collector

Reynolds, Justin, Diver/Collector

Robhins, Gillian. Volunteer

Rohrbaugh. Lynne. Work-study Student. Boston University

Tubbs, Mollie. Work-study Student. Boston University

Zucchini, Mossimo, Volunteer. Universita di Napoli "Federico II," Italy

Laboratory of Roger Hanlon

This laboratory investigates the behavior of cephalopods and other
marine organisms with an integrative biology approach focused at the
organismal level. Molecular, cellular, and ecological approaches are
used to complement this organismal approach, and there is emphasis on
sensory biology and behavioral ecology.

Laboratory studies on the mechanisms and functions of polarized light
sensiiivity in cephalopods are underway. Olfactory sensing by Nautilus
(which functions in food detection and location as well as mate choice)
is being studied in the laboratory. Visual features that octopuses use for
maze learning are also being investigated. Lab experiments in large
indoor seawater tanks are being conducted to determine how male
squids. Laligo pealeii, use visual, then contact, chemical cues in egg
capsules to initiate highly robust agonistic behavior.

The functional morphology and neurobiology of the chromatophore
system of cephalopods are studied on a variety of cephalopod species,
and image analysis techniques are being developed to study crypsis and
the mechanisms that enable cryptic body patterns to be neurally
regulated by visual input. Various aspects of predation, antipredator
defenses, and reproduction are conducted in field sites worldwide.

Sexual selection theory is being tested using squid and cuttlefish.
Field and laboratory studies focus on mechanisms of agonistic behavior.

R56 Annual Report

female mate choice, and sperm competition. The latter studies involve
DNA Fingerprinting to determine paternity and help assess alternative
mating tactics.

Population structure and reproductive success in several highly
valuable squid fisheries (Loligo vulgaris reynaudii in South Africa,
Loligo pealeii in the N.E. United States. Loligo opalescens in
California) are being assessed for fishery management and conservation.
We also culture species of commercial and biomedical importance. For
example, the toadfish Opsanus beta is used in vestibular research related
to human medicine, yet the species is difficult to obtain from nature.
Thus, we are performing the first mariculture experiments to culture
toadfish through the life cycle to provide the biomedical community
with high-quality experimental animals. Such an approach lightens the
impact of collecting toadfish from the natural environment.

Staff

Hanlon. Roger, Senior Scientist

Ament, Seth. Summer Research Assistant, Harvard University
Boal. Jean. Adjunct Scientist
Buresch. Kendra, Research Assistant

Conroy. Lou-Anne. Summer Research Assistant. Dartmouth College
Gilles, Nicole, REU Intern, University of Minnesota, Duluth
Lee. Tony. REU Intern. Duke University
Richmond, Hazel, Research Assistant
Shashar. Nadav. Adjunct Scientist
Sussman, Raquel, Investigator

Vaughan. Katrina. Summer Research Assistant, University of Wales.
Swansea

Visiting Investigators

Baddeley. Roland, University of Sussex, England
Baker. Robert. New York University
Cavanaugh. Joseph. Boston University Marine Program
Chiao, Chuan-Chin. Grass Fellow. University of Maryland.

Baltimore County

Cronin, Thomas. University of Maryland, Baltimore County
Grable, Melissa, Boston University Marine Program
Hall. Karina. University of Adelaide, Australia
Hatfield, Emma, FRS Marine Laboratory, Aberdeen. Scotland
Karson. Miranda. Michigan State University
Kier, William. University of North Carolina
Mensinger. Allen. University of Minnesota, Duluth
Messenger, John, University of Cambridge. England
Osorio, Daniel. Investigator. University of Sussex, England
Saidel, William, Rutgers University
Schmolesky, Matthew, Grass Fellow. University of Utah

Laboratory of Alan M. Ktizirian

Research in the laboratory explores the functional morphology and
ultrastructure of various organ systems in molluscs. The program includes
mariculture of the nudibranch, Hemrissenda crassicomis, with emphasis on
developing reliable culture methods for rearing and maintaining the animal
as a research resource. The process of metamorphic induction by natural
and artificial inducers is being explored in an effort to understand the
processes involved and as a means to increase the yield of cultured animals.
Morphologic studies stress the ontogeny of neural and sensory structures
associated with the photic and vestibular systems which have been the focus
of learning and memory studies, as well as the spatial and temporal
occurrence of regulatory and transmitter neurochemicals. Concurrent studies
detailing the toxic effects of lead on Hennissendu learning and memory.

feeding, and the physiology of cultured neurons are also being conducted.
New studies include cytochemical investigations of the Ca2+/GTP binding
protein, calexcitin, and its modulation with learning and lead exposure.

Collaborative research includes histochemical investigations on
strontium's role in initiating calcification in molluscan embryos (shell
and statoliths), iminunoeytochemical labelling of cell-surface antigens,
neurosecretory products, second messenger proteins involved with
learning and memory, as well as intracellular transport organelles using
mono- and polyclonal antibodies on squid (Loligo pealeii) giant axons
and Hermissenda sensory and neurosecretory neurons. Additional
collaborations involve studying neuronal development of myelin,
myelination defects, as well as nerve regeneration and repair in
phylogenetically conserved nervous systems.

Additional collaborative research includes DNA fingerprinting using
RAPD-PCR techniques in preparation for isogenic strain development of
laboratory-reared Hennissenda and hatchery produced bay scallops
(Argopectin irradians) with distinct phenotypic markers for rapid field
identification and stock assessments. Recently obtained funding will
expand this research to perform population genetic analyses of currently
designated yellowtail flounder ( Limanda ferruginea I stocks occurring in
the Northeast Fisheries Region.

Systematic and taxonomic studies of nudibranch molluscs, to include
molecular phylogenetics, are also of interest.

Staff

Kuzirian. Alan M., Associate Scientist
Kozlowski. Robbin. Research Technician

Visiting Investigators

Chikarmane, Hemant, Investigator

Clay. John R.. NINDS/NIH

Gould. Robert. NYS Institute of Basic Research

Summer Intents

Kingston, Margaret. REU Intern, Wake Forest University
Kuzirian. Mark, REU Intern. University of Rhode Island
Lee, Tony. REU Intern. Duke University

Laboratory of Roxanna Smolowitz

This laboratory investigated the pathogenesis of aquatic animal
diseases using traditional pathological methods combined with in situ
molecular methods. Research conducted during 2000 included 1)
examination of hard-clam-strain susceptibility to a protistan disease
agent named Quahog Parasite Unknown, and the methods of
transmission of that organism between infected and uninfected animals;
2) detection of disease-causing, protozoan organisms (MSX and SSO) in
eastern oysters using PCR and in situ hybridization techniques; and 3)
evaluation of inbred strains of oysters for resistance to disease vs.
productivity as commercial aquaculture stock. Work began on the
determination of possible causes of lobster shell disease in the northeast.

Staff

Smolowitz, Roxanna. MBL Veterinarian
Brothers, Christine, Laboratory Assistant
Cavanaugh. Joseph. Laboratory Assistant
Marks, Ernie. AmeriCorps member
Stukey. Jetley, Laboratory Assistant
Summers, Erin, Laboratory Assistant
Tirrell, Kerri-Ann. AmeriCorps member

Honors

Friday Evening Lectures

June 16
June 23
June 30
July 7
July 14
July 20-21

July 28
August 4
August 1 1

Edward Pearce, Cornell University

"Life-long Enemies — The Relationship Between Schistosomes and Their Hosts"

Stephen Farrand, University of Illinois at Champaign-Urbana

"Agrobacterium lumefaciens: Nature's Own Genetic Engineer"

Judith Eisen. University of Oregon

"From Lobster to Zebrafish: Development of Identified Neurons" (Lang Lecture)

David Anderson, California Institute of Technology

"Stem Cells from the Mammalian Nervous System: Basic Biology and Implications for Tissue Repair"

Sallie Chisholm (Penny). Massachusetts Institute of Technology

"The Invisible Forest: Marine Phytoplankton and Climate"

Eve Marder. Brandeis University

1) "Activity-dependent Timing of Neurons and Synapses in Adult and Developing Circuits" 2)

"Neurotransmitter Modulation of Neural Networks" (Forbes Lectures)

Jean-Pierre Changeux, Institut Pasteur

"Chemical Communications in the Brain: Nicotine, Receptors, and Learning" (Classman Lecture)

Susan Middleton/David Liittschwager

"Paradise Up Close: Hawaii — Endangered Eden"

Titia de Lange. The Rockefeller University

"At the Ends of Our Chromosomes: the Key to Immortality"

Fellowships and Scholarships

In 2( Hill, the MBL awarded research fellowships to 22 scientists from around the world. The MBL awarded scholarships to 77 students in the
MBL's summer courses as well as 4 post-course research awards. Donors provided gifts for endowed and expendable funds amounting to $256.090 in
support of the research fellowships program and an additional $738.107 to provide scholarships to students in MBL courses. Those funds that
received donations in 2000 are listed below. The individuals who received fellowships and scholarships are listed beginning on p. R58.

Robert Day Allen Fellowship
Fund

Drs. Joseph and Jean Sanger

The American Society for Cell
Biology Scholarships

The American Society for Cell Biology

Frederik B. Bang Fellowship Fund

Mrs. Betsy G. Bang

Max Burger Endowed Scholarship
for the Embryology Course

Dr. Max M. Burger

Jean and Katsuma Dan
Fellowship Fund

Drs. Joseph and Jean Sanger
Mrs. Eleanor Steinbach

Bernard Davis Fellowship Fund

Mrs. Elizabeth M. Davis

The Mac V. Edds, Jr. Endowed
Scholarship Fund

Dr. and Mrs. James D.

Dr. and Mrs. Kenneth T. Edds

Dr. Louise M. Luckenbill-Edds

Gerald D. and Ruth L. Fischbach
Endowed Scholarship Fund

Drs. Gerald and Ruth Fischbach
R57

Thomas B. Grave and Elizabeth
F. Grave Scholarship

Estate of Elizabeth F. Grave

Daniel S. and Edith T. Grosch
Scholarship Fund

Mr. Gustav Grosch

Ms. Laura Grosch and Mr. Herb Jackson

Aline D. Gross Scholarship Fund

Dr. and Mrs. Paul R. Gross
Dr. and Mrs. Benjamin Kaminer
Technic, Inc.

E. E. Just Endowed Research
Fellowship Fund

The Cole Memorial Family Fund

R58 Annual Report

Fred Karush Endowed Library
Readership

Dr. and Mrs. Laszlo Lorand

Dr. and Mrs. Arthur M. Silverstein

Keffer Hartline Fellowship Fund

Mrs. Elizabeth K. Hartline

Dr. and Mrs. Edward F. MacNichol. Jr.

Dr. William H. Miller

Dr. Torsten Wiesel and Ms. Jean Stein

Dr. and Mrs. Stephen Yeandle

Kuffler Fellowship Fund

Dr. and Mrs. Edward A. Kravitz

MBL Associates Endowed
Scholarship Fund

MBL Associates

Mrs. Anne L. Meigs-Brown

James A. and Faith Miller
Fellowship Fund

Drs. David and Virginia Miller

Frank Morrell Endowed Memorial
Scholarship

Dr. Leyla deToledo-Morrell

Mountain Memorial Fund

Dr. and Mrs. Dean C. Allard. Jr.
Ms. Brenda J. Bodian
Dr. and Mrs. Benjamin Kaminer
Mr. and Mrs. Thomas H. Roberts
Dr. and Mrs. R. Walter Schlesinger

Neural Systems and Behavior
Scholarship Fund

Anonymous ( 1 )

Mr. Srdjan D. Antic

Drs. Mary Atkisson and Joel White

Bristol-Myers Squibb Corporation

Dr. and Mrs. John Byrne

Ms. Lu Chen

Dr. Warren M. Grill

Dr. Anya C. Hurlbert

Dr. Eve Marder

Dr. Mark W. Miller

Fellowships Awarded

Mr. Rex R. Robison
Ms. M. Jade Zee

Nikon Fellowship

Nikon Instruments. Inc.

The Plum Foundation John E.
Dowling Fellowship Fund

The Plum Foundation

William Townsend Porter

Scholarship Fund for Minority

Students

William Townsend Porter Foundation

Phillip H. Presley Scholarship
Fund

Carl Zeiss, Inc.

Science Writing Fellowships
Program Support

American Society for Biochemistry and

Molecular Biology
American Society for Cell Biology
American Society for Photobiology
FASEB

NASA (Astrobiology Institute)
National Institutes of Health — Office of

Science Education
National Institutes of Health — National Cancer

Institute
National Science Foundation — Biological

Sciences
National Science Foundation — Office of Polar

Programs
Society for Integrative and Comparative

Biology

Times Mirror Foundation
Waksman Foundation for Microbiology
The Washington Post Company

The Catherine Filene Shouse SES
Scholarship Fund

The Catherine Filene Shouse Foundation

The Catherine Filene Shouse
Scholarship Fund

The Catherine Filene Shouse Foundation

The Catherine Filene Shouse
Fellowship Fund

The Catherine Filene Shouse Foundation

The Evelyn and Melvin Spiegel
Fellowship Fund

The Sprague Foundation
Drs. Joseph and Jean Sanger

H. B. Steinbach Fellowship Fund

Mrs. Eleanor Steinbach

Eva Szent-Gyorgyi Scholarship
Fund

Dr. and Mrs. Laszlo Lorand
Drs. Joseph and Jean Sanger
Dr. Andrew and Ms. Ursula Szent-Gyorgyi

Universal Imaging Fellowship Fund

Universal Imaging Corporation

The Irving Weinstein Endowed
Scholarship

The Irving Weinstein Foundation, Inc.

Walter L. Wilson Endowed
Scholarship

Dr. Paul N. Chervm

Dr. Jean R. Wilson

Mr. and Mrs. Ross A. Wilson

Young Scholars/Fellows Program

Drs. Harriet and Alan Bernheimer

Dr. and Mrs. Francis P. Bowles

Dr. and Mrs. Sherwin J. Cooperstein

Mrs. Elizabeth M. Davis

Mrs. James R. Glazebrook

Ms. Jeannie Leonard

Mrs. Barbara C. Little

Dr. and Mrs. Anthony Liuzzi

Drs. Luigi and Elaine Mastroianni

Drs. Matthew and Jeanne Meselson

Dr. and Mrs. Philip Person

Drs. Dorothy Skinner and John Cook

Drs. Ann Stuart and John Moore

Mr. and Mrs. Richard Yoder

MBL Summer Research Fellows

• Srdjan Antic. M.D., is a post-doctoral fellow in the Department
of Cellular and Molecular Physiology at Yale University School of

Medicine. New Haven, CT. The title of his project is "Selective
modulation of the dendritic membrane potential." Dr. Antic is funded
by the Baxter Postdoctoral Fellowship, the Charles R. Crane
Fellowship Fnntl, the MBL A\\Oi'i(ift's Felltni'shift Fund, and the .liiine
A. and Fuitli Miller Memorial Fund.

Honors R59

• Roberto Bru//one. Ph.D.. is ;in Associate Professor at the
Institut Pasteur in Paris. France. The title of this research is "Molecular
analysis of the biophysical properties of connexin channels that mediate
cell- cell communication between neurons of the vertebrate retina and
CNS." Dr. Bruzzone is funded by the Erik R. Fries Endowed
Fellowship, the MBL Associates Fellowship Fund, and the H. B.
Steinbach Fellowship Fund.

• Mario H. Burgos. M.D., is an Emeritus Professor of the
Medical School at the Universidad Nacional de Cuyo and Director of
the Institute de Histologia y Embriologia. National Council of Research
(CONICET). Argentina. His research project is titled. "Mechanism of
release of spermatozoa from the Serroli cells." He is also collaborating
with Dr. Shinya Inoue in the identification of the birefringent zones in
Arbacia eggs after centrifuge polarizing microscopy. Dr. Burgos is
funded by the Chairman 's Fellowship.

• Jean-Pierre Changeux is a Professor at the College de France
and Director of the Unit of Molecular Neurobiology at the Institut
Pasteur in Paris. He is the author of Neunmal Man: The Biology of
Mind (1990). Dr. Changeux has been awarded a Herbert W. Rand
Fellowship for his research.

• Debra Ann Fadool. Ph.D.. is an Associate Professor in the
Biomedical Research Facility at Florida State University. Tallahassee,
FL. The title of her research project is "Chemosensory transduction in
the vomero-nasal organ." Dr. Fadool is funded by the Frederik B. Bang
Fellowship Fund, the Ann E. Kammer Memorial Fellowship Fund, and
the MBL Associates Fellowship Fund.

• Mariano A. Garcia-Bhnco, M.D.. Ph.D.. is Associate Professor
of Genetics. Microbiology, and Medicine at Duke University Medical
Center. Durham. NC. He is a Raymond and Beverly Sackler Scholar
and is a member of the Biochemistry Study Section of the National
Institutes of Health. The Josiah Macy. Jr. Foundation is funding his
research.

• George G. Holz. Ph.D.. has been appointed Associate Professor
at New York University School of Medicine to establish a Diabetes
Research Laboratory at New York University Medical Center. His
summer research project is "Spatial distribution of second messengers
in pancreatic fi-cells." Dr. Holz is funded by the Erik B. Fries
Endowed Fellowship, the Frank R. Lillie Fellowship Fund, and the
MBL Associates Fellowship Fund.

• Peter Koulen. Ph.D.. is a postdoctoral associate in the
Department of Pharmacology at Yale University School of Medicine,
New Haven. CT. The title of his research project is "Calcium signaling
in zebrafish neurons mediated by differentially distributed intracellular
calcium channels." Dr. Koulen has received funding from the Erik B.
Fries Endowed Fellowship and the Litc\ B. Lemann Fellowship Fund.

• George Langford. Ph.D., is the Ernest Everett Just Professor of
Natural Sciences and Professor of Biological Sciences at Dartmouth
College. Hanover. NH. His research project is titled "Actin-based
vesicle transport in the squid giant axon." Dr. Langford is funded by
the Josiah Macy, Jr. Foundation.

• Jennifer LaVail. Ph.D.. is Professor of Anatomy/
Ophthalmology at the University of California. San Francisco. She is
spending her second summer at the MBL. Her research project is tilled
"HSV tegument proteins in axonal transport and microtubule
architecture." Dr. LaVail is funded by an MBL Research Fellowship
and the Evelyn and Melvin Spiegel Fellowship Fund.

• Carolyn Lesser has published eight children's books and
numerous articles. She has also served as a consultant and lecturer. Ms.
Lesser was awarded a Science Writing Fellowship in 1999. and is a
Desk Reader at the MBLAVHO1 Library in 2000. Ms. Lesser is funded
by the Fred Karu.sh Endowed Library Readership.

• Jeffrey Magee. Ph.D., is an Assistant Professor in the
Neuroscience Center at Louisiana State University Medical Center.
New Orleans. Louisiana. The title of his research is "Mechanisms of

Ca~* entry into hippocampal neurons." Dr. Magee is funded by the
MBL Associates Fellowship Fund and the Lnc\ B. Lemann Fellowship
f- mid

• Antonio Malgaroli. Ph.D.. is a Professor in the Unit of
Neurobiology of Learning at the University of San Rafaele. Milan.
Italy. The title of his summer research is "Presynaptically silent
synapses in the hippocampus." Dr. Malgaroli is funded by the Herbert
W. Rand Fellowship and the Frank R. Lillie Fellowship Fund.

* Mark Messerli. Ph.D.. is a Research Associate in the
Department of Biological Sciences at Purdue University. West
Lafayette. IN. The title of his research project is "Regulation of
cytoplasmic pH in eucaryotic acidophiles." Dr. Messerli is funded by a
NASA Life Sciences Program Fellowship.

* Timothy Mitchison. M.D., is a Professor in the Department of
Cell Biology at Harvard Medical School, Boston. MA. His research
project is titled "Optical Approaches to Cell Division." The Universal
Imaging Corporation is funding Dr. Mitchison.

• David Ogden. Ph.D.. is a Principal Investigator at the National
Institute for Medical Research in London. England. The title of his
research is "Central electrosensory processing in the skate." Dr. Ogden
is funded by an MBL Associates Fellowship.

' Oladele A. Ogunseitan, Ph.D.. is an Associate Professor in the
Department of Environmental Analysis and Design at the University of
California, Irvine. Dr. Ogunseitan returns to the MBL to study "Toxic
metal resistance, swarming phenotype. and enzyme polymorphism in
Vibrio alginolyticus." Dr. Ogunseitan is funded by the Josiah Macy. Jr.
Foundation.

' David Paydarfar, Ph.D.. is an Associate Professor at the
Department of Neurobiology at the University of Massachusetts
Medical School in Worcester. The title of his research project is "Can
noise regulate oscillatory state? In nurnero and in vitro analysis of
squid axon membrane." Dr. Paydartar is funded by the Frederick B.
Bang Fellowship Fund, the M. G. F. Fuortes Memorial Fellowship
Fund, the MBL Associates Fellowship Fund, and the John O. Crane
Fellowship Fund.

* Peter Saggau. Ph.D.. is an Associate Professor in the Division
of Neuroscience at Baylor College of Medicine. Houston. Texas. The
title of his summer research project is "Transmission and plasticity at
single hippocampal synapses." Dr. Saggau received the Nikon
Fellowship.

• Miduturu Srinivas. Ph.D.. is a Research Associate at the Albert
Einstein College of Medicine, Bronx, New York. His research project
for the summer is titled "Biophysical properties of gap junctions in
retinal neurons." Dr. Srinivas is lunded by the Erik B. Fries Endowed
Fellowship and the H. Keffer Hartline Fellowship Fund.

• Thomas W. White, Ph.D., is an Instructor in the Department of
Neurobiology at Harvard Medical School. Boston, MA. His research
title is "Gap junctional communication in the retina." Dr. White is
funded by the H. Kefter Hartline Fellowship Fund, the Stephen W.
Kuffler Fellowship Fund, and the Frank R. Lillie Fellowship Fund.

* Iain Stuart Young. Ph.D.. is a Research Associate in the
Department of Biology at the University of Pennsylvania. Philadelphia.
The title of Dr. Young's summer research is "The molecular
mechanisms of relaxation in superfast muscles." Dr. Young is funded
by the Robert Day Allen Fellowship Fund, the MBL Research
Fellowship Fund, the H. Burr Steinbach Memorial Fellowship Fund,
and the Lucy B. Lemann Fellowship Fund.

Grass Fellows

• Leonardo Belluscio. Ph.D.. Duke University Medical Center.
Project: "The role of spontaneous activity in the olfactory system."

• Chuan-Chin Chiao. University of Maryland, Baltimore County.

R60 Annual Report

Project: "Camouflage in cephalopods: visual control and effectiveness
when viewed by predators."

• Melina Hale. Ph.D.. State University of New York at Stony
Brook. Project: "The neural basis of startle behavior and its
development in the toadfish (Opsainis tan}."

• Johann Hofmann. Ph.D.. Stanford University. Project: "The
consequences of socially induced differential growth on the retina."

• Thomas Kuner. M.D.. Duke University Medical Center.
Project: "The timing of NSF action in neurotransmitter release probed
with photolysis of caged peptides."

• Brian Link, Ph.D., Harvard University. Project: "Time-lapse
analysis of zebrafish retinal cells during development: investigation of
lamination mutants."

• Matthew B. McFarlane, Ph.D.. New York University Medical
Center. Project: "Central pathways mediating the horizontal vestibulo-
ocular reflex in an elasmobranch, Scyliorliinus canicula."

' Matthew T. Schmolesky, University of Utah. Project: "Visual
stimulus encoding in the optic lobe of squid Loligo pealei."

• Ava J. Udvadia. Ph.D., Duke University Medical Center.
Project: "Investigation of signaling pathways that activate axon growth-
associated gene expression in regenerating spinal neurons."

• Jing W. Wang. Ph.D.. Lucent Technologies. Project: "Optical
imaging and electrophysiological recording of Drosophila central
nervous system: A search for the significance of synchrony."

• Zachary P. Wills. Harvard Medical School. Project: "The
function of dAbl pathway components in neuronal outgrowth and
growth cone turning in vitro."

MBL Science Writing Fellowships Program

Fellows

Begos, Kevin, Winston-Sulein Journal
Ben-Ari. Elia. BioScience

Berger, Cynthia. Finger Lakes Productions

Borenstein. Seth. Knight Ridder Newspapers

Enright, Leo, BBC

Fagin, Dan, Newsday

Garber, Ken, Freelance

Gorman, Jessica. Discover magazine

Hathaway, William, Hartford Courant

Helmuth, Laura, Science magazine

Mansur. Mike. Kansas City Star

Martin, Roger, University of Kansas

Marx. Vivien, Freelance

Milano, Gianna, Mondadori Publishing Company

Nemecek. Sasha. Scientific American

Poulson, David, Booth Newspapers

Senkowsky, Sonya. Anchorage Daily News

Scott, Bari. SoundVision Productions

Tangley, Laura, U.S. News and World Report

Program Staff

Goldman, Robert D.. Northwestern University, Co-Director

Hinkle, Pamela Clapp, Marine Biological Laboratory. Administrative

Director
Rensberger. Boyce, Knight Science Journalism Program. Co-Director

Hwids-On Laboratory' Fiwulty

Chisholm, Rex, Northwestern University, Biomedical Hands-On

Laboratory Director
Hobbie, John E., Marine Biological Laboratory, Environment Hands-On

Laboratory Co-Director
Melillo. Jerry, Marine Biological Laboratory, Environment Hands-On

Laboratory Co-Director
Palazzo, Robert, University of Kansas, Biomedical Hands-On

Laboratory Associate Director

Scholarships Awarded

The Bruce and Betty Alberts Endowed Scholarship
in Physiology

Fleegal. Melissa. University of Florida

American Society for Cell Biology

Colon-Ramos, Daniel, Duke University
Bradford. Yvonne. University of Oregon
Crespo-Barreto. Juan, University of Puerto Rico
Espinoza, Nora, University of Chicago
Glater. Elizabeth, Brown University
Greenlee, Anne, Marshfield Medical Research Foundation
Hubhard. Aida, University of Texas, San Antonio
Mah, Silvia, Scripps Institution of Oceanography
Powell, Jacqueline, Morehouse School of Medicine
Triggs, Veronica, University of Wisconsin, Madison

Biology Club of the College of the City of New York

Konur. Sila. Columbia University

C. Lalor-Burdick Scholarship

Powell. Jacqueline, Morehouse School of Medicine

Burroughs Wellcome Fund
Biology of Parasitism Course

Andersson. John. Karolinska Institute

D'Angelo, Maximiliano, University of Buenos Aires

Dolezal, Pavel, Charles University

Ferreira. Ludmilu. Universidade Federal de Minas Gerais

Figueiredo. Luisa. University of Porto

Gilk, Stacey. University of Vermont

Lowell, Joanna, Rockefeller University

Martins, Gislaine, Faculdade de Medicina de Ribeirao Preto

Murta, Silvane, Centra e Pesquisas Rene Rachou-FIOCRUZ

Sehgal, Alfica, Tata Institute of Fundamental Research

Ulbert, Sebastian. Netherlands Cancer Institute

Villarino. Alejandro, University of Pennsylvania

Burroughs Wellcome Fund
Frontiers in Reproduction Course

Alberio, Ramiro, Ludwig-Maximilian University
Allegrucci. Cinzia, Universita degli Studi di Perugia

Honors R61

Ashkar, All. University of Guelph

Chong. Kowit-Yu. Oregon Regional Primate Research Cenler

Heit'etz. Yael. Cornell University

Lavoie, Holly. University of South Carolina

Majumdar, Subeer, Primate Research Center

Powell, Jacqueline, Morehouse School of Medicine

Richard, Craig. University of Pittsburgh

Sahgal, Namita, Kansas University Medical Center

Zhang, Gongqiao, University of Virginia

Burroughs Wellcome Fund
Modeling of Biological Systems Course

Ginsberg, Tara, University of Texas Medical School

Immerstrand, Charlotte. Linkoping University

Quinteiro, Guillermo, Comision Nacional de Energia Atomica

Teng, Ching-Ling. University of Virginia

Uppul. Hirdesh. Punjab Agricultural University

Genick, Ulrich, The Salk Institute for Biological Studies

Hershberg, Uri, Hebrew University

Mosavi. Leila. University of Connecticut Health Center

Burroughs Wellcome Fund
Molecular Mycology Course

Austin. W. Lena. Howard University School of Medicine

Ibrahim, Ashraf. Harbor-UCLA Medical Center

Mol, Pieternella, University of Amsterdam

Munro. Carol, University of Aberdeen

Perea, Sofia, The University of Texas Health Science Center

Spellberg. Brad, Harbor-UCLA Medical Center

Spreghini, Elisabetta. Yale University School of Medicine

Toenjes, Kurt. University of Vermont

Wasylnka. Julie. Simon Eraser University

The Ellison Medical Foundation
Molecular Biology of Aging Course

Bailey. Adina, University of California, Berkeley

Baur. Joe. University of Texas Southwestern

Bordone. Laura, University of Minnesota Medical School

Cui, Wei. Roslin Institute

Cypser. James. University of Colorado

Filosa, Stefania, IIGB-CNR

Furfaro, Joyce. Pennsylvania State University College of Medicine

Harper, James, University of Idaho

Huang, Xudong, Massachusetts General Hospital-East

Johnson, Kristen, Purdue University

Konigsberg. Mina. Universidad Autonoma

Kostrominova, Tatiana, University of Michigan

Luo, Yuan, University of Southern Mississippi

Munoz. Demse, University of California

Peel, Alyson, The Buck Center for Research in Aging

Podlutsky. Andrej. National Institute on Aging. NIH

Radulescu, Andreea. Albert Einstein College of Medicine

Srivivsan, Chandra. University of California, Los Angeles

Tong. Jiayuan (James), Cold Spring Harbor Laboratory

Zaid, Ahmed. Stockholm University

Caswell Grave Scholarship Fund

Bates. Damien, Murdoch Institute

Brelidze, Tinatin, University of Miami School of Medicine

Cordeiro, Maria Sofia, Instituto Gulbenkian de Ciencia

Gong. Ying. California Institute of Technology

Menna, Elisabetta, Institute of Neurophysiology. Pisa

Prud'homme. Benjamin, Centre de Genetique Moleculaire

Daniel S. Grosch Scholarship Fund

Neretin, Lev, Shirshov Institute of Oceanography

Gary N. Calkins Memorial Scholarship Fund

Ellertsdottir. Eh'n. University of Freiburg

Aline D. Gross Scholarship Fund

Cordeiro, Maria Sofia, Instituto Gulbenkian de Ciencia
Johansson. Viktoria. Gtiteborgs Universitet

Edwin Grant Conklin Memorial Fund

Jhaveri. Dhanisha. Tata Institute of Fundamental Research

William F. and Irene C. Diller
Memorial Scholarship Fund

Jhaveri. Dhanisha. Tata Institute of Fundamental Research
Menna. Elisabetta. Institute of Neurophysiology. Pisa

The Ellison Medical Foundation
Biology of Parasitism Course

Gilk, Stacey. University of Vermont

Lowell, Joanna, Rockefeller University

O'Donnell, Rebecca, University of Melbourne

Ralph, Stuart. University of Melbourne

Triggs, Veronica, University of Wisconsin, Madison

Villarino. Alejandro, University of Pennsylvania

William Randolph Hearst Foundation Scholarship

Rankin, Kathleen. Oberlin College

Rodeheffer, Carey, Emory University

Rodgers, Erin, University of Alabama, Birmingham

Shatkin-Margolis. Seth. Duke University

Takai, Erica. Columbia University

Howard Hughes Medical Institute

Akay, Turgay. KSIn University

Barak. Yoram. Hebrew University

Ding. Long. University of Pennsylvania

Globerson, Amir. Hebrew University

Imai. Kazushi. Columbia University P&S

Konur, Sila. Columbia University

Krishna. B. Suresh. New York University

Kyrkjebo, Vibeke, Sars International Centre

Lauritzen, Thomas. University of California. San Francisco

Lin, Li-hung, Princeton University

McCance. James, University of Leicester

Menna, Elisabetta. Institute of Neurophysiology, Pisa

R62 Annual Report

Nasevicius. Aidas. University of Minnesota

Paz. Ron, Hebrew University Medical School

Petereit, Christian. Universitat Bielefeld

Prud'homme. Benjamin. Centre de Genetique Moleculaire

Rela. Lorena, Universidad de Buenos Aires

Rokni. Uri. Hebrew University of Jerusalem

Schreiber. Susanne. Humboldt Universitat Berlin

Shi. Songhai, Cold Spring Harbor Laboratory

Singh. Brajesh, Imperial College at Silwood Park

Skromne, Isaac, Princeton University

Szalisznyo, Krisztina, Hungarian Academy of Sciences

Warkman, Andrew, University of Western Ontario

Zaar, Annette. Universitat Freiburg

International Brain Research Organization

Challa, Anil Kumar, Ohio State University

Leung. Fung Ping, Hong Kong University of Science and Technology

Lupo, Giuseppe, University of Pisa

Maldonado. Ernesto, Massachusetts Institute of Technology

Arthur Klorfein Scholarship and Fellowship Fund

Aspock. Gudrun, University of Basel

Gong. Ying, California Institute of Technology

Imai. Kazushi, Columbia University P&S

KyrkjebO. Vibeke. Sars International Centre

Lee, Vivian, Oregon Health Sciences University

Prud'homme, Benjamin, Centre de Genetique Moleculaire

Skromne, Isaac, Princeton University

Frank R. Lillie Fellowship and Scholarship Fund

Carroll, Michael. University of Newcastle

Costa, Patricia. Universidad Federal do Rio de Janeiro

Cotrufo. Tiziana. Institute of Neurophysiology

Dayel. Mark, University of California, San Francisco

Fleegal. Melissa, University of Florida

Marrari. Yannick. Observatoire Oceanographique

Petrie, Ryan. University of Calgary

Rankin. Kathleen. Oberlin College

Shatkin-Margolis. Seth. Duke University

Takai. Erica. Columbia University

Jacques Loeb Founders' Scholarship Fund

Fleegal. Melissa, University of Florida

Petrie, Ryan, University of Calgary

Shilkrut. Mark. Technion-Israel Institute of Technology

Massachusetts Space Grant Consortium

Barbour. Jason, University of California. San Francisco

Handley. Scott. Centers for Disease Control and Prevention

Holland. Brenden. University of Hawaii

Longnecker. Krista, Oregon State University

McMahun, Katherine, University of California. Berkeley

Munroe, Stephen, Marquette University

Nepokroelt. MolK. Smithsonian Institution

Stone. Karen. University of Alaska Museum. Fairbanks

Xie. Gary. Los Alamos National Lab

S.O. Mast Memorial Fund

Marrari. Yannick. Observatoire Oceanographique
Zhong. Haining. Johns Hopkins University

MBL Associates Endowed Scholarship Fund

Rela. Lorena. Universidad de Buenos Aires

MBL Pioneers Scholarship Fund

Ballard. Victoria, Weill Medical College

Bates. Damien, Murdoch Institute

Beckhelling. Clare, Laboratoire de Biologie-Cellulaire

Lee. Vivian. Oregon Health Sciences University

Warkman. Andrew, University of Western Ontario

Merck & Company, Inc. Scholarships

Gilk. Stacey, University of Vermont

Lamb, Tracey, ICAPB

Lowell. Joanna, Rockefeller University

Martins, Gislaine. Faculdade de Medicina de Ribeirao Preto

Murta. Silvane, Centra e Pesquisas Rene Rachou-FIOCRUZ

O'Donnell, Rebecca, University of Melbourne

Ralph. Stuart. University of Melbourne

Villarino, Alejandro. University of Pennsylvania

Charles Baker Metz and William Metz
Scholarship Fund in Reproductive Biology

Keller, Dominique, Texas A&M University
Richard. Craig, University of Pittsburgh

Frank Morrell Endowed Memorial Scholarship

Kettunen, Petronella, Novel Institute for Neurophysiology

Mountain Memorial Fund Scholarship

Carroll, Michael, University of Newcastle

Cordeiro, Maria Sofia, Instituto Gulbenkian de Ciencia

Cotrufo, Tiziana. Institute of Neurophysiology

Sen. Subhojit. Tata Institute of Fundamental Research

Shilkrut. Mark. Technion-Israel Institute of Technology

Zeidner. Gil. The Weizmann Institute of Science

Pfizer Inc Endowed Scholarship

Dayel, Mark. University of California. San Francisco
Lin. Li-hung. Princeton University

Planetary Biology Internship Awards

Kadavy. Dana. University of Nebraska. Lincoln
Ward, Dawn, University of Delaware

William Townsend Porter Fellowship and
Scholarship Fund

Bradford, Yvonne, University of Oregon
Colon-Ramos, Daniel, Duke University
Crespo-Barreto, Juan. University of Puerto Rico
Espinoza, Nora, University of Chicago

Honors R63

Glater, Elizabeth. Brown University

Hubbard. Aida, L'niversity of Texas. San Antonio

Triggs, Veronica, University of Wisconsin. Madison

Phillip H. Presley Scholarship Award
Funded by Carl Zeiss, Inc.

Neretin, Lev. Shirshov Institute of Oceanography

Zaar. Annette. Universitat Freiburg

Livet, Jean, INSERM U.382

McKellar. Claire, Harvard Medical School

Misgeld. Thomas. Institute for Clinical Neuroimmunology

Nelson, Laura. National Institute for Medical Research

Yasuda, Ryohei. Teikyo University Biotech. Research Center

Aspo'ck, Gudrun, University of Basel

Beckhelling. Clare. Laboratoire de Biologie-Cellulaire

Nasevicius. Aidas. University of Minnesota

Herbert \V. Rand Fellowship and Scholarship Fund

Bodelon. Clara. Boston University

Costa. Patricia, Universidad Federal do Rio de Janeiro

Cotrufo. Tiziana, Institute of Neurophysiology

Cruz. Georgina. University of South Florida

Dayel. Mark. University of California, San Francisco

Feinerman, Ofer. Weizmann Institute of Science

Jhaveri, Dhanisha. Tata Institute of Fundamental Research

Johansson, Viktoria, Goteborgs Universitet

Kang, Kukjin, Hebrew University of Jerusalem

Marrari. Yannick, Observatoire Oceanographique

Menna. Elisabetta. Institute of Neurophysiology, Pisa

Sen. Subhojit, Tata Institute of Fundamental Research

Shilkrut. Mark, Technion-Israel Institute of Technology

Ulanovsky, Nachum, Hebrew University

Zeidner, Gil. Weizmann Institute of Science

Ruth Sager Memorial Scholarship

Cheeks, Rebecca, University of North Carolina, Chapel Hill

Howard A. Schneiderman Endowed Scholarship

Misgeld. Thomas. Institute for Clinical Neuroimmunology
Nelson, Laura, National Institute lor Medical Research
Zhong. Haining. Johns Hopkins University

Moshe Shilo Memorial Scholarship Fund

Barak. Yoram. Hebrew University

Post-Course Research Awards

Society of General Physiologists' Scholarship

Ballard, Victoria, Weill Medical College
McKellar. Claire. Harvard Medical School
Menna. Elisabetta. Institute of Neurophysiology, Pisa
Rela, Lorena. Universidad de Buenos Aires

Marjorie W. Stetten Scholarship Fund

Ballard, Victoria, Weill Medical College
Ellertsdottir. Eh'n, University of Freiburg

Horace W. Stunkard Scholarship Fund

Berkowitz, Karen, Hospital of the University of Pennsylvania
Sahgal, Namita, Kansas University Medical Center

Surdna Foundation Scholarship

Brelidze, Tinatin. University of Miami School of Medicine

Cotrufo, Tiziana. Institute of Neurophysiology

Ding. Long. University of Pennsylvania

Johansson. Viktoria. Goteborgs Universitet

Kettunen, Petronella. Nobel Institute for Neurophysiology

Livet, Jean. INSERM U.382

Marrari. Yannick. Observatoire Oceanographique

McKellar. Claire. Harvard Medical School

Misgeld. Thomas, Institute for Clinical Neuroimmunology

Nelson. Laura, National Institute for Medical Research

Sen, Subhojit, Tata Institute of Fundamental Research

Yasuda, Ryohei. Teikyo University Biotech. Research Center

Zeidner, Gil, Weizmann Institute of Science

William Morton Wheeler Family Founders'
Scholarship

Zhong, Haining. Johns Hopkins University

Walter L. Wilson Endowed Scholarship Fund

Brelidze. Tinatin. University of Miami School of Medicine
Cotrufo. Tiziana. Institute of Neurophysiology

World Academy of Arts and Sciences
Emily Mudd Scholarship

Greenlee, Anne, Marshfield Medical Research Foundation
Powell. Jacqueline, Morehouse School of Medicine

World Health Organization

Diaz, Lorenza. Instituto Nacional de la Nutricion

Colon-Ramos, Daniel, Duke University (Physiology)
Costa, Patricia, University of Rio de Janeiro (Physiology)

Dayel. Mark, University of California, San Francisco (Physiology)
Siuda, Eduard. Michigan State University (Neural Systems and Behavior)

Board of Trustees and Committees

Corporation Officers and Trustees

Chairman of the Board of Trustees. Sheldon J. Segal. The Population

Council
Co-Vice Chair of the Board of Trustees. Frederick Bay, Josephine Bay

Paul and C. Michael Paul Foundation

Co-Vice Chair of the Board of Trustees. Mary J. Greer. New York. NY
President of the Corporation, John E. Dowling. Harvard University
Director and Chief Executive Officer. John E. Burris. Marine Biological

Laboratory*
Interim Director and Chief Executive Officer, William T. Speck. Marine

Biological Laboratory*
Treasurer of the Corporation. Mary B. Conrad, Fiduciary Trust

International*

Clerk of the Corporation. Robert E. Mainer, The Boston Company
Chair of the Science Council. Robert B. Barlow. SUNY Health Science

Center*
Chair of the Science Council. Kerry S. Bloom. University of North

Carolina*

Class 0/2004

Jacobson. M. Howard, Bankers Trust
Langford, George M., Dartmouth College
Miller, G. William, G. William Miller & Co.. Inc.
Press, Frank, The Washington Advisory Group
Weld. Christopher M., Sullivan and Worcester. Boston
Wiesel, Torsten N.. The Rockefeller University

Honorary Trustees

Cunningham, Mary Ellen, Grosse Pointe Farms, MI
Ebert, James D.. Baltimore. MD
Golden. William T.. New York. NY
Grass. Ellen R.. The Grass Foundation

Class of 2001

Anderson. Porter W.. North Miami Beach. FL

Bay. Frederick, Josephine Bay Paul and C. Michael Paul

Foundation. Inc.

Cox, Martha W.. Hobe Sound, FL
Greer. Mary J.. New York, NY
Steere, William C. Jr.. Pfizer Inc
Weissmann. Gerald. New York University School of Medicine

Class of 2002

Lakian. John R.. The Fort Hill Group, Inc.
Ruderman, Joan V.. Harvard Medical School
Segal. Sheldon J., The Population Council
Speck, William T., Marine Biological Laboratory
Zeien, Alfred M.. The Gillette Company

Class of 2003

Kelley. Darcy Brisbane, Columbia University

Landeau. Laurie J., Marinetics, Inc.

Lee, Burton J. Ill, Vero Beach, FL

O'Hanley. Ronald P.. Mellon Institutional Asset Management

Pierce. Jean. Boca Grande, FL

Ryan, Vincent J., Schooner Capital LLC

/,< i 'tin a-

Trustees Emeriti

Adelherg, Edward A.. Yale University. New Haven. CT

Buck, John B.. Sykesville. MD

Cohen. Seymour S., Woods Hole, MA

Colwin. Arthur L.. Key Biscayne. FL

Colwin. Laura Hunter. Key Biscayne, FL

Copeland. Donald Eugene, Woods Hole, MA

Crowell, Sears Jr., Indiana University, Bloomington, IN

Hayashi. Teru. Woods Hole, MA

Hubbard, Ruth. Cambridge, MA

Kleinholz. Lewis, Reed College, Portland. OR

Krahl, Maurice, Tucson. AZ

Prosser, C. Ladd. University of Illinois, Urbana. IL

Russell-Hunter, W. D.. Syracuse University. Syracuse. NY

Saunders. John W.. Waquoit. MA

Shepro, David. Boston, MA

Trigg. D. Thomas, Wellesley, MA

Vincent. Walter S.. Woods Hole. MA

Directors Emeriti

Ebert. James D.. Baltimore. MD

Gross. Paul, Falmouth, MA

Halvorson. Harlyn O.. Woods Hole, MA

R64

Trustees and Committees R65

Executive Committee of the Board of
Trustees

Segal, Sheldon J.. Chair
Bay. Frederick. Co-Vice Chair
Greer. Man' J.. Co-Vice Chair
Anderson. Porter W.
li.nK'u. Robert B.
Bloom. Kerry S.
Burris. John E.
Conrad, Mary B., Treasurer
Mainer. Robert E.
O'Hanley. Ronald P.
Speck. William T.
\\ cissmann, Gerald

Science Council

Barlow. Robert B.. Chair (2001 )
Bloom, Kerry S., Chair (2000)
Armstrong. Clay M. (2002)
Armstrong, Peter (2002)
Atema. Jelle (2001)
Burris. John E.*
Dawidowicz. E. A.*
Haimo. Leah (2001)
Hopkinson, Charles (2002)
Jaffe. Laurinda (2001)
Smith. Peter J. S. (2001)
Sogin, Mitchell (2002)
Speck. William T.*
Weeks. Janis C. (2002)

Standing Committees of the Board of Trustees

Development

Speck. William, Chair
Anderson. Porter W,
Barlow. Robert W.
Bay, Frederick
Conrad. Mary B.
Cox, Martha W.
Dowling, John
Ebert. James D.
Grant. Philip
Lakian, John R.
Langford. George
Lee, Burton J.
Miller. G. William
Pierce. Jean
Steere. William C.
Weld, Christopher M.
Wiesel. Torsten

Facilities and Capital Equipment

Anderson, Porter W.. Chair
Bay. Frederick
Boyer. Barbara
Langford. George
Pros. Frank
Ruderman. Joan
Weld. Christopher M.
Wiesel. Torsten

; Ex officio

Investment

Conrad, Mary B.. Chair
Jacobson. M. Howard
Lakian. John R.
Mainer. Robert E.
Miller, G. William
O'Hanley. Ronald P.
Ryan, Vincent J.
Segal, Sheldon J.
Zeien. Alfred M.

Finance

O'Hanley, Ronald. Chan
Conrad. Mary B.
DeHart, Donald
Jacobson, M. Howard
Kelley. Darcy Brisbane
Lakian, John R.
Landeau, Laurie J.
Loewenstein. Werner
Mainer. Robert E.
Manz. Robert
Miller, G. William
Ryan. Vincent J.
Zeien. Alfred M.

Nominating

Weissmann. Gerald. Chair
Barlow. Robert B.
Bloom, Kerry S.
Cox, Martha W.
Greer. Man J.
Pierce. Jean
Segal. Sheldon J.
Speck. William T.

R66 Annual Report

Standing Committees of the Corporation and Science Council

Buildings and Grounds

Boyer, Barbara C.. Chair
Cutler, Richard*
Eckberg. Bill
Fleet, Barry*
Hayes, Joe*
McArthur, Andrew
Peterson, Bruce J.
Tweedell, Kenyon S.
Valiela, Ivan

Beckwith. Mary*
Bloom. Kerry S.*
Browne. Robert*
Cutler, Richard*
Goux, Susan*
Hinkle. Pamela Clapp*
Malchow, Robert P.
McDonough. Stefan
Rastetter, Edward
Stuart. Ann E.
Weeks, Janis C.

Education Committee

Barlow, Robert B.*
Bloom. Kerry S.*
Dawidowicz, E. A.*
Dionne, Vincent, Chair
Dunlap, Paul
Fink, Rachel
Hanlon, Roger
Kiehart, Dan
Madison, Dan
Venuti. Judith
Wadsworth, Patricia
Zottoli. Steve

Fellowships

Pederson, Thoru. Chair

Dawidowicz, E. A.*

Deegan, Linda

Ehrlich. Barbara

Kaufmann, Sandra* (Recording Secretary)

Lemos, Jose

Pipscombe, Diane

Sluder. Greenfield

Smith, Peter J. S.

Treistman, Steven (Guest Member)

Housing, Food Service and Child Care

Browne, Carol, Chair
Barlow, Robert B.*

MBL/WHOI Library Joint Advisory Committee

Shepro, David, Chair, MBL
Ashmore, Judy, MBL*
Dow, David, NMFS
Harbison, G. Richard. WHOI
Hobbie, John. MBL
Hurter, Colleen. WHOI*
Norton, Cathy, MBL*
Robb. James, USGS
Smith, Peter J. S., MBL
Smolowitz, Roxanna, MBL
Tucholke, Brian, WHOI
Warren, Bruce, WHOI

Research Services and Space

Laufer, Hans. Chair
Armstrong, Peter B.
Cutler. Richard*
Dawidowicz. E. A.*
Foreman, Kenneth
Kerr. Louis M.*
Landowne. David
Mattox, Andrew*
Melillo, Jerry
Mizell, Merle
Smith, Peter J. S.
Steudler. Paul
Valiela, Ivan

Discovery: The Campaign for Science at the Marine Biological Laboratory
Steering Committee

Bay. Frederick, Campaign Chair
Golden. William T., Honorary Chair
Grass, Ellen R., Honorary Chair
Clowes, Alexander W., Vice-Chair
Cox. Martha W.. Vice-Chair
Miller. G. William, Vice-Chair
Weissmann. Gerald, Vice-Chair
Anderson. Porter

; Ex officio

Barlow. Robert B.. Jr.
Bernstein. Norman
Cobb, Jewell Plummer
Conrad, Mary B.
Cunningham, Mary-Ellen
Dowling, John E.
Ebert. James D.
Fischbach, Gerald D.
Goldman. Robert D.
Greer, Mary J.
Jacobson. M. Howard

Trustees and Committees R67

Landeau, Laurie J.
Langford. George M.
Lee. Burton J. Ill
Pierce. JeJe
Prendergast. Robert A.
Shepro. David
Speck. William T.
Steere. William C. Jr.
Swope. John F.
Weld. Christopher M.
Zeien, Alfred M.

Council of Visitors

Norman B. Asher, Esq.. Hale and Dorr, Counsellors at Law.

Boston. MA
Mr. Robert W. Ashton. Bay Foundation. New York, NY

Mr. Donald J. Bainton. Continental Can Co.. Boca Raton. FL

Mr. David Bakalar, Chestnut Hill. MA

Mr. Charles A. Baker. The Liposome Company. Inc.. Princeton. NJ

Dr. George P. Baker. Mass General Hospital. Boston, MA

Dr. Sumner A. Barenberg, Bernard Technologies. Chicago. IL

Mr. Mel Barkan. The Barkan Companies, Boston. MA

Mr. Bruce A. Beal, The Beal Companies. Boston. MA

Mr. Robert P. Beech. Component Software International. Inc.,

Mason. OH

Mr. George Berkowitz, Legal Sea Foods, Allston. MA
Jewelle and Nathaniel Bickford. New York, NY
Dr. Elkan R. Blout. Harvard Medical School. Boston. MA
Mr. and Mrs. Philip Bogdanovitch. Lake Clear, NY
Mr. Malcolm K. Brachman. Northwest Oil Company. Dallas, TX
Dr. Goodwin M. Breinin. NY University Medical Center,

New York. NY

Mr. John Callahan, Carpenter. Sheperd & Warden. New London. NH

Mrs. Elizabeth Campanella. West Falmouth. MA

Thomas S. Crane. Esq.. Mintz Levin Cohen Ferris Glovsky & Popeo,

PC. Boston. MA

Dr. Stephen D. Crocker. Cyber Cash Inc.. Reston. VA
Mrs. Lynn W. Piasecki Cunningham. Piasecki Productions.

Brookline. MA
Dr. Anthony J. Cutaia. Anheuser-Busch. Inc.. St. Louis. MO

Dr. Georges de Menil, DM Foundation. New York. NY

Mrs. Sara Greer Dent, Chevy Chase. MD

Mr. D. H. Douglas-Hamilton. Hamilton Thome Research, Beverly, MA

Mr. Benjamin F. du Pont, Du Pont Company, Deepwater, NJ

Dr. Sylvia A. Earle. Deep Ocean Engineering, Oakland. CA

Mr. and Mrs. Hoyt Ecker. Vero Beach. FL

Mr. Anthony B. Evnin, Venrock Associates. New York. NY

Stuart Feiner, Esq., Inco Limited. Toronto. Ontario. Canada
Mrs. Hadley Mack French. Edsel and Eleanor Ford House. Grosse
Pointe Farms. MI

Mr. and Mrs. Huib Geerlings. Boston, MA

Mr. William J. Gilbane. Jr., Gilbane Building Company. Providence. RI
Dr. Michael J. Goldblatt, Defense Sciences Office. Arlington, VA
Mr. Maynard Goldman. Maynard Goldman & Associates. Boston. MA

Ms. Charlotte I. Hall. Edgartown. MA

Dr. Thomas R. Hedges. Jr.. Neurological Institute. PA Hospital,

Philadelphia, PA
Drs. Linda Hirshman and David Forkosh. Brandeis University & FMH

Foundation. Waltham, MA
Mr. Thomas J. Hynes, Jr., Meredith & Grew, Inc., Boston, MA

Mrs. Elizabeth Ford Kontulis. New Canaan, CT

Mr. and Mrs. Robert Lambrecht. Boca Grande. FL

Dr. Catherine C. Lastavica. Tufts University School of Medicine,

Boston, MA

Mr. Joel A. Leavitt. Boston. MA

Mr. Stephen W. Leibholz, TechLabs, Inc.. Huntingdon. PA
Mrs. Margaret Lilly. West Falmouth, MA
Mr. Richard Lipkin, Laird & Co.. LLC, New York, NY
Mr. George W. Logan. Valley Financial Corp., Roanoke. VA

Mr. Michael T. Martin. SportsMark. Inc., New York. NY

Dr. Walter E. Massey, President, Morehouse College. Atlanta. GA

Mrs. Christy Swift Maxwell. Grosse Pointe Farms, MI

Mr. Ambrose Monell. G. Unger Vetlesen Foundation, Palm Beach, FL

Dr. Mark Novitch. Washington. DC

Mr. David R. Palmer. David Ross Palmer & Associates, Waquoit. MA
Dr. Rodenc B. Park. Richmond. CA

Mr. Santo P. Pasqualucci, Falmouth Co-Operative Bank. Falmouth. MA
Mr. Robert Pierce, Jr., Pierce Aluminum Co., Canton, MA

Mr. Richard Reston, Vineyard Gazette, Edgartown, MA

Mr. Marius A. Robinson. Fundamental Investors Ltd., Biscayne, FL

John W. Rowe. M.D., Mt. Sinai School of Medicine & Mt. Sinai

Medical Center, New York. NY
Mr. Edward Rowland. Tucker. Anthony, Inc.. Boston, MA

Mr. Gregory A. Sandomirsky. Mintz Levin Cohen Ferris Glovsky &

Popeo, PC, Boston. MA
Mrs. Mary Schmidek, Marion. MA
Dr. Cecily C. Selby. New York. NY
Mr. Robert S. Shifman. St. Simon's Island. GA
Mr. and Mrs. Gregory Skau. Grosse Pointe Farms, MI
Mr. Malcolm B. Smith. General American Investors Co.,

New York, NY

Mr. John C. Stegeman. Campus Rentals. Ann Arbor. MI
Mr. Joseph T. Stewart. Jr., Skillman, NJ

Mr. John W. Stroh, III, The Stroh Brewery Company, Detroit, MI
Mr. Gerard L. Swope. Washington, DC
Mr. John F. Swope. Concord, NH

Mr. and Mrs. Stephen E. Taylor. Milton. MA

Mrs. Donna Vanden Bosch-Flynn. Spring Lake, NJ
Mrs. Carolyn W. Verbeck, Vineyard Haven. MA

Mr. Benjamin S. Warren III. Grosse Pointe Farms. MI

Nancy B. Weinstein. R.N.. The Hospice, Inc.. Glen Ridge, NJ

Stephen S. Weinstein, Esq.. Morristown. NJ

Mr. Frederick J. Weyerhaeuser. Beverly, MA

Mrs. Robin Wheeler. West Falmouth. MA

Mr. Tony L. White. The Perkin Elmer Corporation. Norwalk, CT

Administrative Support Staff1

Biological Bulletin

Greenberg, Michael J.. Editor-in-Chief
Hinkle, Pamela Clapp, Managing Editor

Bums, Patricia

Gibson, Victoria R.

Marrama, Carol

Schachinger, Carol H.

Director's Office

Burris, John E., Director and Chief Executive Officer
Speck. William T., Interim Director and Chief Executive Officer
Donovan, Marcia H.

Associates Program
Bohr. Kendall B.
Sgarzi, Dorothy J.2

Communications Office

Hinkle, Pamela Clapp, Director

Flynn, Bridget

Hartmann, Kelley2

Joslin. Susan

Kent, Elizabeth F.2

Langill. Christine2

Liles. Beth R.

Mossman, Beverly

Sloboda. Lara N.2

Equal Emplo\mt'ni Opportunity
MacNeil, Jane L.

Veterinarian Services
Dushman, Beth2
O'Shea. Erin2
Smolowitz. Roxanna
St. Pierre, Aimee2
Stukey. Jetley
Watmough, Elizabeth2
Weiss, Erica2

Ecosystems Center Administrative Staff

Berthel, Dorothy J.
Donovan. Suzanne J.
Nunez, Guillermo
Seifert, Mary Ann

External Affairs

Carotenuto, Frank C.. Director
Butcher, Valerie
Faxon, Wendy P.
Johnson, A. Kristine
Martin, Theresa H.
Patch-Wing. Dolores
Quigley, Barbara A.
Shaw, Kathleen L.

1 Including persons who joined or left the staff during 2000.

2 Summer or temporary.

Financial Sen'ices Office

Lane. Jr., Homer W.. Chief Financial Officer

Bowman. Richard. Controller

Mullen. Richard J., Manager, Research Administration

Afonso, Janis

Aguiar, Deborah

Bliss. Casey M.

Crosby, Kenneth

Lancaster, Cindy

Livingstone, Suzanne

McLaughlin. Rebecca Jill

Ran/mger, Laura

Stellrecht. Lynette

Stock Room

Schorer, Timothy M., Supervisor

Burnette, Donald

Olive. Charles W.. Jr.

Treistman. Ethan2

Purchasing

Hall, Lionel E., Jr.. Supervisor
Barkley, Rachel A.2
Hunt. Lisa M.

Housing and Conferences

Beckwith. Mary M.. Director
King. LouAnn D.. Director

Adams. Jessica2

Campbell. Anne T.

Grasso. Deborah

R68

Administrative Support Staff R69

Hanlon, Arlene K.2
Johnson-Horman, Frances N.
Kruger. Sally J.
Masse, Todd C.
Perito, Diana

Human Resources

Goux. Susan P., Director
Houser. Carmen

Josephine Bay Paul Center for Comparative Molecular
Biology and Evolution Administrative Staff

Harris. Marian
Lim, Pauline
Sansone, Rebecca
Schlichter. Mimi

Journal of Membrane Biology

Loewenstein, Werner R., Editor
Fay. Catherine H.
Howard Isenberg. Linda L.
Lynch. Kathleen F.

Marine Resources Center

Hanlon. Roger T., Director
Santore. Gabrielle

Aquatic Resources Department

Enos, Edward G.. Jr.. Superintendent

Dimond. James L.2

Grossman, William M.

Gudas. Christopher N.2

Klimm. Henry W., Ill

Potter. Benjamin2

Reynolds. Justin M.2

Sexton, Andrew W.

Sullivan. Daniel A.

Tassinari. Eugene

Whelan. Sean P.

MRC Life Support System

Mebane. William N., Systems Operator

Carroll, James

Hanley, Janice S.

Kuzirian, Alan

MBL/WHOI Library

Norton. Catherine N.. Director
Ashmore, Judith A.
Costa, Marguerite E.
Deveer, Joseph M.
Farrar. Stephen R L
Lavoie. Amy2
Martel, David2
Monahan. A. Jean
Moniz. Kimberly L.
Moore. Laurel E.
Nelson, Heidi
Person. Matthew

Riley. Jacqueline
Walton, Jennifer

Copy Center

Moumford, Rebecca J.. Supervisor

Brissenden. Roberta2

Clark, Sarah2

Clark. Tamara L.

Cosgrove. Nancy

Douglas, Valerie M.2

Eldridge, Myles2

Jenkins, Sarah2

Mancini, Mary E.

Reuter, Laura

Information Systems Division

Inzina, Barbara, Network Manager

Borst, Douglas T.2

Campbell. David J., Jr.

Cohen, Alex2

Douglas, Valerie2

Houser. Clarissa2

Jones, Patricia L.

Kokmeyer, Remmert2

Lowell, Gregory

Mountford, Rebecca J.

Moynihan. James V.

Purdy, Heather

Remsen. David P.

Renna, Denis J.

Space. David B.

Wheeler, Patrick

Williams, Shelly R.2

NASA Center for Advanced Studies in the Space Life Sciences

Blazis, Diana, Administrator
Golden. Catherine
Oldham. Pamela

Research Administration and Educational Programs

Dawidowicz, Eliezar A.. Director
Brooks, Marilyn2
Hamel. Carol C.
Holzworth, Kelly
Kaufmann, Sandra J.
Mebane, Dorianne C.
White, Laurie

Central Microscopv Facility and General Use Rooms
Kerr, Louis M.. Supervisor
Bennett-Stamper, Christina2
Luther, Herbert
Peterson. Martha B.

Safety Sen'ices

Mattox. Andrew H., Environmental, Health, and Safety Manager
Normand, Danielle2

R70 Annual Report

Satellite/Periwinkle Children 's Programs

Robinson. Paulina H.2
Audran, Chantal2
Beaudoin, Cynthia2
Bothner. Katharine2
Brown, Shannon2
Duncan, Brett M.2
Fitzelle, Annie2
Gallant, Carolyn2
Gallant, Cynthia2
Guiffrida, Beth2
Halter. Sarah2
Karalekas, Nina2
Noonan, Brendan2
Noonan, Ryan2
Pascavage, Leigh2
Shanley. Jennifer2
Shwartz, Cortney2

Sen'ice, Projects and Facilities

Cutler. Richard D.. Director
Enos, Joyce B.
Guarente. Jeffrey2

Apparatus
Baptiste. Michael G.
Barnes, Franklin D.
Haskins, William A.
Pratt, Barry

Building Sen'ices and Grounds

Hayes, Joseph H., Superintendent

Anderson, Lewis B.

Atwood, Paul R.

Baker. Harrison S.

Barnes, Susan M.

Berrios, Jessica

Berube. Douglas T.2

Berthel. Frederick

Billings. Julia2

Boucher. Richard L.

Brereton. Richard S.2

Cameron. Lawrence M.2

Chen, Zhi Xin

Clayton. Daniel

Collins, Paul J.

Cutillo, David

Djelidi. Meriam J.2

Doherty, Garrett2

Fernandez. Peter R.2

Ficher. Jason

Frisk. Maria2

Gibbons, Roberto G.

Gore. Simon J.2

Hannigan, Catherine

Heede, Kelly2

Houle, Michael E.2

Illgen, Robert F.
Kaczmarek. Konrad2
Keefe. Edward C.
Kijowski, Wojciech2
Ledwell. L. Patrick2
MacDonald. Cynthia C.
Massi. Christopher
McGee, Melissa
McHugh, Mary2
McNamara. Noreen M.
McQuillan. Jeffrey2
McVey, Brienna2
Mendoza. Duke R.2
O'Brien, William P.2
Peros. Kristina2
Pratt. Barry
Rana, Saoud2
Robinson, Marva
Ryan. Nicholas P.2
Santiago. Crystal2
Schlemermeyer, Jaan2
Shum. Mei Wah
Sizelove, Robert2
Ullian, Adam2
Wagner, Paul2
Ware, Lynn M.
Waterbury, Matthew2

Plant Operations and Maintenance

Fleet, Barry M.. Manager

Cadose. James W.. Maintenance Supervisor

Barnes. John S.

Blunt, Hugh F.

Bourgoin. Lee E.

Callahan. John J.

Carroll, James R.

Duncan. Brett2

Elias. Michael

Fish, David L.. Jr.

Fuglister. Charles K.

Goehl, George

Gonsalves. Walter W., Jr.

Hathaway, Peter J.

Henderson, Jon R.

Kelley. Kevin

Langill, Richard

Lochhead. William M.

McAdams. Herbert M.. Ill

McHugh. Michael O.

Mills, Stephen A.

Olive. Charles W.. Jr.

Rattacasa. Frank2

Rozum. John

Scanlan. Melanie

Settlemire. Donald

Shepherd, Denise M.

Toner, Michael

Wetzel, Ernest D.2

Members of the Corporation

Life Members

Acheson, George H., 25 Quissett Avenue. Woods Hole, MA 02543

(deceased 2000]
Adelberg, Edward A., 204 Prospect Street, New Haven, CT 0651 1-

2107
Afzelius, Bjorn, University of Stockholm. Wenner-Gven Institute.

Department of Ultrastructure Research. Stockholm. SWEDEN
Amatniek, Ernest, (address unknown)
Arnold, John M., 329 Sippewissett Road, Falmouth. MA 02540

Copeland, D. Eugene, Marine Biological Laboratory, Woods Hole. MA

02543

Corliss, John O., P.O. Box 2729, Bala Cynwyd. PA 19004-21 16
Costello, Helen M., 137 Carolina Meadows, Chapel Hill, NC

27514-8512
Crouse, Helen, Rte. 3, Box 213, Hayesville. NC 28904

DeHaan. Robert L., Emory University School of Medicine, Department
of Anatomy & Cell Biology, 1648 Pierce Drive, Room 108, Atlanta.
GA 30322

Dudley, Patricia L., 3200 Alki Avenue SW, #401, Seattle. WA 981 16

Bang. Betsy G., 76 F. R. Lillie Road. Woods Hole. MA 02543
Bartlett, James H., University of Alabama, Department of Physics, Box

X70324. Tuscaloosa. AL 35487-0324 (deceased 2000)
Berne, Robert M., 1 9 Gardiner Road, Woods Hole, MA 02543
Bernheimer, Alan W., New York University Medical Center,

Department of Microbiology. 550 First Avenue, New York. NY

10016
Bertholf, Lloyd M., Westminster Village. #2114, 2025 East Lincoln

Street. Bloomington, IL 61701-5995

Bosch, Herman F., 163 Elm Road. Falmouth. MA 02540-2430
Brinley, F. J., National Institutes of Health. NINCDS, Neurological

Disorders Program. Room 812 Federal Building. Bethesda, MD 20892
Buck, John B., Fairhaven C-020. 7200 Third Avenue, Sykesville, MD

21784

Burbanck. Madeline P., P.O. Box 15134, Atlanta. GA 30333
Burbanck, William D., P.O. Box 15134, Atlanta. GA 30333

Clark, Arnold M., 53 Wilson Road, Woods Hole, MA 02543
Clark, James M., 258 Wells Road. Palm Beach, FL 33480-3625
Cohen, Seymour S., 10 Carrot Hill Road. Woods Hole, MA

02543-1206
Colwin, Arthur L., 320 Woodcrest Road, Key Biscayne. FL

33149-1322
Colwin. Laura Hunter, 320 Woodcrest Road, Key Biscayne. FL

33149-1322
Cooperstein, Sherwin J., University of Connecticut, School of

Medicine. Department of Anatomy, Farmington, CT 06030-3405

Edwards, Charles, 3429 Winding Oaks Drive. Longboat Key. FL

34228
Elliott, Gerald F., The Open University Research Unit, Foxcombe Hall.

Berkeley Road, Boars Hill, Oxford OX1 5HR, UK

Failla, Patricia M., 2149 Loblolly Lane, Johns Island, SC 29455
Frazier, Donald T., University of Kentucky Medical Center.

Department of Physiology and Biophysics, MS501 Chandler Medical

Center. Lexington, KY 40536

Gabriel, Mordetai L., Brooklyn College, Department of Biology. 2900

Bedford Avenue. Brooklyn. NY 1 1210
Glusman, Murray, New York State Psychiatric Institute. 722 W. 168th

St.. Unit #70, New York. NY 10032
Graham, Herbert, 36 Wilson Road, Woods Hole, MA 02543

Hamburger, Viktor, Washington University, Department of Biology,

740 Trinity Avenue. St. Louis. MO 63130 (deceased 2001)
Hamilton, Howard L., University of Virginia, Department of Biology,

238 Gilmer Hall. Charlottesville. VA 22901

Harding, Clifford V.. Jr., 100 Saconesset Road, Falmouth. MA 02540
Haschemeyer, Audrey E. V., 21 Glendon Road. Woods Hole. MA

02543-1406

Hayashi, Teru, 15 Gardiner Road. Woods Hole, MA 02543-1 1 13
Hisaw, Frederick L., (address unknown)
Hoskin, Francis C. G., do Dr. John E. Walker. U.S. Army Natick

RD&E Center. SAT NC-YSM, Kansas Street, Natick, MA 01760-

5020

R71

R72 Annual Report

Hubbard, Ruth, Harvard University, Biological Laboratories,

Cambridge. MA 02138
Hunter, W. Bruce, 305 Old Sharon Road. Peterborough, NH 03458-

1736
Hurwitz, Charles, Stratton VA Medical Center, Research Service,

Albany, NY 12208

Katz, George, 1636 Brookhouse Drive. Apt. BR131. Sarasota, FL

34731
Kingsbury, John M., Cornell University, Department of Plant Biology,

Plant Science Building, Ithaca. NY 14853
K li iiilini/. Lewis, Reed College, Department of Biology. 3203 SE

Woodstock Boulevard. Portland. OR 97202
Kusano, Kiyoshi, National Institutes of Health, Building 36, Room 4D-

20, Bethesda. MD 20892

Laderman, Ezra, Yale University. New Haven, CT 06520
LaMarche, Paul H., Eastern Maine Medical Center, 489 State Street,

Bangor, ME 04401
Lauffer, Max A., Penn State University Medical Center. Department of

Biophysics & Physiology, Hershey. PA 17033
Levitan, Herbert, National Science Foundation, 4201 Wilson

Boulevard. Room 835, Arlington, VA 22230
Lochhead, John H., 49 Woodlawn Road, London SW6 6PS, UK
Loewus, Frank A., Washington State University. Institute of Biological

Chemistry. Pullman, WA 99164
Loftfield, Robert B., University of New Mexico. School of Medicine,

915 Stanford Drive, Albuquerque, NM 87131
Lorand, Laszlo, Northwestern University Medical School, CMS

Biology, Searle 4-555, 303 East Chicago Avenue, Chicago. 1L 60611-

3008

Mainer, Robert E., The Boston Company, Inc., One Boston Place.

OBP-15-D. Boston, MA 02108

Malkiel, Saul, 174 Queen Street, #9A, Falmouth, MA 02540
Marsh, Julian B., 9 Eliot Street. Chestnut Hill, MA 02467-1407
Martin, Lowell V., 10 Buzzards Bay Avenue, Woods Hole. MA 02543
Mathews, Rita W., East Hill Road, P.O. Box 237. Southfield. MA

01259-0237
Metuzals, Janis, University of Ottawa Faculty of Medicine, Department

of Pathology, 451 Smyth Road, Ottawa, ON K1H 8M5. Canada
Moore, John A., University of California, Department of Biology,

Riverside, CA 92521
Moore, John W., Duke University Medical Center, Department of

Neurobiology. Box 3209, Durham. NC 27710

Moscona, Aron A., 221 West 82nd Street, #8C, New York, NY 10024
Musacchia, X. J., P.O. Box 5054, Bella Vista, AR 72714-0054

Nasatir, Maimon, P.O. Box 379, Ojai, CA 93024

Passano, Leonard M., University of Wisconsin. Department of

Zoology, Birge Hall. Madison. WI 53706
Price, Carl A., 20 Maker Lane. Falmouth, MA 02540
Prosser, C. Ladd, University of Illinois. Department of Physiology. 524

Bumll Hall. Urbana, IL 61801
Prytz, Margaret McDonald, (Address unknown)

Renn, Charles E., (Address unknown)

Reynolds, George T., Princeton University, Department of Physics.

Jadwin Hall, Princeton, NJ 08544
Rice, Robert V., 30 Burnham Drive, Falmouth, MA 02540

Rockstein, Morris, 600 Biltmore Way. Apt. 805. Coral Gables. FL

33134
Roth, Jay S., 26 Huettner Road, P.O. Box 692. Woods Hole. MA

02543-0692
Ronkin, Raphael R., 32 1 2 McKinley Street, NW. Washington, DC

20015-1635
Roslansky, John D., 57 Buzzards Bay Avenue, Woods Hole, MA

02543
Roslansky, Priscilla F., Associates of Cape Cod, Inc., P.O. Box 224.

Woods Hole. MA 02543-0224

Sanders, Howard L., Woods Hole Oceanographic Institution, Woods

Hole. MA 02543 (deceased 2001)
Sato, Hidemi, Nagova University, 3-24-101. Oakimshi Machi. Toba

Mie 517-0023, JAPAN

Schlesinger, R. Walter, 7 Langley Road, Falmouth. MA 02540-1809
Scott, Allan C., Colby College. Waterville. ME 04901
Silverstein, Arthur M., Johns Hopkins University, Institute of the

History of Medicine, 1900 E. Monument Street, Baltimore. MD

21205
Sjodin, Raymond A., 3900 N. Charles Street, Apt. #1301, Baltimore,

MD 21218-1719

Smith, Paul F., P.O. Box 264, Woods Hole, MA 02543-0264
Speer, John W., 293 West Main Road. Portsmouth. RI 0287 1
Sperelakis, Nicholas, University of Cincinnati, Department of

Physiology/Biophysics. 231 Bethesda Avenue, Cincinnati. OH 45267-

0576
Spiegel, Evelyn, Dartmouth College. Department of Biological Sciences,

204 Oilman, Hanover, NH 03755
Spiegel, Melvin, Dartmouth College. Department of Biological

Sciences, 204 Oilman. Hanover, NH 03755
Stephens, Grover C., University of California. School of Biological

Sciences, Department of Ecology and Evolution/Biology, Irvine. CA

92717
Strehler, Bernard L., 31561 Crystal Sands Drive. Laguna Niguel, CA

92677

Sussman, Maurice, 72 Carey Lane, Falmouth. MA 02540
Sussman. Raquel B., Marine Biological Laboratory. Woods Hole. MA

02543
Szent-Gyorgyi, Gwen P., 45 Nobska Road, Woods Hole, MA 02543

Thorndike, W. Nicholas, Wellington Management Company. 200 State

Street, Boston, MA 02109
Trager, William, The Rockefeller University, 1230 York Avenue. New

York. NY 10021-6399
Trinkaus, J. Philip, Yale University, Department of Molecular. Cellular

and Developmental Biology. Osborne Memorial Laboratory. New

Haven. CT 06520

Villee, Claude A., Jr., Harvard Medical School. Carrel L, Countway

Library. 10 Shattuck Street. Boston, MA 021 15
Vincent, Walter S., 16 F.R. Lillie Road. Woods Hole, MA 02543

Waterman, Talbot H., Yale University, Box 208103, 912 KBT Biology

Department, New Haven. CT 06520-8103
Wigley, Roland L., 35 Wilson Road, Woods Hole, MA 02543
Witkovsky, Paul, NYU Medical Center, Department of Ophthalmology.

550 First Avenue, New York. NY 10016

Members

Abt, Donald A., Aquavet. University of Pennsylvania. School of
Veterinary Medicine. 230 Main Street, Falmouth. MA 02540

Members of the Corporation R73

Adams, James A., 3481 Paces Ferry Road, Tallahassee, FL 32308
Adelman. William J., 160 Locust Street. Falmouth. MA 02540
Alkon, Daniel L., National Institutes of Health, Laboratory of Adaptive

Systems. 36 Convent Drive, MSC 4124, 36/4A21, Bethesda, MD

20X92-4124
Allen, Garland E., Washington University, Department of Biology, Box

1 137, One Brookings Drive, St. Louis, MO 63130-4899
Allen. Nina S., North Carolina State University. Department of Botany.

Box 7612. Raleigh. NC 27695
Alliegro, Mark C., Louisiana State University Medical Center.

Department of Cell Biology and Anatomy. 1901 Perdido Street, New

Orleans. LA 70112
Anderson, Everett, Harvard Medical School. Department of Cell

Biology, 240 Longwood Avenue, Boston, MA 021 15-6092
Anderson, John M., 1 10 Roat Street, Ithaca, NY 14850
Anderson, Porter W., 914 Grande Avenue, Key Largo, FL 33037
Armett-Kibel, Christine, University of Massachusetts. Dean of Science

Faculty, Boston, MA 02125
Armstrong, Clay M., University of Pennsylvania School of Medicine,

B701 Richards Building, Department of Physiology, 3700 Hamilton

Walk, Philadelphia, PA 19104-6085

Armstrong, Ellen Prosser, 57 Millfield Street, Woods Hole, MA 02543
Armstrong, Peter B., University of California. Section of Molecular

and Cell Biology. 149 Bnggs Hall. Davis, CA 95616-8755
Arnold, William A., Oak Ridge National Laboratory, Biology Division,

102 Balsalm Road, Oak Ridge, TN 37830
Ashton, Robert W., Bay Foundation. 1 7 West 94th Street. New York,

NY 10025
Atema, Jelle, Boston University Marine Program, Marine Biological

Laboratory. Woods Hole. MA 02543

Baccetti, Baccio, University of Sienna. Institute of Zoology, 53100

Siena. Italy
Baker, Robert G., New York University Medical Center, Department

Physiology and Biophysics, 550 First Avenue, New York, NY 10016
Baldwin, Thomas O., University of Arizona. Department of

Biochemistry. P.O. Box 210088, Tucson, AZ 85721-0088
Baltimore. David, California Institute of Technology. 1200 East

California Boulevard. Pasadena. CA 91 125
Barlow, Robert B., SUNY Upstate Medical University, Center for

Vision Research. 750 East Adams Street, Syracuse, NY 13210
Barry, Daniel T., National Aeronautics and Space Administration, Lyn

B. Johnson Space Center, 2101 NASA Road 1, Houston, TX 77058
Barry, Susan R., Mount Holyoke College, Department of Biological

Sciences, South Hadley, MA 01075
Bass. Andrew H., Cornell University, Department of Neurobiology and

Behavior, Seely Mudd Hall, Ithaca, NY 14853
Battelle, Barbara-Anne, University of Florida. Whitney Laboratory,

9505 Ocean Shore Boulevard. St. Augustine, FL 32086
Bay, Frederick, Bay Foundation, 17 W. 94th Street, First Floor, New

York. NY 10025-7116

Baylor, Martha B., P O Box 93. Woods Hole, MA 02543
Bearer, Elaine L., Brown University, Division of Biology and

Medicine, Department of Pathology, BMC 518. Providence. RI 02912
Beatty, John M., University of Minnesota, Department of Ecology and

Behavioral Biology, 1987 Gortner. Street Paul, MN 55108
Beauge, Luis Alberto, Institute de Investigacion Medica. Department of

Biophysics. Casilla de Correo 389, Cordoba 5000. Argentina
Begenisich, Ted, University of Rochester, Medical Center, Box 642,

601 Elmwood Avenue, Rochester, NY 14642
Begg, David A., University of Alberta. Faculty of Medicine.

Department of Cell Biology and Anatomy, Edmonton. Alberta T6G

2H7, Canada

Bell, Eugene, 305 Commonwealth Avenue, Boston, MA 02115
Benjamin, Thomas L., Harvard Medical School, Pathology. D2-230,

200 Longwood Avenue, Boston. MA 021 15
Bennett, Michael V. L., Albert Einstein College of Medicine.

Department of Neuroscience, 1300 Morris Park Avenue, Bronx, NY
10461
Bennett, Miriam F., Colby College. Department of Biology, Waterville,

ME 04901
Bennett, R. Suzanne, Albert Einstein College of Medicine, Department

of Neuroscience, 1410 Pelham Parkway South. Bronx, NY 10461
Berg, Carl J., Jr., P.O. Box 681, Kilauea, Kauai, HI 96754-0681
Berlin, Suzanne T., 87 Payneton Hill Road, York, ME 03909-5401
Bernstein, Norman, Columbia Realty Venture, 5301 Wisconsin

Avenue, NW, #600, Washington, DC 20015-2015
Bezanilla, Francisco, Health Science Center, Department of Physiology,

405 Hilgard Avenue, Los Angeles, CA 90024
Biggers, John D., Harvard Medical School, Department of Physiology,

Boston. MA 02115

Bishop. Stephen H., 2609 Eisenhower, Ames, I A 50010
Blaustein, Mordecai P., University of Maryland. School of Medicine,

Department of Physiology, Baltimore, MD 21201
Blazis, Diana E. J., Marine Biological Laboratory. Center for Advanced

Studies in the Space Life Sciences, Woods Hole, MA 02543
Blennemann, Dieter, 1117 East Putnam Avenue, Apt. #174, Riverside,

CT 06878-1333
Bloom, George S., The University of Texas Southwestern Medical

Center. Department of Cell Biology and Neuroscience, 5323 Harry

Hines Boulevard, Dallas, TX 75235-9039
Bloom, Kerry S., University of North Carolina. Department of Biology,

623 Fordham Hall CB#3280, Chapel Hill, NC 27599-3280
Bodznick, David A., Wesleyan University. Department of Biology,

Lawn Avenue. Middletown, CT 06497-0170
Boettiger, Edward G., P.O. Box 48, Rochester, VT 05767-0048
Boolootian, Richard A., Science Software Systems, Inc., 3576

Woodcliff Road, Sherman Oaks, CA 91403
Borgese, Thomas A., Lehman College, CUNY. Department of Biology,

Bedford Park Boulevard. West, Bronx, NY 10468
Borst, David W., Jr., Illinois State University. Department of

Biological Sciences. Normal, IL 61790-4120
Bowles, Francis P., Marine Biological Laboratory, Ecosystems Center.

Woods Hole. MA 02543
Boyer, Barbara C., Union College, Biology Department, Schenectady,

NY 12308
Brandhorst, Bruce P., Simon Eraser University, Institute of Molecular

Biology/Biochemistry, Barnaby, B.C. V5A 1S6, Canada
Brinley, F. J., Jr., NINCDS/NIH, Neurological Disorders Program,

Room 812 Federal Building, Bethesda, MD 20892
Bronner-Fraser, Marianne, California Institute of Technology,

Beckman Institute Division of Biology, 139-74, Pasadena, CA 91125
Brown, Stephen C., SUNY, Department of Biological Sciences,

Albany, NY 12222

Brown, William L., 80 Black Oak Road, Weston, MA 02193
Browne, Carole L., Wake Forest University, Department of Biology,

Box 7325 Reynolda Station, Winston-Salem, NC 27109
Browne, Robert A., Wake Forest University. Department of Biology.

Box 7325. Winston-Salem. NC 27109
Bucklin, Anne C., University of New Hampshire. Ocean Process

Analysis Laboratory, 142 Morse Hall, Durham, NH 03824
Bullis, Robert A., Oceanic Institute of Applied Aquaculture, 41-202

Kalanianaole Highway, Waimanalo, HI 96795
Burger, Max M., Friedrich Miescher-Institute, P.O. Box 2543. CH-

4002 Basel, Switzerland

R74 Annual Report

Burgess, David R., Boston College. Department of Biology. Higgins

Hall, 140 Commonwealth Avenue, Chestnut Hill, MA 02167
Burgos, Mario H., IHEM Medical School. UNC Conicet. Casilla de

Correo 56. 5500 Mendoza, Argentina
Burky, Albert, University of Dayton, Department of Biology. Dayton,

OH 45469

Burris, John E., Beloit College. 700 College Street, Beloit. WI 5351 1
Burstyn, Harold Lewis, Air Force Research Laboratory (IFOJ), Office

of the Staff Judge Advocate, 26 Electronic Parkway, Rome. NY

13441-4514
Bursztajn, Sherry, LSU Medical Center, 1501 Kings Highway,

Building BRIF 6-13. Shreveport, LA 71 130

Calabrese, Ronald L., Emory University, Department of Biology. 1510

Clifton Road. Atlanta, GA 30322
Cameron, R. Andrew, California Institute of Technology. Division of

Biology 156-29, Pasadena, CA 91 125
Campbell, Richard H., Bang-Campbell Associates, Eel Pond Place,

Box 402. Woods Hole, MA 02543
Candelas, Graciela C., University of Puerto Rico, Department of

Biology. P.O. Box 23360. UPR Station. San Juan. PR 00931-3360
Cariello, Lucio, Stazione Zoologica "A. Dohrn," Villa Comunale,

80121 Naples, Italy
Case, James F., University of California, Marine Science Institute.

Santa Barbara, CA 93106
Cassidy, Father Joseph D., Providence College. Priory of St. Thomas

Aquinas, Providence, RI 02918-0001
Cavanaugh, Colleen M., Harvard University. Biological Laboratories,

16 Divinity Avenue, Cambridge, MA 02138
Chaet, Alfred B., University of West Florida. Department of Cell and

Molecular Biology. 11000 University Parkway, Pensacola, FL 32514
Chambers, Edward L., University of Miami School of Medicine,

Department of Physiology and Biophysics P.O. Box 016430, Miami,

FL 33101
Chang, Donald C., Hong Kong University, Science and Technology,

Department of Biology, Clear Water Bay. Kowloon. Hong Kong
Chappell, Richard L., Hunter College. CUNY, Department of

Biological Sciences, 695 Park Avenue, New York, NY 10021
Child, Frank M., 28 Lawrence Farm Road. Woods Hole. MA 02543-

1416
Chisholm, Rex Leslie, Northwestern University, Medical School.

Department of Cell Biology. Chicago, IL 6061 1
Citkowitz, Elena, Hospital of St. Raphael, Lipid Disorders Clinic, 1450

Chapel Street, New Haven. CT 065 1 1
Clark, Eloise E., Bowling Green State University. Biological Sciences

Department, Bowling Green, OH 43403
Clark, Hays, 150 Gomez Road, Hobe Sound. FL 33455
Clark, Wallis H., Jr., 12705 NW 1 12th Avenue, Alachua, FL 32615
Claude, Fhilippa, University of Wisconsin. Department of Zoology,

Zoology Research Building 125. 1 1 17 W Johnson Street, Madison,

WI 53706
Clay, John R., National Institutes of Health, NINDS, Building 36.

Room 2-CO2. Bethesda. MD 20892
Clowes. Alexander W., University of Washington. School of Medicine,

Department of Surgery, Box 356410, Seattle, WA 98195-6410
Cobb, Jewel Plummer, California State University, 5151 University

Drive. Health Center 205. Los Angeles. CA 90032-8500
Cohen, Carolyn, Brandeis University. Rosenstiel Basic Medical.

Sciences Research Center. Waltham, MA 02254
Cohen, Lawrence B., Yale University School of Medicine, Department

of Physiology. 333 Cedar Street. New Haven. CT 06520
Cohen, Maynard M., Rush Medical College, Department of

Neurological Sciences. 600 South Paulina. Chicago, IL 60612

Cohen, William D., Hunter College. Department Biological Sciences,

695 Park Avenue. New York. NY 10021
Coleman, Annette W., Brown University, Division of Biology and

Medicine. Providence, RI 02912
Colinvaux, Paul, Marine Biological Laboratory, Woods Hole, MA

02543

Collier, Jack R., 3431 Highway. #107, P.O. Box 139, Effie, LA 71331
Collier, Marjorie McCann, 3431 Highway 107. P.O. Box 139, Effie.

LA 71331
Collier, R. John, Harvard Medical School. Department of Microbiology

and Molecular Genetics, 200 Longwood Avenue, Room 356, Boston,

MA 02 1 1 5
Cook, Joseph A., Edna McConnell Clark Foundation, 250 Park Avenue,

New York, NY 10177-0026
Cornwall, Melvin C., Jr., Boston University School of Medicine.

Department of Physiology L714, Boston. MA 021 18
Corson, D. Wesley, Jr., Storm Eye Institute. Room 537. 171 Ashley

Avenue, Charleston, SC 29425
Corwin. Jeffrey T., University of Virginia, School of Medicine,

Department Otolaryngology and Neuroscience. Box 396,

Charlottesville. VA 22908
Couch, Ernest F., Texas Christian University, Department of Biology.

TCU Box 298930, Fort Worth, TX 76129
Cox. Rachel Llanelly, Woods Hole Oceanographic Institute, Biology

Department, Woods Hole. MA 02543
Crane, Sylvia E., c/o Mr. Thomas Crane, 40 Chestnut Street, Weston,

MA 02493
Cremer-Bartels, Gertrud, Horstmarer Landweg 142. 48149 Muenster.

Germany
Crow, Terry J., University of Texas Medical School. Department of

Neurobiology and Anatomy. Houston. TX 77225
Crowell, Sears, Indiana University, Department of Biology,

Bloomington. IN 47405
Crowther, Robert J., Shriners Hospitals for Children. 51 Blossom

Street. Boston. MA 02114
Cummings, Michael P., Marine Biological Laboratory, Bay Paul

Center. Woods Hole. MA 02543
Cunningham, Mary-Ellen, 62 Cloverly Road, Grosse Pointe Farms, MI

48236-33 13 (deceased 2000)
Cutler, Richard D., Marine Biological Laboratory. Woods Hole. MA

02543

Davidson, Eric H., California Institute of Technology. Division of

Biology 156-29. 391 South Holliston. Pasadena, CA 91125
Davison, Daniel B., Bristol-Myers Squibb PR1. Bioinformatics

Department. 5 Research Parkway. Wallingford. CT 06492
Daw, Nigel W., 5 Old Pawson Road. Brunford. CT 06405
Dawidowicz, Eliezar A., Marine Biological Laboratory. Office of

Research Administration and Education. Woods Hole, MA 02543
De Weer. Paul J., University of Pennsylvania, B400 Richards Building.

Department of Physiology. 3700 Hamilton Walk. Philadelphia. PA

19104-6085
Deegan, Linda A., Marine Biological Laboratory. The Ecosystems

Center. Woods Hole, MA 02543
DeGroof, Robert C., 145 Water Crest Drive, Doylestown. PA 18901-

3267
Denckla, Martha Bridge. Johns Hopkins University. School of

Medicine. Kennedy-Krieger Institute. 707 North Broadway. Baltimore.

MD 2 1 205
DePhillips, Henry A., Trinity College, Department of Chemistry, 300

Summit Street. Hartford. CT 06106

Members of the Corporation R75

DeSimone, Douglas \V., University of Virginia. Department of Cell

Biology. Box 439. Health Sciences Center. Charlottesville, VA 22908
Dettbarn, Wolf-Dietrich, 4422 Wayland Drive, Nashville. TN 37215
Dionne. Vincent E., Boston University Marine Program. Marine

Biological Laboratory. Woods Hole, MA 02543
Dow ling, John E., Harvard University, Biological Laboratories, 16

Divinity Street, Cambridge. MA 02138
Drapeau, Pierre, Montreal General Hospital, Department of Neurology,

1650 Cedar Avenue, Montreal, Quebec H3G IA4. Canada
DuBois, Arthur Brooks, John B. Pierce Foundation Laboratory, 290

Congress Avenue. New Haven, CT 06519
Duncan, Thomas K., Nichols College, Environmental Sciences

Department. Dudley, MA 01571
Dunham, Philip B., Syracuse University. Department of Biology, 130

College Place, Syracuse. NY 13244-1220
Dunlap, Paul V., University of Michigan, Department of Biology. 830

North University Avenue, Ann Arbor, MI 48109-1048

Ehert, James D., The Johns Hopkins University. Department of

Biology. Homewood. 3400 North Charles Street. Baltimore. MD

21218-2685 (deceased 2001)
Eckberg, William R.. Howard University. Department of Biology. P.O.

Box 887, Administration Building, Washington. DC 20059
Edds, Kenneth T., R & D Systems, Inc., Hematology Division, 614

McKinley Place, NE. Minneapolis, MN 55413
Eder, Howard A., Albert Einstein College of Medicine. 1300 Morris

Park Avenue. Bronx, NY 10461

Edstrom, Joan, 53 Two Ponds Road, Falmouth, MA 02540
Egyud, Laszlo G., Cell Research Corporation, P.O. Box 67209,

Chestnut Hill, MA 02167-0209
Ehrlich, Barbara E., Yale University Medical School, Department of

Pharmacology, Sterling Hall of Medicine, B207, 333 Cedar Street.

New Haven, CT 06520-8066
Eisen, Arthur Z., Washington University, Division of Dermatology,

St. Louis. MO 63110
Eisen, Herman N., Massachusetts Institute of Technology. Center for

Cancer Research, El 7- 128, 77 Massachusetts Avenue. Cambridge,

MA 02139-4307
Elder, Hugh Young, University of Glasgow. Institute of Physiology,

Glasgow G12 8QQ. Scotland
Knglund, Paul T., Johns Hopkins Medical School, Department of

Biological Chemistry, 725 North Wolfe Street, Baltimore, MD 21205
Epel, David, Stanford University, Hopkins Marine Station, Ocean View

Boulevard, Pacific Grove, CA 93950
Epstein, Herman T., 18 Lawrence Farm Road, Woods Hole. MA

02543
Epstein, Ray L., 701 Winthrop Street. #311, Taunton, MA 02780-2187

Farb, David H., Boston University School of Medicine. Department of

Pharmacology L603, 80 East Concord Street, Boston, MA 021 18
Earmanfarmaian, A. Verdi, Rutgers University, Department of

Biological Sciences, Nelson Biology Laboratory FOB 1059,

Piscataway. NJ 08855
Feldman, Susan C., University of Medicine and Dentistry. New Jersey

Medical School. 10(1 Bergen Street. Newark, NJ 07103
Festoff, Barry William, VA Medical Center, Neurology Service (151).

4801 Linwood Boulevard, Kansas City, MO 64128
Fink, Rachel D., Mount Holyoke College, Department of Biological

Sciences, Clapp Laboratories, Room 215, South Hadley, MA 01075

Finkelstein, Alan, Albert Einstein College of Medicine. 1300 Morris

Park Avenue, Bronx. NY 10461
Fischbach, Gerald D., Columbia College of Physicians and Surgeons,

630 West 168th Street. R 2-401. New York. NY 10032
Fishman, Harvey M., University of Texas Medical Branch. Department

of Physiology and Biophysics, 301 University Boulevard, Galveston.

TX 77555-0641

Flanagan. Dennis, 12 Gay Street, New York, NY 10014
Fluck, Richard Allen, Franklin and Marshall College. Department of

Biology, Box 3003, Lancaster, PA 17604-3003
Foreman, Kenneth H., Marine Biological Laboratory, Woods Hole,

MA 02543
Fox, Thomas Oren, Harvard Medical School, Division of Medical

Sciences, MEC 435, 260 Longwood Avenue, Boston. MA 021 15
Franzini-Armstrong, Clara, LIniversity of Pennsylvania. School of

Medicine. 330 South 46th Street, Philadelphia, PA 19143
Fraser, Scott, California Institute of Technology. Beckman Institute

139-74, 1201 East California Boulevard, Pasadena, CA 91 125
Frazier, Donald T., University of Kentucky Medical Center.

Department of Physiology and Biophysics, MS501 Chandler Medical

Center. Lexington, KY 40536
French, Robert J., University of Calgary, Health Sciences Centre.

Alberta, T2N 4N1. CANADA
Fulton, Chandler M., Brandeis University. Department of Biology. MS

008, Waltham. MA 02454-91 10
Furie, Barbara C., Beth Israel Deaconess Medical Center, BIDMC

Cancer Center, Kirstein 1, 330 Brookline Avenue, Boston. MA 02215
Furie, Bruce, Beth Israel Deaconess Medical Center, BIDMC Cancer

Center, Kirstein 1. 330 Brookline Avenue, Boston. MA 02215
Furshpan, Edwin J., Harvard Medical School, Department of

Neurobiology, 220 Longwood Avenue, Boston. MA 021 15
Futrelle, Robert P., Northeastern University. College of Computer

Science, 360 Huntington Avenue, Boston, MA 021 15

Gabr, Howaida, Suez Canal University. Department of Marine Science.

Faculty of Science. Ismailia. Egypt
Gadsby, David C., The Rockefeller University, Laboratory of Cardiac

Physiology, 1230 York Avenue. New York, NY 10021-6399
Gainer, Harold, National Institutes of Health. NINDS.BNP.DIR.

Neurochemistry. Building 36. Room 4D20, Bethesda, MD 20892-

4130

Galatzer-Levy, Robert M., 534 Judson Avenue, Evanston, IL 60202
Gall, Joseph G., Carnegie Institution, 1 15 West University Parkway,

Baltimore, MD 21210
Gallo, Michael A., UMDNJ-Robert Wood Johnson Medical School,

EOHSI, Room 408, 170 Frelinghuysen Road, Piscataway, NJ 08854-

8020
Garber, Sarah S., Allegheny University of the Health Sciences,

Department of Physiology. 2900 Queen Lane. Philadelphia. PA 19129
Gelperin, Alan, Bell Labs Lucent. Department Biology Comp., Rm

1C464. 600 Mountain Avenue. Murray Hill. NJ 07974
German, James I.., Ill, Weill Medical College of Cornell University,

1300 York Avenue. New York. NY 10021
Gibbs, Martin, Brandeis University. Institute for Photobiology of Cells

and Organelles. Waltham. MA 02254
(iiblin. Anne E., Marine Biological Laboratory. The Ecosystems

Center. Woods Hole. MA 02543
Gibson, A. Jane, Cornell LIniversity. Department of Biochemistry.

Biotech Building. Ithaca. NY 14850
Gifford, Prosser, Library of Congress, Madison Building LM605,

Washington DC 20540
Gilbert, Daniel L., National Institutes of Health. Biophysics Sec.. BNP,

Building 36, Room 5A-27. Bethesda, MD 20892

R76 Annual Report

Giudice, Giovanni, Universita di Palermo, Dipartimento di Biologia.

Cellulare e Dello Sviluppo, 1-90123 Palermo. Italy
Giuditta, Antonio, University of Naples, Department of General

Physiology, Via Mezzocannone 8, Naples. 80134, Italy
Glynn, Paul, P.O. Box 369, Hampton Falls, NH 03844
Golden, William T., Chairman Emeritus. American Museum of Natural

History, 500 Fifth Avenue. 50* Floor. New York, NY 10110
Goldman, Robert D., Northwestern University Medical School,

Department of Cell and Molecular Biology, 303 E. Chicago Avenue,

Chicago, IL 60611-3008
Goldsmith, Paul K., National Institutes of Health, 9000 Rockville Pike,

Building 10, Room 8C206, Bethesda. MD 20892
Goldsmith, Timothy H., Yale University, Department of Biology, New

Haven. CT06510
Goldstein, Moise H., Jr., The Johns Hopkins University, ECE

Department, Barton Hall, Baltimore, MD 21218
Gould, Robert Michael, NYS Institute of Basic Research, Department

of Pharmacology, 1050 Forest Hill Road, Staten Island, NY 10314-

6399
Govind. C. K., Scarborough College, Life Sciences Division, 1265

Military Trail, West Hill, Ontario MIC 1A4, Canada
Grace, Dick, Doreen Grace Fund, The Brain Center, Promontory Point,

New Seabury, MA 02649
Graf, Werner M., College of France, 1 1 Place Marcelin Berthelot,

75231 Paris Cedex 05, France
Grant, Philip, National Institutes of Health. NINDS\BN\DIR-

Neurochemistry, Building 36, Room 4D20, Bethesda, MD 20892-

4130
Grass, Ellen R., The Grass Foundation, 77 Reservoir Road. Quincy.

MA 02170-3610 (deceased 2001 )
Grassle, Judith P., Rutgers University, Institute of Marine and Coastal

Studies, 71 Dudley Road, New Brunswick. NJ 08901-8521
Graubard, Katherine G., University of Washington. Department of

Zoology, NJ-15, Box 351800, Seattle, WA 98195-1800
Greenberg, Everett Peter, University cf Iowa, College of Medicine,

Department of Microbiology, Iowa City, IA 52242
Greenberg, Michael J., University of Florida, The Whitney Laboratory.

9505 Ocean Shore Boulevard, St. Augustine, FL 32080-8610
Greer, Mary J., 176 West 87th Street. #12A, New York, NY 10024-

2902
Griffin, Donald R., Harvard University, Concord Field Station, Old

Causeway Road, Bedford, MA 01730

Gross, Paul R., 123 Perkins Street, Jamaica Plain, MA 02130
Grossman, Albert, New York University Medical Center, 550 First

Avenue, New York, NY 10016
Grossman, Lawrence, The Johns Hopkins University, Hygien Building,

Room W8306, Baltimore, MD 21205
Gruner, John A., Cephalon, Inc., 145 Brandywine Parkway, West

Chester, PA 19380-4245

Gunning, A. Robert, P. O. Box 165, Falmouth, MA 02541
Gwilliam, G. Francis, Reed College. Department of Biology. Portland.

OR 97202

Haimo, Leah T., University of California. Department of Biology.

Riverside. CA 92521
Hajduk, Stephen L., University of Alabama, School of

Medicine/Dentistry. Department of Biochemistry/Molecular Genetics.

University Station. Birmingham. AL 35294
Hall. Linda M., Shriners Hospital for Children. 2425 Stockton

Boulevard. Sacramento, CA 95817
Halvorson, Harlyn O., University of Massachusetts, Policy Center for

Marine Biosciences and Technology. 100 Morrissey Boulevard,

Boston, MA 02125-3393

Haneji, Tatsuji, The University of Tokushima. Department of Histology

& Oral Histology, School of Dentistry, 18-15, 3 Kuramoto-cho,

Tokushima 770-8504, Japan
Hanlon, Roger T., Marine Biological Laboratory. Woods Hole. MA

02543
Harosi, Ferenc, New College of the USF. Division of Natural Sciences.

5700 North Tamiami Trail, Sarasota, FL 34243-2197
Harrigan, June F.. 7415 Makaa Place, Honolulu. HI 96825
Harrington, Glenn W., Weber State University. Department of

Microbiology, Ogden, UT 84408
Harrington, John P., University of South Alabama, Department of

Chemistry, Mobil. AL 36688
Harrison, Stephen C., Harvard University. Department of Molecular

and Cell Biology, 7 Divinity Avenue, Cambridge. MA 02138
Haselkorn, Robert, University of Chicago, Department of Molecular

Genetics and Cell Biology, Chicago. IL 60637
Hastings, J. Woodland, Harvard University, The Biological

Laboratories, 16 Divinity Avenue, Cambridge, MA 02138-2020
Hayes, Raymond L., Jr., Howard University. College of Medicine, 520

W Street. NW. Washington, DC 20059
Heck, Diane E., Rutgers University, Department of

Pharmacology/Toxicology, 681 Frelinghuysen Road. Piscataway, NJ

08855
Henry, Jonathan Joseph, University of Illinois. Department of Cell and

Structural Biology. 601 South Goodwin Avenue #B107, Urbana, IL

61801-3709
Hepler, Peter K., University of Massachusetts. Department of Biology.

Morrill 111. Amherst. MA 01003
Herndon, Walter R., University of Tennessee, Department of Botany.

Knoxville, TN 37996-1100
Hershko, Avram, Technion-Israel Institute of Technology, Unit of

Biochemistry, The Bruce Rappaport Faculty of Medicine, Haifa

31096, Israel
Herskovits, Theodore T., Fordham University. Department of

Chemistry. John Mulcahy Hall. Room 638. Bronx, NY 10458
Hiatt, Howard H., Brigham and Women's Hospital. Department of

Medicine. 75 Francis Street, Boston, MA 021 15
Highstein, Stephen M., Washington University, Department of

Otolaryngology, Box 8115. 4566 Scott Avenue, St. Louis, MO 631 10
Hildebrand. John G., University of Arizona, ARL Division of

Neurobiology, P.O. Box 210077. Tucson, AZ 85721-0077
Hill, Richard W., Michigan State University, Department of Zoology,

East Lansing, MI 48824
Hill, Susan D., Michigan State University, Department of Zoology, East

Lansing. MI 48824
Hillis, Llewellya W., Marine Biological Laboratory. Woods Hole, MA

02543
Hinchcliffe, Edward H., University of Massachusetts Medical School,

Department of Cell Biology. 377 Plantation Street. Worcester, MA

01605
Hinkle, Gregory J., Bioinformatics Group. Cereon Genomics. 45

Sidney St.. Cambridge. MA 02139
Hinsch, Gertrude W., University of South Florida, Department of

Biology, Tampa. FL 33620

Hinsch, Jan, Leica, Inc.. 110 Commerce Drive. Allendale. NJ 07401
Hobbie, John E.. Marine Biological Laboratory. The Ecosystems

Center. Woods Hole. MA 02543

Hodge, Alan J.. 3843 Mount Blackburn Avenue. San Diego. CA 921 1 1
Hoffman, Joseph F., Yale University School of Medicine. Cellular and

Molecular Physiology, 333 Cedar Street, New Haven, CT 06520-8026
Hollytield, Joe G., The Cleveland Clinic. Opthalmic Research, 9500

Euclid Avenue, Cleveland, OH 44195

Members of the Corporation R77

Holz, George G., IV, New York University Medical Center.

Department of Physiology and Neuroscience, Medical Sciences

Building. Room 442. 550 First Avenue. New York. NY 10016
Hopkinson, Charles S., Jr., Marine Biological Laboratory. Woods

Hole. MA 02543
Houk. James C., Northwestern University Medical School, 303 East

Chicago Avenue. Ward 5-315. Chicago. IL 60611-3008
Hoy. Ronald R., Cornell University. Section of Neurobiology and

Behavior. 215 Mudd Hall. Ithaca. NY 14853
Huang, Alice S., California Institute of Technology, Mail Code 1-9.

Pasadena, CA 91125
Hul'nagel-ZackrotT, Linda A., University of Rhode Island. Department

of Microbiology, Kingston, RI 02881
I liiiiiiiniii. William D., Ohio University. Department of Biological

Sciences, Athens, OH 45701
Humphreys, Susie H., Food and Drug Administration, HFS-308, 200 C

Street, SW, Washington, DC 20204-0001
Humphreys, Tom, University of Hawaii. Kewalo Marine Laboratory,

41 Ahui Street. Honolulu. HI 96813
Hunt. Richard T., ICRF. Clare Hall Laboratories. South Mimms

Potter's Bar. Herts EN6-3LD. England
Hunter, Robert D., Oakland University. Department of Biological

Sciences. Rochester. MI 48309-4401
Huxley, Hugh E., Brandeis University. Rosenstiel Center. Biology

Department. Waltham. MA 02154

Ilan, Joseph, Case Western Reserve University, School of Medicine,

Department of Anatomy. Cleveland, OH 44106
Ingoglia, Nicholas A., New Jersey Medical School, Department of

Pharmacology/Physiology, 185 South Orange Avenue, Newark. NJ

07103
Inoue, Saduyki, McGill University, Department of Anatomy, 3640

University Street, Montreal, PQ H3A 2B2, Canada
Inoue, Shinya, Marine Biological Laboratory. Woods Hole, MA 02543
Isselbacher. Kurt J., Massachusetts General Hospital Cancer Center,

Charlestown. MA 02129
Issidorides, Marietta Radovic, National and Capodistrian University of

Athens, Department of Psychiatry, Eginition Hospital, 74, Vas.

Sophias Avenue. 1 15 28 Athens, Greece
Izzard, Colin S., SUNY-Albany, Department of Biological Sciences,

1400 Washington Avenue. Albany, NY 12222

Jacobs, Neil, Hale and Dorr, 60 State Street, Boston, MA 02109
Jaffe, Laurinda A., University of Connecticut Health Center,

Department of Physiology, Farmington Avenue, Farmington, CT

06032

Jaffe, Lionel, Marine Biological Laboratory, Woods Hole. MA 02543
Jeffery, William R., University of Maryland, Department of Biology,

College Park, MD 20742
Johnston, Daniel, Baylor College of Medicine. Division of

Neuroscience, One Baylor Plaza. Room S740. Houston, TX 77030
Josephson, Robert K., University of California. School of Biological

Science. Department of Psychobioiogy, Irvine, CA 92697

Kaczmarek, Leonard K., Yale University School of Medicine.

Department of Pharmacology. 333 Cedar Street, New Haven, CT

06520
Kaley, Gabor, New York Medical College, Department of Physiology.

Basic Sciences Building, Valhalla, NY 10595
Kaltenbach. Jane, Mount Holyoke College, Department Biological

Sciences. South Hadley, MA 01075

Kaminer, Benjamin, Boston University Medical School. Physiology

Department. 80 East Concord Street. Boston, MA 02 1 1 8
Kaneshiro, Edna S., University of Cincinnati, Biological Sciences

Department, JL 006. Cincinnati. OH 45221-0006
Kaplan, Ehud, Mount Sinai School of Medicine. I Gustave Levy Place.

Box 1 183. New York. NY 10029
Karakashian. Stephen J., Apartment 16-F. 165 West 91st Street. New

York. NY 10024
Karlin, Arthur, Columbia University, Center for Molecular

Recognition, 630 West 168th Street, Room 11-401, New York, NY

10032
Karnovsky, Morris John, Harvard Medical School, Department of

Pathology, 200 Longwood Avenue, Boston, MA 021 15
Keller, Hartmut Ernst, Carl Zeiss, Inc., One Zeiss Drive, Thornwood.

NY 10594
Kelley, Darcy B., Columbia University, Department of Biological

Sciences, 911 Fairchild. Mailcode 2432, New York, NY 10027
Kelly, Robert E., 5 Little Harbor Road, Woods Hole, MA 02543
Kemp, Norman E., University of Michigan, Department of Biology,

Ann Arbor, MI 48109
Kendall, John P., Faneuil Hall Associates, 176 Federal Street, 2nd

Floor, Boston, MA 02110
Kerr, Louis M., Marine Biological Laboratory, Woods Hole, MA

02543
Keynan, Alexander, Israel Academy of Science and Humanity. P.O.

Box 4040, Jerusalem. Israel
Khan, Shahid M. M., Albert Einstein College of Medicine, Department

of Physiology and Biophysics. 1300 Morris Park Avenue, Room

U273, Bronx, NY 10461
Khodakhah, Kamran, University of Colorado School of Medicine,

Department of Physiology and Biophysics, 4200 East 9th Avenue,

C-240. Denver, CO 80262
Kiehart, Daniel P., Duke University Medical Center. Department of

Cell Biology. Box 3709, 308 Nanaline Duke Building, Durham. NC

27710
Kim ill-Ill. David, University of California, Department of Physics, 0319

9500 Gilman Drive, La Jolla, CA 92093
Klessen, Rainer, (address unknown)
Klotz, Irving M., Northwestern University. Department of Chemistry,

Evanston, IL 60201
Knudson, Robert A., Marine Biological Laboratory, Woods Hole, MA

02543
Koide, Samuel S., The Rockefeller University, The Population Council,

1230 York Avenue, New York, NY 10021
Kornberg, Hans, Boston University, The University Professors, 745

Commonwealth Avenue. Boston. MA 02215
Kosower, Edward M., Tel-Aviv University, Department of Chemistry,

Ramat-Aviv. Tel Aviv, 69978, Israel
Krahl. Maurice E., 2783 West Casas Circle, Tucson, AZ 85741

(deceased 2000)
Krane, Stephen M., Massachusetts General Hospital. 55 Fruit Street.

Bulf-165. Boston, MA 02114
Krauss, Robert, P.O. Box 291, Denton, MD 21629
Kravitz, Edward A., Harvard Medical School, Department of

Neurobiology, 220 Longwood Avenue, Boston, MA 021 15
Kriebel, Mahlon E., SUNY Health Science Center, Department of

Physiology. Syracuse, NY 13210
Kristan, William B., Jr., University of California. Department of

Biology 0357, 9500 Gilman Drive. La Jolla, CA 92093-0357
Kropinski, Andrew M., Queen's University, Department of

Microbiology/Immunology, Botterell Hall, Room 74, Kingston,

Ontario K7L 3N6, CANADA

R78 Annual Report

Kuffler, Damien P., Institute of Neurobiology, 201 Boulevard del

Valle. San Juan 00901. PR
knlins. William J., Hospital for Sick Children, Biochemistry Research.

555 University Avenue, Toronto, Ontario M5G 1X8. Canada
Kunkel, Joseph G., University of Massachusetts, Department of

Biology, Amherst. MA 01003
Kuzirian, Alan M., Marine Biological Laboratory. Woods Hole, MA

02543-1015

Laderman, Aimlee D., Yale University, School of Forestry and

Environmental Studies, 370 Prospect Street. New Haven, CT 06511
Landeau, Laurie J., Listowel, Inc.. 2 Park Avenue, Suite 1525, New

York, NY 10016
Landis, Dennis M. D., University Hospital of Cleveland, Department

Neurology, 1 1 100 Euclid Avenue. Cleveland, OH 44106
Landis, Story C., National Institutes of Health, Building 36, Room

5A05, 36 Convent Drive. Bethesda. MD 20892-4150
Landowne. David, University of Miami Medical School, Department of

Physiology and Biophysics, PO Box 016430, Miami. FL 33101
Langford, George M., Dartmouth College, Department of Biological

Sciences, 6044 Oilman Laboratory, Hanover. NH 03755
Laskin, Jeffrey, University of Medical and Dentistry of New Jersey.

Robert Wood Johnson Medical School. 675 Hoes Lane. Piscataway,

NJ 08854
Lasser-Ross, Nechama, New York Medical College. Department of

Physiology. Valhalla, NY 10595
Laster, Leonard, University of Massachusetts Medical School, 55 Lake

Avenue, North, Worcester, MA 01655
Laties, Alan, Scheie Eye Institute, Myrin Circle, 5 1 North 39th Street,

Philadelphia, PA 19104
Laufer, Hans, University of Connecticut, Department of Molecular and

Cell Biology, U-125, 75 North Eagleville Road Storrs, CT 06269-

3125
Lazarow, Paul B., Mount Sinai School of Medicine. Department of

Cell Biology and Anatomy, 1190 Fifth Avenue, Box 1007, New

York, NY 10029-6574
Lazarus, Maurice, Federated Department Stores. Sears Crescent, City

Hall Plaza, Boston, MA 02108
Leadbetter, Edward R., University of Connecticut. Department of

Molecular and Cell Biology. U-131, Beach Hall, Room 249, 354

Mansfield Road. Storrs, CT 06269-2131
Lederberg, Joshua, The Rockefeller University, Suite 400 (Founders

Hall), 1230 York Avenue, New York, NY 10021
Lee, John J., City College of CUNY, Department of Biology, Convent

Avenue and 138th Street, New York. NY 10031
Lehy, Donald B., 35 Willow Field Drive. North Falmouth, MA 02556
Leighton, Stephen B., Beecher Instruments, P.O. Box 8704, Silver

Spring, MD 20910
Lerner, Aaron B., Yale University School of Medicine, Department of

Dermatology. P.O. Box 3333, New Haven, CT 06510
Levin, Jack, Veterans Administration, Medical Center, 1 1 1 H2, 4150

Clement Street. San Francisco. CA 94121
Levine, Michael, University of California. Department MCB. 401

Barker Hall. Berkeley. CA 94720
Levine, Richard B., University of Arizona. Division of Neurobiology,

Room 61 1. Gould Simpson Building. PO Box 210077. Tucson, AZ

85721-0077
Levinthal, Francoise, Columbia University, Department of Biological

Sciences, Broadway and H6th Street, New York, NY 10026
Levitan, Irwin B., University of Pennsylvania. School of Medicine, 218
Stemmler Hall, 3450 Hamilton Walk, Philadelphia, PA 19104-6074

I link, Richard W., University of Minnesota School of Medicine. Cell

Biology and Neuroanatomy Department. 4-135 Jackson Hall, 321

Church Street, Minneapolis. MN 55455
Lipicky. Raymond J., Food and Drug Administration. CDER/ODEI/

HFD-1 10. 5600 Fishers Lane. Rockville, MD 20857
Lisman. John E., Brandeis University. Molecular and Cell Biology, 415

South Street. Waltham, MA 02454-9110
Liuzzi. Anthony, 180 Beacon Street. #8G, Boston. MA 021 16
Llinas, Rodolfo R., New York University Medical Center, Department

of Physiology/Biophysics, 550 First Avenue. Room 442, New York.

NY 10016
Lobel, Phillip S., Boston University Marine Program, Marine Biological

Laboratory. Woods Hole, MA 02543
Loew, Franklin M., Becker College. 61 Sever Street, Worcester, MA

01615-0071

Loewenstein, Birgit Rose, 102 Ransom Rd.. Falmouth, MA 02540
Loewenstein, Werner R., 102 Ransom Rd.. Falmouth. MA 02540
London, Irving M., Harvard-MIT. Division. E-25-551, Cambridge, MA

02139
Longo, Frank J., University of Iowa. Department of Anatomy. Iowa

City, IA 52442
Luckenbill, Louise M., 430 Sippiwissett Road. Falmouth. MA 02540

Macagno, Eduardo R., Columbia University. 109 Low Memorial

Library. Mail Code 4306, New York, NY 10027
MacNichol Edward R., Jr., Boston University School of Medicine.

Department of Physiology. 80 East Concord Street. Boston. MA

02118
Maglott-Duffield, Donna R., American Type Culture Collection, 12301

Parklawn Drive, Rockville, MD 20852-1776
Maienschein, Jane Ann, Arizona State University. Department of

Physiology. P.O. Box 872004. Tempe, AZ 85287-2004
Malbon, Craig C., SUNY, University Medical Center, Pharmacology-

HSC, Stony Brook. NY 11794-8651
Malchow, Robert P., University of Illinois, Department of Biology.

M/C 066, 845 West Taylor Street. Chicago. IL 60607
Manalis. Richard S., Indiana-Purdue University. Department of

Biological Science, 2101 Coliseum Boulevard East, Fort Wayne, IN

46805

Manz, Robert D., P.O. Box 428, Glen Mills. PA 19342
Margulis. Lynn, University of Massachusetts, Department of

Geosciences. Morrill Science Center, Box 35820, Amherst, MA

01003-5820

Marinucci, Andrew C., 102 Nancy Drive, Mercerville, NJ 08619
Martinez, Joe L., Jr., The University of Texas, Division of Life

Sciences, 6900 North Loop 1604 West, San Antonio, TX 78249-0662
Martinez-Palomo, Adolfo, CINVESTAV-IPN, Sec. de Patologia

Experimental. 07000 Mexico. D.F.A.P. 140740. Mexico
Mastroianni, Luigi, Jr., Hospital of University of Pennsylvania, 106

Dulles, 3400 Spruce Street. Philadelphia. PA 19104-4283
Mauzerall, David, Rockefeller University. 1230 York Avenue, New

York, NY 10021
McAnelly. M. Lynne, University of Texas. Section of Neurobiology,

School of Life Sciences, Austin. TX 78712
McCann, Frances V., Dartmouth Medical School. Department of

Physiology. Lebanon. NH 03756
McLaughlin, Jane A., Marine Biological Laboratory1, Woods Hole, MA

022543
McMahon, Robert F., University of Texas, Arlington, Department of

Biology. Box 19498. Arlington, TX 76019
Meedel, Thomas, Rhode Island College, Biology Department. 600

Mount Pleasant Avenue, Providence, RI 02908

Members of the Corporation R79

Meinertzhagen, Ian A., Dalhousie University, Department of

Psychology. Halifax, NS B3H 4J1, Canada
Meiss, Dennis K., Immunodiagnostic Laboratories, 488 McCormick

Street. San Leandro. CA 94577
Melillo, Jerry M., Marine Biological Laboratory, Ecosystems Center.

Woods Hole. MA 02543
Mellon, DeForest, Jr., University of Virginia. Department of Biology,

Gilmer Hall. Charlottesville. VA 22903

Mellon, Richard P., P.O. Box 187, Laughlintown, PA 15655-0187
Mendelsohn, Michael E., New England Medical Center. Molecular

Cardiology Laboratory, NEMC Box 80, 750 Washington Street,

Boston, MA 021 11
Mensinger, Allen F., University of Minnesota. Biology Department,

LSCI 211. Duluth. MN 55812
Merriman, Melanie Pratt, 751 1 Beach View Drive, North Bay Village.

FL 33141
Meselson, Matthew, Harvard University. Fairchild Biochemistry

Building. 7 Divinity Avenue. Cambridge, MA 02138
Miledi. Ricardo, University of California, Irvine. Department of

Psychobiology, 2205 Biology Science II. Irvine. CA 92697-4550
Milkman, Roger D., University of Iowa, Department of Biological

Sciences. Biology Building, Room 318, Iowa City. IA 52242-1324
Miller, Andrew L., Flat 2A, Block 2, Greon Park, Razor Hill,

Clearwater Bay. Kowloon. Hong Kong

Miller, Thomas J., Analogic, 8 Centennial Drive. Peabody, MA 01960
Mills, Robert, 6410 2P1 Avenue W, #311. Brandenton, FL 34210

(deceased 2001)
Misevic, Gradimir, University Hospital of Basel, Department of

Research, Mebelstr. 20, CH-4031 Basel, Switzerland
Mitchell, Ralph, Harvard University, Division of Applied Sciences, 29

Oxford Street, Cambridge, MA 02138
Miyakawa, Hiroyoshi, Tokyo College of Pharmacy, Laboratory of

Cellular Neurobiology, 1432-1 Horinouchi, Hachiouji. Tokyo 192-03,

Japan
Miyamoto, David M., Drew University. Department of Biology.

Madison. NJ 07940
Mi/Hi. Merle, Tulane University, Department of Cell and Molecular,

Biology. New Orleans. LA 70118
Moreira, Jorge E., National Institutes of Health, NICHD. Department

of Cell and Molecular Biol.. Bethesda. MD 20852
Morin, James G., Cornell University, Department of Ecology and

Evolutionary Biology, G14 Stimson Hall, Ithaca. NY 14853-2801
Morrell, Leyla deToledo, Rush-Presbyterian St. Luke's Medical Center,

1653 West Congress Parkway, Chicago, IL 60612
Morse, Stephen S., 275 Central Park West, New York, NY 10024
Mote, Michael L, Temple University. Department of Biology,

Philadelphia, PA 19122
Muller, Kenneth J., University of Miami School of Medicine,

Department of Physiology and Biophysics, 1600 NW 10th Avenue.

R-430. Miami. FL 33136
Murray, Andrew W., University of California, Department of

Physiology. Box 0444, 513 Parnassus Avenue, San Francisco, CA

94143-0444

Nabrit, Samuel M., 686 Beckwith Street. SW. Atlanta. GA 30314
Nadelhoffer, Knute J., Marine Biological Laboratory. 7 MBL Street,

Woods Hole. MA 02543
Nagel, Ronald L., Albert Einstein College of Medicine. 1300 Morris

Park Avenue. Bronx, NY 10461

Naka, Ken-ichi, 2-9-2 Tatumi Higashi, Okazaki, 444. Japan
Nakajima, Yasuko, University of Illinois. College of Medicine.

Anatomy and Cell Biology Department, MAT 512, Chicago, IL 60612

Narahashi, Toshio, Northwestern University Medical School.

Department of Pharmacology. 303 East Chicago Avenue. Chicago. IL

60611
Nasi, Enrico, Boston University School of Medical. Department of

Physiology, R-406. 80 East Concord Street, Boston, MA 02118
Neill, Christopher, Marine Biological Laboratory, 7 MBL Street,

Woods Hole, MA 02543
Nelson, Margaret C., Cornell University, Section of Neurobiology and

Behavior, Ithaca. NY 14850

Nicholls, John G., SISSA, Via Beirut 2, 1-34014 Trieste, Italy
Nickerson, Peter A., SUNY at Buffalo, Department of Pathology.

Buffalo. NY 14214
Nicosia, Santo V., University of South Florida, College of Medicine,

Box 1 1, Department of Pathology, Tampa. FL 33612
Noe, Bryan D., Emory University School of Medicine. Department of

Anatomy and Cell Biology. Atlanta, GA 30322
Norton, Catherine N., Marine Biological Laboratory, 7 MBL Street,

Woods Hole. MA 02543
Nusbaum, Michael P., University of Pennsylvania School of Medicine,

Department of Neuroscience, 215 Stemmler Hall. Philadelphia. PA

19104-6074

O'Herron, Jonathan, Lazard Freres and Company, 30 Rockefeller

Plaza, 59th Floor, New York, NY 10020-1900
Obaid, Ana Lia, University of Pennsylvania School of Medicine,

Neuroscience Department, 234 Stemmler Hall, Philadelphia, PA

19104-6074
Ohki, Shinpei, SUNY at Buffalo, Department of Biophysical Sciences,

224 Cary Hall, Buffalo. NY 14214
Oldenbourg. Rudolf, Marine Biological Laboratory, 7 MBL Street.

Woods Hole, MA 02543
Olds, James L., George Mason University, Krasnow Institute for

Advanced Studies, Mail Stop 2A1, Fairfax. VA 22030-4444
Olins, Ada L., Foundation for Blood. 69 U.S. Route One, P.O. Box

190. Scarborough, ME 04070-0190
Olins, Donald E., Foundation for Blood, 69 U.S. Route One. P.O. Box

190. Scarborough. ME 04070-0190
Oschman, James L., 827 Central Avenue. Dover. NH 03820

Palazzo, Robert E., University of Kansas. Department of Physiology

and Cell Biology. Lawrence, KS 66045
Palmer, John D., University of Massachusetts, Department of Zoology,

221 Morrill Science Center, Amherst, MA 01003
Pant, Harish C., National Institutes of Health, NINCDS, Laboratory of

Neurochemistry, Building 36, Room 4D20, Bethesda, MD 20892
Pappas, George D., University of Illinois, Psychiatric Institute, 1601 W.

Taylor Street, MC 912, Chicago, IL 60612
Pardee, Arthur B., Dana-Farber Cancer Institute, D810. 44 Binney

Street. Boston. MA 02115
Pardy. Rosevelt L., University of Nebraska. School of Life Sciences,

Lincoln, NE 68588
Parmentier. James L., Massachusetts General Hospital,

Partners/Fenway/Shattuck Center for Aids Research, 149 13^ Street,

Room 5219, Charlestown, MA 02129
Pederson, Thoru, University of Massachusetts Medical Center.

Worcester Foundation Campus. 222 Maple Avenue, Shrewsbury, MA

01545

Perkins, Courtland D., 400 Hilltop Terrace, Alexandria. VA 22301
Person, Philip, 137-87 75th Road, Flushing, NY 1 1367
Peterson, Bruce J., Marine Biological Laboratory, 7 MBL Street,

Woods Hole, MA 02543

R80 Annual Report

Pethig, Ronald. University College of North Wales, School of

Electronic Engineering, Bangor, Gwynedd. LL 57 IUT, United

Kingdom
Pfohl, Ronald J., Miami University, Department of Zoology, Oxford.

OH 45056
Pierce, Sidney K., Jr., University of South Florida. Department of

Biology. SCA 110, 4202 East Fowler Avenue, Tampa, FL 33620
Pleasure, David E., Children's Hospital, Neurology Research, 5th

Floor, Ambramson Building, Philadelphia. PA 19104
Poindexter, Jeanne S., Barnard College. Columbia University. 3009

Broadway, New York. NY 10027-6598
Pollard, Harvey B., U.S.U.H.S., 4301 Jones Bridge Road, Bethesda,

MD 20814
Pollard. Thomas D., Salk Institute for Biological Studies, 10010 N.

Torrey Pines Road, La Jolla, CA 92037

Porter, Beverly H., 5542 Windysun Court. Columbia. MD 21045
Porter, Mary E., University of Minnesota, Department of Cell Biology

and Neuroanatomy, 4-135 Jackson Hall, 321 Church Street SE,

Minneapolis, MN 55455
Potter, David D., Harvard Medical School. Department of

Neurobiology. 25 Shattuck Street. Boston, MA 021 15
Potts, William T., University of Lancaster. Department of Biology,

Lancaster, England
Powers, Maureen K., University of California. Department of

Molecular & Cellular Biology, Life Sciences Addition, Berkeley, CA

94720

Prendergast, Robert A., 29 Pondlet Place, Falmouth. MA 02540
Prior, David J., Northern Arizona University, Arts and Sciences Dean's

Office, Box 5621, Flagstaff, AZ 8601 1
Prusch, Robert D., Gonzaga University, Department of Life Sciences,

Spokane, WA 99258
Purves, Dale, Duke University Medical Center, Department of

Neurobiology, Box 3209, 101-1 Bryan Research Building, Durham,

NC 27710

Quigley, James P., The Scripps Research Institute. Department of
Vascular Biology, 10550 N. Torrey Pines Road VB-1, La Jolla, CA
92037

Rabb, Irving W., 1010 Memorial Drive, #20A. Cambridge, MA 02138
Rabin. Harvey, 1 102 Ralston Road, Rockville, MD 20852
Rabinowitz, Michael B., Marine Biological Laboratory, 7 MBL Street,

Woods Hole. MA 02543
Rafferty, Nancy S., Marine Biological Laboratory, 7 MBL Street,

Woods Hole, MA 02543
Rakowski, Robert F., Finch University of Health Sciences, The

Chicago Medical School, Department of Physiology and Biophysics,

3333 Greenbay Road, N. Chicago, IL 60064
Ramon, Fidel, Universidad Nacional Autonoma de Mexico, Division

EStreet Posgrado E Invest., Facultad de Medicina. 04510, D.F.,

Mexico
Rastetter, Edward B., Marine Biological Laboratory, The Ecosystems

Center. Woods Hole. MA 02543
Rebhun, Lionel I., University of Virginia, Department of Biology,

Gilmer Hall 45, Charlottesville. VA 22901
Reddan, John R., Oakland University, Department of Biological

Sciences. Rochester, MI 48309-4401
Reese, Thomas S., National Institutes of Health, NINDS. Department of

Neurobiology, Building 36, Room 2A-21. 36 Convent Drive,

Bethesda, MD 20892

Reinisch, Carol L., Marine Biological Laboratory. 7 MBL Street,

Woods Hole. MA 02543
Rickles, Frederick R., 3910 Highwood Court. N.W., Washington, DC

20007
Rieder, Conly L., Wadsworth Center, Division of Molecular Medicine,

P.O. Box 509, Albany. NY 12201-0509
Riley. Monica, Marine Biological Laboratory. 7 MBL Street, Woods

Hole. MA 02543
Ripps, Harris, University of Illinois at Chicago, Department of

Ophthalmology/Visual Sciences, 1855 West Taylor Street, Chicago,

IL 60612
Ritchie, J. Murdoch, Yale University School of Medicine, Department

of Pharmacology, 333 Cedar Street, New Haven. CT 06510
Rome, Lawrence C., University of Pennsylvania, Department of

Biology, Leidy Labs, Philadelphia. PA 19104
Rosenbluth, Jack, New York University School of Medical.

Department of Physiology and Biophysics, RR 714, 400 East 34th

Street, New York, NY 10016
Rosenbluth, Raja, Simon Fraser University, Institute of Molecular

Biology and Biochemistry. Burnaby, BC V5A 1S6, Canada
Rosenfield, Allan, Columbia University School of Public Health, 600

West 168th Street. New York. NY 10032-3702

Rosenkranz, Herbert S., 130 Desoto Street, Pittsburgh. PA 15213-2535
Ross, William N., New York Medical College, Department of

Physiology, Valhalla, NY 10595
Rottenfusser, Rudi, Marine Biological Laboratory. 7 MBL Street,

Woods Hole, MA 02543
Rowland, Lewis P.. Neurological Institute, 710 West 168th Street, New

York, NY 10032
Ruderman, Joan V., Harvard Medical School, Department of Cell

Biology. C2-428. 240 Longwood Avenue. Boston, MA 021 15
Rummel, John D., NASA Headquarters. Office of Space Science.

Washington. DC 20546
Rushforth, Norman B., Case Western Reserve University, Department

of Biology, Cleveland, OH 44106
Russell-Hunter, William D., 71 1 Howard Street, Easton, MD 21601-

3934

Saffo, Mary Beth, Harvard University, MCZ Labs 408, 26 Oxford

Street, Cambridge, MA 02138
Salama, Guy, University of Pittsburgh, Department of Physiology.

Pittsburgh. PA 15261
Salmon, Edward D., University of North Carolina, Department of

Biology, CB 3280, Chapel Hill, NC 27514
Salyers, Abigail, University of Illinois, Department of Microbiology.

B-103, 601 South Goodwin Avenue, Urbana, IL 61801
Salzberg, Brian M., University of Pennsylvania School of Medicine,

Department of Neuroscience, 215 Stemmler Hall, Philadelphia, PA

19104-6074
Sanger, Jean M., University of Pennsylvania School of Medicine.

Department of Anatomy, 36th and Hamilton Walk, Philadelphia, PA

19104
Sanger, Joseph W., University of Pennsylvania Medical Center,

Department of Cell and Developmental Biology. 36th and Hamilton

Walk. Philadelphia, PA 19104-6058
Saunders, John W., Jr., 118 Metoxit Road. P.O. Box 3381. Waquoit.

MA 02536
Schachman, Howard K., University of California, Molecular and Cell

Biology Department, 229 Stanley Hall, #3206, Berkeley, CA 94720-

3206
Schatten, Gerald P., Oregon Health Sciences University, Oregon

Regional Primate Research Center, 505 N.W. 185th Avenue,

Beaverton, OR 97006

Members of the Corporation R81

Schmeer, Arlene C., Mercenene Cancer Research Institute. 790

Prospect Street. New Haven. CT 06511
Schuel, Herbert. SUNY at Buffalo. Department of Anatomy/Cell

Biology. Buffalo. NY 14214
Schwartz, Lawrence. University of Massachusetts. Department of

Biology. Morrill Science Center. Amherst. MA 01003
Schweitzer. A. Nicola. Brigham and Women's Hospital. Immunology

Division. Department of Pathology. 221 Longwood Avenue, LMRC

521, Boston, MA 02115
Segal, Sheldon J., The Population Council. One Dag Hammarskjold

Plaza, New York, NY 10036
Senft, Stephen Lament, Yale University,

Neuroengineering/Neuroscience Center, P.O. Box 208205, New

Haven. CT 06520-8205
Shanklin. Douglas R., University of Tennessee. Department of

Pathology, Room 576, 800 Madison Avenue, Memphis. TN 381 17
sli.i-.li.ii . Nadav, The Interuniversity Institute of Eilat. P.O. Box 469,

Eilat 88103. Israel
Shashoua, Victor E., Harvard Medical School. Ralph Lowell Labs.

McLean Hospital, 115 Mill Street, Belmont, MA 02178
Shaver, Gaius R., Marine Biological Laboratory, The Ecosystems

Center. Woods Hole. MA 02543
Shaver. John R., Michigan State University. Department of Zoology.

East Lansing. MI 48824
Sheetz, Michael P., Duke University Medical Center. Department of

Cell Biology. Bx 3709. 388 Nanaline Duke Building. Durham. NC

27710
Shepro, David, Boston University. CAS Biology, 5 Cummington Street,

Boston. MA 022 1 5
Shimomura. Osamu, Marine Biological Laboratory, 7 MBL Street,

Woods Hole. MA 02543

Shipley. Alan M., P.O. Box 943. Forestdale. MA 02644
Silver, Robert B., Marine Biological Laboratory. 7 MBL Street. Woods

Hole. MA 02543
Siwicki, Kathleen K., Swarthmore College, Biology Department. 500

College Avenue. Swarthmore. PA 19081-1397
Skinner, Dorothy M., 24 Gray Lane, Falmouth. MA 02540
Sloboda, Roger D., Dartmouth College, Department of Biological

Science, 6044 Gilman, Hanover. NH 03755-1893
Sluder, Greenfield, University of Massachusetts Medical School. Room

324. 377 Plantation Street. Worcester. MA 01605
Smith, Peter J. S., Marine Biological Laboratory, 7 MBL Street.

Woods Hole. MA 02543
Smith, Stephen J., Stanford University School of Medicine, Department

of Molecular and Cellular Physiology. Beckman Center. Stanford, CA

94305
Smolowitz. Roxanna S., Marine Biological Laboratory, 7 MBL Street,

Woods Hole, MA 02543
Sogin, Mitchell L., Marine Biological Laboratory, 7 MBL Street.

Woods Hole, MA 02543
Sorenson, Martha M., Cidade Universitaria-UFRJ, Department

Bioquimica Medica-ICB. 21941-590 Rio de Janerio. Brazil
Speck, William T., Marine Biological Laboratory. 7 MBL Street.

Woods Hole. MA 02543
Spector, Abraham, Columbia University. Department of

Ophthalmology, 630 West 168th Street. New York. NY 10032
Speksnijder, Johanna E., DeMeent 12. 3984JJ Odijk. The Netherlands
Spray, David C., Albert Einstein College of Medicine. Department of

Neuroscience. 1300 Moms Park Avenue. Bronx. NY 10461
Spring, Kenneth R., National Institutes of Health, 10 Center Drive.

MSC 1598, Building 10. Room 6N260, Bethesda, MD 20892-1603
Steele, John H., Woods Hole Oceanographic Institution, Woods Hole,

MA 02543

Steinacker, Antoinette, University of Puerto Rico. Institute of

Neurobiology. 201 Boulevard Del Valle. San Juan. PR 00901
Steinberg, Malcolm. Princeton University, Department of Molecular

Biology, M-18 Moffett Laboratory. Princeton. NJ 08544-1014
Stemmer, Andreas C., Institut filr Robotik. ETH-Center. 8092 Zurich,

Switzerland
Stenflo, Julian. University of Lund, Department of Clinical Chemistry.

Malmo General Hospital. S-205 02 Malmo, Sweden
Stetten, Jane Lazarow, 4701 Willard Avenue, #1413. Chevy Chase.

MD 20815-4627
Steudler, Paul A., Marine Biological Laboratory, The Ecosystems

Center, Woods Hole, MA 02543
Stokes, Darrell R., Emory University, Department of Biology, 1510

Clifton Road NE. Atlanta. GA 30322-1 100
Stommel, Elijah W., Dartmouth Hitchcock Medical Center. Neurology

Department, 1 Medical Drive, Lebanon, NH 03756
Stracher. Alfred, SUNY Health Science Center, Department of

Biochemistry. 450 Clarkson Avenue. Brooklyn. NY 11203
Strumwasser, Felix, P.O. Box 923. East Falmouth. MA 02536-2278
Stuart, Ann E., 1818 North Lakeshore Drive. Chapel Hill, NC 27514
Sugimori, Mutsuyuki. New York University Medical Center.

Department of Physiology and Neuroscience. Room 442, 550 First

Avenue. New York. NY 10016
Summers. William C., Western Washington University. Huxley College

of Environmental Studies, Bellingham, WA 982259181
Suprenant, Kathy A., University of Kansas, Department of Physiology

and Cell Biology. 4010 Haworth Hall, Lawrence. KS 66045
Sydlik, Mary Anne, Hope College. Peale Science Center. 35 East 12th

St./PO Box 9000. Holland, MI 49422
Szent-Gyorgyi, Andrew G., Brandeis University. Molecular and Cell

Biology, 415 South Street. Waltham, MA 02454-91 10

Tamm, Sidney L., Boston University. CAS Biology. 5 Cummington

Street. Boston, MA 02215
Tanzer, Marvin L., University of Connecticut School of Dental

Medicine. Department of Biostructure and Function. Farmington. CT

06030-3705
Tasaki, Ichiji, National Institutes of Health. NIMH. Laboratory of

Neurobiology, Building 36, Room 2B-16. Bethesda. MD 20892
Taylor, D. Lansing, Cellomics. Inc., 635 William Pitt Way. Pittsburgh.

PA 15238
Taylor, Edwin W'., University of Chicago, Department of Molecular

Genetics, 920 E. 58th Street, Chicago. IL 60637
Teal, John M., 567 New Bedford Lane. Rochester, MA 02770
Telfer, William H., University of Pennsylvania. Department of Biology.

Philadelphia. PA 19104
Telzer, Bruce, Pomona College, Department of Biology, Thille

Building. 175 West 6th Street, Claremont, CA 91711
Terasaki, Mark, University of Connecticut Health Center, Department

of Physiology, 263 Farmington Avenue, Farmington. CT 06032
Townsel, James G., Meharry Medical College. Department of Anatomy

and Physiology. 1005 DB Todd Boulevard. Nashville. TN 37208
Travis. David M., 19 High Street. Woods Hole. MA 02543-1221
Treistman. Steven N., University of Massachusetts Medical Center.

Department of Pharmacology, 55 Lake Avenue North. Worcester. MA

01655

Trigg, D. Thomas, One Federal Street. 9th Floor, Boston, MA 022 1 1
Troll, Walter, NYU Medical Center. Department of Environmental

Medicine. 550 First Avenue. New York. NY 10016
Troxler, Robert F., Boston University School of Medicine. Department

of Biochemistry. 80 East Concord Street. Boston. MA 021 18
Tucker, Edward B., Baruch College. CUNY. Department of Natural

Sciences, 17 Lexington Avenue, New York, NY 10010

R82 Annual Report

Turner, Ruth D., Harvard University, Museum of Comparative

Zoology, Mollusk Department, Cambridge. MA 02138 (deceased

2000)
Tweedell, Kenyon S., University of Notre Dame, Department of

Biological Sciences, Notre Dame, IN 46556-0369
Tykocinski. Mark L., Case Western Reserve University, Institute of

Pathology, 2085 Adelbert Road, Cleveland, OH 44106
Tytell, Michael, Wake Forest University, Bowman Gray School of

Medicine. Department of Anatomy and Neurobiology. Winston-

Salem, NC 27157

Ueno, Hiroshi, Kyoto Universily, AGR Chemistry, Faculty of
Agriculture. Sakyo, Kyoto 606-8502, Japan

Valiela, Ivan, Boston University Marine Program, Marine Biological

Laboratory, Woods Hole, MA 02543
Vallee, Richard, University of Massachusetts Medical Center,

Worcester Foundation Campus. 222 Maple Avenue. Shrewsbury. MA

01545

Valois, John J., 420 Woods Hole Road, Woods Hole, MA 02543
Van Dover, Cindy Lee, The College of William and Mary. Biology

Department. 328 Millington Hall, Williamsburg, VA 23187
Van Holde, Kensal E., Oregon State University, Biochemistry and

Biophysics Department. Corvallis, OR 97331-7503
Vogl, Thomas P., Environmental Research Institute of Michigan, 1101

Wilson Boulevard, Arlington. VA 22209

Wainwright, Norman R., Marine Biological Laboratory. 7 MBL Street.

Woods Hole, MA 02543
Waksman, Byron H., New York University Medical Center.

Department of Pathology. 550 First Avenue, New York. NY 10016
Wall, Belly, 9 George Street. Woods Hole, MA 02543
Wangh, Lawrence J., Brandeis University, Department of Biology, 415

South Street. Waltham. MA 02254
Warner, Robert C., 1609 Temple Hills Drive, Laguna Beach, CA

9265 1
Warren, Leonard, Wistar Institute, 36th and Spruce Streets,

Philadelphia. PA 19104
Waterbury, John B., Woods Hole Oceanographic Institution,

Department of Biology, Woods Hole. MA 02543
Waxman, Stephen G., Yale Medical School. Neurology Department.

333 Cedar Street. P.O. Box 208018, New Haven. CT 06510
Weber, Annemarie, University of Pennsylvania School of Medicine,

Department of Biochemistry and Biophysics. Philadelphia, PA 19066
Weeks, Janis C., University of Oregon. Institute of Neuroscience,

Eugene, OR 97403-1254
Weidner, Earl, Louisiana State University, Department of Biological

Sciences, 502 Life Sciences Building, Baton Rouge, LA 70803
Weiss, Alice Sara, 105 University Boulevard West. Silver Spring, MD

20901

Weiss, Dieter G., University of Rostock, Institute of Zoology, D- 18051

Rostock, Germany
Weiss, Leon P., University of Pennsylvania School of Veterinary

Medicine, Department of Animal Biology, Philadelphia. PA 19104
Weiss, Marisa C., Paoli Memorial Hospital, Department of Radiation

Oncology. 255 W. Lancaster Avenue, Paoli, PA 19301
Weissmann, Gerald, New York University Medical Center. Department

of Medicine/Division Rheumatology, 550 First Avenue, New York,

NY 10016
Westerfield, Monte, University of Oregon, Institute of Neuroscience,

Eugene. OR 97403
Whittaker, J. Richard, University of New Brunswick, Department of

Biology, BS 451 1, Fredericton, NB E3B 6E1, Canada
Wiesel, Torsten N., Rockefeller University, 1230 York Avenue, New

York, NY 10021
Wilkens, Lon A., University of Missouri. Department of Biology. 8001

Natural Bridge Road. St. Louis. MO 63121-4499
Wilson, Darcy B., Torrey Pines Institute, 3550 General Atomics Court,

Building 2. Room 138. San Diego. CA 92121
Wilson, T. Hastings, Harvard Medical School. Department of

Physiology, 25 Shattuck Street, Boston, MA 02 1 1 5
Witkovsky, Paul, NYU Medical Center. Department of Ophthalmology.

550 First Avenue, New York, NY 10016
Wittenberg, Beatrice, Albert Einstein College of Medicine. Department

nt Physiology and Biophysics. Bronx, NY 10461
Wittenberg, Jonathan B., Albert Einstein College of Medicine,

Department of Physiology and Biophysics, Bronx, NY 10461
Wonderlin, William F., West Virginia University, Pharmacology and

Toxicology Department, Morgantown, WV 26506
Worden, Mary Kate, University of Virginia, Department of

Neuroscience, McKim Hall Box 230. Charlottesville, VA 22908
Worgul, Basil V., Columbia University, Department of Ophthalmology,

630 West 168 Street. New York. NY 10032
Wu, Chau Hsiung, Northwestern University Medical School,

Department of Pharmacology (S215). 303 East Chicago Avenue,

Chicago. IL 60611-3008
Wyttenbach, Charles R., University of Kansas. Biological Sciences

Department. 2045 Haworth Hall, Lawrence, KS 66045-2106

Zakon, Harold H., University of Texas, Section of Neurobiology,

School of Life Science, Austin. TX 787 1 2
Zigman, Seymour, Marine Park Condominiums, 1 74 Queen Street. Unit

10-F, Falmouth, MA 02540
Zigmond, Michael J., University of Pittsburgh, S-526 Biomedical

Science Tower, 3500 Terrace Street. Pittsburgh, PA 15213
Ziinmerberg, Joshua J., National Institutes of Health, LCMB. NICHD,

Building 10, Room KID 14, 10 Center Drive. Bethesda, MD 20892
Zottoli, Steven J., Williams College. Department of Biology,

Williamstown. MA 01267
Zucker, Robert S., University of California, Neurobiology Division,

Molecular and Cellular Biology Department, Berkeley, CA 94720

Members of the Corporation R83

MBL Associates

Executive Board

Ruth Ann taster. President

Jack Pearce. Vice President

Kitty Brown. Treasurer

Molly N. Cornell, Secretary

Duncan Aspinwall. Membership Chair

Tammy Smith Amon

Barbara At wood

Julie Child

Seymour Cohen

Elizabeth Farnham

Michael Fenlon

Pat Ferguson

Sallie A. Giffen

Alice Knowles

Rebecca Lash

Cornelia Hanna McMurtrie

Joan Pearlman

Virginia Reynolds

Volker Ulbnch

Associates Liaison/Gift Shop
Coordinator

Kendall B. Bohr

Patron

Judge and Mrs. John S. Langford

Sustaining Associate

Mr. and Mrs. G. Nathan Calkins, Jr.
Mrs. Janet F. Gillette

Dr. and Mrs. Edward F. MacNichol, Jr.

Supporting Associate

Mr. and Mrs. William O. Burwell

Mr. and Mrs. Thomas Claflin

Mrs. George H. A. Clowes

Dr. and Mrs. James D. Ebert

Mr. and Mrs. David Fausch

Mr. Mike Fenlon and Ms. Linda Sallop

Dr. and Mrs. James J. Ferguson. Jr.

Mrs. Janet F. Gillette

Mrs. Mary L. Goldman

Mr. and Mrs. Lon Hocker

Mr. and Mrs. Arthur King

Dr. and Mrs Leonard Laster

Drs. Luigi and Elaine Mastroianni

Mr. and Mrs. Walter J. Salmon

Mrs. Anne W. Sawyer

Dr. John Tochko and Mrs. Christina Myles-

Tochko

Mr. and Mrs. John J. Valois
Mr. and Mrs. Leslie J. Wilson

Familv Membership

Dr. and Mrs. Edward A. Adelberg

Mr. and Mrs. David C. Ahearn

Dr. and Mrs. Dean C. Allard. Jr.

Drs. James and Helene Anderson

Dr. and Mrs. Samuel C. Armstrong

Mr. and Mrs. Duncan P. Aspinwall

Mr. and Mrs. Donald R. Aukamp

Mr. and Mrs. John M. Baitsell

Mr. and Mrs. David Bakalar

Dr. and Mrs. Robert B. Barlow, Jr.

Mr. and Mrs. John E. Barnes

Dr. and Mrs. Robert M. Berne

Drs. Harriet and Alan Bemheimer

Mr. and Mrs. Robert O. Bigelow

Dr. and Mrs. Edward G. Boettiger

Mr. and Mrs. Kendall B. Bohr

Dr. and Mrs. Thomas A. Borgese

Dr. and Mrs. Francis P. Bowles

Dr. and Mrs. John B. Buck

Dr. and Mrs. John E. Bums

Mr. and Mrs. D. Bret Carlson

Dr. and Mrs. Richard L. Chappell

Dr. and Mrs. Frank M. Child

Dr. and Mrs. Arnold M. Clark

Mr. and Mrs. James M. Cleary

Dr. and Mrs. Laurence P. Cloud

Drs. Harry Conner and Carol Scott-Conner

Mrs. Neal Cornell

Mr. and Mrs. Norman C. Cross

Dr. and Mrs. John M. Cummings

Mr. and Mrs. Joel P. Davis

Mr. and Mrs. F. Gerald Douglass

Dr. and Mrs. John E. Dowling

Dr. and Mrs. Arthur Brooks DuBois

Dr. and Mrs. Michael J. Fishbein

Mr. and Mrs. Harold Frank

Mr. and Mrs. Howard G. Freeman

Dr. and Mrs. Robert A. Frosch

Dr. and Mrs. John J. Funkhouser

Dr. and Mrs. Mordecai L. Gabriel

Dr. and Mrs. Sydney Gellis

Dr. and Mrs. James L. German, III

Dr. and Mrs. Harold S. Ginsberg

Dr. and Mrs. Murray Glusman

Drs. Alfred and Joan Goldberg

Mr. and Mrs. Charles Goodwin, III

Mr. and Mrs. Anthony D. Green

Dr. and Mrs. Thomas C. Gregg

Dr. and Mrs. Newton H. Gresser

Mr. and Mrs. Peter A. Hall

Dr. and Mrs. Harlyn O. Halvorson

Dr. and Mrs. Richard Bennet Harvey

Dr. and Mrs. J. Woodland Hastings

Dr. Robert R. Haubrich

Mr. and Mrs. Gary G. Hayward

Dr. and Mrs. Howard H. Hiatt

Mr. and Mrs. David Hibbitt

Dr. and Mrs. John E. Hobble

Mr. and Mrs. Gerald J. Holtz

Drs. Francis Hoskin and Elizabeth Farnham

Dr. and Mrs. Robert J. Huettner

Dr. and Mrs. Shinya Inoue

Dr. and Mrs. Kurt J. Isselbacher

Mrs. Mary D. Janney

Dr. and Mrs. Benjamin Kaminer

Mr. and Mrs. Paul W. Knaplund

Mr. and Mrs. A. Sraney Knowles, Jr.

Mr. and Mrs. Walter E. Knox

Sir and Lady Hans Kornberg

Dr. and Mrs. S. Andrew Kulin

Mr. Ezra and Dr. Aimlee Laderman

Mr. and Mrs. Trevor Lambert

Dr. and Mrs. George M. Langford

Dr. and Mrs. Hans Laufer

Dr. and Mrs. Berton J. Leach

Dr. and Mrs. John J. Lee

Mr. and Mrs. Stephen R. Levy

Mr. and Mrs. Robert Livingstone, Jr.

Dr. and Mrs. Laszlo Lorand

Mr. and Mrs. Francis C. Lowell, Jr.

Dr. Isabelle and Mr. Bernard Manuel

Mr. and Mrs. Joseph C. Martyna

Mr. and Mrs. Frank J. Mather, UJ

Dr. and Mrs. Robert T. McCluskey

Dr. and Mrs. William M. McDermott

Dr. and Mrs. Jerry M. Melillo

Mr. and Mrs. Wesley J. Merrill

Mr. and Mrs. Richard Meyers

Mr. and Mrs. Charles A. Mitchell

Dr. and Mrs. Merle Mizell

Dr. and Mrs. Charles H. Montgomery

Mr. and Mrs. Stephen A. Moore

Dr. and Mrs. John E. Naugle

Dr. Pamela Nelson and Mr. Christopher

Olmsted

Mr. and Mrs. Frank L. Nickerson
Dr. and Mrs. Clifford T. O'Connell
Mr. and Mrs. James J. O'Connor
Mr. and Mrs. David R. Palmer
Mr. and Mrs. Robert Parkinson
Mr. and Mrs. Richard M. Paulson, Jr.
Dr. and Mrs. John B. Pearce
Mr. and Mrs. William J. Pechilis
Mrs. Nancy Pendleton
Mr. and Mrs. John B. Peri
Dr. and Mrs. Courtland D. Perkins
Dr. and Mrs. Philip Person
Mr. and Mrs. Frederick S. Peters
Mr. and Mrs. E. Joel Peterson
Mr. and Mrs. Harold Pilskaln
Mr. and Mrs. George H. Plough
Dr. and Mrs. Aubrey Pothier, Jr.
Mr. and Mrs. Allan Putnam
Dr. and Mrs. Lionel I. Rebhun
Dr. and Mrs. George T. Reynolds
Dr. and Mrs. Harris Ripps
Dr. Paul B. Rizzoli
Ms. Jean Roberts
Drs. Priscilla and John Roslansky
Mr. and Mrs. John D. Ross
Dr. and Mrs. John W. Saunders, Jr.
Dr. and Mrs. R. Walter Schlesinger

R84 Annual Report

Mr. and Mrs. Harold H. Sears

Dr. and Mrs. Sheldon J. Segal

Mr. and Mrs. Daniel Shearer

Dr. and Mrs. David Shepro

Mr. and Mrs. Bertram R. Silver

Mr. and Mrs. Jonathan O. Simonds

Drs. Frederick and Marguerite Smith

Dr. and Mrs. Alan B. Stembach

Dr. and Mrs. William K. Stephenson

Mr. and Mrs. E. Kent Swift, Jr.

Mr. and Mrs. Gerard L. Swope. Ill

Mr. Norman N. Tolkan

Dr. and Mrs. Walter Troll

Prof, and Mrs. Michael Tytell

Mr. and Mrs. Volker Ulbrich

Dr. and Mrs. Gerald Weissmann

Dr. and Mrs. Paul S. Wheeler

Dr. and Mrs. Martin Keister White

Mr. and Mrs. Geoffrey G. Whitney, Jr.

Mr. and Mrs. Lynn H. Wilke

Dr. and Mrs. T. Hastings Wilson

Mrs. Sumner Zacks

Dr. Linda and Mr. Erik Zettler

Dr. and Mrs. Seymour Zigman

Individual Membership

Drs. Fred and Peggy Alsup
Mrs. Tammy Smith Amon
Mr. Dean N. Arden
Mrs. Ellen Prosser Armstrong
Mrs. Kimball C. Atwood, III
Mr. Everett E. Bagley
Dr. Millicent Bell
Mr. C. John Berg
Dr. Thomas P. Bleck
Ms. Avis Blomberg
Mr. Theodore A. Bonn
Mr. James V. Bracchitta
Mrs. Jennie P. Brown
Mrs. M. Kathryn S. Brown
Dr. Robert H. Broyles
Mrs. Barbara Gates Burwell
Dr. Graciela C. Candelas
Mr. Frank C. Carotenuto
Dr. Robert H. Carrier
Mrs. Patricia A. Case
Ms. Mia D. Champion
Dr. Sallie Chisholm
Mrs. Octavia C. Clement
Mr. Allen W. Clowes
Dr. Jewel Plummer Cobb
Mrs. Margaret H. Coburn
Dr. Seymour S. Cohen
Dr. Alan Robert Cole
Ms. Anne S. Concannon
Prof. D. Eugene Copeland
Dr. Vincent Cowling
Mrs. Marilyn E. Crandall
Ms. Dorothy Crossley
Ms. Helen M. Crossley
Mrs. Villa B, Crowell
Mrs. Alexander T. Daignault
Dr. Morton Davidson

Mrs. Elizabeth M. Davis

Ms. Maureen Davis

Ms. Carol Reimann DeYoung

Ms. Shirley Dierolf

Mrs. Juliette G. Dively

Mr. David L. Donovan

Ms. Suzanne Droban

Mr. Roy A. Duffus

Ms. Maureen J. Dugan

Mrs. Charles Eastman

Dr. Frank Egloff

Ms. Judy Ernst

Dr. Stephen L. Estabrooks

Mrs. Eleanor B. Faithorn

Mrs. Ruth Alice Fitz

Ms. Sylvia M. Flanagan

Mr. John W. Folino. Jr.

Mrs. Kathryn W. Foster

Dr. Krystyna Frenkel

Mr. Paul J. Freyheit

Mrs. Ruth E. Fye

Mrs. Lois E. Galvin

Miss Eleanor Garneld

Mrs. Ruth H. Garland

Mr. John Garnett

Ms. Sallie A. Giffen

Mrs. James R. Glazebrook

Mr. Michael P. Goldring

Mrs. Phyllis Goldstein
Mrs. DeWitt S. Goodman
Ms. Muriel Gould
Mrs. Rose Grant
Ms. Janet M. Gregg
Mrs. Jeanne B. Griffith
Mrs. Barbara Grossman

Mrs. Valerie A. Hall

Ms. Mary Elizabeth Hamstrom

Dr. Carol W. Hannenberg

Ms. Elizabeth E. Hathaway

Mrs. Elizabeth Heald

Mrs. Jane G. Heald

Mrs. Betty G. Hubbell

Miss Elizabeth B. Jackson

Mr. Raymond L. Jewett

Mrs. Barbara W. Jones

Mrs. Joan T. Kanwisher

Mrs. Sally Karush

Ms. Patricia E. Keoughan

Dr. Peter N. Kivy

Dr. Annlee D. Laderman

Mrs. Janet W. Larcom

Ms. Rebecca Lash

Mr. William Lawrence

Dr. Marian E. LeFevre

Mr. Edwin M. Libbin

Mr. Lennart Lindberg

Mrs. Barbara C. Little

Mrs. Sarah J. Loessel

Mr. Richard C. Lovenng

Mrs. Margaret M. MacLeish

Ms. Anne Camille Maher

Mrs. Nancy R. Malkiel

Ms. Diane Maranchie

Dr. Miriam Jacob Mauzerall

Mrs. Mary Hartwell Mavor

Mr. Paul McGonigle

Dr. Susan Gerbi Mcllwam

Ms. Mary W. McKoan

Ms. Jane A. McLaughlm

Ms. Louise McManus

Ms. Cornelia Hanna McMurtne

Mrs. Anne L. Meigs-Brown

Mr. Ted Melillo

Dr. Martin Mendelson

Ms. Carmen Merryman

Mrs. Grace S. Metz

Mrs. Mary G. Miles

Mrs. Florence E. Mixer

Mr. Lawrence A. Monte

Mrs. Mary E. Montgomery

Ms. Cynthia Moor

Mr. James V. Moynihan

Mrs. Eleanor M. Nace

Mrs. Anne Nelson

Ms. C. Marie Newman

Dr. Eliot H. Nierman

Mr. Edmund F. Nolan

Ms. Catherine N. Norton

Dr. Renee Bennett O' Sullivan

Dr Arthur B. Pardee

Ms. Carolyn L. Parmenter

Ms. Joan Pearlman

Mr. Raymond W. Peterson

Ms. Elizabeth T. Price

Ms. Dianne Purves

Mrs. Julia S. Rankin

Dr. Margaret M. Rappaport

Mr. Fred J. Ravens, Jr.

Ms. Mary W. Rianhard

Dr. Mary Elizabeth Rice

Dr. Monica Riley

Mrs. Lola E. Robertson

Mrs. Arlene Rogers

Ms. Jean Rogers

Mrs. Wendy E. Rose

Mrs. Atholie K. Rosett

Dr. Virginia F. Ross

Dr. John D. Rummel

Mr. Raymond A. Sanbom

Mr. Claude Schoepf

Ms. Elaine Schott

Ms. Emily Schwartz-Clark

Mrs. Elsie M. Scott

Dr. Cecily C. Selby

Mrs. Deborah G. Senft

Ms. Dorothy Sgarzi

Mrs. Charlotte Shemin

Ms. Enid K. Sichel

Dr. Jeffrey D. Silberman

Mrs. Cynthia C. Smith

Mr. Sean W. Smith

Mrs. Louise M. Specht

Dr. Guy L. Steele, Sr.

Dr. Robert E. Steele

Mrs. Eleanor Steinbach

Mrs. Judith G. Stetson

Mrs. Jane Lazarow Stetten

Mrs. Elizabeth Stommel

Members of the Corporation R85

Mr. Albert H. Swain

Elisabeth Buck

Barbara Thomson

Mrs. Belle K. Taylor

Jewel Cobb

Alice Todd

Mr. James K. Taylor

Janet Daniels

Elaine Troll

Mr. Emil D. Tietje, Jr.

Carol DeYoung

Natalie Trousof

Mrs. Alice Todd

Fran Eastman

Barbara Van Holde

Mr. Arthur D. Traub

Alma Ebert

Doris Van Keuren

Mr. D. Thomas Trigg

Jane Foster

Susan Veeder

Ms. Natalie Trousof

Becky Glazebrook

Carol Ann Wagner

Ms. Ciona Ulbrich

Muriel Gould

Mabel Whelpley

Ms. Sylvia Vatuk

Barbara Grossman

Clare Wilber

Ms. Susan Veeder

Jean Halvorson

Betty Wilson

Mr. Lee D. Vincent

Hanna Hastings

Grace Witzell

Mr. Arthur D. Voorhis

Sally Karush

Bunnie Rose Zigman

Mrs. Eve Warren

Marcella Katz

Mr. John T. Weeks

Alice Knowles

Mr. Michael S. Weinstein

Donna Kornberg

MBL Summer Tour Guides

Ms. Lillian Wendorff

Evelyn Laufer

Ms. Mabel Y. Whelpley

Barbara Little

Gloria Borgese

Mrs. Barbara Whitehead

Winnie Mackey

Nancy Campana

Mrs. Ava Whittemore

Diane Maranchie

Frank Child

Mrs. Joan R. Wickersham

Miriam Mauzerall

Julie Child

Mrs. Clare M. Wilber

Mary Mavor

Nancy Fraser

Mrs. Helen Wilson

Jane McCormack

Sallie Giffen

Ms. Nancy Woitkoski

Louise McManus

Nichole Graham

Ms. Marion K. Wright

Mary Miles

Lois Harvey

Mrs. Dorothy M. York

Florence Mixer

Lincoln Kraeuter

Mrs. Margery P. Zinn

Lorraine Mizell

Barbara Little

Helen Murphy

Jennifer Machado

Bertha Person

Charles Mahoney

MBL Gift Shop Volunteers

Margareta Pothier

Francis X. Mahoney

Liz Price

Julie Rankin

Marion Adelberg

Julie Rankin

Howard Redpath

Barbara Atwood

Arlene Rogers

Beth Berne

Lil Saunders

Pucky Roslansky

Harriet Bernheimer

Cynthia Smith

Suzanne Thomas

Avis Blomberg

Peggy Smith

Mary Ulbnch

Gloria Borgese

Louise Specht

John Valois

Kitty Brown

Jane Stetten

Margery Zinn

Certificate of Organization
Articles of Amendment
Bylaws

Certificate of Organization

Articles of Amendment

(On File in the Office of the Secretary of the Commonwealth)

No. 3170

We, Alpheus Hyatt. President. William Stanford Stevens, Treasurer, and William T.
Sedgwick, Edward G. Gardiner. Susan Mims and Charles Sedgwick Minot being a
majority of the Trustees of the Marine Biological Laboratory in compliance with the
requirements of the fourth section of chapter one hundred and fifteen of the Public
Statutes do hereby certify that the following is a true copy of the agreement of
association to constitute said Corporation, with the names of the subscribers thereto:

We, whose names are hereto subscribed, do, by this agreement, associate ourselves
with the intention to constitute a Corporation according to the provisions of the one
hundred and fifteenth chapter of the Public Statutes of the Commonwealth of Mas-
sachusetts, and the Acts in amendment thereof and in addition thereto.

The name by which the Corporation shall be known is
THE MARINE BIOLOGICAL LABORATORY

The purpose for which the Corporation is constituted is to establish and maintain a
laboratory or station for scientific study and investigations, and a school for instruc-
tion in biology and natural history.

The place within which the Corporation is established or located is the city of Boston
within said Commonwealth.
The amount of its capital stock is none.

In Witness Whereof, we have hereunto set our hands, this twenty seventh day of
February in the year eighteen hundred and eighty-eight, Alpheus Hyatt, Samuel Mills,
William T. Sedgwick. Edward G. Gardiner, Charles Sedgwick Minot, William G.
Farlow, William Stanford Stevens. Anna D. Phillips, Susan Minis, B. H. Van Vleck.
That the first meeting of the subscribers to said agreement was held on the thirteenth
day of March in the year eighteen hundred and eighty-eight.

In Witness Whereof, we have hereunto signed our names, this thirteenth day ot March
in the year eighteen hundred and eighty-eight. Alpheus Hyatt. President. William
Stanford Stevens. Treasurer. Edward G. Gardiner. William T. Sedgwick. Susan Mims,
Charles Sedgwick Minot.
(Approved on March 20. 1888 as follows:

I hereby certify that it appears upon an examination of the within written certificate
and the records of the corporation duly submitted to my inspection, that the require-
ments of sections one. two and three of chapter one hundred and fifteen, and sections
eighteen, twenty and twenty-one of chapter one hundred and six, of the Public
Statutes, have been complied with and I hereby approve said certificate this twentieth
day of March A.D. eighteen hundred and eighty-eight.

Charles Endicott
Commissioner of Corporations)

(On File in the Office of the Secretary of the Commonwealth)

We, James D. Ebert, President, and David Shepro. Clerk of the Marine Biological
Labor.iior>. located at Woods Hole. Massachusetts 02543. do hereby certify that the
following amendment to the Articles of Organization of the Corporation was duly
adopted at a meeting held on August 15. 1975. as adjourned to August 29. 1975. by
vote of 444 members, being at least two-thirds of its members legally qualified to vote
in the meeting of the corporation:

Voted: That the Certificate of Organization of this corporation be and it hereby is

amended by the addition of the following provisions:

"No Officer. Trustee or Corporate Member of the corporation shall be personally
liable for the payment or satisfaction of any obligation or liabilities incurred as a result
of. or otherwise in connection with, any commitments, agreements, activities or
affairs of the corporation.

"Except as otherwise specifically provided by the Bylaws of the corporation, meet-
ings of the Corporate Members of the corporation may be held anywhere in the United
States.

"The Trustees of the corporation may make, amend or repeal the Bylaws of the
corporation in whole or in part, except with respect to any provisions thereof which
shall by law. this Certificate or the bylaws of the corporation, require action by the
Corporate Members."

The foregoing amendment will become effective when these articles of amendment
are filed in accordance with Chapter 180. Section 7 of the General Laws unless these
articles specify, in accordance with the vote adopting the amendment, a later effective
date not more than thirty days after such filing, in which event the amendment will
become effective on such later date.

In Witness whereof and Under the Penalties of Perjury, we have hereto signed our

names this 2nd day of September, in the year 1975. James D. Ebert. President; David

Shepro, Clerk.

(Approved on October 24, 1975, as follows:

I hereby approve the within articles of amendment and, the filing fee in the amount

of $10 having been paid, said articles are deemed to have been filed with me this 24th

day of October. lc)75

Paul Guzzi

Secretary of the Commonwealth )

Bylaws

(Revised August 7. 1992 and December 10. 1992)
ARTICLE I— THE CORPORATION

A. Name an,/ fiirpan: The name of the Corporation shall be The Marine Biolog-
ical Laboratory. The Corporation's purpose shall be to establish and maintain ,i

R86

Bylaws of the Corporation R87

laboratory or station tor scientific study and investigation and a school lor instruction
in biology and natural history.

B. Nondiscrimination, The Corporation shall not discriminate on the basis of age,
religion, color, race, national or ethnic origin, sex or sexual preference in its policies
on employment and administration or in its educational and other programs.

ARTICLE n— MEMBERSHIP

A. Memhcr\. The Members of the Corporation ("Members") shall consist of
persons elected by the Board of Trustees (the "Board"), upon such terms and
conditions and in accordance with such procedures, not inconsistent with law or these
Bylaws, as may be determined by the Board. At any regular or special meeting of the
Board, the Board may elect new Members. Members shall have no voting or other
rights with respect to the Corporation or its activities except as specified in these
Bylaws, and any Member may vote at any meeting of ihe Members in person only and
not by proxy. Members shall serve until their death or resignation unless earlier
removed with or without cause by the affirmative vote of two-thirds of the Trustees
then in office. Any Member who has retired from his or her home institution may,
upon written request to the Corporation, be designated a Life Member. Life Members
shall not have the right to vote and shall not be assessed for dues.

B. Meetings. The annual meeting of the Members shall be held on the Friday
following the first Tuesday in August of each year, at the Laboratory of the Corpo-
ration in Woods Hole, Massachusetts, at 9:30 a.m. The Chairperson of the Board shall
piusulL- at meetings of the Corporation. If no annual meeting is held in accordance
with the foregoing provision, a special meeting may be held in lieu thereof with the
same effect as the annual meeting, and in such case all references in these Bylaws,
except in this Article II. B., to the annual meeting of the Members shall be deemed to
refer to such special meeting. Members shall transact business as may properly come
before the meeting. Special meetings of the Members may be called by the Chair-
person or the Trustees, and shall be called by the Clerk, or in the case of the death,
absence, incapacity or refusal by the Clerk, by any other officer, upon written
application of Members representing at least ten percent of the smallest quorum of
Members required for a vote upon any matter at the annual meeting of the Members,
to be held at such time and place as may be designated.

C. Quorum. One hundred ( 100) Members shall constitute a quorum at any meeting.
Except as otherwise required by law or these Bylaws, the affirmative vote of a
majority of the Members voting in person at a meeting attended by a quorum shall
constitute action on behalf of the Members.

D. Notice of Meetings. Notice of any annual meeting or special meeting of
Members, if necessary, shall be given by the Clerk by mailing notice of the time and
place and purpose of such meeting at least 15 days before such meeting to each
Member at his or her address as shown on the records of the Corporation.

E. Waiver of Notice. Whenever notice of a meeting is required to be given a
Member, under any provision of the Articles or Organization or Bylaws of the
Corporation, a written waiver thereof, executed before or after the Meeting by such
Member, or his or her duly authorized attorney, shall be deemed equivalent to such
notice.

F. Adjournments. Any meeting of the Members may be adjourned to any other
time and place by the vote of a majority of those Members present at the meeting,
whether or not such Members conslitute a quorum, or by any officer entitled to preside
at or to act as Clerk of such meeting, if no Member is present or represented. It shall
not be necessary to notify any Members of any adjournment unless no Member is
present or represented at the meeting which is adjourned, in which case, notice of the
adjournment shall be given in accordance with Article II. D. Any business which could
have been transacted at any meeting of the Members as originally called may he
transacted at an adjournment thereof.

ARTICLE III— ASSOCIATES OF THE CORPORATION

Associates of the Corporation. The Associates of the Marine Biological Laboratory
shall be an unincorporated group of persons (including associations and corporations)
interested in the Laboratory and shall be organized and operated under the general
supervision and authority of the Trustees. The Associates of the Marine Biological
Laboratory shall have no voting rights.

ARTICLE IV— BOARD OF TRUSTEES

A. Powers. The Board of Trustees shall have the control and management of the
affairs of the Corporation. The Trustees shall elect a Chairperson of the Board who
shall serve until his or her successor is elected and qualified. They shall annually elect
a President of the Corporation. They shall annually elect a Vice Chairperson of the
Board who shall be Vice Chairperson of the meetings of the Corporation. They shall

annually elect a Treasurer. They shall annually elect a Clerk, who shall be a resident
of Massachusetts. They shall elect Trustees-at-Large as specified in this Article IV.
They shall appoint a Director of the Laboratory for a term not to exceed rive years,
provided the term shall not exceed one year if the candidate has attained the age of
65 years prior to the date of the appointment. They shall choose such other officers
and agents as they shall think best. They may fix the compensation of all officers and
agents of the Corporation and may remove them at any time. They may till vacancies
occurring in any of the offices. The Board shall have the power to choose an
Executive Committee from their own number as provided in Article V, and to
delegate to such Committee such of their own powers as they may deem expedient in
addition to those powers conferred by Article V. They shall, from time to time, elect
Members to the Corporation upon such terms and conditions as they shall have
determined, not inconsistent with law or these Bylaws.

B. Composition and Election.

i 1 ) The Board shall include 24 Trustees elected by the Board as provided below;

(a) At least six Trustees {"Corporate Trustees") shall be Members who are
scientists, and the other Trustees ("Trustees-at-Large") shall be individuals who need
not be Members or otherwise affiliated with the Corporation.

(b) The 24 elected Trustees shall be divided into four classes of six Trustees
each, with one class to be elected each year to serve for a term of four years, and with
each such class to include at least one Corporate Trustee. Such classes of Trustees
shall be designated by the year of expiration of their respective terms.

(2) The Board shall also include the Chief Executive Officer, Treasurer and the
Chairperson of the Science Council, who shall be ex officio voting members of the
Board.

(3) Although Members or Trustees may recommend individuals for nomination
as Trustees, nominations for Trustee elections shall be made by the Nominating
Committee in its sole discretion. The Board may also elect Trustees who have not
been nominated by the Nominating Committee.

C. Eligibility. A Corporate Trustee or a Trustee-at-Large who has been elected to
an initial four-year term or remaining portion thereof, of which he/she has served at
least two years, shall be eligible for re-election to a second four-year term, but shall
be ineligible for re-election to any subsequent term until one year has elapsed after
he/she has last served as a Trustee.

D. Removal. Any Trustee may be removed from office at any time with or without
cause, by vote of a majority of the Members entitled to vote in the election of
Trustees; or for cause, by vote of two-thirds of the Trustees then in office. A Trustee
may be removed for cause only if notice of such action shall have been given to all
of the Trustees or Members entitled to vote, as the case may be, prior to the meeting
at which such action is to be taken and if the Trustee to be so removed shall have been
given reasonable notice and opportunity to be heard before the body proposing to
remove him or her.

E. Vacancies. Any vacancy in the Board may be filled by vote of a majority of the
remaining Trustees present at a meeting of Trustees at which a quorum is present. Any
vacancy in the Board resulting from the resignation or removal of a Corporate Trustee
shall be filled by a Member who is a scientist.

F. Meetings. Meetings of the Board shall be held from time to time, not less
frequently than twice annually, as determined by the Board. Special meetings of
Trustees may be called by the Chairperson, or by any seven Trustees, to be held at
such time and place as may be designated. The Chairperson of the Board, when
present, shall preside over all meetings of the Trustees. Written notice shall be sent to
a Trustee's usual or last known place of residence at least two weeks before the
meeting. Notice of a meeting need not be given to any Trustee if a written waiver of
notice executed by such Trustee before or after the meeting is filed with the records
of the meeting, or if such Trustee shall attend the meeting without protesting prior
thereto or at its commencement the lack of notice given to him or her.

G. Quorum and Action by Trustees. A majority of all Trustees then in office shall
constitute a quorum. Any meeting of Trustees may be adjourned by vote of a majority
of Trustees present, whether or not a quorum is present, and the meeting may be held
as adjourned without further notice. When a quorum is present at any meeting of the
Trustees, a majority of the Trustees present and voting (excluding abstentions) shall
decide any question, including the election of officers, unless otherwise required by
law, the Articles of Organization or these Bylaws.

H. Transfers of Interests in Land. There shall be no transfer of title nor long-term
lease of real property held by the Corporation without prior approval of not less than
two-thirds of the Trustees. Such real property transactions shall be finally acted upon
at a meeting of the Board only if presented and discussed at a prior meeting of the
Board. Either meeting may be a special meeting and no less than four weeks shall
elapse between the two meetings. Any property acquired by the Corporation after
December 1. 1989 may be sold, any mortgage or pledge of real property (regardless

R8S Annual Report

of when acquired) to secure borrowings by the Corporation may be granted, and any
transfer of title or interest in real property pursuant to the foreclosure or endorsement
of any such mortgage or pledge of real property may be effected by any holder of a
mortgage or pledge of real property of the Corporation, with the prior approval of not
less than two-thirds of the Trustees (other than any Trustee or Trustees with a direct
or indirect financial interest in the transaction being considered for approval) who are
present at a regular or special meeting of the Board at which there is a quorum.

ARTICLE V— COMMITTEES

A. Executive Committee. There shall be an Executive Committee of the Board of
Trustees which shall consist of not more than eleven (11) Trustees, including c\
officio Trustees, elected by the Board.

The Chairperson of the Board shall act as Chairperson of the Executive Committee
and the Vice Chairperson as Vice Chairperson. The Executive Committee shall meet
at such times and places and upon such notice and appoint such subcommittees as the
Committee shall determine.

The Executive Committee shall have and may exercise all the powers of the Board
during the intervals between meetings of the Board except those powers specifically
withheld, from time to time, by vote of the Board or by law. The Executive
Committee may also appoint such committees, including persons who are not Trust-
ees, as it may, from time to time, approve to make recommendations with respect to
matters to be acted upon by the Executive Committee or the Board.

The Executive Committee shall keep appropriate minutes of its meetings, which
shall be reported to the Board. Any actions taken by the Executive Committee shall
also be reported to the Board.

B. Nominating Committee. There shall be a Nominating Committee which shall
consist of not fewer than four nor more than six Trustees appointed by the Board in
a manner which shall reflect the balance between Corporate Trustees and Trustees-
at-Large on the Board. The Nominating Committee shall nominate persons for
election as Corporate Trustees and Trustees -at -Large, Chairperson of the Board. Vice
Chairperson of the Board. President, Treasurer, Clerk, Director of the Laboratory and
such other officers, if any, as needed, in accordance with the requirements of these
Bylaws. The Nominating Committee shall also be responsible for overseeing the
training of new Trustees. The Chairperson of the Board of Trustees shall appoint the
Chairperson of the Nominating Committee. The Chairperson of the Science Council
shall be an e\ officio voting member of the Nominating Committee.

C. Science Council. There shall he a Science Council (the "Council") which shall
consist of Members of the Corporation elected to the Council by vote of the Members
of the Corporation, and which shall advise the Board with respect to matters con-
cerning the Corporation's mission, its scientific and instructional endeavors, and the
appomtmenl and promotions of persons or committees with responsibility for mailers
requiring scientific expertise. Unless otherwise approved by a majority of the mem-
bers of the Council, the Chairperson of the Council shall be elected annually by the
Council. The chief executive officer of the Corporation shall be an ex officio voting
member of the Council.

D. Board of Overseers. There shall be a Board of Overseers which shall consist of
not fewer than five nor more than eight scientists who have expertise concerning
matters with which the Corporation is involved. Members of the Board of Overseers
may or may not be Members of the Corporation and may be appointed by the Board
of Trustees on the basis of recommendations submitted from scientists and scientific
organizations or societies. The Board of Overseers shall be available to review and
offer recommendations to the officers. Trustees and Science Council regarding
scientific activities conducted or proposed by the Corporation and shall meet from
time to time, not less frequently than annually, as determined by the Board of
Trustees.

E. Board Committees Generally. The Trustees may elect or appoint one or more
other committees (including, but not limited to, an Investment Committee, a Devel-
opment Committee, an Audit Committee, a Facilities and Capital Equipment Com-
mittee and a Long-Range Planning Committee) and may delegate to any such
committee or committees any or all of their powers, except those which by law, the
Articles of Organization or these Bylaws the Trustees are prohibited from delegating;
provided that any committee to which the powers of the Trustees are delegated shall
consist solely of Trustees. The members of any such committee shall have such tenure
and duties as the Trustees shall determine. The Investment Committee, which shall
oversee the management of the Corporation's endowment funds and marketable
securities shall include as e\ officio members, the Chairperson of the Board, the
Treasurer and the Chairperson of the Audit Committee, together with such Trustees
as may be required for not less than two-thirds of the Investment Committee to consist
of Trustees. Except as otherwise provided by these Bylaws or determined by the
Trustees, any such committee may make rules for the conduct of its business, but.

unless otherwise provided by the Trustees or in such rules, its business shall be
conducted as nearly as possible in the same manner as is provided by these Bylaws
for the Trustees.

F. Actions Without a Meeting. Any action required or permitted to be taken at any
meeting of the Executive Committee or any other committee elected by the Trustees
may be taken without a meeting if all members of such committees consent to the
action in writing and such written consents are filed with the records of meetings.
Members of the Executive Committee or any other committee elected by the Trustees
may also participate in any meeting by means of a telephone conference call, or
otherwise lake action in such a manner as may. from time to time, be permitted by
law.

G. Manual oj Procedures. The Board of Trustees, on the recommendation of the
Executive Committee, shall establish guidelines and modifications thereof to be
recorded in a Manual of Procedures. Guidelines shall establish procedures for: (1)
Nomination and election of members of the Corporation, Board of Trustees and
Executive Committee; (2) Election of Officers; (3) Formation and Function of
Standing Committees.

ARTICLE VI— OFFICERS

A. Enumeration. The officers of the Corporation shall consist of a President, a
Treasurer and a Clerk, and such other officers having the powers of President.
Treasurer and Clerk as the Board may determine, and a Director of the Laboratory.
The Corporation may have such other officers and assistant officers as the Board may
determine, including (without limitation) a Chairperson of the Board, Vice Chairper-
son and one or more Vice Presidents, Assistant Treasurers or Assistant Clerks. Any
two or more offices may be held by the same person. The Chairperson and Vice
Chairperson of the Board shall be elected by and from the Trustees, but other officers
of the Corporation need nol be Trustees or Members. If required by the Trustees, any
officer shall give the Corporation a bond for the faithful performance of his or her
duties in such amount and with such surety or sureties as shall be satisfactory to the
Trustees.

B. Tenure. Except as otherwise provided by law, by the Articles of Organization
or by these Bylaws, the President. Treasurer, and all other officers shall hold office
until the first meeting of the Board following the annual meeting of Members and
thereafter, until his or her successor is chosen and qualified.

C. Resignation. Any officer may resign by delivering his or her written resignation
to the Corporation al its principal office or to the President or Clerk and such
resignation shall be effective upon receipt unless it is specified to be effective at some
other lime or upon the happening of some other event.

D. Removal. The Board may remove any officer with or without cause by a vote
of a majority of the entire number of Trustees then in office, at a meeting of the Board
called for lhat purpose and for which notice of the purpose thereof has been given,
provided that an officer may be removed for cause only after having an opportunity
to be heard by the Board at a meeting of the Board at which a quorum is personally
present and voting.

E. Vacancy. A vacancy in any office may be filled for the unexpired balance of the
term by vote of a majority of the Trustees present at any meeting of Trustees at which
a quorum is present or by written consent of all of the Trustees, if less than a quorum
of Trustees shall remain in office.

F. Chairperson. The Chairperson shall have such powers and duties as may be
determined by the Board and. unless otherwise determined by the Board, shall serve
in that capacity for a term coterminous with his or her term as Trustee.

G. Vice Chairperson. The Vice Chairperson shall perform the duties and exercise
the powers of the Chairperson in the absence or disability of the Chairperson, and
shall perform such other duties and possess such other powers as may be determined
by the Board. Unless otherwise determined by the Board, the Vice Chairperson shall
serve for a one-year term.

H. Director. The Director shall be the chief operating officer and, unless otherwise
voted by the Trustees, the chief executive officer of the Corporation. The Director
shall, subject to the direction of the Trustees, have general supervision of the
Laboratory and control of the business of the Corporation. Al the annual meeting, the
Director shall submit a report of the operations of the Corporation for such year and
a statement of its affairs, and shall, from lime to time, report to the Board all matters
within his or her knowledge which the interests of the Corporation may require to be
brought lo its notice.

I. Deputy Director. The Deputy Director, if any, or if there shall be more than one,
the Deputy Directors in the order determined by (he Trustees, shall, in the absence or
disability of the Director, perform ihe duties and exercise the powers of the Director
and shall perform such other duties and shall have such other powers as the Trustees
may, from time to time, prescribe.

Bylaws of the Corporation R89

J. President. The President shall have the powers and duties as may be vested in
him or her by the Board.

K. Treasurer and Assistant Treasurer. The Treasurer shall, subject to the direction
of the Trustees, have general charge of the financial affairs of the Corporation,
including its long-range financial planning, and shall cause to be kept accurate books
of account. The Treasurer shall prepare a yearly report on the financial status of the
Corporation to be delivered at the annual meeting. The Treasurer shall also prepare or
oversee all filings required by the Commonwealth of Massachusetts, the Internal
Revenue Service, or other Federal and State Agencies. The account of the Treasurer
shall be audited annually by a certified public accountant.

The Assistant Treasurer, if any, or if there shall be more than one, the Assistant
Treasurers in the order determined by the Trustees, shall, in the absence or disability
of the Treasurer, perform the duties and exercise the powers of the Treasurer, shall
perform such other duties and shall have such other powers as the Trustees may, from
time to time, prescribe.

L. Clerk and Assistant Clerk. The Clerk shall be a resident of the Commonwealth
of Massachusetts, unless the Corporation has designated a resident agent in the
manner provided by law. The minutes or records of all meetings of the Trustees and
Members shall be kept by the Clerk who shall record, upon the record books of the
Corporation, minutes of the proceedings at such meetings. He or she shall have
custody of the record books of the Corporation and shall have such other powers and
shall perform such other duties as the Trustees may, from time to time, prescribe.

The Assistant Clerk, if any, or if there shall be more than one. the Assistant Clerks
in the order determined by the Trustees, shall, in the absence or disability of the Clerk,
perform the duties and exercise the powers of the Clerk and shall perform such other
duties and shall have such other powers as the Trustees may, from time to time,
prescribe.

In the absence of the Clerk and an Assistant Clerk from any meeting, a temporary
Clerk shall be appointed at the meeting.

M. Other Powers and Duties. Each officer shall have in addition to the duties and
powers specifically set forth in these Bylaws, such duties and powers as are custom-
arily incident to his or her office, and such duties and powers as the Trustees may,
from time to time, designate.

ARTICLE VII— AMENDMENTS

These Bylaws may be amended by the affirmative vote of the Members at any
meeting, provided that notice of the substance of the proposed amendment is stated
in the notice of such meeting. As authorized by the Articles of Organization, the
Trustees, by a majority of their number then in office, may also make, amend or repeal
these Bylaws, in whole or in part, except with respect to (a I the provisions of these
Bylaws governing (i) the removal of Trustees and (n) the amendment of these Bylaws
and (b) any provisions of these Bylaws which by law, the Articles of Organization or
these Bylaws, requires action by the Members.

No later than the time of giving notice of meeting of Members next following the
making, amending or repealing by the Trustees of any Bylaw, notice thereof stating
the substance of such change shall be given to all Members entitled to vote on
amending the Bylaws.

Any Bylaw adopted by the Trustees may be amended or repealed by the Members
entitled to vote on amending the Bylaws.

ARTICLE Vm— INDEMNITY

Except as otherwise provided below, the Corporation shall, to the extent legally
permissible, indemnify each person who is, or shall have been, a Trustee, director or
officer of the Corporation or who is serving, or shall have served at the request of the
Corporation as a Trustee, director or officer of another organization in which the
Corporation directly or indirectly has any interest as a shareholder, creditor or
otherwise, against all liabilities and expenses (including judgments, fines, penalties,
and reasonable attorneys' fees and all amounts paid, other than to the Corporation or
such other organization, in compromise or settlement) imposed upon or incurred by
any such person in connection with, or arising out of. the defense or disposition of any
action, suit or other proceeding, whether civil or criminal, in which he or she may be
a defendant or with which he or she may be threatened or otherwise involved, directly
or indirectly, by reason of his or her being or having been such a Trustee, director or
officer.

The Corporation shall provide no indemnification with respect to any matter as to
which any such Trustee, director or officer shall be finally adjudicated in such action,
suit or proceeding not to have acted in good faith in the reasonable belief that his or
her action was in the best interests of the Corporation. The Corporation shall provide
no indemnification with respect to any matter settled or comprised unless such matter

shall have been approved as in the best interests of the Corporation, after notice thai
indemnification is involved, by (i) a disinterested majonty of the Board of the
Executive Committee, or (ii) a majority of the Members.

Indemnification may include payment by the Corporation of expenses in defending
a civil or cnminal action or proceeding in advance of the final disposition of such
action or proceeding upon receipt of an undertaking by the person indemnified to
repay such payment if it is ultimately determined that such person is not entitled to
indemnification under the provisions of this Article VIII, or under any applicable law .

As used in the Article VIII, the terms "Trustee." "director," and "officer"
include their respective heirs, executors, administrators and legal representatives, and
an "interested" Trustee, director or officer is one against whom in such capacity the
proceeding in question or another proceeding on the same or similar grounds is then
pending.

To assure indemnification under this Article VIII of all persons who are determined
by the Corporation or otherwise to be or to have been "fiduciaries" of any employee
benefits plan of the Corporation which may exist, from time to time, this Article VIII
shall be interpreted as follows: (i) "another organization" shall be deemed to include
such an employee benefit plan, including without limitation, any plan of the Corpo-
ration which is governed by the Act of Congress entitled "Employee Retirement
Income Secunty Act of 1974," as amended, from time to time, ("ERISA"); (ii)
"Trustee" shall be deemed to include any person requested by the Corporation to
serve as such for an employee benefit plan where the performance by such person of
his or her duties to the Corporation also imposes duties on. or otherwise involves
services by, such person to the plan or participants or beneficiaries of the plan; (iii)
"fines" shall be deemed to include any excise tax plan pursuant to ERISA; and (iv)
actions taken or omitted by a person with respect to an employee benefit plan in the
performance of such person's duties for a purpose reasonably believed by such person
to be in the interest of the participants and beneficiaries of the plan shall be deemed
to be for a purpose which is in the best interests of the Corporation.

The right of indemnification provided in this Article VIII shall not be exclusive of
or affect any other rights to which any Trustee, director or officer may be entitled
under any agreement, statute, vote of Members or otherwise. The Corporation's
obligation to provide indemnification under this Article VIII shall be offset to the
extent of any other source of indemnification of any otherwise applicable insurance
coverage under a policy maintained by the Corporation or any other person. Nothing
contained in the Article shall affect any rights to which employees and corporate
personnel other than Trustees, directors or officers may be entitled by contract, by
vote of the Board or of the Executive Committee or otherwise.

ARTICLE IX— DISSOLUTION

The consent of every Trustee shall be necessary to effect a dissolution of the Marine
Biological Laboratory. In case of dissolution, the property shall be disposed of in such
a manner and upon such terms as shall be determined by the affirmative vote of
two-thirds of the Trustees then in office in accordance with the laws of the Com-
monwealth of Massachusetts.

ARTICLE X— MISCELLANEOUS PROVISIONS

A. Fiscal Year. Except as otherwise determined by the Trustees, the fiscal year of
the Corporation shall end on December 31st of each year.

B. Seal. Unless otherwise determined by the Trustees, the Corporation may have
a seal in such form as the Trustees may determine, from time to time.

C. Execution of Instruments. All checks, deeds, leases, transfers, contracts, bonds,
notes and other obligations authorized to be executed by an officer of the Corporation
in its behalf shall be signed by the Director or the Treasurer except as the Trustees
may generally or in particular cases otherwise determine. A certificate by the Clerk or
an Assistant Clerk, or a temporary Clerk, as to any action taken by the Members.
Board of Trustees or any officer or representative of the Corporation shall as to all
persons who rely thereon in good faith be conclusive evidence of such action.

D. Corporate Records. The original, or attested copies, of the Articles of Organi-
zation, Bylaws and records of all meetings of the Members shall be kept in Massa-
chusetts at the principal office of the Corporation, or at an office of the Corporation's
Clerk or resident agent. Said copies and records need not all be kept in the same office.
They shall be available at all reasonable times for inspection by any Member for any
proper purpose, but not to secure a list of Members for a purpose other than in the
interest of the applicant, as a Member, relative to the affairs of the Corporation.

E. Articles of Organization. All references in these Bylaws to the Articles of
Organization shall be deemed to refer to the Articles of Organization of the Corpo-
ration, as amended and in effect, from time to time.

F. Transactions with Interested Parties. In the absence of fraud, no contract or other

R90 Annual Report

transaction between this Corporation and any other corporation or any firm, association,
partnership or person shall be affected or invalidated by the fact that any Trustee or officer
of this Corporation is pecuniarily or otherwise interested in or is a director, member or
officer of such other corporation or of such firm, association or partnership or in a party
to or is pecuniarily or otherwise interested in such contract or other transaction or is in any
way connected with any person or person, firm, association, partnership, or corporation
pecuniarily or otherwise interested therein; provided that the fact that he or she individ-
ually or as a director, member or officer of such corporation, firm, association or
partnership in such a party or is so interesied shall he disclosed to or shall have been

known by the Board of Trustees or a majority of such Members thereof as shall be present
at a meeting of the Board of Trustees at which action upon any such contract or
transaction shall be taken; any Trustee may be counted in determining the existence of a
quorum and may vote at any meeting of the Board of Trustees for the purpose of
authon/ing any such contract or transaction with like force and effect as if he/she were not
so interested, or were not a director, member or officer of such other corporation, firm,
association or partnership, provided that any vote with respect to such contract or
transaction must be adopted by a majority of the Trustees then in office who have no
interest in such contract or transaction.

Publications

Abenavoli A., L. Forti, and A. Malgaroli. 2000. Mechanisms of spon-
taneous miniature activity at CA3-CA1 synapses: evidence for a diver-
gence from a random Poisson process. Biol. Bull. 199: 184-186.

Ahrens, T. D., and P. A. Siver. 2000. Trophic condition and water
chemistry of lakes on Cape Cod. Massachusetts. USA. Lake Reservoir
Manage. 16(4|: 268-280.

Alvarez, J., A. Giuditta. and E. Koenig. 2000. Protein synthesis in
axons and terminals: signiticance for maintenance, plasticity and reg-
ulation of phenotype. With a critique of slow transport theory. Progr.
Neurobiol. 62: 1-62.

Amaral Zettler, L. A., T. A. Nerad. C. J. O'Kelly, M. T. Peglar, P. M.
Gillevet, J. D. Silberman, and M. L. Sogin. 2000. A molecular
reassessment of the Leptomyxid amoebae. Protist 151: 275-282.

Armstrong, P. B., and R. Asokan. 2000. A Ca ^-independent cytolytic
system from the blood of the marine snail, Busvcon canaliculum. Bin/.
Bull. 199: 194-195.

Asokan, R., M. T. Armstrong, and P. B. Armstrong. 2000. Associa-
tion of os-macroglobulin with the coagulin clot in the American
horseshoe crab. Limulus polyphemus: A potential role in stabilization
from proteolysis. Biol. Bull. 199: 190-192.

Atkins. M. S., A. G. McArthur, and A. P. Teske. 2000. Ancyromona-
dida: A new phylogenetic lineage among the protozoa closely related
to the common ancestor of metazoans. fungi, and choanoflagellates
(Opisthokonta). J. Mol. Evol 51: 278-285.

Basil, J. A., R. T. Hanlon, S. I. Sheikh, and J. Atema. 2000. Three-
dimensional odor tracking by Nautilus pompiliu.i. J. Exp. Biol. 203(91:
1409-1414.

Bearer. E. L.. X. O. Breakeheld, D. Schuback, T. S. Reese, and J. H.
La Vail. 2000. Retrograde axonal transport of herpes simplex virus:
Evidence for a single mechanism and a role for tegument. Prm: Nail.
Acad. Sci. USA 97(14): 8146-8150.

Begley. G. S., B. C. Furie, E. Czerwiec, K. L. Taylor, G. L. Furie, L.
Bronstein, J. Stenflo, and B. Furie. 2000. A conserved motif within
the vitamin K-dependent carboxylase gene is widely distributed across
animal phyla. J. Biol. Chem. 275: 36245-36249.

Bittner, G. D.. and H. M. Fishman. 2000. Axonal sealing following
injury. Pp. 337-370 in /Verve Regeneration. N. Ingoglia and M. Mur-
ray, eds. Marcel Dekker. New York.

Blazquez, P., A. Partsalis, N. Gerrits, and S. M. Highstein. 2000.
Input of the anterior and posterior semicircular canals via interneurons
carrying head velocity information to the dorsal Y group of the ves-
tibular nuclei. J. Neurophysiol. 83: 2891-2904.

Boal, J. G., A. W. Dunham, K. T. Williams, and R. T. Hanlon. 2000.

Experimental evidence for spatial learning in octopuses (Octopus bi-
macitloides). J. Comp. Psychol. 114(3): 246-252.

Bose, C. M., D. Qiu, A. Bergamaschi, B. Gravante, M. Bossi, A. Villa,
F. Rupp. and A. Malgaroli. 2000. Agrin controls synaptic differen-
tiation in hippocampal neurons. /. Neurosci. 20: 9086-9095.

Bouzat, J. L.. L. K. McNeil, H. M. Robertson, L. F. Solter, J. Nixon,
J. E. Beever, H. R. Gaskins, G. Olsen, S. Subramaniam. M. L.
Sogin, and H. A. Lewin. 2000. Phylogenomic analysis of a protea-
some gene family from early-diverging eukaryotes. J. Mol. Evol. 51:
532-543.

Breton, S., N. N. Nsumu, T. Galli, I. Sabolic, P. J. S. Smith, and D.
Brown. 2000. Tetanus toxin-mediated cleavage of cellubrevin inhib-
its proton secretion in the male reproductive tract. Am. J. Phvsiol. Renal
Physinl. 278: F7 17-725.

Brothers, C., E. Marks, and R. Smolowitz. 2000. Conditions affecting
growth and zoosporulation of protistan parasite QPX in culture. Biol.
Bull. 199: 200-201.

Burgos, M. H., M. Goda, and S. Inoue. 2000. Fertilization-induced
changes in the fine structure of stratified Arbacia eggs. II. Observations
with electron microscopy. Biol. Bull. 199: 213-214.

Bush, M. B., M. C. Miller, P. E. De Oliveira, and P. A. Colinvaux. 2000.
Two histories of environmental change and human disturbance in
eastern lowland Amazonia. The Holocene 10: 543-554.

Buzby, K. M., and L. A. Deegan. 2000. Inter-annual fidelity to summer
feeding sites in Arctic grayling. Em-iron. Biol. Fishes 59: 319-327.

Buzby, K. M., and S. A. Perry. 2000. Modeling the potential effects of
climate change on leaf pack processing in central Appalachian streams.
Can. J. Fish. Aauat. Sci. 57: 1773-1783.

Canadell, J. G., H. A. Mooney, D. D. Baldocchi. J. A. Berry. J. R.
Ehleringer, C. B. Field, S. T. Gower, D. Y. Hollinger, J. E. Hunt,
R. B. Jackson, S. W. Running, G. R. Shaver, W. Steffen, S. E.
Trumbore, R. Valentini. and B. Y. Bond. 2000. Carbon metabo-
lism of the terrestrial biosphere: A multi-technique approach for im-
proved understanding. Ecosystems 3: 1 15-130.

Chatterjee, A.. D. M. Porterfield, P. J. S. Smith, and S. J. Roux. 2000.
Gravity-directed calcium current in germinating spores of Ceratopseris
richardii. Planta 210: 607-610.

Chikarmane, H. M., A. M. Kuzirian, R. Kozlowski, M. Kuzirian, and
T. Lee. 2000. Population genetic structure of the goosefish, Laphius
amencanm. Biol. Bull. 199: 227-228.

Clark, M. A., N. A. Moran. P. Baumann, and J. J. VVernegreen. 2000.

R91

R92 Annual Report

Cospeciation between bacterial endosymbionts (Buc/inera} and a recent
radiation of aphids ( Uroleucon) and pitfalls of testing for phylogenetic
congruence. Evolution 54: 517-525.

Clay, J. R., and A. M. Kuzirian. 2000. Localization of voltage-gated
K* channels in squid giant axons. J. Neurobiol 45: 172-184.

Clein, J. S., B. L. Kwiatkowski, A. D. McGuire, J. E. Hobbie, E. B.
Rastetter, J. M. Melillo, and D. W. Kicklighter. 2000. Modeling
carbon responses of tundra ecosystems to historical and project climate:
A comparison of a plot- and a global-scale ecosystem model to identify
process-based uncertainties. Global Change Butl. 6(Suppl. I): 127-
140.

Colinvaux, P. A., and P. E. De Oliveira. 2000. Paleoecology and
climate of the Amazon basin during the last glacial cycle. J. Quat. Sci.
15: 347-356.

Colinvaux, P. A., P. E. De Oliveira, and M. B. Bush. 2000. Amazon
and neotropical plant communities on glacial time scales: The failure of
the aridity and refuge hypotheses. Quat. Sci. Re\: 19: 141-169.

Creton, R., J. A. Kreiling, and L. F. Jaffe. 2000. Presence and roles of
calcium gradients along the dorsal-ventral axis in Drosophila embryos.
De\: Bid. 217: 375-385.

Crump, B. C., and J. A. Baross. 2000. Archaeaplankton in the Colum-
bia River, its estuary and the adjacent coastal ocean. USA. FEMS
Microbiol. Ecol. 31: 231-239.

Danuser. G., and R. Oldenbourg. 2000. Probing f-actin flow by track-
ing shape fluctuations of radial bundles in lamellipodia of motile cells.
Biophysics J. 79: 191-201.

Danuser, G., P. T. Iran, and E. D. Salmon. 2000. Tracking differential
interference contrast diffraction line images with nanometer sensitivity.
J. Microsc. 198(1): 34-53.

Deegan, L. A., J. E. Hughes, and R. A. Rountree. 2000. Salt marsh
ecosystem support of marine transient species. Pp. 333-365 in Con-
cepts and Controversies in Tidal Marsh Ecology, M. P. Weinstein and
D. A. Kreeger, eds. Kluwer Academic. Boston, MA. 864 pp.

Delgado-Viscogliosi, P., E. Viscogliosi, D. Gerbod, J. Kulda, M. L.
Sogin, and V. P. Edgcomb. 2000. Molecular phylogeny of paraba-
salids based on small subunit rRNA sequences, with emphasis on the
Trichomonadinae subfamily. J. Eukaryol. Microbiol. 47: 70-75.

Detrait, E., C. S. Eddleman, S. M. Yoo, M. Fukuda, M. P. Nguyen,
G. D. Bittner, and H. M. Fishman. 2000. Axolemmal repair re-
quires proteins that mediate synaptic vesicle fusion. J. Neurobiol 44:
382-391.

Detrait, E. R., S. Yoo, C. S. Eddleman, M. Fukuda, G. D. Bittner, and
H. M. Fishman. 2000. Plasmalemmal repair of severed neurites of
PC 12 cells requires Ca2"1" and synaptotagmin. J. Neurosci. Res. 62:
566-573.

De Weer, P.. D. C. Gadsby, and R. F. Rakowski. 2000. The Na/K-
ATPase: A current-generating enzyme. Pp. 27-34 in Na/K A TPase and
Related ATPases, K. Taniguchi and S. Kaya, eds. Excerpta Medica
International Congress Series 1207. Elsevier, Amsterdam.

DiPolo, R., G. Berberian, and L. Beauge. 2000. In squid nerves intra-
cellular Mg2 + promotes deactivation of the ATP-upregulated Na"1"/
Ca2* exchanger. Am. J. Physio/. Cell Physiol. 279: C1631-C1639.

Dosemeci, A., T. S. Reese, J. Petersen, and J-H. Tao-Cheng. 2000. A
novel paniculate form of Ca2+/CaMKJI-dependent protein kinase II in
neurons. J. Neurosci. 20: 3076-3084.

Doussau, F., and G. J. Augustine. 2000. The actin cytoskeleton and
neurotransmitter release: An overview. Biochemie 82: 353-363.

Eddleman, C. S., G. D. Bittner, and H. M. Fishman. 2000. Barrier
permeability at cut axonal ends progressively decreases until an ionic
seal is formed. Binphys. J. 79: 1883-1890.

Epstein, D. A., H. T. Epstein, F. M. Child, and A. M. Kuzirian. 2000.
Memory consolidation in Hennissenda crassicomis. Biol. Bull. 199:
182-183.

Fadool, D. A.. K. Tucker, J. J. Phillips, and J. A. Simmen. 2000. Brain

insulin receptor causes activity-dependent current suppression in the
olfactory bulb through multiple phosphorylation of Kvl.3. J. Neuro-
physiot. 83: 2332-2348.

Filoso, S., and M. R. Williams. 2000. The hydrochemical influence of
the Branco River on the Negro River and Anavilhanas archipelago.
Amazonas, Brazil. Arch. Hydrobiol 148: 563-585.

Fisher, T. R., D. Correll, R. Costanza, J. T. Hollibaugh, C. S. Hopkin-
son, Jr., R. W. Howarth, N. N. Rabalais, J. E. Richey, C. J.
Vbrosmarty, and R. Wiegert. 2000. Synthesizing drainage basin
inputs to coastal systems. Pp. 81-105 in Estuarine Science: A Synthetic
Approach to Research and Practice, J. Hobbie, ed. Island Press, Wash-
ington, D.C.

Forster, H., M. P. Cummings, and M. D. Coffey. 2000. Phylogenetic
relationships of Phytophthora species based on ribosomal ITS I DNA
sequence analysis with emphasis on Waterhouse groups V and VI.
Mycol. Res. 104: 1055-1061.

Fukui, Y., T. Q. P. Uyeda, C. Kitayama, and S. Inoue. 2000. How well
can an amoeba climb? Proc. Natl. Acad. Sci. USA 97: 10020-10025.

Funk, D. J., L. Helbling, J. J. Wernegreen, and N. A. Moran. 2000.
Perfect evolutionary congruence among multiple symbiont genomes in
an aphid species. Proc. R. Soc. Land. B 657: 2517-2521.

Garcia-Montiel, D., C. Neill, J. M. Melillo, S. M. Thomas, P. A.
Steudler, and C. C. Cerri. 2000. Soil phosphorus transformations
after forest clearing for pasture in the Brazilian Amazon. Soil Sci. Soc.
Am. J. 64: 1792-1804.

Garcia-Verela, M., G. Perez-Ponce de Leon, P. de la Torre, M. P.
Cummings, S. S. S. Sarma, and J. P. Laclette. 2000. Phylogenetic
analysis of Acanthocephala based on 18S ribosomal gene sequences. J.
Mol. Evol. 50: 532-540.

Gerbod, D., V. P. Edgcomb, C. Noel, P. Delgado-Viscogliosi, and E.
Viscogliosi. 2000. Phylogenetic position of parabasalid symbionts
from the termite Kalolermes flavicollis based on small subunit rRNA
sequences. Int. Microbiol. 3: 165-172.

Gleeson, R. A., K. Hammar. and P. J. S. Smith. 2000. Sustaining
olfaction at low salinities: Mapping ion flux associated with the olfac-
tory sensilla of the blue crab Callinectes sapidus. J. Exp. Biol. 203:
3145-3152.

Gleeson, R. A., L. M. McDowell, H. C. Aldrich, K. Hammar, and
P. J. S. Smith. 2000. Sustaining olfaction at low salinities: Evidence
for a paracellular route of ion movement from the hemolymph to the
sensillar lymph in the olfactory sensillar of the blue crab, Callinecles
sapidus. Cell Tissue Res. 301: 423-431.

Goda, M., M. H. Burgos, and S. Inoue. 2000. Fertilization-induced
changes in the fine structure of stratified Arbacia eggs. I. Observations
on live cells with the centrifuge polarizing microscope. Biol. Bull. 199:
212-213.

Gough, L., G. R. Shaver, J. Carroll, D. Rover, and J. A. Laundre. 2000.
Vascular plant species richness in Alaskan arctic tundra: The impor-
tance of soil pH. J. Ecol. 88: 54-66.

Gould, R. M., C. M. Freund, J. Engler, and H. G. Morrison. 2000.
Optimization of homogenization conditions used to isolate mRNAs in
processes of myelinating oligodendrocytes. Biol. Bull. 199: 215-217.

Hanselmann, R., R. Smolowitz, and D. Gibson. 2000. Identification of
proliferating cells in hard clams. Biol. Bull. 199: 199-200.

Harasewych, M. G., and A. G. McArthur. 2000. A molecular phylog-
eny of the Patellogastropoda (Mollusca: Gastropoda). Mar. Biol. 137:
183-194.

Harrington, J. M., and P. B. Armstrong. 2000. Initial characterization
of a potential anti-fouling system in the American horseshoe crab,
Limulus polyphemus. Biol. Bull. 199: 189-190.

Head, J. F., S. Inouye, K. Teranishi, and O. Shimomura. 2000. The
crystal structure of the photoprotein aequorin at 2.3 A resolution.
Nature 405: 372-376.

Hendricks, J. J., J. D. Aber, K. J. Nadelhoffer, and R. D. Hallett. 2000.

Publications R93

Nitrogen controls on tine root substrate quality in temperate forest
ecosystems. Ecosvstems 3: 57-69.

Henry, J. Q., M. Q. Martindale, and B. C. Boyer. 2(100. The unique
developmental program of the acoel flatworm Neochildia fiisca. De\:
Biol. 220: 285-295.

Herak-Kramberger. C.. I. Sabolic, M. Blanusa. P. J. S. Smith. 1).
Brown, and S. Breton. 2000. Cadmium inhibits vacuolar H+
ATPase-mediated acidification in rat epididymis. Biol. Retrod. 63:
599-606.

Hobbie. E. A., S. A. Macko, and M. Williams. 2000. Correlations
between foliar delta 15N and nitrogen concentrations may indicate
plant-mycorrhizal interactions. Oecologia 122: 273-283.

Hobbie, J. E., ed. 2000. Estuarine Science: A Synthetic Approach to
Research ami Practice. Island Press. Washington. D.C. 539 pp.

Holmes, R. M. 2000. The importance of ground water to stream ecosys-
tem function. Pp. 137-148 in Streams and Ground Waters, J. B. Jones
and P. J. Mulholland, eds. Academic Press. San Diego, CA.

Holmes, R. M., B. J. Peterson, L. Deegan, J. Hughes, and B. Fry. 2000.
Nitrogen biogeochemistry in the oligohaline zone of a New England
estuary. Ecology 81: 416-432.

Holmes, R. M., B. J. Peterson, V. V. Gordeev. A. V. Zhulidov, M.
Meybeck, R. B. Lammers, and C. J. Vorosmarty. 2000. Flux of
nutrients from Russian rivers to the Arctic Ocean: Can we establish a
baseline against which to judge future changes? Water Resources Res.
36: 2309-2320.

Holmgren, M., J. Wagg, F. Bezanilla, R. F. Rakowski, P. De Weer, and
D. C. Gadsby. 2000. Three distinct sequential steps in extracellular
release of three Na+ ions by the Na.K-ATPase. Nature 403: 898-901.

Hughes, J. E., L. A. Deegan, B. J. Peterson, R. M. Holmes, and B. Fry.
2000. Nitrogen flow through the food web in the oligohaline zone of
a New England estuary. Ecology 81: 433-452.

Inouye, S., K. Watanabe, H. Nakamura, and O. Shimomura. 2000.
Secretional luciferase of the luminous shrimp Oplopliorus graciliros-
tris: cDNA cloning of a novel imidazopyrazinone luciferase. FEBS
Lett. 481: 19-25.

Jackson, R. B., H. J. Schenk, E. G. Jobbagy, J. Canadell, G. D. Colello,
R. E. Dickinson, C. B. Field, P. Friedlingstein, M. Heimann, K.
Hibbard, D. VV. Kicklighter, A. Kleidon, R. P. Neilson, W. J.
Parton, O. E. Sala, and M. T. Sykes. 2000. Belowground conse-
quences of vegetation change and their treatment in models. Ecol. Appl.
10: 470-483.

Johnson, L. C., G. R. Shaver, D. H. Cades, E. Rastetter, K. Nadelhof-
fer, A. Giblin, J. Laundre, and A. Stanley. 2000. Plant carbon-
nutrient interactions control CO, exchange in Alaskan wet sedge tundra
ecosystems. Ecology 81: 453-469.

Jonasson, S., T. V. Callaghan, G. R. Shaver, and L. A. Nielsen. 2000.
Arctic terrestrial ecosystems and ecosystem function. Pp. 275-313 in
The Arctic: Environment, People, Policy, M. Nuttall and T. V. Cal-
laghan, eds. Harwood Academic Publishers, Amsterdam.

Jung, S.-K., K. Hammar, and P. J. S. Smith. 2000. Development of
self-referencing oxygen microsensor and its application to single pan-
creatic HIT cells: Effects of adenylate cyclase activator forskolin on
oxygen consumption. Biol. Bull. 199: 197-198.

Kalume, D. E., J. Stenflo, E. Czerwiec, B. Hambe, B. C. Furie, B. Furie,
and P. Roepstorff. 2000. Determination of the covalent structure of
two conotoxins from Coitus textile by MALDI-TOF and ESI MS. J.
Mass Spectrom. 35: 145-156.

Kaplan, I. M., and H. Kite-Powell. 2000. Safety at sea and fisheries
management: Fishermen's attitudes and the need for co-management.
Mar. Policy 24: 493-497.

Kreiling, J. A., R. E. Stephens, A. M. Kuzirian, K. Jessen-Eller, and
C. L. Reinisch. 2000. Polychlorinated biphenyls are selectively neu-
rotoxic in the developing Spisula solidissima embryo. J. Toxicol. En-
viron. Health A 61: 101-119.

Kremer, J. N., W. M. Kemp, A. E. Giblin, 1. Valiela, S. P. Seitzinger,
and E. E. Hofmann. 2000. Linking biogeochemical processes to
higher trophic levels. Pp. 299-341 in Estuarine Science: A Synthetic
Approach to Research ami Practice, i. Hobbie. ed. Island Press. Wash-
ington, D.C.

Kuhns. VV. J., M. M. Burger, M. Sarkar, X. Fernandez-Busquets, and
T. Simpson. 2000. Enzymatic biosynthesis of N-linked glycan by the
marine sponge Microciona pro/ifera. Biol. Bull. 199: 192-194.

Kuner, T., and G. J. Augustine. 2000. A genetically encoded ratiomet-
ric indicator for chloride: Capturing chloride transients in cultured
hippocampal neurons. Neuron 27: 447-459.

Lam, Y.-w., L. B. Cohen, M. Wachowiak, and M. R. Zochowski. 2000.
Odors elicit three different oscillations in the turtle olfactory bulb.
J. Neurosci. 20: 749-762.

Landowne, D. 2000. Heavy water (D2O) alters the sodium channel
gating current in squid giant axons. Biol. Bull. 199: 164-165.

Langford, G. M. 2000. Video-enhanced microscopy for analysis of
cytoskeleton structure and function. Pp. 31-43 in Methods in Molec-
ular Biology, Vol. 161: Cytoskeleton Methods and Protocols, Ray H.
Gavin, ed. Humana Press. Totowa. NJ.

Lichstein, J. W., M. L. Ballinger, A. R. Blanehette, H. M. Fishman, and
G. D. Bittner. 2000. Structural changes at cut ends of earthworm
giant axons in the interval between dye barrier formation and neuritic
outgrowth. J. Coinp. Neurol. 416: 143-157.

Liu, L., R. Oldenbourg, J. R. Trimarchi, and D. L. Keefe. 2000. A
reliable, noninvasive technique for spindle imaging and enucleation of
mammalian oocytes. Nature Biotechnol. 18: 223-225.

Liu, L., J. R. Trimarchi, and D. L. Keefe. 2000. Involvement of
mitochondria in oxidative stress-induced cell death in mouse zygotes.
Biol. Reprod. 62: 1745-1753.

Liu, L., J. R. Trimarchi, R. Oldenbourg, and D. L. Keefe. 2000.
Increased birefringence in the meiotic spindle provides a new marker
for the onset of activation in living oocytes. Biol. Reprod. 63: 25 1-258.

MacKenzie, R., D. Newman, M. M. Burger, R. Roy, and W. J. Kuhns.
2000. Adhesion of a viral envelope protein to a non-self-binding
domain of the aggregation factor in the marine sponge Microciona
prolifera. Biol. Bull. 199: 209-21 1.

Magill, A. H., J. D. Aber, G. M. Berntson, W. H. McDowell, K. J.
Nadelhoffer. J. M. Melillo, and P. Steudler. 2000. Long-term ad-
ditions and nitrogen saturation in two temperate forests. Ecosystems 3:
238-253.

Mandile, P., S. Vescia, P. Montagnese, S. Piscopo, M. Cotugno, and A.
Giuditta. 2000. Post-trial sleep sequences including transition sleep
are involved in avoidance learning of adult rats. Behav. Brain Res. 112:
23-31.

Maxwell, M. R., and R. T. Hanlon. 2000. Female reproductive output
in the squid Loligo pea/eii: Multiple egg clutches and implications for
a spawning strategy. Mar. Ecol. Prog. Ser. 199: 159-170.

Maxwell, M. R., K. M. Buresch, and R. T. Hanlon. 2000. Pattern of
inheritance of microsatellite loci in the squid Lo/igo pealeii (Mollusca:
Cephalopoda). Mar. Biotechnol. 2: 517-521.

McArthur, A. G., H. G. Morrison, J. E. J. Nixon, N. Q. E. Passama-
neck, U. Kim, G. Hinkle, M. K. Crocker, M. E. Holder, R. Farr,
C. I. Reich, G. J. Olsen, S. B. Aley, R. D. Adam, F. D. Gillin, and
M. L. Sogin. 2000. The Giardia genome project database. FEMS
Microbiol. Lett. 189: 271-273.

McCulloh. D. H., P. I. Ivinnet, D. Landowne, and E. L. Chambers.
2000. Calcium influx mediates the voltage-dependence of sperm en-
try into sea urchin eggs. Dev. Biol. 223: 449-462.

McGuire, A. D., J. M. Melillo, J. T. Randerson, W. J. Parton, M.
Heimann, R. A. Meier, J. S. Clein, D. W. Kicklighter, and S. Sauf.
2000. Modeling the effects of snowpack on heterotrophic respiration
across northern temperate and high latitude regions: Comparison with

R94 Annual Report

measurements of atmospheric carbon dioxide in high latitudes. Biogeo-
chemistry 48: 94-114.

McGuire, A. D., J. S. Clein, J. M. Melillo, D. W. Kicklighler, R. A.
Meier, C. J. Vorosmarty, and M. C. Serreze. 2000. Modelling
carbon responses of tundra ecosystems to historical and projected
climate: Sensitivity of pan-Arctic carbon storage to temporal and
spatial variation in climate. Global Change Biol. 6: 141-159.

McNeil. P. L., S. S. Vogel, K. Miyake. and M. Terasaki. 2000. Patch-
ing plasma membrane disruptions with cytoplasmic membrane. J. Cell
Sci. 113: 1891-1902.

Mensinger, A. F., D. J. Anderson, C. J. Buchko, M. A. Johnson, D. C.
Martin, P. A. Tresco, R. B. Silver, and S. M. Highstein. 2000.
Chronic recording of regenerating VIHth nerve axons with a sieve
electrode. J. Neiirophysioi. 83(1): 611-615.

Messerli, M. A., R. Creton, L. F. Jaffe, and K. R. Robinson. 2000.
Periodic increases in elongation rate precede periodic increases in
cytosolic Ca2+ during tip growth of Lilium longiflorum pollen tubes.
Dev. Biol. 222: 84-98.

Molina, A. J. A., P. J. S. Smith, and R. P. Malchow. 2000. Hydrogen
ion fluxes from isolated retina] horizontal cells: Modulation by gluta-
mate. Biol. Bull. 199: 168-170.

Molyneaux, B. J., M. K. Mulcahey, P. Stafford, and G. M. Langt'ord.
2000. Sequence and phylogenetic analysis of squid myosin V: a
vesicle motor in nerve cells. Cell Molil. Cytoskeleton 46: 108-1 15.

Moran. N. A., and J. J. Wernegreen. 2000. Are mutualism and para-
sitism irreversible evolutionary alternatives for endosymbiotic bacte-
ria? Insights from molecular phylogenetics and genomics. Trends Ecol.
Evol. 15: 321-326.

Morgan, J. R., K. Prasad, W. Hao, G. J. Augustine, and E. Lafer. 2000.
A conserved clathrin assembly motif essential for synaptic vesicle
endocytosis. J. Neurosci. 20: 8667-8676.

Mulholland, P. J., J. L. Tank, D. M. Sanzone, W. M. Wollheim, B. J.
Peterson, J. R. Webster, and J. L. Meyer. 2000. Food resources of
stream macroinvertebrates determined by natural-abundance stable C
and N isotopes and a super ( 15)N tracer addition. J. North Am. Benihol.
Soc. 19(1): 145-157.

NadelholTer, K. J. 2000. The potential effects of nitrogen deposition on
fine root production in forest ecosystems. New Phvtol. 137: 131-139.

Nadelhoffer, K. J., R. D. Bowden, R. D. Boone, and K. Lajtha. 2000.
Controls on forest soil organic matter development and dynamics:
Chronic litter manipulation as a potential international LTER activity.
Pp. 3-9 in Cooperation in Long Tenn Ecological Research in Central
and Eastern Europe: Proceedings of the ILTER Regional Workshop,
22-25 June, Budapest, Hungary. K. Lajtha and K. Vanderbilt, eds.
Oregon State University. Corvallis, OR.

Neill, C., and E. A. Davidson. 2000. Soil carbon accumulation or loss
following deforestation for pasture in the Brazilian Amazon. Pp. 197-
21 1 in Glohal Climate Change ami Tropical Ecosystems, R. Lai. J. M.
Kimble. and B. A. Stewart, eds. CRC Press. New York.

Ng, W. V., P. Liang, M. Riley, L. Hood, and S. DasSarma. 2000.
Genome sequence of Halobacte riiim species NRC-1. Proc. Natl. Acad.
Sci. USA 97: 12176-12181.

Ogunseitan, O. A., S. Yang, and J. Ericson. 2000. Microbial delta-
aminolevulinate dehydratase as a biosensor for lead (Pb) bioavailabihty
in contaminated environments. Soil Biol. Biochem. 32: 1899-1906.

Ohara, P. T., M. S. Chin, and J. H. La Vail. 2000. The spread of herpes
simplex virus type 1 from trigeminul neurons to the murine cornea: an
immunoelectron microscopy study. J. Virol. 74(10): 4776-4786.

Oldenbourg, R., and P. Toro'k. 2000. Point spread functions of a
polarizing microscope equipped with high numerical aperture lenses.
.4/v>/. Optics 39: 6325-6331.

Oldenbourg. R., K. Katoh, and G. Danuser. 2000. Mechanism of
lateral movement of filopodia and radial actin bundles across neuronal
growth cones. Biophys. J. 78: 1176-1182.

Palazzo, R. E., and B. J. Schnackenberg. 2000. Centrosome matura-
tion. Cnir. Top. Dc\: Biol. V 49: 449-470.
Pan, Y., J. M. Melillo, D. \V. Kicklighter, X. Xiao, and A. D. McGuire.

2000. Modeling structural and functional responses of terrestrial eco-
systems in China to changes in climate and atmospheric CO2. Acta
Phytoecol. Sinica 24(4): 513-526.

Pollock, D. D., J. A. Eisen, N. A. Doggett, and M. P. Cummings. 2000.
A case for evolutionary genomics and the comprehensive examination
of sequence biodiversity. Mol. Biol. Evol. 17: 1776-1788.

Porterh'eld, D. M., and P. J. S. Smith. 2000. Single-cell, real-time
measurements of extracellular oxygen and proton fluxes from Spiro-
f>\ra grevilleana. Protoplasma 212: 80-88.

Porterfield, D. M., R. F. Corkey, R. H. Sanger, K. Tornheim, P. J. S.
Smith, and B. E. Corkey. 2000. Oxygen consumption oscillates in
single clonal pancreatic beta-cells (HIT). Diabetes 49: 151 1-1516.

Prahlad, V.. B. T. Helfand, G. M. Langford, R. D. Vale, and R. D.
Goldman. 2000. Fast transport of neurotilament protein along mi-
crotubules in squid axoplasm. J. Cell Sci. 113: 3939-3946.

Raymond, P. A., and J. E. Bauer. 2000. Bacterial consumption of DOC
during transport through a temperate estuary. Aunat. Microb. Ecol. 22:
1-12.

Raymond, P. A.. J. E. Bauer, and J. J. Cole. 2000. Atmospheric CO2
evasion, dissolved inorganic carbon production, and net heterotrophy in
the York River estuary. Limnol. Oceanogr. 45: 1707-1717.

Riley, M., and M. Serres. 2000. Escherichia coli. In The Prokaryotes:
An Evolving Electronic Resource for the Microbiological Community.
[Online]. Springer- Verlag, New York. Available: http://www.
prokaryotes.com.

Riley, M., and M. Serres. 2000. Interim report on genomics of E. coli.
.A/»i». Rev. Microhiol. 54: 341-41 I.

Sagar, S.. B. Pazdur, E. Indyk, and R. F. Rakowski. 2000. The beta
subunit of the Na+,K*-ATPase in Xenopus laevis. Pp. 261-264 in
Na/K ATPase and Related ATPases, K. Taniguchi and S. Kaya. eds.
Excerpta Medica International Congress Series 1207. Elsevier. Amster-
dam.

Sandberg, L., P. Stafford, and G. M. Langford. 2000. Effects of
myosin-II antibody on actin-dependent vesicle transport in extracts of
clam oocytes. Biol. Bull. 199: 202-203.

Schimel, D., J. Melillo, H. Tian, A. D. McGuire, D. Kicklighter, T.
Kittel, N. Rosenbloom, S. Running, P. Thorton, D. Ojima, W.
Parton, R. Kelly, M. Sykes, R. Neilson, and B. Rizzo. 2000. Con-
tribution of increasing COi and climate to carbon storage by ecosys-
tems in the United States. Science 287: 2004-2006.

Schmidt, I. K., L. Ruess, E. Baath, A. Michelsen, F. Ekelund, and S.
Jonasson. 2000. Long-term manipulation of the microbes and micro-
fauna of two subarctic heaths by addition of fungicide, bactericide,
carbon and fertilizer. Soil Biol. Biochem. 32: 707-720.

Schmolesky, M. T., Y-C. Wang, D. J. Creel, and A. G. Leventhal. 2000.
Abnormal retinotopic organization of the dorsal lateral geniculate nu-
cleus of the tyrosinase negative albino cat. J. Comp. Neural. 427:
209-219.

Schmolesky, M. T., Y-C. Wang, M. Pu, and A. G. Leventhal. 2000.
Degradation of stimulus selectivity of visual cortical cells in senescent
rhesus monkeys. Nat. Neurosci. 3(4): 384-390.

Schnackenberg, B. J., D. R. Hull, R. D. Balczon, and R. E. Palazzo.
2000. Reconstitution of microtubule nucleation potential in centro-
somes isolated from Sriisula so/iilissima oocytes. J. Cell Sci. 113:
943-953.

Serikawa, K. A., D. M. Porterfield, P. J. S. Smith, and D. F. Mandoli.
2000. Calcification and measurements of net proton and oxygen flux
reveal subcellular domains in Acetahularia acctahtilmn. Plantu 211(4):
474-483.

Shashar, N., R. Hagan, J. G. Boal, and R. T. Hanlon. 2000. Cuttlefish

Publications R95

use polarization sensitivity in predation on silvery tish. Vision AY.v
40(11: 71-75.

Shaver, G. R., J. Canadell, F. S. Chapin, III, J. Gurevitch, J. Harte, G.
Henry, P. Ineson, S. Jonasson, J. Melillo, L. Pitelka. and L. Rustad.
200(1. Global warming and terrestrial ecosystems: a conceptual
framework for analysis. BioScience 50: 871-882.

Shinummra, ().. and K. Teranishi. 2000. Light-emitters involved in the
luminescence of coelenterazine. Luminescence 15: 51-58.

Simenstad, C., S. Brandt, A. Chalmers, R. Dame. L. Deegan, R.
Hodson. and K. Houde. 2000. Habitat-biotic interactions. Pp. 427-
455 in Estmirine Science: .4 Synthetic Approach to Research unit
Practice, J. Hohbie. ed. Island Press. Washington. D.C.

Stafford, P., J. Brown, and G. M. Langford. 2000. Interaction of actin-
and microtubule-based motors in squid axoplasm probed with antibod-
ies to myosin V and kinesin. Biol. Bull. 199: 203-205.

Stieglitz, M., A. Giblin, J. Hobbie, M. Williams, and G. Kling. 2000.
Simulating the effects of climate change and climate variability on
carbon dynamics in Arctic tundra. Global Biogeochem. Cycles 14(Part
4): 1123-1136.

Suddith, A. W., E. A. Vaisberg, S. A. Kuznetsov, W. Steffen, C. L.
Rieder, and R. E. Palazzo. 2000. Centriole duplication, centrosome
maturation and spindle assembly in lysates of Spisula solidissima
oocytes. Pp. 215-228 in Methods in Molecular Biology, Vol. 161:
Cytoskeleton Methods ami Protocols, R. Gavin, ed. Humana Press.
Totowa, NJ.

Swarnakar. S., R. Asokan, J. P. Quigley, and P. B. Armstrong. 2000.
Binding of alpha 2-macroglobulin and limulin: Regulation of the
plasma hemolytic system of the American horseshoe crab, Liinulus.
Biochem. J. 347: 679-685.

Tank, .1. L., J. L. Meyer, D. M. Sanzone, P. J. Mulholland, J. R.
Webster, B. J. Peterson, W. M. Wollhcim, and N. E. Leonard. 2000.
Analysis of nitrogen cycling in a forest stream during autumn using a
l5N-tracer addition. Limnol. Oceanogr. 45(5): 1013-1029.

Tian, H., J. M. Melillo, I). W. Kicklighter, A. D. McGuire, J. Helfrich,
III, B. Moore, III, and C. J. VSrosmarty. 2000. Climatic and biotic
controls on annual carbon storage in Amazonian ecosystems. Global
Ecol. Biogeography 9: 315-336.

Trimarchi, J. R., L. Liu, D. M. Porterfield. P. J. S. Smith, and D. L.
Keefe. 2000. A non-invasive method for measuring pre-implantutiun
embryo physiology. Zvgute 8: 15-24.

Trimarchi, J. R., L. Liu. D. M. Porterfield, P. J. S. Smith, and D. L.
Keefe. 2000. Oxidative phosphorylation-dependent and -independent
oxygen consumption by individual preimplantation mouse embryos.
Biol. Rcproil. 62: 1866-1874.

Trimarchi, J. R., L. Liu, P. J. S. Smith, and D. L. Keefe. 2000.
Non-invasive measurement of potassium efflux as an early indicator of
cell death in mouse embryos. Biol. Reprod. 63: 851-857.

Vallino, J. J. 2000. Improving marine ecosystem models: Use of data
assimilation and mesocosm experiments. J. Mar. Res. 58: 117-164.

Vallino, J. J., and G. Stephanopoulos. 2000. Metabolic flux distribu-
tions in Corynehcicleriiini glutamicum during growth and lysine over-
production. Biotechnol. Bioeng. 67: 872-885.

Vinade, L., and A. Dosemeci. 2000. Regulation of the phosphorylation
state of the AMPA receptor GluRl subunit in the postsynaptic density.
Cell Mol. Neurobiol. 20: 451-463.

Vincent, W. F., and J. E. Hobbie. 2000. Ecology of Arctic lakes and
rivers. Pp. 197-232 in The Arctic: Environment, People, Policies, M.
Nuttall and T. V. Callaghan, eds. Harwood Academic Publishers,
Amsterdam.

Vorb'smarty, C. J., and B. J. Peterson. 2000. Macro-scale models of
water and nutrient flux to the coastal zone. Pp. 43-79 in Estiiarine
Science: A Synthetic Approach to Research and Practice. J. Hobbie, ed.
Island Press. Washington. D.C.

Wachowiak, M., M. Zochowski, L. B. Cohen, and C. X. Falk. 2000.
The spatial representation of odors by olfactory receptor neuron input
to the olfactory bulb is concentration invariant. Biol. Bull. 199: 162-
163.

Wernegreen. J. J., and N. A. Moran. 2000. Decay of mutualistic
potential in aphid endosymbionts through silencing of biosymhetic
loci: Buchnera of Diuraphis. Proc. R. Soc. Land. B 267: 1423-1431.

Wernegreen, J. J.. H. Ochman, I. B. Jones, and N. A. Moran. 2000.
The decoupling of genome size and sequence divergence in a symbiotic
bacterium. J. Bacterial. 182: 3867-3869.

White, T. W., H. Ripps, M. Srinivas, and R. Bruzzone. 2000. Voltage
gating properties of channels formed by a skate retinal connexin. Biol.
Bull. 199: 165-168.

Williams, M., W. Eugster. E. B. Rastetter, J. P. McFadden, and F. S.
Chapin, III. 2000. The controls on net ecosystem productivity along
an arctic transect: A model comparison with flux measurements. Global
Change Biol. 6(Suppl. I): 1 16-126.

Wollmuth. L. P., T. Kuner, C. Jatzke, P. H. Seeburg, N. Heintz, and J.
Zuo. 2000. The Lurcher mutation identifies 82 as an AMPA/kamate
receptor-like channel that is potentiated by Ca2*. J. Neurosci. 20:
5973-5980.

Wu, G.. A. G. McArthur, A. Fiser, A. Sali, M. L. Sogin, and M. Muller.
2000. Core histones of the amitochondriate protist Giardia lamblia.
Mol. Biol. Evol. 17: I 156-1 163.

Zochowski, M., L. B. Cohen, G. Fuhrmann, and D. Kleinfeld. 2000.
Distributed and partially separate pools of neurons are correlated with
two different components of the gill withdrawal reflex in Aplvsia.
J. Neurosci. 20: 8485-8492.

Zottoli, S. J., and D. S. Faber. 2000. The Mauthner cell: What has it
taught us? The Neiiroscientist 6: 25-37.

ARINE IXESOURCES CENTER

MARINE BIOLOGICAL LABORATORY • WOODS HOLE. MA 02543 • (508)289-7700
WWW.MBL.EDU/SERVICES/MRC/INDEX.HTML

Animal and Tissue Supply for
Education & Research

• 150 aquatic species available for shipment via
online catalog: <http://www.mbl.edu/animals/
index.htmb; phone: (508)289-7375; or
e-mail: specimens@mbl.edu

• zebrafish colony containing limited mutant strains

• custom dissection and furnishing of specific organ
and tissue samples

zebrafish facilities

MRC Services Available

• basic water Quality analysis

• veterinary services (clinical, histopathologic,
microbial services, health certificates, etc.)

• aquatic systems design (mechanical, biological,
engineering, etc.)

• educational tours and collecting trips aboard
the RA/ Gemma

Using the MRC for Your Research

• capability for advanced animal husbandry (temperature, light control, etc.)

• availability of_year-round, developmental life stages

• adaptability of tank system design for live marine animal experimentation

The Ultimate
Oecon volution

Machine.

Original Widefieltl Image.

When image quality is essential and capturing lost
information is a must, Carl Zeiss 3D Deconvolution
system is the answer.

Our fast, intuitive software allows reassignment
of non-focal haze, noise and blur to provide
image restoration to its brilliant best. Choose
between theoretically calculated or measured PSF,
iterative or non-iterative method, and the Zeiss
3D Deconvolution will do the rest.

Iterative DCV (measured PSF):

Maximum Likelihood: 15 CG Iterations: Regularized.

Whether conventional widefield or confocal
fluorescence imaging, your own 3D deconvolution
system from Zeiss takes you to new heights of
image fidelity.

Let us show you how!

For immediate demonstration, call 800-233-2343.

Carl Zeiss, Inc.

Microscopy & Imaging Systems

One Zeiss Drive
"rhornwood, NY 10594

800.233.2343
Fax 914.681.7446
micro@zeiss.com
www.zeiss.com/micro

ZEISS

Volume 201

Number 2

LETT

Pud»hed by the Marjjje Biological Isfcoratory

THE BIOLOGICAL BULLETIN

ONLINE

The Marine Biological Laboratory is

pleased to announce that the full text of

The Biological Bulletin is available online at

http://www.biolbull.org

The Biological Bulletin publishes outstanding
experimental research on the full range
of biological topics and organisms, from the
fields of Neurobiology, Behavior, Physiology,
Ecology, Evolution, Development and
Reproduction, Cell Biology, Biomechanics,
Symbiosis, and Systematics.

Published since 1897 by the Marine
Biological Laboratory (MBL) in Woods Hole,
Massachusetts, The Biological Bulletin is one
of America's oldest peer-reviewed scientific
journals.

The journal is aimed at a general readership,
and especially invites articles about those
novel phenomena and contexts characteristic
of intersecting fields.

The Biological Bulletin Online contains the
full content of each issue of the journal,
including all figures and tables, beginning
with the February 2001 issue (Volume 200,
Number 1). The full text is searchable by
keyword, and the cited references include
hyperlinks to Medline. PDF files are available
beginning in February 2000 (Volume 198,

Number 1), some abstracts are available
beginning with the October 1976 issue
(Volume 151, Number 2), and some Tables of
Contents are online beginning with the
October 1965 issue (Volume 129, Number 2).

Each issue will be placed online
approximately on the date it is mailed to
subscribers; therefore the online site will be
available prior to receipt of your paper copy.
Online readers may want to sign up for the
eTOC (electronic Table of Contents) service,
which will deliver each new issue's table of
contents via e-mail. The web site also
provides access to information about the
journal (such as Instructions to Authors, the
Editorial Board, and subscription
information), as well as access to the Marine
Biological Laboratory's web site and other
publications.

There is currently a free trial period for
access to The Biological Bulletin. Once the free
trial period ends on January 31, 2002,
individuals and institutions who are
subscribers to the journal in print or are
members of the Marine Biological
Laboratory Corporation will be able to
activate an online subscription. All other
access (e.g., to Abstracts, eTOCs, searching,
Instructions to Authors) will remain freely
available. Online access will be included in
the print subscription price.

http :// www.biolbull.org

Made to my exact
specifications.

Here's how the BX2's modular design came
through for me. First, the 8 position universal
condenser offers the flexibility to choose
from brightfield, darkfield and phase. Next,
it's assortment of DIC prisms makes it possi-
ble to match the optical image shear to the
specimen, achieving the optimal balance of
contrast and resolution. Finally, the motor-
ized nosepiece, Z-drive, condenser, illuminator
and filter wheels are fully integrated
through the user-friendly software package.
So digital images can now be acquired,
processed and analyzed faster than before.
Now let's move on.

And yours.

Picture yourself sitting here, looking into your
Olympus BX2 research microscope, your
fluorescence requirements having been met.
Specifically: The aspherical collector lens
produces a fluorescence intensity that's
twice as bright as others. The unique excita-
tion balancers improve visualization of
multiple labels. The six-position filter turret
makes single and multiband imaging faster
and simpler. And the rectangular field stop,
another Olympus exclusive, protects the
specimen by exposing only the precise area
being imaged. With all this modularity and
flexibility, my BX2 microscope is also your
BX2 microscope.

And)

Now modularity re
Olympus FLUOVIE
laser scanning mici
5 imaging channels
intuitive operation
productivity is grea
the BX2 is the only
a Metal Matrix Con
static and thermal i
use of 3D microsco
time-lapse observat
high-end digital ini£
resulting in a comp
confocal system. It
optimal solution.

Cover

The composite image on the cover shows, in the
background, a scattering of ovoid embryos of the
squid Loligo pealeii; each is encased in a chorionic
membrane (about 25 /nm thick all around; 45-50
/xm at the micropyle, where sperm enter). The size
of these early embryos is 1 .6 X 1 .0 mm, and they
have been developing for about 24 hours since their
fertilization. The mature eggs from which they de-
veloped were fertilized in a petri dish and, shortly
thereafter, the egg cytoplasm streamed toward the
animal pole and formed a clear lenticular cap called
the blastodisc, which underwent meroblastic cleav-
age, as in birds. The blastodisc is clearly visible as
a low. flat projection at the end of the embryo under
the micropyle. Also shown on the cover, in the
foreground, are two mature 21 -day embryos, or
hatchlings, one still in its chorion. In life, the hatch-
lings (in or out of the chorion) would be about the
same size (2 mm); the chorion swells to accommo-
date the growing embryo. [The cover images were
produced by Karen Crawford, St. Mary's College of
Maryland.)

The embryos on the cover are unusual in that they
were cultured /// vitro; that is why they are all
separate and clean. In nature, squid eggs are re-
leased from the female's oviduct in batches of about
180, packaged in elongated, jellylike capsules, or
egg strings. Fertilization and development occur
within the egg string, which is deposited, with those
of other females, in a communal egg mass attached
to a suitable benthic surface. The reproduction, re-
productive behavior, and development of Loligo
pealeii are set out, online, at http://www.mbl.edu/
publications/Loligo/squid.

Embryos within egg strings are readily cultured;
moreover, they can be snipped out of their matrix
periodically and examined, providing a means of
following and describing squid development. But

the development of embryos that are removed from
the egg string soon fails, so the ability to manipulate
an early embryo and then to culture it through to
hatching is precluded. Thus, many methods of ex-
perimental and comparative embryology become
difficult or impossible with squid: e.g., the effect on
later stages of manipulating earlier ones; classical
chemical treatments that perturb axis formation;
isolation of large numbers of specific stages of
embryos for molecular analysis; and even time-
lapse microscopy. The result is that squid embryos,
being difficult to work with, have been neglected.

In the summer of 1984, at the General Scientific
Meetings of the Marine Biological Laboratory,
Karen Crawford (Klein) and Laurinda A. Jaffe de-
scribed a method of fertilizing squid eggs in vitro
and culturing them through organogenesis to cho-
rionated hatchlings. Now, 17 years later and at the
same venue, Crawford shows us that the embryos
can be made to hatch on their own. More important,
she reports (p. 25 1 ) that treatment of fertilized eggs
of Loligo pealeii with colchicine, but not cytocha-
lasin D, interferes with ooplasmic segregation and
blastodisc formation, suggesting that microtubules
participate in these processes — in contrast to the
process as it is known in zebrafish.

This short report is signaling that squid embryogen-
esis is now accessible and may be applicable and
informing to other aspects of physiology currently
being studied with hatchlings or adult animals.
Some of these aspects are represented in this issue:
e.g., neuronal development (J. P. H. Burbach el ui,
p. 252); morphological and functional ontogeny of
squid mantle (J. T. Thompson and W. M. Kier, p.
136; p. 154); polarization patterns in squid and
cuttlefish skin (N. Shashar et al, p. 267): vesicle
transport in giant axon (J. R. Brown et al. [p. 240]
and J. R. Clay and A. M. Kuzirian [p. 243]); and
excitability (J. R. Clay and A. Shrier, p. 186).

THE

BIOLOGICAL BULLETIN

OCTOBER 2001

Editor
Associate Editors

Section Editor
Online Editors

Editorial Board

Editorial Office

MICHAEL J. GREENBERG

Louis E. BURNETT
R. ANDREW CAMERON
CHARLES D. DERBY
MICHAEL LABARBERA

SHINYA INDUE, Imaging and Microscopy

JAMES A. BLAKE, Keys to Marine
Invertebrates of the Woods Hole Region
WILLIAM D. COHEN, Marine Models
Electronic Record and Compendia

PAMELA CLAPP HINKLE
VICTORIA R. GIBSON
CAROL SCHACHINGER
WENDY CHILD

The Whitney Laboratory, University of Florida

Grice Marine Biological Laboratory, College of Charleston
California Institute of Technology
Georgia State University
University of Chicago

Marine Biological Laboratory

ENSR Marine & Coastal Center, Woods Hole

Hunter College, City University of New York

University of California, Davis

Bodega Marine Lab., University of California, Davis

Louisiana State University

Oregon Institute of Marine Biology, Univ. of Oregon

Hopkins Marine Station, Stanford University

Cereon Genomics, Cambridge, Massachusetts

Hiroshima University of Economics, Japan

University of North Carolina Greensboro

University of Southern California

Kewalo Marine Laboratory, University of Hawaii

Institute of Neurobiology, University of Puerto Rico

Tokyo Institute of Technology, Japan

National Institute for Basic Biology, Japan

Marine Biomed. Inst., Univ. of Texas Medical Branch

University of California, Santa Barbara

University of California, Los Angeles

Managing Editor

Staff Editor

Editorial Associate

Subscription & Advertising Secretary

Published by

MARINE BIOLOGICAL LABORATORY
WOODS HOLE, MASSACHUSETTS

OCT 3 0 2001

http://www.biolbull.org

Genomic Research Leaders Choose
Microway8 Scalable Clusters

Eos Biotechnology, Marine Biological Laboratory, Millennium Pharmaceu-
ticals, Mount Sinai Medical School, NIH, Pfizer, and Rockefeller University
All Choose Microway Custom Clusters and Workstations for Reliability,
Superior Technical Support and Great Pricing.

1.4 GHz Dual Athlon, 1.7 GHz Pentium 4,
or 1 GHz Dual Pentium III in 1U or 2U
Clusters

Dual Alpha 833 MHz Clusters and Towers

For maximum price/performance choose our
Alpha 1U 833 MHz, 4 MB DDK Cache CS20,
4U UP2000+ or 4LI 264DP RnggedRack™

Myrinet, Gigabit Ethernet or Dolphin Wulfkit
High Speed Low Latency Interconnects

RAID and Fibre Channel Storage Solutions

Microway1 Screamer™
Dual Alpha UP2000'
833 MHz. 4MB Cache in
RuggedRack™ Chassis
with RRR™ Redundant
Power Supply

Microway has delivered high-performance computing prod-
ucts since 1982, when our pioneering software made it pos-
sible to use an 8(187 in the IBM-PC. In 1987 we created the
world's first PC parallel processing systems. Since then, our
QuadPuter™ architecture has migrated from Transputers to
i86()s and finally to Alphas in 1995. Over the past three years,
we engineered and delivered over 300 clusters that utilized
MPI running on Linux. As a software developer and hard-
ware manufacturer, we know the value of extensive testing
and validation. We are experts at configuring and validating
the low latency interconnects we employ in our clusters.
Oui technii I support is legendary — the systems we sell
arrive at yom- e and WORK! LosAlamos chose Microway to
maintain ami ;-><;rade its 144 node Alpha Avalon Cluster
because of ou; r. nutation. Large clusters we have sold
include 400+ nodes at the University of Wisconsin and 250+
nodes at Rockefeller University.

Microwav

Technology you can count on- m

Microway offers three Athlon/
Pentium enclosures— 1U, 2U and
tower, plus five Alpha configura-
tions— 1U, 3U, 4LI RuggedRack™,
QuadPuter™ and full tower. Our
26-iDP includes two 21264's with
up to -i GB of memory in our cus-
tom 4U RuggedRack, which features
front accessible redundant power
supplies and hard disks. This rugged
configuration was chosen by the
U.S. Navy for onboard use. We also
offer a dual Alpha UP2000+ running
at 833MHz with 2GB of memory.
Our QuadPuter chassis holds 4
Alpha processors and up to 4GB
memory. The 1U CS20 dual Alpha
(at right) is the highest density
computational platform available.

Microway is API-Networks' Top
North American

Channel
Partner.

"Most Powerful, Highest Density
Computational Platform On the Planet"

Microway Scalable 25 Node
50 Processor Cluster Using CS20 Dual
833 MHz Alphas and Myrinet Interconne
Yielding Peak Throughput of 82.5 Gigafloi

'l have ordered numerous Alpha and Intel-based servers and work-
stations from Microway running both Tru64 UNIX and Linux. We
have been very happy with both the performance and great value
of Microway's products. The major UNIX vendors don't come close
to Microway in this regard, and we have also found that Microway
provides better value than other Linux hardware vendors. I have
also used Microway's tech support and was pleased with their
response. We've been using their systems for over a year and
have had only a couple of minor incidents which were dealt with
promptly."

— David Kristofferson, Ph.D., MBA,
Director of Information Systems, Eos Biotechnology, Inc.

Find out why over 75% of Microway's

sales come from repeat customers.

Please call 508-746-7341 for a

technical salesperson who

speaks your language!

Visit us at www.mlcroway.com

Research Park Box 79, Kingston, MA 0236* • 508-7-*6-73-*l • info@microway.coi

CONTENTS

VOLUME 201. No. 2: OCTOBER 2001

RESEARCH NOTE

ECOLOGY AND EVOLUTION

Maier, Ingo, Christian Hertweck, and Wilhelm Boland

Stereochemical specificity of lamoxirene, the sperm-
releasing pheromone in kelp (Laminariales, Phaeo
phyceae) 121

Rondeau, Amelie, and Bernard Sainte-Marie

Variable mate-guarding time and sperm allocation by
male snow crabs (Cluonoecetes opilio) in response to
sexual competition, and their impact on the mating
success of females 204

PHYSIOLOGY AND BIOMECHANICS

BIOGRAPHY

Johnson, Amy S.

Drag, drafting, and mechanical interactions in cano-
pies of the red alga Chninlnt.\ rri.tpus 126

Thompson, Joseph T., and William M. Kier

Ontogenetic changes in fibrous connective tissue or-
ganization in the oval squid, Sepioteuthii leswniana
Lesson, 1830 136

Thompson, Joseph T., and William M. Kier

Ontogenetic changes in mantle kinematics during
escape jet locomotion in the oval squid. Sepioteuthis
lessoniana Lesson, 1830 154

Martinez, Anne-Sophie, Jean-Yves Toullec, Bruce Shillito,

Mireille Charmantier-Daures, and Guy Charmantier
Hydrominera] regulation in the hydrothermal \'ent
crab Bythograea thermydron 167

NEUROBIOLOGY AND BEHAVIOR

Campbell, A. C., S. Coppard. C. D'Abreo, and
R. Tudor-Thomas

Escape and aggregation responses of three echino-

derms to conspecific stimuli 175

Clay, John R., and Alvin Shrier

Action potentials occur spontaneously in squid giant
axons with moderately alkaline intracellular pH ... 186

SYSTEMATICS

Dahlgren, Thomas G., Bertil Akesson, Christoffer Schander,
Kenneth M. Halanych, and Per Sundberg

Molecular phylogeny of the model annelid Ophryotro-

cha . . 193

Zottoli, Steven J.

The origins of The Grass Foundation .

218

SHORT REPORTS FROM THE 2001 GENERAL

SCIENTIFIC MEETINGS OF THE MARINE

BIOLOGICAL LABORATORY

FEATURED REPORT

The Editors

Introduction to the featured report, green fluores-
cent protein: enhanced optical signals from native

crystals 231

Inoue, Shinya, and Makoto Goda

Fluorescence polarization ratio of GFP crystals 231

CELL BIOLOGY

Knudson, Robert A., Shinya Inoue, and Makoto Goda

Centrifuge polarizing microscope with dual speci-
men chambers and injection ports 234

Tran, P. T., and Fred Chang

Transmitted light fluorescence microscopy revisited. . . . 235

Hernandez, R. V., J. M. Garza, M. E. Graves,

J. L. Martinez, Jr., and R. G. LeBaron

The process of reducing G\l long-term potentiation
by the integrin binding peptide, GRGDSP, occurs
within the first few minutes following theta-burst
stimulation 236

Kuhns, William J., Dario Rusciano, Jane Kaltenbach,

Michael Ho, Max Burger, and Xavier Fernandez-Busquets
L'p-regulation of integrins a., j3, in sulfate-starved ma-
rine sponge cells: functional correlates 238

Brown, Jeremiah R., Kyle R. Simonetta, Leslie A. Sandberg,
Phillip Stafford, arid George M. Langford

Recombinant globular tail fragment of myosin-V blocks
vesicle transport in squid nerve cell extracts 240

Wollert, Torsten, Ana S. DePina, Leslie A. Sandberg,

and George M. Langford

Reconstitution of active pseudo-contractile rings and
myosin-II-mediated vesicle transport in extracts of
clam oocytes 241

Clay, John R., and Alan M. Kuzirian

A novel, kinesin-rich preparation derived from squid
giant axons 243

Weidner, Earl

Microsporidian spore/sporoplasm dynactin in Spra-
guea 245

Conrad, Mara L., R. L. Pardy, and Peter B. Armstrong
Response of the blood cell of the American horse-
shoe crab, Limulus pol\phemus, to a lipopolysaccha-
ride-like molecule from the green alga Chlorella. . . . 246

Silver, Robert

LtB4 evokes the calcium signal that initiates nuclear
envelope breakdown through a multi-enzyme net-
work in sand dollar (Echinaracnius parma) cells .... 248

DEVELOPMENTAL BIOLOGY

Crawford, Karen

Ooplasm segregation in the squid embryo, Loligo
pealfii 251

Burbach, J. Peter H., Anita J. C. G. M. Hellemons,

Marco Hoekman, Philip Grant, and Harish C. Pant
The stellate ganglion of the squid Loligo pealeii as a
model for neuronal development: expression of a
POU Class VI homeodomain gene, Rpf-1 252

Link, Brian A.

Evidence for directed mitotic cleavage plane reorien-
tations during retinal development within the ze-
brafish 254

Smith, Ryan, Emma Kavanagh, Hilary G. Morrison,

and Robert M. Gould

Messenger RNAs located in spiny dogfish oligoden-
drocyte processes 255

Hill, Susan D., and Barbara C. Boyer

Phalloidin labeling of developing muscle in embryos

of the polychaete Capid-lla sp. 1 257

Rice, Aaron N., David S. Portnoy, Ingrid M. Kaatz,

and Phillip S. Lobel

Differentiation of pharvngeal muscles on the basis of
enzyme activities in the cichlid Tramitichmmis interme-
dia . 258

NEUROBIOLOGY

Twig, Gilad, Sung-Kwon Jung, Mark A. Messerli,
Peter J. S. Smith, and Orian S. Shirihai

Real-time detection of reactive oxygen intermediates
from single microglial cells 261

Silver, Robert B., Mahlon E. Kriebel, Bruce Keller,
and George D. Pappas

Porocytosis: Quanta! synaptic secretion of neuro-
transmitter at the neuromuscular junction through
arrayed vessicles 263

Chappell, Richard L , and Stephen Redenti

Endogenous zinc as a neuromodulator in vertebrate
retina: evidence from the retinal slice 265

Shashar, Nadav, Douglas Borst, Seth A. Ament,

William M. Saidel, Roxanna M. Smolowitz,

and Roger T. Hanlon

Polarization reflecting iridophores in the arms of the
squid Loligo pealeii 267

Chiao, Chuan-Chin, and Roger T. Hanlon

Cuttlefish cue visually on area — not shape or aspect
ratio — of light objects in the substrate to produce
disruptive body patterns for camouflage 269

Errigo, M., C. McGuiness, S. Meadors, B. Mittmann,

F. Dodge, and R. Barlow

Visually guided behavior of juvenile horseshoe crabs ... 271

Meadors, S., C. McGuiness, F. A. Dodge,

and R. B. Barlow

Growth, visual field, and resolution in the juvenile
Limulus lateral eye 272

Kozlowski, Corinne, Kara Yopak, Rainer Voigt,

and Jelle Atema

An initial study on the effects of signal intermittency
on the odor plume tracking behavior of the Ameri-
can lobster, Homarus americamis 274

Hall, Benjamin, and Kerry Delaney

Cholinergic modulation of odor-evoked oscillations

in the frog olfactory bulb 276

Zottoli, S. J., D. E. W. Arnolds, N. O. Asamoah,

C. Chevez, S. N. Fuller, N. A. Hiza, J. E. Nierman,

and L. A. Taboada

Dye coupling evidence for gap junctions between
supramedullary/ dorsal neurons of the cunner, Tau-
togolabrui adspersiu 277

Kaatz, Ingrid M., and Phillip S. Lobel

A comparison of sounds recorded from a catfish
(Orinocodoras eigenmanni, Doradidae) in an aquarium
and in the field 278

Fay, R. R., and P. L. Edds-Walton

Bimodal units in the torus semicircularis units of the
toadfish (Opsatnit, tan) 280

MARICULTURE

Mensinger, Allen F., Katherine A. Stephenson,
Sarah L. Pollema, Hazel E. Richmond, Nichole Price,
and Roger T. Hanlon

Mariculture of the toadfish Opsanus tau 282

Rieder, Leila E., and Allen F. Mensinger

Strategies for increasing growth of juvenile toadfish. . . . 283
Chikarmane, Hemant M., Alan M. Kuzirian, Ian Carroll,
and Robbin Dengler

Development of genetically tagged bay scallops for

evaluation of seeding programs 285

ECOLOGY A\D Porn .\i/o\ BIOU>I;Y

Williams, Libby, G. Carl Noblitt FV, and
Robert Buchsbaum

The effects of salt marsh having on bcnthic algal
biomass 287

Hinckley, Eve-Lyii S., Christopher Neill, Richard McHorney,

and Ann Lezberg

Dissolved nitrogen dynamics in gronndwater under a
coastal Massachusetts forest 288

Haiixwell, Alyson M., Christopher Neill, Ivan Valiela,

and Kevin D. Kroeger

Small-scale heterogeneity of nitrogen concentrations
in groundwater at the seepage face of Edgartown
Great Pond 290

Novak, Melissa, Mark Lever, and Ivan Valiela
Top down i>5. bottom-up controls of microphytobenthic
standing crop: role of mud snails and nitrogen supply
in the littoral of Waquoit Bay estuaries 292

1 il.i. Laurie, Ruth Herrold Cannichael, Andrea Shriver,
and Ivan Valiela

Stable N isotopic signatures in bay scallop tissue,
feces, and pseudofeces in Cape Code estuaries sub-
ject to different N loads 294

Grady, Sara P., Deborah Rutecki, Ruth Cannichael,

and Ivan Valiela

Age structure of the Pleasant Bay population of Crep-
iduln fomicata: a possible tool for estimating horse-
shoe crab age 296

Kuzirian, Alan M., Eleanor C. S. Terry,

Deanna L. Bechtel. and Patrick I. James

Hydrogen peroxide: an effective treatment for ballast
water . 297

ORAL PRESENTATIONS
Published bv title only. .

300

ANNOUNCEMENT

The Marine Biological Laboratory is pleased to announce that it has entered into an agreement with HighWire Press of Stanford
University to publish The Biological Bulletin electronically. The online journal was launched on 23 August 2001. It will be available
free of charge to subscribers and the general public for six months after that date. Subsequently, subscribers to The Biological
Bulletin will receive both the print and electronic versions of the journal.

We invite you to visit The Biological Bulletin online at
http://www.biolbull.org

THE BIOLOGICAL BULLETIN

THE BIOLOGICAL BULLETIN is published six times a year by the Marine Biological Laboratory. 7 MBL Street,
Woods Hole, Massachusetts 02543.

Subscriptions and similar matter should be addressed to Subscription Manager, THE BIOLOGICAL BULLETIN,
Marine Biological Laboratory, 7 MBL Street, Woods Hole, Massachusetts 02543. Subscription per year (six
issues, two volumes): $235 for libraries; $95 for individuals. Subscription per volume (three issues): $1 17.50 for
libraries; $47.50 for individuals. Back and single issues (subject to availability): $40 for libraries; $20 for
individuals.

Communications relative to manuscripts should be sent to Michael J. Greenberg, Editor-in-Chief, or Pamela
Clapp Hinkle, Managing Editor, at the Marine Biological Laboratory, 7 MBL Street. Woods Hole. Massachusetts
02543. Telephone: (508) 289-7428. FAX: 508-289-7922. E-mail: pclapp@mbl.edu.

http://www.biobull.org

THE BIOLOGICAL BULLETIN is indexed in bibliographic services including Index Medicus and MED-
LINE, Chemical Abstracts, Current Contents, Elsevier BIOBASE/Current Awareness in Biological
Sciences, and Geo Abstracts.

Printed on acid free paper,
effective with Volume 180, Issue 1, 1991.

POSTMASTER: Send address changes to THE BIOLOGICAL BULLETIN, Marine Biological Laboratory,

7 MBL Street, Woods Hole, MA 02543.

Periodicals postage paid at Woods Hole, MA, and additional mailing offices.

ISSN 0006-3185

INSTRUCTIONS TO AUTHORS

Very short, especially topical papers (less than 9 manuscript
pages including tables, figures, and bibliography) will be published
in a separate section entitled "Research Notes." A Research Note
in The Biological Bulletin follows the format of similar notes in
Nature. It should open with a summary paragraph of 150 to 200
words comprising the introduction and the conclusions. The rest of
the text should continue on without subheadings, and there should
be no more than 30 references. References should be referred to in
the text by number, and listed in the Literature Cited section in the
order that they appear in the text. Unlike references in Nature,
references in the Research Notes section should conform in
punctuation and arrangement to the style of recent issues of The
Biological Bulletin. Materials and Methods should be incorpo-
rated into appropriate figure legends. See the article by Loh-
mann et al. (October 1990, Vol. 179: 214-218) for sample
style. A Research Note will usually appear within two months
after its acceptance.

The Editorial Board requests that regular manuscripts con-
form to the requirements set below; those manuscripts that do
not conform will be returned to authors for correction before
review.

1. Manuscripts. Manuscripts, including figures, should be
submitted in quadruplicate, with the originals clearly marked.
(Xerox copies of photographs are not acceptable for review pur-
poses.) The submission letter accompanying the manuscript should
include a telephone number, a FAX number, and (if possible) an
E-mail address for the corresponding author. The original manu-
script must be typed in no smaller than 12 pitch or 10 point, using
double spacing (including figure legends, footnotes, bibliography,
etc.) on one side of 16- or 20-lb. bond paper. 8 by 1 1 inches.
Please, no right justification. Manuscripts should be proofread
carefully and errors corrected legibly in black ink. Pages should be
numbered consecutively. Margins on all sides should be at least 1
inch (2.5 cm). Manuscripts should conform to the Council of
Biology Editors Style Manual, 5th Edition (Council of Biology
Editors, 1983) and to American spelling. Unusual abbreviations
should be kept to a minimum and should be spelled out on first
reference as well as defined in a footnote on the title page.
Manuscripts should be divided into the following components:
Title page. Abstract (of no more than 200 words). Introduction,
Materials and Methods, Results, Discussion, Acknowledgments,
Literature Cited, Tables, and Figure Legends. In addition, authors
should supply a list of words and phrases under which the article
should be indexed.

2. Title page. The title page consists of a condensed title or
running head of no more than 35 letters and spaces, the manuscript
title, authors' names and appropriate addresses, and footnotes

listing present addresses, acknowledgments or contribution num-
bers, and explanation of unusual abbreviations.

3. Figures. The dimensions of the printed page. 7 by 9
inches, should be kept in mind in preparing figures for publication.
We recommend that figures be about 1 limes the linear dimensions
of the final printing desired, and that the ratio of the largest to the
smallest letter or number and of the thickest to the thinnest line not
exceed 1:1.5. Explanatory matter generally should be included in
legends, although axes should always be identified on the illustra-
tion itself. Figures should be prepared for reproduction as either
line cuts or halftones. Figures to be reproduced as line cuts should
be unmounted glossy photographic reproductions or drawn in
black ink on white paper, good-quality tracing cloth or plastic, or
blue-lined coordinate paper. Those to be reproduced as halftones
should be mounted on board, with both designating numbers or
letters and scale bars affixed directly to the figures. All figures
should be numbered in consecutive order, with no distinction
between text and plate figures and cited, in order, in the text. The
author's name and an arrow indicating orientation should appear
on the reverse side of all figures.

Digital an: The Biological Bulletin will accept figures sub-
mitted in electronic form; however, digital art must conform to the
following guidelines. Authors who create digital images are
wholly responsible for the quality of their material, including color
and halftone accuracy.

Format. Acceptable graphic formats are TIFF and EPS. Color
submissions must be in EPS format, saved in CMKY mode.

Resolution. The minimum requirements for resolution are
1200 DPI for line art and 300 for halftones.

Size. All digital artwork must be submitted at its actual
printed size so that no scaling is necessary.

Multipanel figures. Figures consisting of individual parts
(e.g., panels A, B, C) must be assembled into final format and
submitted as one file.

Hard copy. Files must be accompanied by hard copy for use
in case the electronic version is unusable.

Disk identification. Disks must be clearly labeled with the
following information: author name and manuscript number; for-
mat (PC or Macintosh); name and version of software used.

Color: The Biological Bulletin will publish color figures and
plates, but must bill authors for the actual additional cost of
printing in color. The process is expensive, so authors with more
than one color image should — consistent with editorial concerns,
especially citation of figures in order — combine them into a single
plate to reduce the expense. On request, when supplied with a copy
of a color illustration, the editorial staff will provide a pre-publi-
cation estimate of the printing cost.

4. Tables, footnotes, figure legends, etc. Authors should
follow the style in a recent issue of The Biological Bulletin in

preparing table headings, figure legends, and the like. Because of
the high cost of setting tabular material in type, authors are asked
to limit such material as much as possible. Tables, with their
headings and footnotes, should be typed on separate sheets, num-
bered with consecutive Roman numerals, and placed after the
Literature Cited. Figure legends should contain enough informa-
tion to make the figure intelligible separate from the text. Legends
should be typed double spaced, with consecutive Arabic numbers,
on a separate sheet at the end of the paper. Footnotes should be
limited to authors' current addresses, acknowledgments or contri-
bution numbers, and explanation of unusual abbreviations. All
such footnotes should appear on the title page. Footnotes are not
normally permitted in the body of the text.

5. Literature cited. In the text, literature should be cited by
the Harvard system, with papers by more than two authors cited as
Jones et al., 1980. Personal communications and material in prep-
aration or in press should be cited in the text only, with author's
initials and institutions, unless the material has been formally
accepted and a volume number can be supplied. The list of
references following the text should be headed Literature Cited,
and must be typed double spaced on separate pages, conforming in
punctuation and arrangement to the style of recent issues of The
Biological Bulletin. Citations should include complete titles and
inclusive pagination. Journal abbreviations should normally follow
those of the U. S. A. Standards Institute (USASI), as adopted by
BIOLOGICAL ABSTRACTS and CHEMICAL ABSTRACTS, with the minor
differences set out below. The most generally useful list of bio-
logical journal titles is that published each year by BIOLOGICAL
ABSTRACTS (BIOSIS List of Serials; the most recent issue). Foreign
authors, and others who are accustomed to using THE WORLD LIST
OF SCIENTIFIC PERIODICALS, may find a booklet published by the
Biological Council of the U.K. (obtainable from the Institute of
Biology, 41 Queen's Gate. London, S.W.7, England, U.K.) useful,
since it sets out the WORLD LIST abbreviations for most biological
journals with notes of the USASI abbreviations where these differ.
CHEMICAL ABSTRACTS publishes quarterly supplements of addi-
tional abbreviations. The following points of reference style for
THE BIOLOGICAL BULLETIN differ from USASI (or modified WORLD
LIST) usage:

A. Journal abbreviations, and book titles, all underlined (for
italics)

B. All components of abbreviations with initial capitals (not
as European usage in WORLD LIST e.g.. J. Cell. Comp. Physiol.
NOT J. cell. comp. Physiol. )

C. All abbreviated components must be followed by a period,
whole word components must not (i.e., J. Cancer Res.)

D. Space between all components (e.g., J. Cell. Comp.
Physiol., not J.Cell.Comp.Physiol.)

E. Unusual words in journal titles should be spelled out in
full, rather than employing new abbreviations invented by the
author. For example, use Kit Visindqfielags Islendinga without
abbreviation.

F. All single word journal titles in full (e.g., Veliger, Ecol-
ogv. Brain).

G. The order of abbreviated components should be the same 6. Reprints, page proofs, and charges. Authors may pur-

as the word order of the complete title (i.e., Proc. and Trans. chase reprints in lots of 100. Forms for placing reprint orders are

placed where they appear, not transposed as in some BIOLOGICAL sent with page proofs. Reprints normally will be delivered about 2

ABSTRACTS listings). to 3 months after the issue date. Authors (or delegates for foreign

authors) will receive page proofs of articles shortly before publi-

H. A few well-known international journals in their preferred cation. They will be charged the current cost of printers' time for

forms rather than WORLD LIST or USASI usage (e.g., Nature, corrections to these (other than corrections of printers' or editors'

Science, Evolution NOT Nature, Land., Science, N.Y.; Evolution. errors). Other than these charges for authors' alterations, The

Lancaster, Pa. ) Biological Bulletin does not have page charges.

Reference: Rial. Riill. 201: 121-125. (October 2001)

Stereochemical Specificity of Lamoxirene,

the Sperm-Releasing Pheromone in Kelp

(Laminariales, Phaeophyceae)

INGO MAIER1 •*. CHRISTIAN HERTWECK2, AND WILHELM BOLAND2

lFachbereich Biologic, Universitat Konstanz, 78457 Konstanz, Germany: and 2Max-Planck-Institute
for Chemical Ecology, Carl-Zeiss-Promenade 10, 07745 Jena, Germany

Sexual reproduction in the large brown seaweeds of the
Laminariales, commonly called kelp, involves signaling
chemicals, or "pheromones. " that induce sperm release
from antheridia and subsequent chemotactic orientation of
sperm towards the luring eggs. Lamoxirene (cis-2-cyclo-
hepta-2' ,5' -dienyl-3-vinyloxirane) has been identified as the
sperm-releasing pheromone in the largest and most ad-
vanced group of the Laminariales, comprising the Lami-
nariaceae, Alariaceae, and Lessoniaceae. Recently, a ste-
reoselective synthesis of lamoxirene has yielded pure
substances for biological studies. Here, we used closed-loop
stripping and chiral gas chromatography to establish which
of the four possible stereoisomers of lamoxirene functions
as the naturallv occurring sperm-releasing pheromone in
Undaria pinnatitida. In addition, the relationship between
absolute configuration and sperm-releasing bioactivity in
Laminaria, Alaria, Undaria, and Macrocystis was clarified
in bioassays with lamoxirene stereoisomers. Our experi-
ments established (l'R,2S,3R)-lamoxirene as the most bio-
active compound in all species tested and as the main
component in egg secretions. Thus no species specificity in
the stereochemistry of the sperm-releasing pheromone of
the Laminariales has yet been found.

Sexuality in the Laminariales is strictly controlled by
environmental cues and is coordinated by chemical interac-
tion. Pheromones secreted by eggs produced on micro-
scopic female gametophytes (Fig. 1) induce the release of
sperm from antheridia on nearby male gametophytes and
subsequently attract the sperm to the eggs (1-4). The spe-
cies in the Laminariales belonging to the Laminariaceae.

Received 18 December 2000; accepted 28 June 2001
* To whom correspondence should be addressed. E-mail: Ingo.Maier@
uni-konstanz.de

Alariaceae, and Lessoniaceae are characterized by the ex-
clusive possession of lamoxirene (r;'.s-2-cyclohepta-2',5'-
dienyl-3-vinyloxirane. Fig. 2) as the sperm-releasing pher-
omone (3, 5, 6). Lamoxirene may exist in four spatial
variations (stereoisomers: enantiomers and diastereomers.
Fig. 2), which may or may not occur in nature or function as
pheromones, respectively. Supernatant taken from mature
female cultures induces sperm release in many, if not all.
interspecific combinations within this group (Maier. unpub-
lished). These observations support the idea that phero-
monal cross-reaction and even competition between kelp
gametes may also occur in the field. In particular, this holds
for sympatric species with overlapping reproductive peri-
ods, as exemplified by the four European Laminaria species
listed in Table 1. However, experiments using culture su-
pernatants or racemic synthetic pheromone do not rule out
the possibility that kelps use specific mixtures of stereoiso-
mers for differentiation of sympatric species (4, 7). Similar
strategies, with different enantiomers or diastereomers used
by different species, or specific mixtures being more active
than single components, are well-known in insects (8). It
was shown that two of the lamoxirenes, compound 1 and its
diastereomer 3 (Fig. 2). are secreted in a ratio of 2.45:1 by
eggs of Laminaria digitata, and preliminary bioassays re-
vealed the highest biological activity in pheromone samples
enriched in lamoxirene 1 (9).

Recently, an effective stereoselective synthesis of lamox-
irene has become available (10), offering for the first time
the chance of quantitative bioassays with pure stereoisomers
of lamoxirene. Unfortunately, bioassay results on lamox-
irene 1 and 3 were exchanged in the previous publication
( 10). In the present paper, we report the results of bioassays
that correct this mistake and broaden the original data on
Laminaria spp. to a number of other genera and families in

121

I. MAIER ET AL.

Table 1

Classification and geographic origin of species used in hioassayx

Figure 1. Female gametophyte of Laminaria hyperborea with several
released eggs secreting pheromone. Scale bar = 50 fiin.

the Laminariales. In addition to bioassays, the abso-
lute configuration of naturally occurring lamoxirene 1 in
Undaria pinnatifida is confirmed by chiral gas chroma-
tography.

Clonal male gametophytes of several species of Lami-
nariales (Table 1) were kept vegetatively in red fluorescent
light (about 15 jimol m~: s ', 16:8 h light/dark cycle) at
10°C with ASM-1 ( 1 1 ) as a culture medium. For the induc-
tion of gametogenesis, they were fragmented using a tissue
homogenizer and transferred into white fluorescent light (30
jamol m~2 s"1, 16 h light cycle) at 10°C. The gametophytes
were used in sperm-release assays 12 days later, when
numerous mature antheridia had been formed.

Lamoxirene stereoisomers 1^4 (Fig. 2) with high config-
urational purity (enantiomeric excess, e.e.) were available in
dimethylsulfoxide (DMSO, puriss. p. a., FLUKA): lamox-

Classification

Origin

Family Lammariaceae

Laminaria Jigituta (Huds.) Lamour.

L. hyinrhorea (Gunn.) Fosl.
L. sacchurina (L.I Lamour.
L ochroleuca Pyl.
Family Alariaceae

Alaria esculenta (L.) Grev.
A. esculenta (L.) Grev.

Umlaria pimwlifida (Harv.) Sunngar
Family Lessoniaceae

Macrocystis pyrifera (L.) C. Ag.

Helgoland. German Bight.

North Sea
Helgoland
Helgoland
Roscoff, Brittany. France

Tjornes, Iceland

St. John's, Newfoundland.

Canada
Hokkaido, Japan

Santa Barbara, California

Figure 2. The four stereoisomers of lamoxirene: 1 = (1R,2'S,3'R);

2 = (1S,2'R.3'S>;3 = (1S,2'S,3'R): 4 = (1R.2'R,3'S).

irene 1: 97% e.e.; lamoxirene 2: 90% e.e.; lamoxirene 3:
95% e.e.; lamoxirene 4: 96% e.e. (10). The solutions were
serially diluted (1 X 10~3 to 1 x 10" " M, 5 X 10~4 M
to 5 X 10"" M, 2 X 10~4 M to 2 X 10"" M) in
DMSO. In addition, an equimolar mixture of lamoxirenes 1
and 3 was prepared and diluted accordingly.

For the stereochemical analysis of natural lamoxirene,
female gametophytes of Undaria pinnatifida from Hok-
kaido, Japan, were cultivated as described above for the
male gametophytes. Upon massive egg release, volatile egg
secretions from the gametophyte suspension were adsorbed
onto a bed of 1.5 ing activated carbon by closed loop
stripping (2, 3), followed by elution with 20 jul diethylether.
Chiral capillary gas chromatography (FS-Lipodex E, Mach-
erey-Nagel, Diiren. Germany; 25 m X 0.32 mm, carrier H2)
revealed a single lamoxirene peak (e.e. > 97%), identified
as lamoxirene 1 by comparison with synthetic samples (12).
To compare the biological activity of the different ste-
reoisomers in the induction of sperm release, a new semi-
quantitative assay was employed. Instead of preparing aque-
ous pheromone solutions by distribution from a water-
immiscible fluorocarbon phase as in the original assay (13),
lamoxirene was directly diluted from stock solutions in
DMSO into seawater. 0.5% of a lamoxirene solution was
added to culture medium in 3-ml or 5-ml glass vials, care-
fully mixed, equilibrated to 12°C, and used immediately.
All experiments were earned out in a climate-controlled
culture room at I2"C. In the bioassay, 100 /xl of a suspen-
sion of male gametophytes were mixed with 100 /xl of an
aqueous pheromone solution in a concavity slide. The final
concentration of DMSO in all assays was 0.25%. DMSO
alone did not induce sperm release, except in Alaria escu-
lenta from Iceland (Table 1 ). This strain was thus excluded
from all further experiments. The reaction to DMSO was
not observed in A. esculenta from Newfoundland, which
was used instead. Release of sperm was observed in a
stereomicroscope at 40X magnification under dark-field

STEREOCHEMISTRY OF LAMOXIRENE

123

Figure 3. A male gumetophyte of Maria escnlenta in the bioassa\
before (a) and with massive sperm release 60 s (b) and 90 s (c) after the
addition of lamoxirene (dark-field illumination, microflash exposure).
Scale bar = 1 mm.

illumination (Fig. 3). At saturating concentrations, release
of an estimated several hundred sperm occurred within 7-15
s. A pheromone solution was considered inactive when no
release occurred within 1 min. At the threshold, several
spermatozoids were reliably released within 1 min. and
stronger release occurred at the next higher concentration
step. Assays were performed at least in triplicate and re-
peated in an independent experimental series. The results of
bioassays on lamoxirene-induced sperm release in Lamina-
riu digitata are shown in Table 2, and threshold concentra-
tions for all species investigated are summarized in Table 3.

The response of the algae to pheromone was very repro-
ducible (Table 2), and threshold concentrations were
sharply defined. They were somewhat higher than those
reported earlier (0.01 nM with racemic lamoxirene in L.
digitata, L. hyperborea, and Macrocystis pyrifera (6); 0.002
nM in L. digitata with lamoxirene 1 (9)). The bioassay used
here is simple and avoids inaccuracies introduced by sol-
vent/water-distribution, but is probably more prone to pher-
omone loss by adsorption and volatilization than the origi-
nal assay. The given thresholds should thus be regarded as
conservative approximations.

In the species tested, lamoxirene 1 was generally the most
active stereoisomer, followed by lamoxirene 3. L. saccha-
rina was the only exception, with compounds 1 and 3 being
equally effective. The relative biological activity of lamox-
irene stereoisomers matches the composition of egg secre-
tions, with lamoxirene 1 being the main or the only stereo-
isomer produced in L. digitata (9) and U. pinnatifida (this
study), respectively. Lamoxirene 3 occurs as a by-product in
L. digitata, while lamoxirene 2 and 4 have not yet been
identified as natural products. Lamoxirene 4 is virtually
inactive, which underlines the central importance of the
spatial orientation of the epoxy group in relation to the
terminal double bond in the side chain and the ring system.
This has already been indicated in earlier studies on receptor
specificity (13). In L. digitata. L. livperborea, and L. och-

Table 2
Results of bioassays on pheromone-induced sperm release in Laminaria digitata

Pheromone concentration (nA/l

Lamoxirene 0.025 0.05 0.1 0.25 0.5

2.5 10

1 0.0.0 +. 0. + +, +. + + + ,+,++, + + + ,+ + +

0.0 (1. +, + +. +. + ++, ++ + + +

2 0. 0. 0 0. 0. 0 0. 0. 0 0.

0 +, +, + ++. + +

. 0 +. +. + + +

3 0. 0. 0 0. 0. 0 +, +, () +. +, +

._ + ++1 ++ + + +. + + +

0, 0. 0 0, +, +

+ +_ +_ ++ + + +

4 0,

0 0.0.0 +. +, +

0 0. +. 0 0. +, +

The thresholds are printed in bold.
-: not tested: 0. +. ++. + + + : no or max. 2. 3-20. 20-100. several hundred sperm released, respecmeh

124

I. MAIER ET AL.
Table 3

Threshold concentrations in seawater for the induction of spermatozoid release by lamoxirene stereoisomers

Threshold concentrations (nM)

Species

Lamoxirene 1

Lamoxirene 2

Lamoxirene 3

Lamoxirene 4

Laminciria digitata

0.05

2.5

0.25

L. hyperborea

0.01

0.1

0.02

2.5

L. saccharina

2.5

>20*

L. ochroleuca

0.02

0.25

Alaria escu/enta

0.1

0.25

Undaria pinnatifda

0.1

>20*

Macrocystis pyrifera

0.1

* Highest test concentration available.

roleuca, the diastereomer mixture of lamoxirenes 1 and 3
had no higher biological activity in sperm release than
lamoxirene 1 alone (Table 4, A compared with B and C). On
the contrary, weak competitive effects indicated by slightly
increased thresholds were observed in L. digitaia and L.
hyperborea (Table 4, A compared with B).

In conclusion, the stereochemical specificity of lamox-
irene action in pheromone-induced sperm release is con-
served among all species tested. This holds not only for the
sympatric European Laminaria and Alaria, but also for
Undaria and Macrocystis from the North Pacific and thus
probably for all species belonging to the Laminariaceae.
Alariaceae, and Lessoniaceae. These families comprise the
monophyletic "core group" of the most advanced Laminari-
ales (14, 15). The origin of their pheromone system and its
stereochemistry reaches back to at least the divergence of
the group from a common ancestor, which dates to between
16 and 40 million years ago based on molecular clock
estimations and various other considerations, including bio-
geography and paleooceanography (3, 14. 16-18).

Species specificity in gamete interaction in the Laminari-
ales is thus not achieved by pheromone specificity in the
induction of sperm release, but must be mediated by sub-
sequent mechanisms. These may include differential sperm
attraction to the egg, which is also under pheromonal con-
Table 4

Threshold concentrations of total lamoxirene (M) in sperm release

Experiment

Laminaria digitata

L- hyperborea
L. iicliialeuca

2.3 x 1(1 "'
1.1 • 10""'
1.1 X 10~'°

1.1 X 10"'"
5.7 X 10""
1.1 • 10~'°

1.2 x 10"'"
2.5 x 10""
1.2 x 10"'"

Experiment A: equimolar combination of lamoxirenes 1 and 3; B: same
concentration of lamoxirene 1, lamoxirene 3 replaced by pure DMSO; C:
lamoxirene 1 only, hut total lamoxirene as in A. The total solvent concen-
tration was identical in all tests.

trol, and gamete surface recognition. In L. digitata, it was
previously shown that different pheromone receptors are
involved in sperm release and chemotaxis, and that des-
marestene (f>-(cis-\ ',3'-butadienyl)-cyclohepta-1.4-diene),
not lamoxirene, is the most potent chemoattractant in this
species (19). The possibility thus exists that a species-
specific diversification of complex egg secretions and
pheromone receptors is operative on the chemoattraction
level.

Literature Cited

1. Maier, I. 1987. Environmental and pheromonal control of sexual
reproduction in Laminaria. Pp. 66-74 in Algal Development — Molec-
ular and Cellular Aspects, W. Wiessner. D. G. Robinson, and R. C.
Starr, eds. Springer- Verlag. Berlin.

2. Maier, I., and D. G. Miiller. 1986. Sexual pheromones in algae.
Biol. Bull. 170: 145-175.

3. Maier, I.. D. G. Miiller, G. Gassmann, W. Boland, and L. Jaenicke.
1987. Sexual pheromones and related egg secretions in Laminariales
(Phaeophyta). Z. Naturforsch. Sect. C 42: 948-954.

4. Maier, I. 1995. Brown algal pheromones. Prog. Phvcoi Res. 11:
51-102.

5. Marner, F.-J., B. Miiller, and L. Jaenicke. 1984. Lamoxirene.
Structural proof of the spermatozoid releasing and attracting phero-
mone of Laminariales. Z. Naturforsch. Sect. C 39: 689-691.

6. Miiller, D. G., I. Maier, and G. Gassmann. 1985. Survey on sexual
pheromone specificity in Laminariales (Phaeophyceae). Phycologia
24: 475-477.

7 Boland. W., U. Flegel, G. Jordt, and D. G. Miiller. 1987. Absolute
configuration and enantiomer composition of hormosirene. Naturwis-
senschaften 74: 448-449.

8. Mori, K. 1997. Pheromones: synthesis and bioactivity. Chem. Com-
imm. 1997: 1153-1158.

9. Maier, I., G. Pohnert, S. Pantke-Bocker, and W. Boland.
1996. Solid-phase microextraction and determination of the absolute
configuration of the Laminaria digitata (Laminariales. Phaeophyceae)
spermatozoid-releasing pheromone. Naturwissenschaften 83: 378-
379.

10. Hertweck, C., and W. Boland. 2000. Tandem reduction-chlo-
roallylboration of esters: asymmetric synthesis of lamoxirene, the
spermatozoid releasing and attracting pheromone of the Laminariales
(Phaeophyceae). J. Org. Chem. 65: 2458-2463.

STEREOCHEMISTRY OF LAMOXIRENE

125

11. Maier, I., and M. Calenberg. 1994. Effect of extracellular Ca2 +
and Ca24 -antagonists on the movement and chemoorientation of male
gametes of Ectocarpus siticulosus (Phaeophyceae). But. Acta 107:
451-460.

12. Hertweck, C. 1999. Funktionalisierte Vinylmirane durch Reduktivc
Allylierung von Eslern: Stereoselektive Synlhesen von Lamoxiren und
Sphingoidbasen. Dissertation, University of Bonn, Germany.

13 Maier, I., D. G. Miiller, C. Schmid, W. Boland, and L. Jaenicke.

1988. Pheromone receptor specificity and threshold concentrations

for spermatozoid release in Laminaria tiigitata. Naturwissenschaften

75: 260-263.
14. Maier, I. 1984. Sc.\milita't hci Braunalgen aus tier Ordnung Lami-

nariales und die Phylogenie der Ordnung. Dissertation. University of

Konstanz, Germany.
1? Boo, S. M., W. J. Lee, H. S. Yoon, A. Kato, and H. Kawai. 1999.

Molecular phylogeny of Laminariales (Phaeophyceae) inferred from
small suhunit ribosomal DNA sequences. Phvcol. Res. 47: 109-114.

16. Estes, J. A., and P. D. Steinberg. 1988. Predation, herbivory. and
kelp evolution. Puleohiology 14: 19-36.

17. Liining, K., and I. torn Dieck. 1990. The distribution and evolution
of the Laminariales: North Pacific-Atlantic relationships. Pp. 187-204
in Evolutionary Biogeography of the Marine Algae of the North
Atlantic. D. J. Garbary. and G. R. South, eds. Springer- Verlag, Berlin.

18. Saunders, G. W., and L. D. Druehl. 1992. Nucleotide sequences of
the small-subunit ribosomal RNA genes from selected Laminariales
(Phaeophyta): implications for kelp evolution. J. Phvcol. 28: 544-549.

19. Maier, I., D. G. Miiller, and W. Boland. 1994. Spermatozoid
chemotaxis in Laminaria Jigitata (Phaeophyceae). III. Pheromone
receptor specificity and threshold concentrations. Z Naturforsch. Sect.
C49: 601-606.

Reference: Biol. Bull. 201: 126-135. (October 20011

Drag, Drafting, and Mechanical Interactions in
Canopies of the Red Alga Chondrus crispus

AMY S. JOHNSON
Department of Biology, Bowdoin College, Brunswick, Maine 04011

Abstract. Dense algal canopies, which are common in the
lower intertidal and shallow subtidal along rocky coastlines,
can alter flow-induced forces in their vicinity. Alteration of
flow-induced forces on algal thalli may ameliorate risk of
dislodgement and will affect important physiological pro-
cesses, such as rates of photosynthesis. This study found
that the force experienced by a thallus of the red alga
Chondrus crispus (Stackhouse) at a given flow speed within
a flow tank depended upon ( 1 ) the density of the canopy
surrounding the thallus, (2) the position of the thallus within
the canopy, and (3) the length of the stipe of the thallus
relative to the height of the canopy. At all flow speeds, a
solitary thallus experienced higher forces than a thallus with
neighbors. A greater than 65% reduction in force occurred
when the thallus drafted in the region of slower velocities
that occurs in the wake region of even a single upstream
neighbor, similar to the way racing bicyclists draft one
behind the other. Mechanical interactions between thalli
were important to forces experienced within canopies. A
thallus on the upstream edge of a canopy experienced 6%
less force than it did when solitary, because the canopy
physically supported it. A thallus in the middle of a canopy
experienced up to 83% less force than a solitary thallus, and
forces decreased with increasing canopy density. Thus, a
bushy morphology that increases drag on a solitary thallus
may function to decrease forces experienced by that thallus
when it is surrounded by a canopy, because that morphology
increases physical support provided by neighbors.

Introduction

Algal canopies dominate space in the intertidal and shal-
low subtidal along rocky coastlines and provide secondary
habitat for encrusting organisms as diverse as bryozoans.

Received 16 January 2001; accepted 18 June 2001.
E-mail: ajuhnsonfe1 bowdoin.edu

hydroids, sponges, and tunicates. Algal canopies alter flow
in their vicinity (Koehl and Alberte, 1988; Eckman et al,
1989) in ways that determine such important phenomena as
algal production and physiology (Taylor and Hay, 1984;
reviewed by Hurd, 2000), the recruitment of algal prop-
agules (Johnson and Brawley, 1998) and invertebrate larvae
(Duggins et al., 1990), the flux of gases and nutrients to the
surface of algal thalli (Koehl and Alberte, 1988), and the
potential for breakage of thalli due to flow-induced forces
(Koehl and Wainwright, 1977; Dudgeon and Johnson, 1992:
Gaylord et al., 1994; Johnson and Koehl, 1994; Blanchette,
1997; Koehl, 1999; Gaylord, 2000).

Flow forces may limit the size of some algae (Carrington,
1990; Gaylord et al.. 1994; Denny, 1999: Gaylord, 2000),
either by dislodgment of entire thalli or by pruning
(Blanchette, 1997; Dudgeon et ai, 1999). However, most
measurements of flow-induced forces on algal thalli have
examined thalli only in isolation from their canopy (Char-
ters et ill.. 1969; Gerard, 1987; Koehl and Alberte, 1988;
Sheath and Hambrook. 1988; Armstrong, 1989; Dudgeon
and Johnson, 1992; Johnson and Koehl, 1994; Gaylord et
ai. 1994; Shaughnessy et al., 1996; but see Carrington,
1990; Holbrook et al., 1991). A surrounding canopy, how-
ever, is likely to mediate flow-induced forces experienced
by constituent thalli. For example, a canopy probably slows
flow within (Koehl and Alberte, 1988; Eckman et al.. 1989),
thus decreasing forces experienced by constituent algal
thalli. Total force on an individual thallus, however, will be
due both to direct fluid dynamic forces, such as drag, and to
forces resulting from mechanical interactions with neigh-
boring thalli (Holbrook et ai. 1991 ). This paper quantifies
the dynamics of the interactions among adjacent thalli due
to flow forces on individual thalli and on a canopy of the red
alga Chondrus crispus.

C. crispus occurs in dense canopies (up to 4 stipes per
cnr; S. Dudgeon, unpubl. data) on intertidal and shallow

126

FLOW-INDUCED FORCKS IN ALGAL CANOPIES

127

Mihtidal rocky shores along the northeast coast of the United
States and Canada. In the Gulf of Maine. C. crixpux can be
found from about 1 m above mean low water to about 15 m
below mean low water (Mathieson and Burns. 1971; Dud-
geon et <;/., 1999). It occurs in distributions that range from
intermittent patches of thalli arising from a single holdfast
to areas where the substratum is covered with a dense,
uniform canopy (as tall as —0.07 m in height when emersed
at low tide: S. Dudgeon. California State University at
Northridge. unpubl. data). A large specimen of C. crixpux
generally consists of a relatively long, narrow stipe topped
by a bushy, bifurcated thallus (Fig. 1 ). Multiple stipes arise

from a persistent encrusting holdfast; thalli seldom occur in
isolation from neighboring thalli.

Flow-induced dislodgment of subtidal C. crixpux thalli
has not been quantified. However, winter dislodgment of
thalli from tall intertidal canopies (>4 cm tall) can be as
great as 30% (Dudgeon and Johnson. 1992); and the sea-
sonal decrease in biomass of the largest thalli in an intertidal
population (which ranged in si/.e from 2.5 cnr to 250 cm2
in planform area) can be as great as 75% (M. Pratt and A.
Johnson, unpubl. data). Because thalli regenerate quickly
from persistent encrusting holdfasts, dislodgment is not
typically a selective death for the genet. New thalli of C.

Figure 1. Long-exposure photographs of Cliiwdnui IT/.V/JH.V taken at an ambient flow speed of 0.10 m s '
showing (a) a solitary thallus and (b) a pair of thalli. where the thallus shown in (a) is in the downstream position.
Flow direction is from left to right. For scale, the distance between stipes was 0.09 m. Only the middle section
of the tank is illuminated by a narrow slit of light. Longer streaks indicate faster components of velocities in the
downstream direction. Streaks at the top of the photograph are uniform in length, indicating treestream flow.
Flow was slowed in the wake of thalli; drag was consequently less on the thallus shown in (a) when it drafted
(b) within the wake of an upstream thallus. The lower drag is reflected by the decrease in bend of the stipe in
(b) relative to (a).

128

A. S. JOHNSON

crispus grow rapidly, outcompeting other species in the
lower rocky intertidal of New England (Lubchenco, 1980;
Dudgeon and Johnson. 1992: Dudgeon et ai, 1999).

In this study I examine how flow-induced forces on algal
thalli depend on both the individual morphology of the thalli
and on the density and morphology of the surrounding
canopy. I specifically determine how flow-induced forces
experienced by a thallus are influenced by ( 1 ) the density of
the canopy, (2) the position of the thallus within the canopy,
and (3) the length of stipe of a thallus relative to the height
of the canopy.

Materials and Methods

Collection and maintenance of algae

Thalli of Chondnts crispus were collected at a shallow
subtidal site at 9 m in depth located 0.2 km northeast of
Canoe Beach. East Point, Nahant. Massachusetts (42° 25 'N:
70° 54' W). Horizontal surfaces in the collection area, which
was protected from extreme wave action, were dominated
by C. crispus. Individual thalli, still attached to the holdfast
at the base of the stipe, were maintained within circulating
seawater tables at 15°C and were used in experiments
within 2 weeks after their removal. Thalli remained healthy
for the duration of experiments.

Quantification of flow and force

The downstream forces exerted by the stipe of a thallus
on the substratum were measured by attaching the end of the
stipe that was originally attached to the holdfast onto a force
beam. That beam protruded downward through a hole in a
flat, horizontally oriented, clear acrylic plastic plate located
0.2 m above the floor of a recirculating seawater flow tank
(two flow tanks were used, each 0.2 m wide by 2 m long,
similar to that described in Vogel and LaBarbera. 1978).
There was freestream flow adjacent to the fronds of the
canopy, as is evident in Figure I . thus indicating that bound-
ary effects from the bottom of the flow tank were negligible.
Forces measured represented those due to drag but not due
to acceleration: this is reasonable, as Gaylord (2000) found
that forces due to acceleration contribute negligibly to
wave-induced forces measured on algal thalli.

Experimental flow speeds were constrained by the max-
imum flow speeds attainable within each flow tank (0.21 m
s~' in the flow tank used for the quantification of the C n
and E of solitary thalli: 0.45 m s ' in the flow tank used for
canopy experiments), but were similar to monthly maxima
measured over the period of a year at the collection site by
an Interocean S4 recording electromagnetic flow meter. The
mean flow speed each month was between 0.023 and 0.042
m s~', and the maximum flow speed each month was
between 0.28 and 0.61 m s~' measured at 0.5 m off the
substratum (K. Sebens, University of Maryland, unpubl.

data). Thalli used in these experiments were from this
subtidal habitat. Maximum flow speeds experienced by in-
tertidal specimens of C. crispus in breaking waves will be
faster.

Experimental flow speeds (U) were calculated from the
measured drag on a flat, circular disk (diameter = 3.62 X
10~2 m) oriented perpendicular to flow, using the standard
empirical drag equation

I
D = - pU2CDS ( I )

where D = drag, p = fluid density, U = flow speed, CD =
coefficient of drag, and S = projected area of the disk. The
disk was attached to a force beam that projected 0.05 m
below the water surface in the working section of the tank.
The drag on the beam alone was subtracted from each
measurement. Disks have a constant coefficient of drag,
Cn = 1.17, over the range of Reynolds numbers used in
this study (Hoerner, 1965): therefore the standard empirical
drag equation (Eqn. 1) applies (Vogel, 1994).

In all treatments described below, total force on a thallus
was quantified as the force that the thallus exerted on the
force beam. Drag accounts for the total force acting on an
individual thallus only when that thallus is not mechanically
interacting with other thalli within the canopy. Therefore, I
call the force exerted on the beam "drag" when there were
no mechanical interactions between thalli, and "total force"
when there were also mechanical interactions between
thalli.

Coefficient of drag

Drag measurements (at U = 0.21 m s~') were used to
calculate the coefficient of drag (C D) for eight solitary thalli
using Eqn. 1, where 5 = planform area of the thallus. The
plan form area of each thallus was measured to the nearest
0.01 cm2 by digitizing the outlines of a photograph of a
thallus that had been pressed flat between two plates of
glass, such that the branches of each thallus did not overlap.
This measurement of planform area is equivalent to the
"planform area" (Carrington. 1990), the "actual planform
area" (Johnson and Koehl. 1994). the "maximal projected
blade area" (Gaylord et ai. 1994), the "total projected blade
area" (Gaylord and Denny, 1997), the "maximum projected
blade area" (Denny et ai, 1997). and the "real area" (Koehl,
2000) quantified by other researchers. For C. crispus. which
has a complex three-dimensional morphology, this mea-
surement of planform area represents the most reliable and
repeatable measure of S. The change in frontal area that
occurs as a function of flow speed is accounted for by
changes in the C D. The eight solitary thalli used for these
measurements ranged in mass from 2.6 to 7.2 g (mean =
4.5 g, SE = 0.6) and in planform area from 0.003 1 to 0.0092
nr (mean = 0.0061 nr, SE = 0.0008).

FLOW-INDUCED FORCES IN ALGAL CANOPIES

129

Reconfiguration in //(in-
Flexible structures such us algae reconfigure in flow as
velocity increases such that their relative drag is reduced at
higher flow speeds. For solitary thalli, a useful measure of
velocity-dependent relative drag reduction is the E-value
(Vogel, 1984). which quantities this relative reduction in
drag (i.e.. the decrease in C „ with increase in velocity).

IT-

(2)

where D = drag at a particular flow speed ( U). A value for
E is determined as the slope of a linear regression of log
(D/U~) versus log U for regions of this graph without
inflection points: KE is the antilog of the intercept of this
line. The magnitude of E is zero for a structure, such as a
rigid sphere, that does not reconfigure in flow. The steeper
the negative slope (i.e., the greater the absolute value of the
negative slope), the greater the relative drag reduction ex-
perienced with an increase in velocity as a consequence of
reconfiguration.

E and KE were determined for the same eight solitary
thalli of C. crispus for which the CD was quantified (de-
scribed above).

Canopy experiments

For all treatments in the canopy experiments, force ex-
erted by one thallus (mass = 7.9 g. planform area = 0.01
m ) on the force beam was determined at flow speeds of
0.09. 0.18, 0.27. 0.36, and 0.45 m s"1. All measurements
were repeated three times (sufficient sampling given the low
variance observed). All statistical comparisons between
treatments, using ANOVA, are for force determined at the

highest experimental flow speed (0.45 m s '). Scheffe
/•"-tests were used for a posteriori comparisons between
treatments.

At all experimental velocities and for all treatments,
forces on the stipe of the experimental thallus were quan-
tified when it was 0.05 m long (i.e., only half the length of
the stipe protruded into flow, which was the same as the
length of the stipes of the rest of the canopy), and 0.10 m
long (i.e., the full length of the stipe protruded into flow,
which was twice the length of the stipes of the rest of the
canopy). Forces on the experimental thallus were quantified
for the following treatments: (1) in isolation (Fig. la); (2) in
the presence of one other thallus (of approximately the same
size and shape as the experimental thallus) located 0.09 m
upstream (Fig. Ib); (3) on the upstream edge, middle, and
downstream edge of a lower density canopy (0.08 thalli per
cm2; Fig. 2): and (4) in the middle of a higher density
canopy (0.16 thalli per cm2).

The lower density experimental canopy, which consisted
of 32 thalli, mimicked the observed maximum density of the
bushy tops of C. crispus in a typical shallow subtidal zone
where they were collected (0.08 thalli per cm2: determined
by counting the bushy tops within 20. 100 cm2, quadrats).
For experimental simplicity, the higher density experimen-
tal canopy, which consisted of 64 thalli, was chosen to
double that of the lower density experimental canopy. That
density is similar to that of large thalli (those with more than
five branches) that occurred in a low intertidal habitat (num-
ber of 5 X 5 cm quadrats = 5; mean density = 0.2 thallus
per cm2, SE = 0.07; S. Dudgeon, unpubl. data).

Canopies were created by fastening individual thalli to a
flat plate and suspending the plate upside down in the flow
tank. Canopy thalli were positioned into regularly spaced.

Figure 2. Sketch from a long-exposure photograph of a low-density canopy of Climitlrus crispus (O.OX thalli
per cm2) at an ambient flow speed of 0. 1 m s ~ ' . Flow direction is from left to right. For scale, the distance from
the stipe at the leading edge to the stipe on the trailing edge was 0.2 m. Streaks between the thalli of the canopy
were shorter, indicating that flow was slowed within the canopy. Forces were less on thalli associated with a
canopy not only because flow was slowed (i.e.. drag was reduced), but also because the canopy provides
mechanical support: thalli were most bent over at the upstream edge of the canopy but were more erect than the
more isolated thalli shown in Figure I.

130

A. S. JOHNSON

staggered arrays by inserting the narrow end of the stipe
through 1-mm holes in the plate and holding the stipes in
place by means of soft modeling clay. Every other row of
thalli was offset from the one before it so that any given
thallus within the canopy was directly downstream of an-
other thallus two rows in front of it. The length of the stipes
of the thalli within the canopy was always 0.05 m from the
surface of the plate.

Results

Tluilli reorient in fio\v

When exposed to flow, a solitary thallus of Chondrus
crispits immediately flopped over close to the substratum
with the stipe reoriented parallel to flow (Fig. la). A thallus
also reoriented when downstream of a single other thallus
but bent over less than when solitary at the same ambient
flow speed (Fig. lb). In contrast, thalli within the canopies
bent over less than did solitary thalli (Fig. 2).

Coefficient of drag and E of ' soli turn thalli

The C n of eight solitary thalli (measured at 0.21 m s~';
range = 0.46 to 0.83. mean = 0.60, SE = 0.046) was
independent of thallus size; linear regression analysis: ( 1 )
Cn by mass (g): F(l 7l --•• 1.8. P == 0.22. (2) C,, by
planform area (m2): F(] 7) = 2.6. P = 0.18.

The E of those eight thalli (range = -0.46 to -0.92,
mean = —0.64, SE = 0.06) was independent of thallus size
(linear regression analysis: F, , 7) = 2.1. P = 0.19 \E by
thallus mass]; F, K7) = 1.2. P = 0.31 [£ by thallus area]).
The magnitude of KE increased with increasing thallus
mass (linear regression analysis: F, , 7) = 6.8. P = 0.04,
r = 0.53):

KE= 0.132 M1"4.
where the units for the coefficient were ka °04 m"

(3)

0.20

~. 0.15

g 0.10-1

0.05

0.00

SOLITARY

TRAILING EDGE

0.0 0.1 0.2 0.3 0.4

FLOW SPEED (ms-1)

0.5

Figure 3. Drag (N) as a function of flow speed (m s ') for an
experimental thallus of ChunJnix ITI.V/JM.V when solitary (circles, solid line).
0.09 m downstream of a single other thallus (squares, long dashed line) and
on the trailing edge of a low-density canopy (triangles, short dashed line).
Bars represent two standard errors about the mean; where not visible, these
bars were smaller than the symbols.

(mean force,,, 45 ms ', = 0.055 N. SE = 0.00087 N) or of an
entire canopy of thalli (mean forcel043 ms-i( = 0.05 1 N.
SE = 0.0014 N). Drag on the thallus at this flow speed was
independent of whether there was only a single thallus or an
entire canopy of thalli upstream (Scheffe F-test). Doubling
the length of the stipe on this thallus when it was located on
the downstream edge of a canopy increased the drag it
experienced by 19% (mean force,,, 45 ms -i, == 0.051 N
[short]: 0.063 N [SE == 0.0026 N. long]); r(4) = 4.3,
P = 0.01 ).

Mechanical interactions between thalli

Total force on the experimental thallus decreased when
the thallus was placed in the middle of an algal canopy

s"cw. By substituting the values for E and K, into Eqn. 2. (ANOVA: F(, S) = 561. P g 0.0001; Fig. 4) and de-
creased more with increasing density of the canopy (Scheffe
F-tests; mean force, a45 ms-., = 0.089 N [SE = 0.0046 N,
low density]: 0.028 N [SE = 0.00054. high density]). Thus,
there was an 83% decrease in total force for this thallus in
the middle of a dense canopy. Surprisingly, total force on
the thallus when surrounded by a low density of neighbor-
ing thalli was greater than when it drafted in the wake of a
single upstream neighbor (Scheffe F-tests; compare Fig. 3
"Pair" with Fig. 4 "Low density"). Doubling the length of
the stipe did not significantly alter the total force the
thallus experienced in the mid-canopy position (mean
force

it can be seen that drag for these thalli can be modeled as:
D = 0.132 MnuU[-b (4)

Drafting behind upstream thalli

The drag on the solitary experimental thallus of C. cris-
pits used in the canopy experiments was 0.16 N (SE =
0.0015 N; measured at a flow speed of 0.45 m s" '; Fig. 3).
Doubling the length of the stipe on this thallus increased
drag by only 6% (mean force(@04?ms i, = 0.16 N (short);
0.17 N (SE = 0.00017 N. long); /,_,, = 3.2. P = 0.03).
The Cn<&(> 45 ms-i, of the thallus used in the canopy exper-
iments was 0.16.

At 0.45 m s , drag on the experimental thallus decreased
by more than 65% (ANOVA: F,2 Sl = 2360. P <S 0.0001;
Fig. 3) whether it was downstream of only a single thallus

(0.45ms ., = 0.089 N [short]; 0.079 N [SE = 0.0071 N,
long]; /,4l = -1.2. P = 0.31).

Forces on the experimental thallus varied with position
in the canopy ( ANOVA(045 ms .,: F(3>11) == 244. P «
0.0001 : Fig. 5), decreasing with increasing distance down-

FLOW-INDUCED FORCES IN ALGAL CANOPIES

131

o
cr
o

0.20

0.15

0.10

0.05

0.00

SOLITARY

HIGH DENSITY

0.0 0.1 0.2 0.3 0.4

FLOW SPEED (ms-1)

0.5

Figure 4. Force (N) as a function of flow speed (m s ') for an
experimental thai Ins of Chondrus crispus when solitary (circles, solid line),
in the middle of a low -density canopy (squares, long dashed line) and in the
middle of a higher density canopy (triangles, short dashed line). Bars
represent two standard errors about the mean; where not visible, these bars
were smaller than the symbols.

stream of the upstream edge (Scheffe F-tests). When the
experimental thullus was placed at the upstream edge of the
low density canopy it experienced only 6% lower total force
than when solitary (Scheffe F-test; mean force, 0 4S ms i, =
0.16 N [solitary]; 0.15 N [SE = 0.0042 N. upstream edge];
Fig. 5). Doubling the length of the stipe of this thallus in this
upstream position increased the total force experienced by
the thallus by 6*7r (mean force(045ms-i, = 0. 15 N [short];
0.16 N [SE = 0.0028 N. long]; /(4l = 3.65, P = 0.02).

1983, 1987; Jackson, 1986; Duggins et «/., 1990), algal
propagules (Johnson and Brawley, 1998), and surfgrass
seeds (Blanchette el ui, 1999). Canopies can also influence
the subsequent growth of both invertebrates (Eckman, 1987;
Eckman and Duggins. 1991) and plants (Holbrook et ui,
1991; Johnson and Brawley. 1998; Koch, 1999); and the
risk of flow-induced dislodgment can be altered by living in
dense conspecific populations as diverse as mussels (Harger
and Landenberger, 1971; Bell and Gosline, 1997) and kelp
(Koehl and Wainwright, 1977).

Experiments presented here show that flow-induced
forces on thalli of the red alga Chondrus crispus must be
considered in the context of interactions with neighboring
thalli. The following discussion first examines how a soli-
tary thallus of C. crispus orients in flow as velocity in-
creases, and then goes on to examine the consequences of
canopies to the reorientation of. and forces experienced by.
a thallus.

Drag in isolation: Ho\\ much do thalli reconfigure in flow?

Drag reduction is the most common mechanism consid-
ered when examining force reduction in flow. The E for
solitary thalli of C. crispus (mean = —0.64) indicates that
flexibility of the thalli resulted in a lower drag than the thalli
would have experienced had they not reconfigured as ve-
locity increased. Although this E is less negative (i.e.,
represents a more shallow slope) than that of many other
species of large macroalgae (e.g.. Sargassum filipenduki:
-1.06 to -1.47, Pentcheff. value given in Vogel, 1984:
Hedophyllum sessile: -0.57 to -1.2, Armstrong. 1989:
Nereocvstis luetkeana: —0.75 to — 1.2, Johnson and Koehl.

Discussion

Understanding the consequences of flow to organisms
entails not only examining their flow-related characteristics
in isolation, but also, where appropriate, in the presence of
surrounding neighbors. In marine environments, interac-
tions among closely spaced neighbors can alter feeding
currents around suspension-feeders such as sea anemones
(Koehl. 1976). sabellid polychaetes (Merz, 1984), bryozo-
anslOkamura. 1988). and phoronids (Johnson. 1990. 1997),
and can influence the productivity of algae (Taylor and Hay,
1984; Holbrook et «/., 1991: Dudgeon et «/.. 1999) and
seagrass (Koch. 1994). Effects on feeding and productivity
occur, in part, because the presence of a canopy can alter
turbulent mixing (reviewed in Worcester. 1995) and slow
flow in seagrass (Fonseca et <//., 1982: Eckman. 1987;
Gambi mi/.. 1990; Worcester, 1995; Koch and Gust, 1999).
kelp (Koehl and Alberte. 1988; Eckman et id.. 1989; Dug-
gins et ai. 1990; Jackson, 1998). and intertidal macroalgae
(this study, see Fig. 2). Alteration of flow within canopies
also influences recruitment of planktonic larvae (Eckman.

0.20

<
O

0.10

0.05

0.00

UPSTREAM
EDGE

MIDDLE

TRAILING EDGE

0.0 0.1 0.2 0.3 0.4

FLOW SPEED (m s'1)

0.5

Figure 5. Force (N) as a function of flow speed (m s ') tor an
experimental thallus of Chumlnn, crn/na when on the upstream edge
(circles, solid line), middle (squares, long dashed line), and trailing edge
(triangles, short dashed line) of a low-density canopy. Bars represent two
standard errors about the mean; where not visible, these bars were smaller
than the symbols.

132

A. S. JOHNSON

1994). it is within the range of that determined for fresh-
water red algae (—0.33 to —1.27; Sheath and Hambrook,
1988), as well as for seven other species of intertidal mac-
roalgae that are more similar in size to C. crispits ( —0.28 to
-0.76, Carrington, 1990).

A small absolute value for a negative E can occur for
thalli that are initially well-streamlined (low Cn over all
flow speeds) such that additional rearrangement of the thal-
lus has little effect on relative drag reduction with increas-
ing flow speed (Armstrong, 1989; Johnson and Koehl,
1994). This is not the case for C. crispus: the coefficient of
drag for C. crispus is relatively high at low flow speeds (this
study: mean C D = 0.60 at 0.21 m s~'; Dudgeon and
Johnson, 1992: mean CD = 0.48 at 0.21 m s" ', n = 33)
even for a small intertidal macroalga (Carrington, 1990).
Thus, drag reduction, either by built-in streamlining (low
CD over all velocities, small absolute value of E) or by
rearrangement into a more streamlined shape (high CD at
low velocities, but large absolute value of E), appears to be
relatively unimportant to C. crispus. Perhaps drag reduction
is a relatively unimportant source of force reduction when
thalli are within a dense canopy of surrounding thalli.

Reduction of forces in canopies: The role of drafting and
mechanical interactions between thalli

The response of the experimental thallus of C. crispus to
flow differed dramatically between the solitary, paired, and
within-canopy treatments. Differences in response were re-
flected in the degree to which thalli reoriented and by the
magnitude of the forces experienced by the stipe in a given
flow. The solitary experimental thallus, which experienced
the greatest reorientation, also experienced the greatest
forces; the presence of even a single upstream neighbor
decreased the reorientation of. as well as the force on. that
thallus. These changes occurred because the downstream
(experimental) thallus was within the area of slowed water
movement in the wake of the upstream thalli (Fig. Ib). I call
this phenomenon "drafting" by analogy to the strategy rac-
ing bicyclists use, whereby a bicyclist rides in the wake of
the bicycle in front. Thus, the higher drag on a downstream
thallus with a longer stipe probably occurred because the
longer stipe placed that thallus into a faster region of the
wake of the upstream thallus.

Since flow speed within the canopy is expected to de-
crease with increasing canopy density (Gambi et a I., 1990),
it is tempting to conclude that the decrease in force expe-
rienced by the thallus within a canopy was also due to a
concomitant decrease in its drag. Just drafting in the wake of
a single upstream neighbor, however, reduced force on the
experimental thallus more than being surrounded by the
lower density canopy. Why might the presence of a sur-
rounding canopy result in a higher force than just a single
neighbor?

The experimental thallus in the middle of a canopy could
not collapse and reorient in the same way as a solitary
thallus (compare Fig. 1 with Fig. 2). It was mechanically
supported by the surrounding canopy, as well as being
physically pushed and pulled by its surrounding neighbors.
Thus, forces within algal canopies are due not only to
hydrodynamic drag on specific individual thalli, but are also
a result of physical interactions within the surrounding
canopy. Upstream thalli were also mechanically supported
by the canopy (Fig. 2), as seen by the reduced force on a
thallus in this position when compared with that on the
solitary thallus.

Similarly. Holbrook et al. ( 1991 ) found that dense stands
of the sea palm Postelsia palmaeformis provided mechani-
cal support for central members, which drooped over when
the surrounding neighbors were removed. In contrast, Koehl
and Wainwright ( 1977) suggested that mechanical entangle-
ments between thalli of the giant kelp Nereocystis luetkeana
increase loads on unbroken stipes in a tangled group of
broken and unbroken thalli, thereby increasing the proba-
bility of breakage of the unbroken stipes within the canopy.
A critical difference between these two species is that P.
palmaeformis resists gravitational forces in air with short,
wide stipes, whereas N. luetkeana resists hydrodynamic
forces in pure tension by means of long, slender stipes. C.
crispus is more similar to N. luetkeana in that the stipes
resist hydrodynamic forces in tension, but is dissimilar in
that downstream individuals of C. crispus can provide me-
chanical support to upstream thalli and in that the smaller
size of C. crispus is likely to make any specific entangle-
ments between thalli easier to untangle and less likely to
promote dislodgement.

My results are in contrast to those of Carrington ( 1990),
who found only minor drag reduction among groups of up
to six thalli of Mastocarpus papillatus, a similar species of
intertidal red alga (both species are in the order Gigartina-
les). There are several reasons for the differences in our
results. Firstly, there were methodological differences be-
tween the studies. The canopies that I mimicked, which
consisted of 32 and 64 thalli, were larger than that of
Carrington (1990). The more extensive canopies used in my
study better mimic those in which C. crispus naturally
occurs. Furthermore, Carrington (1990) measured drag on
the entire group of thalli (not on an individual thallus within
the group) and then divided the total drag for that group by
the sum of the drags measured for each individual thallus.
While this method will give an estimate of drag reduction
experienced by the entire group (e.g.. Vogel, 1989), it will
fail to reveal much about forces experienced by individual
thalli in different positions within the group. The latter is
more important because it is the individual stipes of the
thalli that typically break, not the holdfasts (which can be
shared by multiple thalli).

Secondly, the differences between the results of our stud-

FLOW-INDUCED FORCES IN ALGAL CANOPIES

133

ies could be due to differences in the morphology of the
species we studied. For example, unlike the majority of
intertidul seaweeds, including M. papillcitns, large C. cris-
pus thalli do not lay flattened on the substratum when
emersed during low tides but instead are supported by the
three-dimensional branches of their thalli. Furthermore,
comparisons of Cn between these studies indicate that C.
cris/nis (this study: subtidal C0(02|ms >, = 0.46-0.83;
mean Cl)(<,2\ ms"1) = 0.60: Dudgeon and Johnson [1992]:
intertidal Cn(^2l ms-i, = 0.19-1.1. mean C/)(02I ms ., =
0.48. // == 33; M. Pratt and A. Johnson |unpubl. data):

intertidal C

0(0.55 m-

= 0.1 4-0.91, mean CD

(0.55 m s ')

0.39. /; =: 149) has an overall higher drag morphology
than M. pupillntiis (CD(lms-i) == 0.02-0.27; predicted
Q>«> 21 ms S = 0.28; Carrington, 1990). These results sug-
gest that intertidal M. papillatns is a more streamlined alga
(relatively low CD over all velocities) than C. crispns and
would therefore be less subject to mechanical interlocking
of thalli within a canopy.

A streamlined or streamlining morphology typically re-
duces the drag on macroalgae (Vogel, 1984; Johnson and
Koehl. 1994: Koehl. 1986: Gerard. 1987; Koehl and Al-
berte. 1988; Armstrong, 1989). However, for smaller mac-
roalgae that live in dense canopies, a morphology that
enhances mechanical interactions between thalli (small ab-
solute value for a negative E, high CD) may be more
important than a low-drag morphology to the mediation of
forces ultimately experienced at the stipe. Furthermore,
increases in density of the canopy cause decreases in the
forces experienced, which may be important to dislodg-
ment.

Thus, the density of a canopy of C. crispns as well as the
bushiness and morphology of constituent thalli are impor-
tant ecophysiological variables in the population dynamics
of C. crispns. There is considerable morphological variation
in C. crispns (Chopin and Floc'h, 1992), which is associated
with differences in flow habitat and tidal height (e.g., more
dichotomies per unit length at less exposed, high intertidal
sites; Gutierrez and Fernandez, 1992), and with differences
in water temperature (e.g., faster growth rates and more
branches per unit length produced at higher temperatures;
Kiibler and Dudgeon. 1996). The increased photosynthetic
area associated with greater branching is likely to increase
productivity of those thalli (Kiibler and Dudgeon. 1996). An
increase in photosynthetic area, as well as decreased shad-
ing from neighbors (which might be associated with a more
bushy morphology), could be particularly important to sub-
tidal populations where light is often limiting. Increased
size and more extensive branching will also increase the
drag of individual thalli, but might, via mechanical interac-
tions with adjacent thalli, increase the protection conferred
by canopies.

Do canopies reduce risk of dislodgment'.'

For C. crispns. the presence of a canopy clearly decreases
the forces on, and increases the upright orientation of,
constituent thalli. If such forces were an important source of
thallus loss, one might reasonably conclude that canopies
reduce risk of dislodgment of thalli within the canopy.
However, for C. crispns thalli growing subtidally, the drag
determined on the solitary thallus in this study (0.16 N
measured at 0.45 m s"1) was more than an order of mag-
nitude less than that required to break healthy, undamaged
stipes of C. crispns (breaking force = 3 to 12 N; Dudgeon
and Johnson, 1992). An order of magnitude difference per-
sists even if the standard drag equation (Eqn. 1) is used to
overestimate the force on a stipe at the highest flow speed
measured in the field in the subtidal habitat where C. crispns
was collected for this study (0.3 N; 0.61 m s~ ' ). This result
indicates that subtidal thalli of C. crispns have an environ-
mental stress factor (ESF; calculated as the ratio of breaking
force to the force due to drag) of at least 10. ESF is a safety
factor calculated over a specific time period (e.g., a season)
or a life-history stage rather than over a lifetime; sensn
Johnson and Koehl, 1994. High values of ESF imply rela-
tive safety, whereas low values of ESF imply higher risk of
dislodgement. Thus, only thalli otherwise compromised by
damage are likely to break in this subtidal habitat (see
Biedka et «/.. 1987. and Denny et ai, 1989, for a discussion
of the contribution of cracks to the fracture mechanics of
macroalgae).

In contrast, the largest thalli of intertidal C. crispns from
the summer populations are typically dislodged during fall
and winter storms (Dudgeon and Johnson, 1992: M. Pratt
and A. Johnson, unpubl. data). So, intertidal canopies do not
prevent thallus dislodgment. They probably do. however,
increase the flow speed at which a thallus of a given size can
persist in the intertidal. An example calculation will illus-
trate this point. Although subtidal thalli can sometimes have
longer stipes, larger size, and greater branching than do
intertidal thalli (pers. obs.), thalli similar in size (m2) and
shape (CD) to those used in the present experiments occur
at intertidal sites: the experimental thalli used in the present
study overlap in terms of both Cn (f-test, Pl 4t) = 0.41)
and planform area (f-test. Pl 4ft = 0.79) with the largest
thalli found in two dense intertidal canopies in Maine in the
autumn (M. Pratt and A. Johnson, unpubl. data). Because of
the similarity in size and shape of the subtidal and intertidal
thalli from these two studies, the E-value from this study
(Eqn. 4) can be used to estimate the drag of the intertidal
thalli. and thereby their ESF, at the site-relevant maximum
flow speeds for these intertidal sites in the autumn (M. Pratt
and A. Johnson, unpubl. data). This method might still
overestimate drag (underestimate ESF) if the E of intertidal
thalli were more negative than those of the subtidal thalli
(i.e., the intertidal thalli were more flexible). Counterbal-

134

A. S. JOHNSON

uncing this possibility is that this method of estimation (as
used by Gaylord ft til., 1994; Denny, 1995; Bell. 1999)
tends to underestimate drag (overestimate ESF) because E
tends to get less negative at higher flow speeds as thalli
reach their maximum ability to reconfigure (Bell, 1999).
Even though this method tends to underestimate drag at
higher flow speeds, 83% of the largest thalli found at these
intertidal sites in the autumn are predicted to have a maxi-
mum drag greater than the maximum breaking force for
their stipes. Thus, 83% of thalli have an ESF < 1 (mean

ESF = 0.62 [SE = 0.06], f-test: P

< 0.001 that the

mean ESF is equal to 1; range ESF = 0.1-1.5).

The unexpected presence of these thalli in the autumn
intertidal could result in part from differences in local flow
microhabitat; however, this seems unlikely as the flow mea-
surements were made in the middle of the algal canopy.
Thalli might also persist if their CD values at these high flow
speeds were lower than those predicted from the E-value
used; however, this is also unlikely as the method used
already tends to give a low estimate for the CD (Bell. 1999).
Instead, thalli may persist at higher flows than predicted
from estimates on individual thalli because of the mediation
of those forces by the surrounding canopies.

Canopies matter

Measurements of drag, C ,,. and E of isolated thalli must
be considered in the context of the forces that thalli expe-
rience within canopies. This is because the morphology of
thalli may influence breakage not only because of their
individual drag characteristics, but also because of the way
that morphology influences the forces that they experience
within canopies. Even in the absence of breakage, canopy-
induced changes in forces on thalli are important. The
consequent reorientation and reconfiguration of thalli are
likely to affect important processes, such as rates of photo-
synthesis (Greene and Gerard, 1990: Norton, 1991; Wing
and Patterson. 1993; Kiibler and Raven. 1994) or the prob-
ability of fertilization (Brawley and Johnson. 1992). For
algae that live in canopies, an understanding of the conse-
quences of the interaction of their morphology with flow
requires information not just in isolation, but also within the
canopies they compose.

Acknowledgments

Thanks to T. Joseph Bradley, S. Dudgeon. O. Filers. J.
Gosline. M. Koehl, J. Miles, and D. Ritchie for helpful
discussions and assistance and to B. Lindsay for the drawing
of the canopy. Thanks also to S. Dudizeon. M. Pratt, and K.
Sebens for use of unpublished data and to two anonymous
reviewers.

Literature Cited

Armstrong, S. I>. 1989. The behavior in flow of the morphologically

variable seaweed Hedophyliim sessile (CAG) Setchell. Hydmbiologia

183: 115-122.
Bell, E. C. 1999. Applying flow tank measurements to the surf zone:

predicting dislodginent of the Gigartinaceae. Phycol. Rex. 47: 159-

166.
Bell, E. C., and J. M. Gosline. 1997. Strategies for life in flow: tenacity,

morphometry. and probability of dislodgment of two Mytilus species.

Mm: Ecol. Prog. Sei: 159: 197-208.
Biedka. R. F., J. M. Gosline, and R. E. De Wreede. 1987. Biomechani-

cal analysis of wave-induced mortality in the marine alga Ptei-ygoplioru

ciilifornicu. Mai: Ecol. Prog. Set: 36: 163-170.
Blanchette, C. A. 1997. Size and survival of intertidal plants in response

to wave action: a case study with Fitcus gardneri. Ecnlogv 78: 1563-

1578.
Blanchette. C. A., S. E. Worcester, D. Reed, and S. J. Holbrook. 1999.

Algal morphology, flow, and spatially variable recruitment of surfgrass

Phyllospudix tnrreyi. Mm: Ecol. Prog. Sei: 184: 119-128.
Brawley, S. H., and I,. Johnson. 1992. Gametogenesis. gametes and

zygotes: an ecological perspective on sexual reproduction in the algae.

Bi: Phycol. J. 27: 233-252.
Carrington, E. 1990. Drag and dislodgment of an intertidal macroalga:

consequences of morphological variation in Mastocarpus papillani.\

Kut/mg. J. Exp. Mar. Biol. Ecol. 139: 185-200.
Charters, A. C., M. Neushul, and C. Barilotti. 1969. The functional

morphology of Eisenia arborea. Proc. Int. Seaweed Symp. 6: 89-105.
Chopin, T., and J.-Y. Floc'h. 1992. Eco-physiological and biochemical

study of two of the most contrasting forms of Chondrus crispus

(Rhudophyta, Gigartinales). Mar. Ecol. Prog. Sei: 81: 185-195.
Denny, M. W. 1995. Predicting physical disturbance: mechanistic ap-
proaches to the study of survivorship on wave-swept shores. Ecol.

Monogi: 65: 371-418.
Denny, M. \V. 1999. Are there mechanical limits to size in wave-swept

organisms'' ./. E.\p. Biol. 202: 3463-3467.
Denny. M. W., V. Brown, E. Carrington, G. Kraemer, and G. Miller.

1989. Fracture mechanics and the survival of wave-swept macroal-

gae. J. Exp. Mar. Bio/. Ecol. 127: 21 1-228.
Denny. M. W., B. P. Gaylord, and E. A. Cowen. 1997. Flow and

flexibility. II. The roles of size and shape in determining wave forces on

the hull kelp M'iv<«-v.vm luctkcami. ./. Exp. Biol. 200: 3165-3183.
Dudgeon, S. R., and A. S. Johnson. 1992. Thick versus thin: thallus

morphology and tissue mechanics influence differential drag and dis-

lodgement of two co-dominant seaweeds. J. Exp. Mar. Biol. Ecol. 165:

23-43.
Dudgeon, S. R., R. S. Steneck, I. R. Davison, and R. L. Vadas. 1999.

Coexistence of similar species in a space-limited intertidal zone. Ecol.

Monogi: 69: 331-352,
Duggins, D. O., J. E. Ecknian, and A. T. Sewell. 1990. Ecology of

understory kelp environments. II. Effects of kelps on recruitment of

henthic invertebrates. J. Exp. Mai: Biol. Ecol. 143: 27-45.
Ecknian. J. E. 1983. Hydrodynamic processes affecting benthic recruit-
ment. Linuiol. Oceanogi: 28: 241-257.
Ecknian, J. E. 1987. The role of hydrodynamics in recruitment, growth.

and survival of Aigopeclcn irradiuns (L.I and Anomia simplex

(D'Orbigny) within eelgrass meadows. J. Exp. Mar. Biol. Ecol. 106:

165-191.
Ecknian, J. E., and D. O. Duggins. 1991. Life and death beneath

macrophyte canopies: effects of understory kelps on growth rates and

survival of marine, benthic suspension feeders. Oecologiu 87: 473-

487.
Ecknian, J. E., D. O. Duggins, and A. T. Sewell. 1989. Ecology of

understory kelp environments. I. Effects of kelps on flow and particle

transport near the bottom. J. Exp. Mar. Biol. Ecol. 129: 173-187.

FLOW-INDUCED FORCES IN ALGAL CANOPIES

135

Fonseca, M. S., J. S. Fisher, J. C. Zieman, and G. W. Thayer. 1982.

Influence of the seagrass. Zoxtera marina L.. on currcnl How. F.xinar.
Coast. Shelf Sci. 15: 351-364.

Gamhi, M. C., A. R. M. Nowell, and P. A. Jumars. 19911. Hume
observations on flow dynamics in Zostem marina (eelgrass) beds. Mar.
Eci'l. Prog. Set: 61: 159-169.

Gay lord, B. 2000. Biological implications of surf-zone flow complexity.
l.iniiiol. Oceanoar. 45: 174-188.

Gaylord. B., and M. W. Denny. 1997. Flow and flexibility. I. Effects of
si/e. shape and stiffness in determining wave forces on the stipitate
kelps Eixenia arlwrea and Pterygophora californica. J. Exp. Bio/. 2(1(1:
3141-3164.

Gaylord, B., C. Blanchette, and M. W. Denny. 1994. Mechanical
consequences of size in wave-swept algae. Ecol. Monogr. 64: 287-313.

Gerard, \. A. 1987. Hydrodynamic streamlining of Laniinaria xaccha-
rina Lamoui . in response to mechanical stress. 7. Exp. Mar. Bid/. Ecol.
107: 237-244.

Greene, R. M., and V. A. Gerard. 1990. Effects of high-frequency light
fluctuations on growth and photoacclimation of the red alga Chondrux
crispus. Mar. Bio/. 105: 337-344.

Gutierrez, L. M., and C. Fernandez. 1992. Water motion and morphol-
ogy in C/ioiK/im crixpux (Rhodophyta). ./. Phycol. 28: 156-162.

Harger. J. R. E., and D. E. Landenberger. 1971. The effect of storms
as a density dependent mortality factor on populations of sea mussels.
Veliger 14: 195-201.

Hoerner, S. F. 1965. Fluid Dynamic Drag. Hoerner Fluid Dynamics.
Bricktown. NJ.

Holbrook, N. M., M. W. Denny, and M. A. R. Koehl. 1991. Intertidal
"trees": consequences of aggregation on the mechanical and photosyn-
thetic properties of sea palms Postelsia palmaeformis Ruprecht. ./. Exp.
Mar. Bio/. Ecol. 146: 39-67.

Hurd, C. L. 2000. Water motion, marine macroalgal physiology, and
production. J. Phycol. 36: 453-472.

Jackson, G. A. 1986. Interaction of physical and biological processes in
the settlement of planktonic larvae. Bull. Mar. Sci. 39: 202-212.

Jackson, G. A. 1998. Currents in the high drag environment of a coastal
kelp stand off California. Coin. Shelf Res. 17: 1913-1928.

Johnson. A. S. 1990. Flow around phoronids: consequences of a neigh-
bor to suspension feeders. Limnol. Oceanogr. 35: 1395-1401.

Johnson, A. S. 1997. Flow is genet and ramet blind: consequences of
individual, group and colony morphology on filter feeding and flow.
Proceeding i/l the .S'l/i International Coral Reef Symposium 2: 1093-
1096.

Johnson, A. S., and M. A. R. Koehl. 1994. Maintenance of dynamic
strain similarity and environmental stress factor in different flow hab-
itats: thallus allometry and material properties of a giant kelp. J. Exp.
Bio/. 195: 381-410.

Johnson. L. E., and S. H. Brawley. 1998. Dispersal and recruitment of
a canopy-forming intertidal alga: the relative roles of propagule avail-
ability and post-settlement processes. Oecologia 117: 517-526.

Koch, E. \V. 1994. Hydrodynamics, diffusion-boundary layers and pho-
tosynthesis of the seagrasses Tlialassia te.inulinnm and Cymodocea
nodosa. Mar. Bio/. 118: 767-776.

Koch, E. \V. 1999. Preliminary evidence on the interdependent effect of
currents and porewater geochemistry on Thalassiu testudinum Banks
ex Konig seedlings. At/nat. Hot. 63: 95-102.

Koch, E. W., and G. Gust. 1999. Water flow in tide- and wave-
dominated beds of the seagrass Thalaxsia testudinum. Mar. Ecol. Prog.
Ser. 184: 63-72.

Koehl, M. A. R. 1976. Mechanical design in sea anemones. Pp. 23-3 1 in

Coelenterate Ecolog\ <ni(/ Behavior, G. O. Mackie. ed. Plenum Pub-
lishing. NY.

Koehl. M. A. R. 1986. Seaweeds in moving water: form and mechanical
function. Pp. 603-734 in On llic Economy of Plant Form and Function.
T. J. Givnish. ed. Cambridge University Press, Cambridge.

Koehl. M. A. R. 1999. Ecological biomechanics of benthic organisms:
life history, mechanical design, and temporal patterns of mechanical
stress. J. Exp. Bio/. 202: 3469-3476.

Koehl, M. A. R. 2000. Mechanical design and hydrodynamics of blade-
like algae: Chondracanthus exasperatus. Pp. 295-308 in Proceedingx
of the Third International Plant Biomechanics Conference. H. C. Spatz
and T. Speck, eds. Thieme Verlag. Stuttgart.

Koehl, M. A. R., and R. S. Alberte. 1988. Flow, Happing and photo-
synthesis of Nereocyxtix liictkecina: a functional comparison of undu-
late and flat blade morphologies. Mar. Bio/. 99: 435-444.

Koehl, M. A. R., and S. A. Wainwright. 1977. Mechanical adaptations
of a giant kelp. LimnoL Oceanoxr. 22: 1067-1071.

Kiibler, J. E., and S. R. Dudgeon. 1996. Temperature dependent change
in the complexity of form of Clumdrus crispus fronds. J. Exp. Mar.
Bio/. Ecol. 207: 15-24.

Kiibler, J. E., and J. A. Raven. 1994. Consequences of light limitation
for carbon acquisition in three rhodophytes. M<»'. £c~o/. Prog. Ser. 110:
203-209.

Lubchenco, J. 1980. Algal zonation in the New England rocky intertidal
community: an experimental analysis. Ecology 61: 333-344.

Mathieson, A. C., and R. L. Burns. 1971. Ecological studies of eco-
nomic red algae. 1. Photosynthesis and respiration of Chondrnx cr/xptn
Stackhouse and Gigurtina xtellatu (Stackhouse) Batters. J. Exp. Mar.
Bio/. Ecol. 7: 197-206.

Merz, R. A. 1984. Self-generated versus environmentally produced feed-
ing currents: a comparison for the sabellid polychaete Eudistylia van-
couveri. Bio/. Bull. 167: 200-209.

Norton, T. A. 1991. Conflicting constraints on the form of intertidal
algae. Br. Phycol. J. 26: 203-218.

Okamura, B. 1988. The influence of neighbors on the feeding of an
epifaunal bryozoan. J. Exp. Mar. Bio/. Ecol. 120: 105-123.

Shaughnessy, F. J., R. E. De Wreede, and E. C. Bell. 1996. Conse-
quences of morphology and tissue strength to blade survivorship of two
closely related Rhodophyta species. Mar. Ecol. Prog. Ser. 136: 257-
266.

Sheath, R. G., and J. A. Hambrook. 1988. Mechanical adaptations to
How in freshwater red algae. ./. Phycol. 24: 107-1 1 1.

Taylor, P. R., and M. E. Hay. 1984. Functional morphology of inter-
tidal seaweeds: adaptive significance of aggregate vs. solitary forms.
Mar. Ecol. Prog. Ser. 18: 295-302.

Vogel, S. 1984. Drag and flexibility in sessile organisms. Am. Zoo/. 24:
37-44.

Vogel, S. 1989. Drag and reconfiguration of broad leaves in high winds.
J. Exp. Bounty 40: 941-948.

Vogel, S. 1994. Life in Moving Fluids. Princeton University Press,
Princeton.

Vogel, S., and M. LaBarbera. 1978. Simple flow tanks for research and
teaching. BiaScience 28: 638-643.

Wing, S. R., and M. R. Patterson. 1993. Effects of wave-induced light
flecks in the intertidal /one on the photosynthesis in the macroalgae
Postelsia palmaeformis and Hedophyllum sessile (Phaeophyceae).
Mar. Bio/. 116: 519-525.

Worcester, S. E. 1995. Effects of eelgrass beds on advection and turbu-
lent mixing in low current and low shoot density environments. Mar.
Ecol. Prog. Ser. 126: 223-232.

Reference: Bid/. Bull. 201: 136-153. (October 21101)

Ontogenetic Changes in Fibrous Connective Tissue

Organization in the Oval Squid, Sepioteuthis

lessoniana Lesson, 1830

JOSEPH T. THOMPSON* AND WILLIAM M. KIER

Department of Biology. CB#3280 Coker Hall, University of North Carolina,
Chapel Hill, North Carolina 27599-3280

Abstract. Ontogenetic changes in the organization and
volume fraction of collagenous connective tissues were
examined in the mantle of Sepioteuthis lessoniana, the oval
squid. Outer tunic fiber angle (the angle of a tunic collagen
fiber relative to the long axis of the squid) decreased from
33.5° in newly hatched animals to 17.7° in the largest
animals studied. The arrangement of intramuscular collagen
fiber systems 1 (IM-1) and 2 (IM-2) also changed signifi-
cantly during ontogeny. Because of the oblique trajectory of
the IM- 1 collagen fibers, two fiber angles were needed to
describe their organization: (1) IM-1SAG, the angle of an
IM-1 collagen fiber relative to the squid's long axis when
viewed from a sagittal plane and (2) IM-1TAN, the angle of
an IM- 1 collagen fiber relative to the squid's long axis when
viewed from a plane tangential to the outer curvature of the
mantle. The sagittal component (IM-1SAG) of the IM-1
collagen fiber angle was lowest in hatchling squid (32.7°)
and increased exponentially during growth to 43° in squid
with a dorsal mantle length (DML) of 15 mm. In squid
larger than 15 mm DML, IM-1SAG fiber angle did not
change. The tangential component (IM-1TAN) of IM-1 col-
lagen fiber angle was highest in hatchling squid (39°) and
decreased to 32° in the largest squid examined. IM-2 col-
lagen fiber angle (the angle of an IM-2 collagen fiber
relative to the outer surface of the mantle) was lowest in
hatchling squid (34.6°) and increased exponentially to about
50° in 15-mm DML animals. In squid larger than 15 mm

Received 13 December 2000; accepted 8 May 2001.

* To whom correspondence should be addressed. E-mail:
joethonip@eniail.unc.edu

Abbreviations: DML, dorsal mantle length; IM-1, intramuscular fiber sys-
tem 1: 1M-1SAO, sagittal component of IM-1 fiber angle; IM-1TAN. tangential
component of IM-1 fiber angle; IM-2, intramuscular fiber system 2.

DML, IM-2 fiber angle increased slightly with size. The
volume fraction of collagen in IM-1 and IM-2 increased 68
and 36 times, respectively, during growth. The Ontogenetic
changes in the organization of collagen fibers in the outer
tunic, IM-1, and IM-2 may lead to ontogenetic differences
in the kinematics of mantle movement and in elastic energy
storage during jet locomotion.

Introduction

In the hydrostatic skeletons of soft-bodied invertebrates,
the organization of connective tissue fibers is crucial for
providing structural reinforcement, controlling shape, trans-
mitting stresses, and storing elastic energy (e.g., Harris and
Crofton. 1957; Chapman, 1958; Clark and Cowey, 1958;
Clark, 1964; Wainwright, 1970; Wainwright et ai. 1976;
Wainwright and Koehl, 1976; Koehl, 1977; Gosline and
Shadwick, 1983a). Though not particularly well docu-
mented for invertebrates, the organization of connective
tissue fibers can change substantially during ontogeny (Cas-
sada and Russell, 1975; Cox et at., 1981 ). Such ontogenetic
changes in the arrangement of connective tissue fibers may
alter the functions and properties of the hydrostatic skeleton.
The goal of this study is to examine the functional impli-
cations of ontogenetic changes in connective tissue fiber
organization in a soft-bodied invertebrate.

Squid mantle morphology

Squid are soft-bodied molluscs that combine a hydro-
static skeleton with an uncalcified, chitinous gladius ( =
pen) to provide shape and structural support for the mantle.
The mantle lacks the large, fluid-filled spaces characteristic
of the hydrostatic skeleton of many worms and polyps.
Instead, the muscle fibers and connective tissue fibers of the

136

ONTOGENY OF SQUID MANTLE

137

mantle are packed into a dense, three-dimensional array.
Water contained within the muscle fibers and the connective
tissue fibers themselves serves as the incompressible fluid.
In such a system of structural support, termed a "muscular
hydrostat" by Kier and Smith (1985). the volume of the
mantle remains constant, such that a change in one dimen-
sion must result in a change in at least one of the other
dimensions of the mantle.

The mechanical support for the mantle arises from a
complex, three-dimensional arrangement of muscle fibers,
connective tissue fibers, and the gladius. The muscle fibers
in squid mantle are arranged primarily in two orientations:
circumferentially and radially. Contraction of the circum-
ferential muscles decreases mantle circumference and ex-
pels water from the mantle cavity through the funnel during
the exhalant phase of jet locomotion (Young. 1938). Con-
traction of the radial muscle fibers thins the mantle wall and
increases the mantle circumference, filling the mantle cavity
during the inhalant phase of jet locomotion (Young, 1938).

The fibrous connective tissues of the squid mantle are
arranged into five networks (Fig. 1): the inner and outer
tunics, which sandwich the circumferential and radial mus-
cles, plus three networks of intramuscular collagen fibers
(Ward and Wainwright, 1972; Bone et ai, 1981 ). Intramus-
cular fiber system 1 (IM-1 ) consists of collagen fibers (Cos-
line and Shadwick. 1983b; MacGillivray et ai. 1999) that
originate and insert on the inner and outer tunics (Ward and
Wainwright. 1972). Viewed in sagittal section, the IM-1
collagen fibers are arranged at a low angle (28° in Lolli 1^1111-
cnla brevis) relative to the long axis of the mantle (Ward
and Wainwright, 1972) (Fig. 1 ). In sections tangential to the
surface of the mantle, the collagen fibers in IM-1 are also
arranged at low angles (10° to 15° in Alloteuthis subiilaun
relative to the long axis of the mantle (Bone et ul.. 1981 )
(Fig. 1). Thus, the IM-1 fibers actually follow an oblique
path through the mantle wall, relative to both tangential and
sagittal planes.

Intramuscular fiber system 2 (IM-2) is composed of col-
lagen fibers (MacGillivray et ai, 1999) localized to the
radial muscle bands (Bone et ai, 1981) (Fig. 1). Collagen
fibers in IM-2 originate and insert on the inner and outer
tunics and are arranged at an angle of about 55° to the
mantle surface in Alloteuthis subulata (Bone et ai, 1981).

The final connective tissue fiber system in squid mantle is
intramuscular fiber system 3 (IM-3). Collagen fibers
(MacGillivray et ai, 1999) in IM-3 are arranged parallel to
the circumferential muscle fibers and are not attached to the
tunics (Bone et ai. 1981).

Mantle connective tissue function

The tunics and intramuscular collagen fibers serve im-
portant roles in controlling shape change in the mantle. The
low fiber angles reported for tunic and IM-1 fibers in Lul-

1 k

T
> '

/*. A

1 \
1 \
/ \

V/C

Figure 1. A schematic diagram illustrating mantle organization. The
block of mantle tissue at the bottom left is from the ventral portion of the
squid mantle at the upper left. The section planes are indicated immediately
to the right of the block of tissue. Note that the IM-1 collagen fibers follow
an oblique trajectory through the mantle. Thus, the fibers are seen in both
the sagittal and tangential planes. IM-2 collagen fibers are restricted to
radial muscle bands in the transverse plane, a, IM-2 fiber angle; 8, outer
tunic fiber angle; QSAG- sagittal component of the IM-1 fiber angle; 8TAN,
tangential component of the IM-1 fiber angle; CMB, circumferential mus-
cle band; CMF. circumferential muscle fibers; IT, inner tunic; OT. outer
tunic; RMF. radial muscle fibers; S. skin. INSET. The inset at the top right
of the figure is the polygon used to model the effect of ontogenetic changes
in collagen fiber orientation on mantle kinematics and fiber strain. The
solid gray line denotes an IM-2 collagen fiber, and the dashed gray line
represents a single IM-1 collagen fiber. C, L. and 7" represent the circum-
ferential direction, longitudinal direction, and thickness of the mantle wall,
respectively. The circumference of the model (side C) was varied to
simulate jet locomotion. See Discussion for additional details.

liguncula brevis suggest strongly that the tunics and IM-1
help prevent mantle elongation during contraction of the
circumferential muscles (Ward and Wainwright. 1972).
This putative role is corroborated by Ward's (1972) obser-
vation that mantle length in L. brevis does not change
measurably during jetting, though Packard and Trueman
( 1974) report small (i.e.. <5%) increases in Loligo vulgaris
and Sepia officinalis.

The collagen fibers in IM-1 and IM-2 may resist the
substantial increase in mantle thickness that occurs during
circumferential muscle contraction. In addition, these col-
lagen fibers are thought to store elastic energy during the
exhalant phase of the jet and return that energy to help
restore mantle shape and refill the mantle cavity (Ward and
Wainwright. 1972; Bone et ai. 1981; Gosline et ai. 1983:
Gosline and Shadwick, 1983a, b; Shadwick and Gosline.

138

J. T. THOMPSON AND W. M. KIER

1985; MacGillivray et ui, 1999; Curtin el al.. 2000). The
IM-1 and IM-2 collagen fibers may also help restore mantle
shape during the low-amplitude movements that occur dur-
ing respiration (Gosline et til., 1983).

Specific problem addressed

Virtually all the published work on squid mantle mor-
phology and function is on adult loliginid squid. The few
studies of hatchling or juvenile loliginid squid reveal dra-
matic changes in mantle function during ontogeny. For
example, the morphology and physiology of the mantle
musculature in Photololigo sp. and Loligo opalescens
(Moltschaniwskyj. 1994: Preuss et ul.. 1997) and the neu-
romuscular physiology underlying the escape response in L.
opalescens (Gilly et al., 1991) change significantly during
growth from hatchlings to adults. Importantly for this study.
the range of mantle movement during jet locomotion in
newly hatched L. opalescens and Lolit>o vulgaris is greater
than in adult animals (Packard, 1969: Gilly et ul.. 1991;
Preuss et al.. 1997). Given the link between the mantle
connective tissue arrangement and mantle kinematics, it is
likely that the orientation, the mechanical properties, or both
the orientation and mechanical properties of squid mantle
collagen change during ontogeny. Here, we examine onto-
genetic changes in the arrangement and amount of connec-
tive tissue in the mantle in the oval squid, Sepioteuthis
lessoniana (Cephalopoda: Loliginidae). The effect of onto-
genetic changes in the collagen fiber arrangement of the
outer tunic, IM-1. and IM-2 on mantle kinematics and
elastic energy storage during jet locomotion is also ana-
lyzed.

Materials and Methods

Animals

We obtained an ontogenetic series of Sepioteuthis lessoni-
ana Lesson, 1830. Wild embryos collected from three loca-
tions (Gulf of Thailand; Okinawa Island. Japan: Tokyo
region, East Central Japan) over a 2-year period were reared
(Lee et al.. 1994) by the National Resource Center for
Cephalopods (NRCC) at the University of Texas Medical
Branch (Galveston, TX). Each of the three cohorts consisted
of thousands of embryos from six to eight different egg
mops. Thus, it is likely that the sample populations were not
the offspring of a few closely related individuals, but were
representative of the natural population at each collection
site.

Commencing at hatching, and at weekly intervals there-
after, live squid were sent via overnight express shipping
from the NRCC to the University of North Carolina. The
squid, which ranged from 5 mm to 70 mm in dorsal mantle
length (DML). were killed by over-anesthesia in a solution
of 7.5% MgCK mixed 1:1 with artificial seawater (Messen-

ger et al., 1985). The MgCU solution relaxed the mantle
musculature of nearly all of the squid. Animals in which the
mantle musculature was contracted noticeably were not
used for the histological study. The MgCl-, solution did not
distort the shape of the mantle. The resting mantle diameter
of an anesthetized squid was always 80% to 90% of the
peak mantle diameter measured during jet locomotion in the
same, unanesthetized animal (for details of the kinematics
measurements, see Thompson and Kier, 2001 ).

Histology

The mantle tissue was examined using standard histolog-
ical methods. Immediately after death, the squid were fixed
whole in 10% formalin in seawater for 48 to 96 h at 20° to
23 °C. In the larger animals (>25 mm DML), the animal
was decapitated to permit unrestricted flow of fixative into
the mantle cavity. The mantles were fixed whole, rather than
dissected into smaller blocks of tissue, to help minimize
shape changes (e.g., curling or bending of the tissue block)
that could affect connective tissue fiber angle.

Following fixation, the tissue was dehydrated in a graded
series of ethanol and cleared in Histoclear (National Diag-
nostics. Atlanta. GA) or Hemo-D (Fisher Scientific. Pitts-
burgh, PA). There was no discernible scale-related distor-
tion of the mantle during dehydration and clearing. After
clearing, the mantle was dissected into smaller pieces and
embedded in paraffin (Paraplast Plus. Oxford Labware. St.
Louis. MO; melting point 56 °C). To minimize shrinkage
artifacts, infiltration with molten paraffin was limited to a
total of 90 min (30-min baths X 3 changes) instead of the
180 min (60-min baths X 3 changes) recommended by Kier
(1992).

Following clearing, most of the squid smaller than 30 mm
DML were sliced in half along the sagittal plane using a fine
razor blade. One half of the animal was oriented in the
embedding mold to permit the cutting of sagittal sections;
the other half was oriented for cutting of transverse sections.
Many of the smaller squid were sliced in half along the
frontal plane. The dorsal and ventral halves were oriented in
the embedding molds to allow grazing sections to be cut.
For the squid larger than 30 mm DML. large blocks of tissue
of about 5 mm by 3 mm by the thickness of the mantle were
dissected from several locations along the length and around
the circumference of the mantle. These tissue blocks were
oriented in the embedding molds to permit the cutting of
sagittal, transverse, and tangential sections.

The tissue blocks were sectioned using a rotary mic-
rotome. The sections were mounted on slides coated with
Mayer's albumin and stained with picrosirius stain (Sweat
et al.. 1964: protocol adapted from Lopez-DeLeon and
Rojkind, 1985). Other connective tissue stains were used
successfully (e.g.. Milligan trichrome. picro-ponceau, and
van Gieson's stain) but picrosirius stain provided the best

ONTOGENY OF SQUID MANTLE

139

contrast between the collagenous and non-collagenous com-
ponents of the tissue sections and made identification of
intramuscular collagen fibers straightforward.

Several additional attributes made picrosirius an excellent
choice for this study. First, the sirius red F3B dye molecules
attach with their long axes parallel to the long axes of the
collagen fibrils, enhancing the natural birefringence of col-
lagen fibers (Monies and Junqueira. 1988). Second, picro-
sirius is an outstanding stain for resolving the smallest
collagen fibers and fibrils. The stain has been used previ-
ously to visualize the fine reticular collagen fibers present in
embryonic mammalian skin and organs, the thin type-II
collagen fibrils present in mammalian cartilage, and the
extremely fine type-IV collagen fibrils present in mamma-
lian basal laminae (Montes and Junqueira, 1988). Third,
there is a strong correlation between the collagen volume
fraction estimated from paraffin-embedded human liver sec-
tions using the picrosirius stain and the collagen volume
fraction from the same tissue sections measured by hy-
droxyproline content analysis (Lopez-DeLeon and Rojkind,
1985). Thus, picrosirius stain is ideal for both visualizing
collagen fibers and making precise estimates of collagen
volume fraction.

The stained sections were viewed using brightfield and
polarized light microscopy. Fiber angles were measured
from digital photomicrographic images using image analy-
sis software (SigmaScan Pro, SPSS Science, Chicago. IL).

Initial survey

We made an initial survey of the mantle intramuscular
fiber (IM) networks 1 and 2 in five squid (25 mm to 70 mm
DML) to help develop a protocol for measuring IM fiber
angles. In this survey, IM-1 and IM-2 fiber angles from
different regions along the length and around the circum-
ference of the mantle were examined. IM-1 and IM-2 fiber
angles were measured at four positions along the length of
the mantle (1/10, 1/4, 1/2, and 3/4 DML) and at three
positions around the circumference of the mantle (ventral,
lateral, dorsal). Given the potential for regional differences
in IM fiber angle (see Results for details), all the compari-
sons among the squid were made at the same location: the
ventral portion of the mantle between 1/3 and 2/3 DML.

IM-1 fiber angle measurements

IM-1 collagen fibers are arranged obliquely to the sagittal
plane (Fig. 1). Therefore, to describe accurately the trajec-
tory of these fibers, two fiber angles must be measured. The
first angle, called IM-1SAO here, is the angle of IM-1 col-
lagen fibers relative to the long axis of the mantle in the
sagittal plane (Fig. 1). The second angle, which we call
IM-1TAN, is the IM-1 fiber angle relative to the long axis of
the mantle in a plane tangential to the outer surface of the
mantle and perpendicular to the sagittal plane (Fig. 1).

IM-1 SAG fiber angle measurements

IM-1SAG fiber angles were measured from sagittal sec-
tions (thickness 10-15 /u,m) of the mantle. Criteria were
developed to ensure consistency across all squid in the
ontogenetic series. First, all measurements of fiber angles
were made from the ventral portion of the mantle between
1/3 and 2/3 DML. This eliminated errors due to variation in
fiber angle along the length and around the circumference of
the mantle.

Second, because the apparent fiber angle depends on the
viewer's perspective. IM-1SAG fiber angles were measured
only from tissue sections in which the circumferential mus-
cle fibers of the mantle were cut in nearly perfect cross
section. This restriction ensured that the perspective was
similar for all the squid examined. Adjusting the orientation
of the tissue block relative to the microtome knife made it
possible, through trial and error, to meet this criterion, and
conformance was determined by examining test sections 20
jiun thick.

Third, sagittal tissue sections contained IM-1 fibers of
varying lengths. It was difficult to obtain accurate angle
measurements of the shortest fibers in each section. There-
fore, IM-1SAG fiber angle measurements were made only on
IM-1 fibers longer than the width of one circumferential
muscle band. A circumferential muscle band was defined as
a region of circumferential muscle fibers bounded by radial
muscle fibers (Fig. I ).

Fourth, in all animals larger than about 15 mm DML.
IM-1SAG fiber angles were measured only from crossed
IM-1 fibers. The angle between the two fibers was measured
and the half angle reported as the IM-1SAG fiber angle (Fig.
1 ). In squid smaller than 15 mm DML. IM-1 fibers were so
scarce that there were few instances of crossed fibers. In
these small squid. IM-1SAG fiber angles were measured
relative to the outer or the inner tunic (Fig. 1). In areas
where the tunics were folded due to histological artifact.
IM-1SAG fiber angles were not measured.

Finally, the fiber angle of every IM-1SAG fiber in a given
microscope field that conformed to the criteria was mea-
sured. A minimum of 20 measurements was made from
each squid larger than about 15 mm DML. Because IM-1
fibers were sparse in animals smaller than 15 mm DML. the
minimum number of fiber angle measurements was eight in
these animals.

1M-1TAN fiber angle measurements

IM-1TAN fiber angles were measured from relatively
thick (10 to 15 jLtm) tangential sections of the mantle. To
ensure consistency in fiber angle measurements among all
squid, the criteria listed previously were used with two
exceptions. First, IM-1TAN fiber angles were measured only
in those sections in which the radial muscle fibers were cut
in nearly perfect cross section (determined from 20-/.IP:

140

J. T. THOMPSON AND W. M. KIER

thick test sections). Second, for squid larger than about 15
mm DML, fiber angles were measured only from crossed
IM-1 fibers. In the smallest squid (<15 mm DML), there
were few IM-1 fibers and virtually no crossed fibers. In
these squid. IM-1TAN fiber angles were measured relative to
a band of radial muscle fibers. Subtracting the measured
angle from 90° gave the angle of the IM-1TAN fiber relative
to the long axis of the squid.

IM-2 fiber angle measurements

IM-2 fiber angles were measured from 5 /u,m thick trans-
verse sections of the mantle. As with the IM-1 measure-
ments, IM-2 fiber angles were measured only from the
ventral portion of the mantle between 1/3 and 2/3 DML.
Fiber angles were measured only from sections that were
nearly perfect transverse sections of the mantle. Sections
oblique to the transverse plane show circumferential muscle
fibers in closely spaced bands separated by a few radial
muscle fibers. Nearly perfect transverse sections exhibited
uninterrupted circumferential muscle fibers. IM-2 fiber an-
gle was measured only from crossed fibers (Fig. 1 ). Given
the scarcity of IM-2 fibers in squid smaller than about 15
mm DML, it was not always possible to measure crossed
fibers. In these small squid. IM-2 fiber angle was also
measured relative to nearby radial muscle fibers. Finally, the
fiber angle of every IM-2 fiber in the microscope field was
measured. No fewer than 20 fiber angle measurements were
made from each squid longer than about 15 mm DML. The
relative paucity of IM-2 fibers in squid smaller than 15 mm
DML reduced the minimum number of fiber angle measure-
ments to eight per squid.

Outer tunic fiber angle measurements

Outer tunic fiber angles were measured from 5-^im-thick
grazing sections of the mantle. Tunic fiber angles were
measured only from the ventral portion of the mantle be-
tween 1/3 and 2/3 DML and only from crossed fibers. The
half angle between the crossed tunic fibers, relative to the
long axis of the squid, was reported as the fiber angle (Fig.
1 ). A minimum of 20 fiber angle measurements was made
from each squid.

Stereolog\

Stereological methods were used to estimate the volume
fraction of IM-1 and IM-2 collagen fibers relative to the
volume of the mantle musculature. To obtain an accurate
estimate of the volume fraction of a particular tissue com-
ponent, stereology requires that the tissue of interest be
sectioned in randomly oriented planes (Weibel, 1979). Be-
cause it is difficult to positively identify a collagen fiber in
a random section plane as an IM-1 or IM-2 fiber, it was not
possible to use random section planes. IM-1 collagen fiber

volume fraction was therefore determined from sagittal sec-
tions of the ventral mantle in which fiber identity could be
verified. Likewise, IM-2 collagen fiber volume fraction was
measured from transverse sections of the ventral mantle.
Although this violates an assumption of stereology. it al-
lows accurate comparison of the relative volume fraction of
collagen fibers among squid in the ontogenetic series. How-
ever, this method is inappropriate for estimation of the
absolute volume fraction of collagen fibers in the mantle
(Weibel. 1979).

The procedure for collagen volume fraction determina-
tion was similar for both IM-1 and IM-2. The ventral
portion of the mantle between 1/3 and 2/3 DML was exam-
ined. A slide containing either sagittal (IM-1) or transverse
(IM-2) 10-/Ltm-thick tissue sections was placed on the stage
of a compound microscope, and the tissue positioned under
a 40 X objective lens without observation through the ocu-
lars. The tissue section was brought into focus, and an
image of the section was captured by a digital camera. The
image was expanded to fill the screen of the monitor, and a
transparent plastic overlay with a grid of 24 lines X 24 lines
(Weibel. 1979) was taped to the screen. The intersection of
a collagen fiber in IM-1 (sagittal sections only) or IM-2
(transverse sections only) with the junction of two lines ( =
a point; there were 24 lines X 24 lines = 576 points on the
grid) was counted as a "hit." After the image was sampled,
the microscope stage was moved haphazardly without ob-
serving the image through the microscope. In all cases, the
stage was moved sufficiently far to ensure that the same
portion of the mantle tissue was not examined twice. An-
other digital image was then captured, and the procedure
was repeated at 2 or 3 different locations within the same
tissue section and on between 3 and 10 different tissue
sections per squid. The average volume fraction of collagen
in IM-1 and IM-2 relative to the average volume of the
mantle musculature was calculated by dividing the total
number of hits by the total number of points counted for
each squid (Weibel. 1979).

In stereology. both the acceptable standard error of the
volume fraction estimate and the volume fraction of the
item of interest determine the total number of points that
must be counted (Weibel, 1979). The total number of points
(Pc) was determined by

pc=

- VV/VV) (Weibel. 1979) (1)

where m, is the number of tissue sections examined per
squid, el is the confidence interval. ta is the acceptable error
probability (the chance that the true volume fraction will be
outside the confidence interval), and Vv is the volume
fraction of the item of interest. To determine Pc, the volume
fraction ( Vr) of collagen in both IM-1 and IM-2 was
estimated for four squid of various sizes (5 mm, 15 mm, 27
mm. and 69 mm DML) using the procedure outlined in the

ONTOGENY OF SQUID MANTLE 141

Table 1

Comparison of the relative volume fraction of collagen in IM-I and IM-2 among squid divided into the life-history stages of Segawa (1987}

Life-history stage

IM-1 points counted

IM-I volume traction

IM-2 points counted

IM-2 volume fraction

Hatchling (n = 4)
Juvenile 1 (» = 4)
Juvenile 2 (n = 4)
Young 2 (n = 4)

14,985(14,265)

6.516(616)
4,344 (606)
3,801 (975)

0.00095 ± 0.0002
0.015 ± 0.0036
0.032 ±0.012
0.065 ± 0.036

10.414(4.504)
5.340(647)
3,258 (436)
3,258(746)

0.0027 ± 0.0018
0.027 ± 0.024
0.057 ±0.021
0.097 ± 0.024

The mean volume fraction of collagen is listed in boldface type ± the standard deviation of the mean. The total number of points counted for each
individual squid is listed. The adjacent numbers in parentheses indicate the number of points that need to be counted (=PC, see equation 1 ) to obtain an
error probability of 5% and a confidence interval of ±10%. Within IM-1 and within IM-2. the volume fraction of collagen differed significantly among
all life-history stages (one-way ANOVA on ranks. P < 0.05).

previous paragraph. Using the initial estimate of Vv, an
error probability of 5%, and a confidence interval of ± 10%,
the total number of points to be counted (Pc) was calculated
(Table 1 ). The Vv of collagen was strongly correlated with
squid size. Thus, the total number of points counted per
squid varied with size. Note that the actual number of points
counted per squid was much greater than the minimum
required to obtain an error probability of 5% and a confi-
dence interval of ±10%. Thus, the actual error probability
and confidence interval were smaller than the predicted
values.

Statistical analysis

The sample population used in this study was subdivided
into the life-history stages described by Segawa ( 1987). The
life-history stages were selected as an independent organi-
zation scheme upon which to base the statistical analysis.
Segawa ( 1987) studied the life cycle of S. lessoniuim from
embryo to adult and divided the life cycle into seven stages
on the basis of morphological and ecological characters.
These seven stages are hatchling (up to 10 mm DML).
juvenile 1 (11-25 mm DML), juvenile 2 (26-40 mm
DML). young 1 (41-60 mm DML), young 2 (61-100 mm
DML), subadult (100-150 mm DML), and adult (> 150 mm
DML). The sample population of S. lessoniana used in the
current investigation included the hatchling, juvenile 1,
juvenile 2, and young 2 stages.

Nonparametric statistics were used for most of the anal-
yses because the sample population was not normally dis-
tributed. For comparisons among the life-history stages.
Kruskal-Wallis one-way analysis of variance on ranks was
used with Dunn's method of pairwise multiple comparisons
(Zar, 1996). All statistical analyses were completed using
SigmaStat 1.01 (SPSS Science).

Results

General description of mantle morphology

The mantle of Sepioteuthis lessoniana is similar to that
described for other loliginid squid (Young. 1938; Ward and

Wainwright, 1972; Bone el ai, 1981). The outer tunic is
located underneath the collagen-rich skin. The fibers within
the outer tunic are robust and closely packed. The outer
tunic serves as the insertion for the radial muscle fibers, the
IM-1 collagen fibers, and the IM-2 collagen fibers. The
fairly low-resolution microscopic methods used in this
study did not permit a detailed examination of the connec-
tions between the outer tunic and the IM collagen fibers or
the radial musculature.

The majority of the mantle is composed of circumferen-
tial muscle fibers. These muscle fibers are bordered by the
outer and inner tunics and are partitioned by regularly
spaced bands of radial muscle fibers. Consistent with the
trend for Photololigo sp. (Moltschaniwskyj, 1994). the cir-
cumferential muscle fibers increased in diameter during
ontogeny from 2.5 /urn ± 0.49 /j.m (mean ± standard
deviation, n = 46 from four specimens) in newly hatched
squid to 3.9 /im ± 0.66 /u,m (;i = 43 from four individuals)
in the largest animals examined. In addition, the ratio of
mitochondria-rich to mitochondria-poor (Bone et «/.. 1981;
Mommsen et al.. 1981) circumferential muscle fibers de-
creased from 1:5 in newly hatched squid to 1:7 in young 2
stage squid. The number of mitochondria-rich fibers adja-
cent to the inner tunic is twice that of the mitochondria-rich
muscle fibers adjacent to the outer tunic in S. lessoniana.

The inner tunic is adjacent to the mantle musculature and
to the thin epithelial lining of the mantle cavity. The radial
muscle fibers and the collagen fibers in IM-1 and IM-2
insert on the inner tunic.

Initial snn'ey

An initial survey of the mantle revealed that IM-1 fiber
angle and IM-2 fiber angle differ both along the length and
around the circumference of the mantle in an individual
squid. In the ventral portion of the mantle, there were no
significant differences in IM-1 fiber angle or in IM-2 fiber
angle between 1/4 and 3/4 DML. However, IM-1 fiber angle
was about 10° higher at 1/10 DML and about 10° lower
between 3/4 DML and the posterior tip of the mantle.
Similar differences in IM-1 fiber angle along the length o

142

J. T. THOMPSON AND W. M. KIF.R

the mantle were also noted in the lateral and dorsal regions.
There was no correlation between mantle thickness at either
1/10 or 3/4 DML and the fiber angle at that location.

Between 1/4 and 3/4 DML. both IM-1 fiber angle and
IM-2 fiber angle were about 10° lower in the dorsal region
of the mantle than in either the lateral or ventral portions.
Within an individual squid, there were no significant differ-
ences between the average IM-1 fiber angles or IM-2 fiber
angles in the lateral or ventral portion of the mantle between
1/4 and 3/4 DML. Again, there was no apparent correlation
between mantle thickness and fiber angle at a particular
location around the circumference of the mantle.

The implications of these differences in collagen fiber
arrangement are unclear. It is interesting, however, that
MacGillivray et al. (1999) did not report significant differ-
ences in mantle mechanical properties either along the
length or around the circumference of the mantle in Loligu
pealei. It is possible that the differences in IM-1 and IM-2
fiber angle reported here for S. lessoniana are not present in
L. pealei. Alternatively, such differences, if present, may
not translate into significant differences in mantle mechan-
ical properties.

IM-1 fiber ontogeny

IM-1 collagen fibers were scarce in newly hatched squid
relative to older, larger animals (Fig. 2). Dozens of sagittal
tissue sections from a hatchling squid could be searched
without encountering a single IM-1 fiber. As the squid grew
during ontogeny, IM-1 collagen fibers became increasingly
numerous and robust (Fig. 2). The diameter of IM-1 fibers
increased during ontogeny from 0.58 /nm ± 0.060 /u.m
standard deviation (SD; n = 24 from four individuals) in
newly hatched squid to 0.68 /im ± 0.052 ^.m SD (n = 37
from three animals) in the young 2 stage squid.

IM-1 fiber angle changed dramatically during ontogeny
(Fig. 2). IM-1SAG fiber angle was between 26° and 33° in
newly hatched animals and increased exponentially during
growth from hatching to about 15 mm DML (Fig. 3 A).
IM-1SAG fiber angle remained fairly constant (about 43") in
squid larger than 15 mm DML (Fig. 3 A). A one-way
ANOVA on ranks showed that while hatchling stage IM-
ISAG fit>er ang'e was significantly lower than the fiber angle
of squid in the other life-history stages examined (P <
0.05. Table 2). there were no significant differences in fiber
angle among the juvenile 1, juvenile 2, and young 2 stage
animals (Table 2).

IM-1TAN fiber angle also changed substantially during
ontogeny. IM-1TAN fiber angle was highest in newly
hatched animals (between 35° and 46°) and declined to
about 28° in the largest squid examined (Fig. 3B). A one-
way ANOVA on ranks indicated that hatchling stage IM-
'TAN noer angle was significantly higher than the fiber
angle in all older, larger animals (P < 0.05. Table 2).

There were no significant differences in IM-1TAN fiber
angle among the squid in the juvenile 1 , juvenile 2, and
young 2 life-history stages (Table 2).

IM-2 fiber ontogeny

IM-2 collagen fibers were scarce in newly hatched ani-
mals when compared to the older, larger squid in the study
(Fig. 4). As with IM-1 fibers, many mantle tissue sections
could be observed without locating a single IM-2 collagen
fiber. However. IM-2 fibers increased in abundance and
diameter as the squid grew during ontogeny (Fig. 4). The
diameter of IM-2 collagen fibers increased from an average
of 0.54 jum ± 0.080 /urn SD (n = 28 from four animals) in
newly hatched squid to an average of 0.7 1 /urn ± 0.087 /u,m
SD (n = 31 from three specimens) in the young 2 stage
animals.

IM-2 fiber angle changed significantly during ontogeny
(Fig. 4). IM-2 fiber angle was between 27° and 36° in newly
hatched squid and rose exponentially until the squid grew to
15 mm DML (Fig. 3C). In squid larger than 15 mm DML,
IM-2 fiber angle ranged between 48° and 58° (Fig. 3C). A
one-way ANOVA on ranks showed that IM-2 fiber angle
was lower in hatchling stage squid than in all older, larger
animals (P < 0.05. Table 2). The one-way ANOVA on
ranks also revealed that the IM-2 fiber angle in the juvenile
2 stage squid was marginally higher than in the juvenile 1
stage animals (P = 0.05, Table 2). There were no signif-
icant differences in IM-2 fiber angle between juvenile 1 and
young 2 stage animals.

Outer tunic fiber ontogeny

Regardless of size, all the squid possessed a robust outer
tunic (Fig. 5). The collagen fibers constituting the outer
tunic changed in orientation during ontogeny. The outer
tunic fiber angle was highest in newly hatched animals
(between 27° and 36°) and declined during ontogeny (Fig.
3D). In squid larger than about 15 mm DML, outer tunic
fiber angle decreased slightly with size from about 26° to
16° (Fig. 3D). A one-way ANOVA on ranks showed that
the outer tunic fiber angle was higher in hatchling stage
animals than in all older, larger squid (P < 0.05. Table 2).
In addition, outer tunic fiber angle was slightly higher in
juvenile 2 stage animals than in either juvenile 1 or young
2 stage animals (P -- 0.05. Table 2). There were no
significant differences in outer tunic fiber angle between
juvenile 1 and young 2 stage squid.

Volume fraction of collagen in IM-1 and IM-2

Relative to the volume of mantle musculature, the vol-
ume fraction (Vr) of collagen in both IM-1 and IM-2
increased nearly 2 orders of magnitude during ontogeny
(Fig. 6. Table 1). The volume fraction of collagen in IM-1

ONTOGENY OF SQUID MANTLE

143

Figure 2. Photomicrographs (polarized light microscopy) of 5-/j,m-thick sagittal sections ot the ventral
mantle that illustrate ontogenetic differences in IM-I collagen fibers. Sections were stained with picrosirius. The
orientation of the mantle in both panels is identical. IM-1. intramuscular fiber system 1 collagen fiber: IT. inner
tunic; MC. mantle cavity: OT. outer tunic: RMF. radial muscle band. Scale bar in A and B. 20 fim. (A) The
ventral mantle of a newly hatched squid (DML. 5.5 mm) with a single IM-1 collagen fiber. The section is oblique
to the sagittal plane. (B) The ventral mantle of a young 2 stage squid (DML. 65 mm). Note the low IM-1 collagen
fiber angle and the absence of other IM-1 collagen fibers in the field of view in the hatchling squid. In the larger
squid, IM-1 fiber angles are higher, and IM-1 collagen fibers are abundant.

144

J. T. THOMPSON AND W. M. KIER

55-
50-

45 —

40 <

i T F

•

1 * T

40-

il } 1 $

• J

f , J

35 -S

* *

!•

]

20 §

on,

1 [.-'"Nb1

"§501
"2 "5

-g 40
[i,
(N 35

S 30
25

40
°

35 oo

30 <

.25 |

•20 |

•15 §

•10 v-

•5 3

' O

•0

0 10 20 30 40 50 60 70 0 10 20 30 40 50 60 70

Dorsal Mantle Length (mm)

Figure 3. Ontogenetic changes in organization of mantle connective
tissue. In all panels, each data point represents the mean of between 8 and
20 fiber angle measurements for one squid. The bars indicate the standard
error of the mean. (A) Sagittal component of the IM-1 fiber angle versus
dorsal mantle length (DML). IM-1SAG is lowest in newly hatched squid and
rises exponentially during growth up to 15 mm DML. In squid larger than
15 mm DML. IM-1SAO does not change significantly. The block at the
lower left illustrates the lower IM-1SAG and higher IM-1TAN fiber angles of
a hatchling (see inset in Fig. 1 for orientation). The block at upper right
illustrates the higher IM-1SAO and lower IM-1TAN fiber angles of an older,
larger squid. (B) Tangential component of IM-1 fiber angle versus DML.
IM-1TAN is highest in newly hatched squid and declines during growth. (C)
IM-2 fiber angle versus DML. IM-2 fiber angle is lowest in hatchlings and
rises exponentially during growth up to 1 5 mm DML. In squid larger than
15 mm DML, IM-2 fiber angle increases slightly. The block at the lower
left illustrates the lower IM-2 fiber angle in hatchlings. The block at the
upper right illustrates the higher IM-2 fiber angle of larger squid. (D) Outer
tunic fiber angle versus DML. Outer tunic fiber angle is highest in hatch-
lings and declines during ontogeny.

increased 68 times, from an average of 0.00095 in newly
hatched squid to an average of 0.065 in the largest animals
examined in this study (Table 1 ). A one-way ANOVA on
ranks indicated that the volume fraction of collagen in IM-1
was significantly different among all the life history stages
(P < 0.05, Table 1).

The volume fraction of collagen in IM-2 increased 36
times, from an average of 0.0027 in newly hatched animals
to an average of 0.097 in the largest squid studied (Fig. 6,
Table 1). A one-way ANOVA on ranks showed that the
volume fraction of collagen was significantly different
among all the life history stages examined (P < 0.05,
Table 1).

Discussion

Connective tissue fibers limit the range of movement in
many soft-bodied, cylindrical animals that rely upon a hy-
drostatic skeleton for support (e.g., Harris and Crofton.
1957: Chapman, 1958; Clark and Cowey, 1958; Clark.
1964). The collagen fibers in the outer tunic, IM-1. and
IM-2 may also affect the limits of mantle movement during
jet locomotion. Because the fiber angles in all of the con-
nective tissue fiber networks examined here change signif-
icantly during ontogeny, the kinematics of mantle move-
ment probably change significantly as well.

The outer tunic

The tunics of squid are hypothesized to restrict mantle
lengthening during jet locomotion (Ward and Wainwright.
1972). This important function ensures that the mechanical
work performed by the circumferential musculature is used
to decrease mantle cavity volume, thereby forcing water out
of the funnel and producing thrust, instead of lengthening
the mantle. The average outer tunic fiber angle of 17.7° from
young 2 stage Sepioteuthis lessoniana was substantially

Table 2

Comparison

of mantle collagen fiber organization among squid divided into the life-history stages of Segawa

(1987)

Life-history

stage

1M-1SAO

fiber angle

IM-1TA

N fiber

angle

IM-2

fiber angle

Tunic fiber angle

Hatchling
Juvenile 1
Juvenile 2
Young 2

32.7 ±
43.7 ±
43.2 ±
42.3 ±

9.22(6)
7.33(7)
6.29(5)
6.50(3)

39.0
33.2
32.8
31.9

±6.37(5)
± 6.74(5)
± 6.59(4)
± 3.65(31

34.6
49.7
53.9
53.3

6.76(5)
6.52(5)*
6.00(5)*
5.40(3)

33.5
20.6

22.4

17.7

± 6.37(3)
± 6.74(5)
± 6.59(5)*
± 3.65 (3)*

The mean fiber angle is listed in boldface type in each column ± the standard deviation of the mean. The number of squid in the sample is in parentheses.
Each mean fiber angle was calculated from between 8 and 25 measurements of fiber angle for each squid in the sample. All the fiber angle measurements
for each squid in a life-history stage were pooled to calculate the mean and the standard deviation. In each column, the mean fiber angle tor the hatchling
stage squid was significantly different from the mean fiber angle for the juvenile 1. juvenile 2. and young 2 life-history stages (one-way ANOVA on ranks,
P < 0.05). The asterisks in the IM-2 fiber angle column indicate significant differences in mean fiber angle between the juvenile 1 and juvenile 2
life-history stages (one-way ANOVA on ranks. P = 0.05 ). The asterisks in the tunic fiber angle column indicate a significant difference in mean fiber angle
between the juvenile 2 and young 2 life-history stages (one-way ANOVA on ranks, P = 0.05). Other within-column comparisons of fiber angle were not
significantly different.

ONTOGENY OF SQUID MANTLE

145

Figure 4. Photomicrographs (brightfield microscopy) of 10-/Mm-thick transverse sections of the ventral
mantle that illustrate ontogenetic differences in IM-2 collagen fibers. Sections were stained with picrosirius. The
orientation of the mantle is identical in both images. IM-2, intramuscular fiber system 2 collagen fiber; IT. inner
tunic: MC. mantle cavity: OT. outer tunic. Scale bar in A and B. 60 /nm. (Al A single IM-2 collagen fiber in a
newly hatched squid (DML. 5 mm). Note the low IM-2 fiber angle and the scarcity of other IM-2 collagen fibers
in the field of view. (B) IM-2 collagen fibers in the ventral mantle of a young 2 stage squid (DML. 69 mm). The
faint vertical fibers near the center of the image are radial muscle fibers. Note that the IM-2 fiber angle is higher
and IM-2 collagen fibers are abundant.

146

J. T. THOMPSON AND W. M. KIER

Figure 5. Photomicrographs of 5-/im-thick grazing sections of ventral squid mantle that illustrate the
ontogenetic change in outer tunic collagen tiber angle. Black lines overlay a pair of collagen fibers in A and B
to help illustrate the fiber angle. CMF. circumferential muscle fibers. Scale bars. 20 p.m. (A) Outer tunic collagen
fibers in a newly hatched squid (DML, 6 mm). Brightiield microscopy with picrosirius stain. The small arrow
indicates additional outer tunic collagen fibers. (B) Outer tunic collagen fibers in a juvenile 2 stage squid I DML.
38 mm). Polarized light microscopy with picrosirius stain. Note the higher fiber angle in the hatchling animal.

lower than the 27° average outer tunic fiber angle reported
for Lolliguncula hreris and Loli^o peulei by Ward and
Wainwright (1972). Indeed, the outer tunic fiber angle mea-

sured for mature L. hrevis and L. peulci by Ward and
Wainwright (le)72) is much closer to the hatchling stage
outer tunic fiber angle of 33.5° in S. lessoniana.

ONTOGENY OF SQUID MANTLE

147

'5 o.io

£ 0.08 •

E 0.06 •

£ 0.04 •

v 0.02

= 0.00 •

•

IM-1
IM-2

0 10 20 30 40 50 60 70 80

Dorsal Mantle Length (mm)

Figure 6. Collagen volume fraction in the ventral mantle versus dorsal
mantle length. Each point represents the volume fraction of collagen for
one squid. Circles indicate the volume fraction of collagen in IM-I, and
triangles indicate the volume fraction of collagen in IM-2. The IM-2 data
points obscure the IM-1 data for the hatchling stage squid.

The outer tunic fibers of all the S. lessoniana studied are
oriented appropriately to resist lengthening of the mantle
during jet locomotion (see Chapman, 1958, and Clark and
Cowey, 1958). Both the mechanical properties and fiber
angle of outer tunic collagen fibers affect mantle lengthen-
ing during jet locomotion. If the mechanical properties of
the outer tunic collagen fibers do not change during ontog-
eny, the ability of the outer tunic to resist increases in
mantle length during jet locomotion will depend on fiber
angle. For example, if the maximum extensibility of an
outer tunic collagen fiber is 0.13, a realistic assumption
based on mechanical tests of squid mantle collagen (Gosline
and Shadwick. 1983b). the mantle length of a hatchling
stage squid may increase up to 23% during a jet whereas the
mantle length of a young 2 stage animal may increase up to
17%. Thus, the ontogenetic variation in outer tunic fiber
angle may allow greater mantle length increases during jet
locomotion in newly hatched squid than in older, larger
squid.

The possible increases in mantle length calculated above
for S. lessoniana probably represent maximal values. The
force balance between the outer tunic collagen fibers, other
networks of connective tissue fibers, the chitinous gladius,
and perhaps, the collagen-rich skin may all serve to limit
changes in mantle length. The purpose of the calculation is
simply to highlight the influence of outer tunic fiber angle
on the potential for increases in mantle length during jet
locomotion. Indeed. Ward (1972) did not observe increases
in mantle length during jet locomotion in L. brevis. Packard
and Trueman (1974), however, did notice small increases
(—5%) in the ventral mantle length of subadult Loligo
vulgaris and adult Sepia officinalis. As the next section
illustrates, small increases in mantle length during the jet
may facilitate elastic energy storage in the IM-1 collagen
fibers of newly hatched S. lessoniana.

IM-1

Previous reports of IM-1SAG fiber angle from Lolligun-
cula hrevis and Loligo pealei are about 15 lower than the
fiber angle reported here for the young 2 stage S. lessonianu
(Ward and Wainwright, 1972). The discrepancy may be due
to species differences or to the histological methods selected
for the analysis. It is also possible that age or size differ-
ences may account for the disparity because mature squid
were analyzed in the previous study. It is not possible to
compare the IM-1SAG hatchling fiber angle because, to our
knowledge, there are no published values for newly hatched
squid.

Bone el al. ( 1981 ) reported IM-ITAN fiber angle data for
Alloteiithis subidata and Loligo forbesi. In both species, the
IM-1TAN fiber angle was 15° to 20° lower than the angle
measured here for young 2 stage S. lessoniana. The fiber
angle reported by Bone et al., however, was measured in
partially contracted specimens. Contraction of the mantle
results in a decrease of the IM-1TAN fiber angle. Histolog-
ical methodology, species differences, or age/size differ-
ences may also account for the disparity between the pub-
lished fiber angle data and this study.

The significant ontogenetic change in IM-1SAG and IM-
1TAN fiber angle may affect the kinematics of mantle move-
ment during jet locomotion. To explore this idea, we devel-
oped a three-dimensional geometric model to evaluate the
influence of changes in fiber angle on mantle kinematics.
The model consists of a right rectangular polygon of mantle
tissue (inset in Fig. 1 ). A single IM-1 collagen fiber extends
from the front lower right corner to the rear upper left corner
of the polygon (the gray dashed line in Fig. 1). The long axis
of the polygon is parallel to the long axis of the mantle, and
the short axis (side C) is parallel to the circumferential
muscle fibers. The height of the polygon (side T) represents
the thickness of a portion of the mantle wall. The dimen-
sions of the polygon are in arbitrary units.

The polygon is assumed to have constant volume, and the
IM-1 fiber is free to reorient as the polygon changes in
dimensions. The short axis (side C, Fig. 1) of the polygon
was decreased to simulate circumferential muscle contrac-
tion during jet locomotion. IM-1SAG and IM-1TAN average
fiber angles from hatchling and young 2 stage S. lessoniana
(see Table 2) were used as the initial condition (i.e.. "rest-
ing" mantle circumference) in which strain in the IM-1 fiber
was assumed to be zero. Initially, the mantle length was
held constant during the simulations. The strain on the IM-1
fiber and the IM-1SAG and IM-1TAN fiber angles were then
calculated for a range of mantle circumference changes (see
Appendix for a sample calculation).

The model predicts the effects of changes in IM-1 fiber
angle on mantle kinematics. If mantle length is held con-
stant during the jet cycle and if an IM-1 collagfi liner
extensibility of 0.13 is assumed (Gosline and Sli:>>!

148

J. T. THOMPSON AND W. M. KIER

IM-1

IM-2

-50 -40 -30 -20 -10 0 10 -50 -40 -30 -20 -10 0

Circumference Change (%)

Figure 7. Predicted ontogenetic differences in collagen fiber strain and fiber angle during jet locomotion.
The plots in the left column are for model IM-1 collagen fibers: the plots in the right column are for model IM-2
collagen fibers. The horizontal axis in each plot is the mantle circumference change that occurs during a
simulated jet. Zero indicates the resting mantle circumference in an anesthetized squid. Negative numbers
indicate mantle contraction, and positive numbers denote expansion of the mantle. The left vertical axis in each
plot indicates strain on the model collagen fiber. The positive strain values above the horizontal zero line indicate
lengthening of the model collagen fiber; the negative values below the zero line indicate compression. The right
vertical axis in plots B. C. E, and F indicate the fiber angle of the model collagen fiber. The dashed lines represent
strain data for the hatchling stage model and solid lines represent strain data for the young 2 stage model. Lines
with symbols indicate the fiber angle predictions. (A and D) IM-1 fiber strain and IM-2 fiber strain, respectively,
during a simulated jet. Mantle length was held constant during the simulated jet. Strain is lower at a given mantle
circumference in the IM-1 and IM-2 hatchling stage model collagen fibers during the simulated jet than in the
young 2 stage collagen fibers. In both A and D. if the maximum extensibility of the model collagen fibers remains
unchanged during ontogeny, hatchling stage squid will experience much greater mantle contraction during the
simulated jet than the young 2 squid. The hatchling fiber is compressed during the initial 17% of the mantle
contraction in the IM- 1 model and during the initial 327r of mantle contraction in the IM-2 model. Consequently,
storage of strain energy in the model IM-1 and IM-2 collagen fibers is not possible unless the mantle contracts
more than 17% and 32%, respectively. (B and C) Predicted changes in the sagittal (0SAG) and tangential (6TAN)
components of the IM-1 fiber angle for a young 2 and a hatchling stage squid, respectively. In each case, note
that SSAG increases while 9TAN decreases during the simulated jet. (E and F) Predicted changes in the IM-2 fiber
angle (a) for a young 2 and a hatchling stage squid, respectively. The fiber angle increases during the simulated
jet in both cases.

1983b), the range of possible mantle movements changes
substantially during ontogeny. In the model of the hatchling
stage S. lessoniana, mantle circumference may decrease
about 45% (Fig. 7A) during jet locomotion, whereas the
mantle circumference of a young 2 stage animal may de-
crease only about 30% (Fig. 7A). In both hatchling and
young 2 stage squid, the models show that IM-1SAG fiber

angle increases during the jet. while IM-1TAN fiber angle
decreases (Fig. 7B, 1C). It is also interesting to note that the
low initial IM-1SAG fiber angle and the high initial IM-1TAN
fiber angle in hatchling stage squid result in IM-1 fiber
compression during the initial - 1 lc/c mantle circumference
change (Fig. 7A). If mantle length is constant during the jet,
low-amplitude movements (i.e., less than a 17% decrease in

ONTOGENY OF SQUID MANTLE

149

Hatchlinu

Young 2

0.20

-50 -40 -30 -20 -10 0

0.20

'« 0,15

^ 0.10'

•2 0.05

«M 0.00

-0.05

-0.10

0.00

020
0.15
0.10
0.05
0.00

.50 -40 -30 -20 -10 0 -40 -30 -20 -1(1 0

Circumference Change (%)

Figure 8. The effect on collagen liber strain of an increase in mantle length during simulated jet locomotion.
The plots in the left column are for the hatchling stage models; plots in the right column are for the young 2 stage
models. The horizontal axis in each plot is the change in mantle circumference (hat occurs during a simulated
jet. Zero indicates the resting mantle circumference in an anesthetized squid. Negative numbers indicate mantle
contraction. The vertical axis in each plol indicates strain on the model collagen fiber. The positive strain values
above the horizontal zero line indicate elongation of the model collagen fiber; the negative values below the zero
line indicate compression. The amount of mantle elongation during the simulated jet in each graph is indicated
in the legend for graph A. (A) Predicted 1M-1 fiber strain in a hatchling stage squid. Increases in mantle length
result in higher strain on the model collagen fiber early in the simulated jet stroke but do not greatly influence
the possible range of mantle kinematics. (B) Predicted IM-1 fiber strain in a young 2 stage squid. Increases in
mantle length do not affect strain on the model collagen fiber early in the simulated jet but do increase the
possible range of mantle kinematics. (C and D) Predicted IM-2 fiber strain in a hatchling stage and a young 2
stage squid, respectively. Increases in mantle length during the simulated jet stroke substantially affect the
possible range of mantle kinematics and the strain of the model IM-2 collagen fibers.

mantle circumference) of the mantle will not store strain
energy in IM-1 collagen fibers in newly hatched squid.

We also examined the effect of mantle length increase
during jet locomotion on mantle kinematics using the
model. For both the hatchling and young 2 stage models,
mantle length was allowed to increase 5%, 10%, and 15%
during the jet cycle. In the hatchling stage model, mantle
length was increased steadily until a mantle circumference
change of —45% was reached (i.e., the point in the previous
IM-1 hatchling model where strain on the IM-1 collagen
fiber was 0.13). In the young 2 stage model, mantle length
was increased gradually until a mantle circumference
change of —30% was reached (i.e., the point in the previous
IM-1 young 2 model where the strain on the IM-1 collagen
fiber was 0.13). After mantle circumference changes of
—45% for the hatchling stage model and —30% for the
young 2 stage model were reached, mantle length was held
constant. Note that incorporating the increases in mantle
length earlier in the jetting cycle (e.g., up to a mantle
circumference change of —20%, then holding mantle length
constant) does not affect the predicted maximum range of

mantle contraction, though it does result in higher strain in
the hatchling stage model IM-1 collagen fiber early in the
jetting cycle. In all the simulations, the strain on the IM-1
fiber was calculated for a range of mantle circumference
changes.

Incorporating mantle length increase during jet locomo-
tion into the model results in several interesting predictions.
In hatchling stage S. lessoniaiui. modest increases in mantle
length during the jet did not substantially affect the maxi-
mum possible amplitude of mantle contraction, but they did
result in an increase in IM-1 collagen fiber strain during
low-amplitude mantle movements (Fig. 8A). The poten-
tially important consequence of small increases in mantle
length during jetting for newly hatched squid is that strain
energy storage in IM-1 collagen fibers is possible during
low-amplitude movements of the mantle (e.g., respiration
and slow jet locomotion). In the young 2 stage squid, only
the 15% increase in mantle length during jet locomotion had
any noticeable effect on IM-1 collagen fiber strain (Fig. 8B).
The maximum possible mantle circumference change dur-
ing the jet, however, increased slightly with modest in-

150

J. T. THOMPSON AND W. M. KIER

creases in mantle length (Fig. 8B). Thus, the model predicts
that small increases in mantle length during low-amplitude
mantle movements (e.g.. slow jetting or respiration) in
newly hatched S. lessoniana may result in increased energy
storage in IM-1 collagen fibers; small increases in mantle
length in older, larger squid do not. The predicted increase
in elastic energy storage comes at the expense of a decrease
in thrust. For slow jetting or respiration, this cost may not
outweigh the benefits of elastic energy storage.

IM-:

The average IM-2 fiber angle of young 2 stage S. lesso-
niana was about the same as the 55° reported for Alloteutliix
subulatu and Loligo forbe si by Bone et al. ( 1981 ). It is not
possible to compare the average IM-2 fiber angle of the
hatchling stage S. lessoniana with the literature because we
are not aware of any published IM-2 fiber angles in newly
hatched squid.

The significant change in IM-2 fiber angle during ontog-
eny may contribute to substantial ontogenetic changes in
mantle kinematics during jet locomotion. We examined the
implications of a fiber angle change on mantle kinematics
using a model similar to the one used to predict the effect of
IM-1 collagen fiber angle on mantle kinematics. The IM-2
model consists of the same polygon mentioned above, ex-
cept there is a single IM-2 fiber running from the lower right
corner to the upper left corner of plane CT (solid gray fiber
in plane CT. Fig. 1 ). The assumptions are the same for both
the IM-1 and IM-2 models.

The model predicts the potential effects of an ontogenetic
change in IM-2 fiber angle on mantle kinematics. If the
extensibility of the IM-2 collagen fiber in the model is
limited to 0.13 (Gosline and Shadwick. 1983b), a mantle
circumference change of about —45% is possible in hatch-
ling stage S. lessoniana during jet locomotion (Fig. 7D).
During mantle contraction, the hatchling IM-2 fiber angle
will increase from about 35° to about 67° (Fig. 7F). Due to
the low initial fiber angle, the IM-2 collagen fiber is com-
pressed during the first —32% change in mantle circumfer-
ence (Fig. 7D). The model predicts that strain energy stor-
age in the IM-2 collagen fibers will occur only during
vigorous jet locomotion that results in large decreases
(>32%) in mantle circumference. Interestingly, this sug-
gests that if hatchling stage S. lessoniana use elastic mech-
anisms to restore mantle shape during respiratory move-
ments of the mantle, as hypothesized for mature Loligo
opalescens by Gosline et al. (1983). strain energy storage
can only take place in the IM-1 collagen fibers.

The model predicts substantially different mantle kine-
matics for young 2 stage S. lessoniana. Again, if IM-2
collagen fiber extensibility is assumed to be 0.13. mantle
circumference changes during jet locomotion of up to about
—25% are possible (Fig. 7D). The fiber angle will increase

from the initial value of 53° to about 65° at the end of the
jet (Fig. 7E). Given the high initial fiber angle, the strain
experienced by the IM-2 collagen fibers will increase rap-
idly during the jet (Fig. 7D).

Because the mantle tissue is probably constant in volume
over the brief period of a single mantle contraction (Ward,
1972), increases in mantle length during jet locomotion will
influence the strain experienced by the IM-2 collagen fibers.
Therefore, we also examined the effect of mantle length
increase during jet locomotion on mantle kinematics. Man-
tle length in both the hatchling and young 2 stage models
was allowed to increase 5%, 10%, and 15% during the jet
cycle. In the hatchling stage model, mantle length was
increased gradually until a mantle circumference change of
—45% was reached (i.e., the point in the previous hatchling
IM-2 model where strain on the IM-2 collagen fiber was
0.13). In the young 2 stage model, mantle length was
increased progressively until a mantle circumference
change of —25% was reached (i.e., the point in the previous
young 2 IM-2 model where strain on the IM-2 collagen fiber
was 0.13). Mantle length was held constant after mantle
circumference changes of —45% and —25%, for the hatch-
ling and young 2 stage models respectively, were reached.
The strain on the IM-2 fiber was calculated for a range of
changes in mantle circumference.

The models predict substantial effects on both mantle
kinematics and elastic energy storage when mantle length
increases during jet locomotion. The hatchling stage model
predicts that modest 5% or 10% increases in mantle length
during jetting increase proportionately the range of possible
mantle circumference changes (Fig. 8C). Increases in the
maximum amplitude of mantle movements during jet loco-

Figure 9. The imaginary polygon used to calculate strain on the model
IM-1 and IM-2 collagen fibers during simulated jet locomotion. Only the
model IM-1 collagen fiber (solid gray line) is shown. See the Appendix for
more detail.

ONTOGENY OF SQUID MANTLE

151

motion, however, result in the compression of IM-2 colla-
gen fibers for a longer portion of the jet cycle (Fig. 8C).
Thus, if the mantle of a hatchling stage squid lengthens even
a small amount, the IM-2 collagen fibers will store elastic
energy only during vigorous jet locomotion.

The young 2 model predicts similar increases in the range
of possible changes in mantle circumference with increases
in mantle length during jetting (Fig. 8D). In contrast to the
hatchling stage model, the young 2 stage model predicts that
only increases in mantle length greater than 10% will result
in compression of the IM-2 collagen fiber during jet loco-
motion (Fig. 8D). Thus, modest increases in mantle length
will permit elastic energy storage in IM-2 collagen fibers
during low-amplitude movements of the mantle.

Can IM-1 and IM-2 fiber angles predict mantle
kinematics?

The models for IM-1 and IM-2 described above both
suggest that a much greater range of mantle circumference
change is possible during jet locomotion in newly hatched 5.
lessoniana than in older, larger squid. The models predict
that mantle circumference changes up to —45% are possible
in hatchling stage squid compared with —25% and —30% in
young 2 stage squid. The few published accounts of mantle
kinematics support these predictions. In Loligo opalescens
the range of circumference change during vigorous jet lo-
comotion is at least 10% greater in hatchling animals than in
older, larger squid (Gilly et ui, 1991). Maximum mantle
circumference changes are about -40% to -42% in hatch-
lings and about —30% in mature animals (Gilly et ai,
1991). In Loligo vidgaris. the maximum mantle circumfer-
ence changes are about —45% and —30% in hatchling and
adult animals, respectively, during escape-jet locomotion
(calculated from Packard, 1969). During escape-jet locomo-
tion in 5. lessoniana, mantle circumference changes —45%
and —33% in hatchling stage and young 2 stage squid,
respectively (Thompson. 2000; Thompson and Kier. 2001 ).
Even during less vigorous jet locomotion and respiratory
mantle movements, the range of mantle circumference
change is considerably greater in newly hatched L. opal-
escens (Preuss et ai, 1997). Given the assumptions of the
geometric models, the measurement errors inherent in the
histological methods, the cross-species comparisons, and
the lack of consideration for the role of circumferential
muscle mechanics in mantle contraction, it is striking that
the models accurately predict the maximum amplitude of
actual mantle kinematics. Thus, we hypothesize that the
arrangement of intramuscular collagen fibers likely plays a
crucial role in determining the mechanical properties of
squid mantle.

Predictions of elastic energy storage from the IM
collagen fiber models

The mathematical models of IM-1 and IM-2 predict on-
togenetic changes in the potential of the intramuscular col-
lagen fibers to store elastic energy. During low-amplitude
mantle contraction in hatchling stage S. lessoniana. the
models predict that IM-1 and IM-2 collagen fibers are
compressed and, thus, are unable to store elastic energy.
This prediction holds if mantle length remains constant
during the simulated jet or if mantle length increases (only
up to 10% for IM-1 fibers) during the jet. Only during
large-amplitude mantle movements (e.g., during vigorous
jet locomotion) do the hatchling stage models predict that
IM-1 and IM-2 collagen fibers are stretched and. thus, able
to store elastic energy. Conversely, the models predict that
IM- 1 and IM-2 collagen fibers in young 2 stage S. lessoni-
ana are stretched as the mantle contracts and, thus, can store
elastic energy during jet locomotion.

The hatchling stage models also predict a difference in
the relative elongation of IM-1 and IM-2 collagen fibers
during jet locomotion. If mantle length remains constant
during the jet, a mantle circumference decrease of 18% is
required to elongate an IM-1 collagen fiber, whereas a
mantle circumference decrease of 33% is necessary to ex-
tend an IM-2 collagen fiber. Thus, the hatchling stage mod-
els predict that IM-2 collagen fibers may not contribute to
elastic energy storage in the mantle except during vigorous
jet locomotion.

Volume fraction of collagen

Isolated portions of squid and cuttlefish mantle store
elastic energy in experiments that simulate mantle move-
ment during the exhalant phase of jet locomotion (Gosline
and Shadwick, 1983a. b; MacGillivray et «/., 1999; Curtin
et nl.. 2000). The amount of elastic energy stored in the
mantle depends on the volume of collagen in the tissue, the
strain experienced by the collagen fibers, and the mechan-
ical properties of the mantle collagen fibers. The volume
fraction of collagen in IM-1 and IM-2 increased 68 times
and 36 times, respectively, during ontogeny. If IM-1 and
IM-2 collagen fibers are strained comparably during loco-
motion in squid of all ages, and if the mechanical properties
of collagen do not change with growth, the elastic energy
storage capacity of the mantle is likely to increase dramat-
ically during ontogeny.

Future directions

We have described ontogenetic changes in IM-1 and
IM-2 fiber angle and in mantle collagen volume fraction. To
analyze these changes further, and to begin testing the
predictions of the models, accurate measurement of mantle
length during jet locomotion is needed. In addition, the

152

J. T. THOMPSON AND W. M. KIER

predictions of the models assume that the mechanical prop-
erties of squid mantle collagen do not change during ontog-
eny. This important assumption needs to be tested, partic-
ularly because the mechanical properties of collagen change
during the growth of many animals (e.g.. Parry and Craig,
1988). Therefore, mechanical tests of intact portions of
squid mantle, in combination with data on mantle length
during jet locomotion, are necessary to test the predictions
of the models and to understand better the ecological and
evolutionary implications of ontogenetic changes in the
morphology of mantle connective tissue.

Acknowledgments

This research was supported by NSF grants to W.M.K.
(IBN-9728707 and IBN-92 19495). Grants and fellowships
to J.T.T. from the Wilson Fund, the American Malacologi-
cal Association, and Sigma Xi helped defray research ex-
penses. We thank L. Walsh at the NRCC for her expertise in
shipping squid cross-country and E. Burgin for his help in
sectioning and staining. We are grateful to the Duke-UNC
biomechanics group for discussion of several of the ideas in
this paper and to D. Pfennig for statistical advice. Finally,
we thank S. A. Wainwright. J. Taylor, and two anonymous
reviewers for constructive comments and suggestions on an
earlier version of the paper.

Literature Cited

Bone, Q., A. Pulsford, and A. D. Chubh. 1981. Squid mantle muscle

J. Mar. Bwl. Assoc. U.K. 61: 327-342.

Cassada, R. C., and R. L. Russell. 1975. The dauerlarsu. a post-
embryonic developmental variant of the nematode Caenorhabditis el-

egans. De\: Biot. 46: 326-342.
Chapman, G. 1958. The hydrostatic skeleton in the invertebrates. Biol.

Rev. Ciimh. Plulos. Sue. 33: 338-371.
Clark, R. B. 1964. Dynamics in Melazoan Evolution. Clarendon Press.

Oxford.
Clark, R. B., and J. B. Cowey. 1958. Factors controlling the change of

shape of certain nemertean and turbellanan worms. J. Exp. Bio/. 35:

731-74X.
Cox, G. N., S. Staprans, and R. S. Edgar. 1981. The cuticle of Cae-

norhabditis e/egans. II. Stage-specific changes in ultrastructure and

protein composition during postembryonic development. l)c\: Biol. 86:

456-47(1.
Curtin, N. A., R. C. Woledge, and Q. Bone. 2000. Energy storage by

passive elastic structures in the mantle of Sepia officinalis. J. Exp. Biol.

203: 869-878.
Gilly, W. F., B. Hopkins, and G. O. Mackie. 1991. Development of

giant motor axons and neural control of escape responses in squid

embryos and hatchlings. Biol. Bull. 180: 209-220.
Gosline, J. M., and R. E. Shadwick. 1983a. The role of elastic energy

storage mechanisms in swimming: an analysis of mantle elasticity in

escape jetting in the squid. Loligo opule.scen.s. Can. J. Zool. 61:

1421-1431.
Gosline, J. M., and R. E. Shadwick. 1983b. Molluscun collagen and its

mechanical organization in squid mantle. Pp. 37 1-398 in The Molhisca.

Vol. I: Metabolic Biochemistry anil Molecular Biomechanics. P. W.

Hochachka, ed. Academic Press. New York.

Gosline. J. M., J. D. Sleeves, A. D. Harman. and M. E. DeMont. 1983.

Patterns of circular and radial mantle muscle activity in respiration and
jetting of the squid Loligo opalescens. J. Exp. Biol. 104: 97—109.

Harris, J. E., and H. D. Crofton. 1957. Structure and function in the
nematodes: Internal pressure and cuticular structure in Ascaris. J. Exp.
Biol. 34: 116-130.

Kier, VV. M. 1992. Hydrostatic skeletons and muscular hydrostats. Pp.
205-23 1 in Biomechanics: Structures and Systems. A Practical Ap-
proach. A. A. Biewener. ed. IRL Press at Oxford University Press,
New York.

Kier, W. M., and K. K. Smith. 1985. Tongues, tentacles, and trunks:
The biomechanics of movement in muscular hydrostats. Zool. J. Linn.
Soc. 83: 307-324.

Koehl, M. A. R. 1977. Mechanical diversity of connective tissue of the
body wall of sea anemones. J. Exp. Biol. 69: 107-125.

Lee, P. G., P. E. Turk, W. T. Yang, and R. T. Hanlon. 1994. Biolog-
ical characteristics and biomedicul applications of the squid Sepioteu-
tliis lessoniana cultured through multiple generations. Biol. Bull. 186:
328-341.

Lopez-DeLeon, A., and M. Rojkind. 1985. A simple micromethod for
collagen and total protein determination in formalin-fixed paraffin-
embedded sections. J. Histochem. Cytochem. 33(8|: 737-743.

MacGillivray, P. S., E. J. Anderson, G. M. Wright, and M. E. DeMont.
1999. Structure and mechanics of the squid mantle. J. Exp. Biol 202:
683-695.

Messenger, J. B., M. Nixon, and K. P. Ryan. 1985. Magnesium chlo-
ride as an anaesthetic for cephalopods. Cinnp. Biochem. Physiol. 82C:
203-205.

Moltschanivvskyj, N. A. 1994. Muscle tissue growth and muscle fibre
dynamics in the tropical loliginid squid Photololigo sp. (Cephalopoda:
Loliginidae). Can. J. Fish. Aquut. Sci. 51: 830-835.

Mommsen, T. P., J. Ballantyne, D. MacDonald, J. Gosline, and P. \V.
Hochachka. 1981. Analogues of red and white muscle in squid
mantle. Proc. Nat. ACM!. Sci. USA 78(5): 3274-3278.

Monies, G. S.. and Junqueira, L. C. U. 1988. Histochemical localiza-
tion of collagen and of proteoglycans in tissues. Pp. 41-72 in Collagen.
Vol. II. Biochemistry and Bioinechanics, M. E. Nimni, ed. CRC Press.
Boca Raton. FL.

Packard, A. 1969. Jet propulsion and the giant fibre response of Loligo.
Nature 221: 875-877.

Packard, A., and E. R. Trueman. 1974. Muscular activity of the mantle
of Sepia and Loligo (Cephalopoda) during respiratory movements and
jetting, and its physiological interpretation. ./. E\/>. Biol. 61: 41 1-419.

Parry, D. A. D., and A. S. Craig. 1988. Collagen fibrils during devel-
opment and maturation and their contribution to the mechanical at-
tributes of connective tissue. Pp. 1-23 in Collagen. Vol. II: Biochem-
istry mill Biomechanics. M. E. Nimni. ed. CRC Press, Boca Raton. FL.

Preuss, T., Z. N. Lebaric, and W. F. Gilly. 1997. Post hatching devel-
opment of circular mantle muscles in the squid Loligo opalescens. Biol.
Bull. 192: 375-387.

Segawa, S. 1987. Life history of the oval squid. Sepioleuthis lesson/ana
m Kominato and adjacent waters central Honshu, Japan. J. Tokyo Univ.
Fish. 74(2): 67-105.

Shadwick, R. E., and J. M. Gosline. 1985. The role of collagen in the
mechanical design of squid mantle. Pp. 299-304 in Biology of Inver-
tebrate and Lower Vertebrate Collagens. A. Bairati and R. Garrone.
eds. Plenum Press. New York.

Sweat. F., H. Puchtler. and S. I. Rosenthal. 1964. Sirius red F3BA as
a stain for connective tissue. Arch. Pathol. 78: 69-72.

Thompson, J. T. 2000. The ontogeny of mantle structure and function in
the o\al squid, Sepioleulhis lessoniana (Cephalopoda: Loliginidae).
Ph.D. dissertation. University of North Carolina at Chapel Hill.

ONTOGENY OF SQUID MANTLE

153

Thompson, J. T., and W. M. Kier. 2001. Ontogenetic changes in
mantle kinematics during escape-jet locomotion in the oval squid,
.Vr/x.iri'iir/ii.v U^simiiimi Lesson. 1830. Bio/. Bull. 201: 154 -1 6ft.

Wainvvright, S. A. 1970. Design in hydraulic organisms. Miri<nn.v.sf;j-
xclhiften 57: 321-326.

Wain« right, S. A., and M. A. R. Koehl. 1976. The nature of flow and
the reaction of benthic cnidaria to it. Pp. 5-21 in Coelenterate Ecolog\
ami Behavior. G. O. Mackie. ed. Plenum Press. New York.

Wainwright, S. A., W. D. Biggs, J. D. Currey, and J. M. Gosline. 1976.
Mechanical Design in Organisms. Princeton University Press. Prince-
ton.

Ward, D. V. 1972. Locomotory function of the squid mantle. J. Zool.

(Lnniil 167: 4K7-4W.
Ward. D. \ ., and S. A. Waimvright. 1972. Locomotory aspects of

squid mantle structure. J. Zool. {Lond) 167: 437-449.
Weibel, E. R. 1979. Stereological Methods. Vol. I: Practical Methods

for Biological Murphometry. Academic Press, New York.
Young, J. Z. 1938. The functioning of the giant nerve fibres of the squid.

J. Exp. Biol. IS: 170-185.

Zar, J. H. 1996. Bimtaiixticul Analysis. Prentice Hall, Upper Saddle
River. NJ.

Appendix

The following is a brief summary of the variables, assumptions, and steps used in the calculation of strain on the IM-1
model collagen fiber. The procedure for calculating strain on the model IM-2 collagen fiber is similar to that outlined below
and is not shown. The only difference between the two calculations is that the model IM-2 collagen fiber is restricted to plane
CT (Fig. 1 ). That simplifies the calculation, in that the IM-2 projected fiber length is the same as the actual fiber length.

Variables (see Fig. 9 in text)

0SAG = IM-JSAG »ber
0TAN = IM-1TAN fiber angle

L = Polygon length (parallel to mantle length)

C = Polygon width (parallel to mantle circumference)

T = Polygon thickness (parallel to mantle wall thick-

ness)
FLSAG = Projected fiber length of the IM-1 collagen fiber

in the sagittal plane
FLTAN = Projected fiber length of the IM-1 collagen fiber

in the tangential plane
FLACT = Actual length of IM-1 collagen fiber

Initial conditions
0<;

"SAG Hatch

= 39°

(I A TO

"SAG Young2 ~ 4-

rAN Hatch ~ "TAN Young2 '

FLSAG = 1.0 (arbitrary units)

L = FL
T= FL

SAG
SAG

COS 0

sin 0,

SAG

Steps

( 1 ). Calculate polygon width, C:
C = L tan 0TAN

(2). Calculate polygon volume:

Volume = LTC

Note, the volume of the model polygon was assumed
to remain constant during the simulation.

(3). Keeping volume constant, vary side C (from 1 ,OC to
0.5C) to mimic circumferential muscle contraction
during jet locomotion. Solve for polygon thickness,
T:

T = Volume - LC

Note, for the calculations in which mantle length
was constant, side L was held constant as side C was
varied. For the calculations in which mantle length
was allowed to increase, side L was increased 5%,
10%, or 15% in length as side C was varied.
(4). Calculate the projected fiber length of the IM-1TAN
collagen fiber, FLTAN:

FL-,

= V(C2 + L2)

(5). Calculate the actual length of the IM-1 collagen
fiber, FLACT:

(6). Calculate 0SAG and 0TAN:
tan 0SAG = T + L
tan 0TAN = C + L

Reference: Bid. Bull. 201: 154-166. (October 2001 >

Ontogenetic Changes in Mantle Kinematics During

Escape-Jet Locomotion in the Oval Squid,

Sepioteuthis lessoniana Lesson, 1830

JOSEPH T. THOMPSON* AND WILLIAM M. KIER

Department of Biology. CB#3280 Coker Hall, University of North Carolina,
l Hill. Nonh Carolina 275W-3280

Abstract. We investigated the kinematics of mantle
movement during escape jet behavior in an ontogenetic
series of Sepioteuthis lessoniana. the oval squid. Changes in
mantle diameter during the jet were measured from digi-
tized S-VHS video fields of tethered animals that ranged in
age from hatchlings to 9 weeks. The amplitude of both
mantle contraction and mantle hyperinflation (expressed as
percent change from the resting mantle diameter) during an
escape jet was significantly greater in hatchlings than in
older, larger squid (P < 0.05). The maximum amplitude of
mantle contraction during the escape jet decreased from an
average of —40% in hatchlings to —30% in the largest
animals studied. The maximum amplitude of mantle hyper-
inflation decreased from an average of 18% in hatchlings to
9% in the largest squid examined. In addition, the maximum
rate of mantle contraction decreased significantly during
ontogeny (P < 0.05), from a maximum of 8.6 mantle
circumference lengths per second (L/s) in hatchlings to 3.8
L/s in the largest animals studied. The ontogenetic changes
in the mantle kinematics of the escape jet occurred concom-
itantly with changes in the organization of collagenous
connective tissue fiber networks in the mantle. The alter-
ation in mantle kinematics during growth may result in
proportionately greater mass flux during the escape jet in
newly hatched squid than in larger animals.

Received 13 December 2000; accepted 8 May 2001.

* To whom correspondence should be addressed. E-mail: joethompCs1
email.unc.edu

Abbreviations: DML, dorsal mantle length: IM-1, intramuscular fiber
system I; IM-2. intramuscular fiber system 2; IM-3. intramuscular fiber
system 3.

Introduction

Post-embryonic change in morphology is a common fea-
ture of most organisms (e.g., Werner, 1988). Such ontoge-
netic modifications may affect the ecology of the organism
(Calder, 1984; Werner, 1988; Stearns, 1992) and may pro-
vide insight into the evolution of form and function, yet they
are often neglected in studies of functional morphology and
comparative biomechanics. Significant effects on the life
cycle of an organism need not involve dramatic alterations
of morphology during ontogeny. For example, at hatching,
cephalopod molluscs are broadly similar in form to adults
(Boletzky. 1974; Sweeney et ai. 1992). Yet these tiny
hatchlings grow several orders of magnitude in size, shift
from the neuston or plankton to the benthos or nekton
(Marliave, 1980; Hanlon et ai, 1985). and may use different
mechanisms to capture prey (O'Dor et ai. 1985; Chen et ai,
1996; Kier, 1996) and to locomote ( Villanueva et ai. 1995).
In many cases, these life-cycle changes are correlated with
morphological alterations that, while not always as drastic
as the wholesale changes that occur during the metamor-
phosis of some other marine molluscs, may be equally
important in their effect on the performance or ecology of
the animal.

Cephalopods depend upon a hydrostatic skeleton for sup-
port during locomotion and movement. In the mantle of
loliginid squid, skeletal support for locomotion is provided
by a complex arrangement of fibers of muscle and of col-
lagenous connective tissue (Ward and Wainwright, 1972;
Bone et ai. 1981 ). The connective tissue fibers are arranged
in five highly organized networks: the inner and outer
tunics, and three distinct systems of intramuscular fibers
(Ward and Wainwright. 1972: Bone et ai. 1981: for review,
see Gosline and DeMont. 1985). These networks of colla-

154

ONTOGENY OF SQUID MANTLE KINEMATICS

155

gen fibers help control changes in mantle shape during
contraction of the muscles that power locomotion. In addi-
tion, the intramuscular collagen fibers store elastic energy
during the exhulunt phase of the jet and return the energy to
help restore mantle shape and refill the mantle cavity (Ward
and Wainwright. 1972: Bone el nl.. 1981; Gosline et nl..
1983: Gosline and Shadwick. 1983a; Shadwick and Gos-
line. 1985: MacGillivray el ai, 1999).

The organization of mantle connective tissue changes
significantly during ontogeny in Sepioteuthis lessoniana.
the oval squid. In hatchlings, the arrangement of outer tunic
and intramuscular collagen fibers is hypothesized to permit
large-amplitude movements of the mantle (Thompson.
2000: Thompson and Kier, 2001). In early ontogeny, the
fiber angle of the collagen fiber networks changes exponen-
tially, potentially limiting the amplitude of movement as the
squid grow (Thompson. 2000: Thompson and Kier. 2001).
Although these changes in connective tissue organization do
not constitute a discrete metamorphosis, their influence on
the mechanical properties of the mantle and the mechanics
of jet locomotion may be considerable.

To explore the implications of changes in the organiza-
tion of mantle connective tissue for the mechanics of jet
locomotion, we studied the kinematics of the escape jet in
an ontogenetic series of S. lesstmiana. The escape jet is a
distinct form of locomotion that typically involves a brief
initial hyperinflation of the mantle (i.e.. the mantle is ex-
panded radially beyond its resting diameter; see Gosline et
nl.. 1983) followed by a rapid contraction that expels water
from the mantle cavity via the muscular funnel. In tethered
S. lessoniatui. we measured ontogenetic changes in the
following kinematic parameters during the escape jet: the
amplitude of mantle hyperinflation and mantle contraction,
the rate of mantle contraction, and the frequency of escape
jetting. In addition, we used measurements of mantle radius,
mantle wall thickness, and mantle cavity volume to calcu-
late the relative mass flux produced during the escape jet.
Finally, we examined the relationship between mantle con-
nective tissue morphology and mantle kinematics during the
escape jet.

Materials and Methods

Animals

We obtained an ontogenetic series of Sepioteuthis lesso-
ninnn Lesson. 1830. We chose S. lessoniana for the exper-
iments because members of this species hatch at a large size
relative to other squid (5-7 mm dorsal mantle length and
0.01-0.03 g body weight) and. like other squid in the family
Loliginidae. they are capable of escape -jet locomotion im-
mediately upon hatching (Fields. 1965; Choe, 1966: Pack-
ard. 1969: Moynihan and Rodaniche, 1982; Segawa, 1987;
Gilly er al.. 1991).

We used S. lessoniana embryos that were collected from

three locations (Gulf of Thailand: Okinawa Island, Japan;
Tokyo Region, East Central Japan) over a 2-year period and
reared (Lee et al.. 1994) by the National Resource Center
for Cephalopods (NRCC) at the University of Texas Med-
ical Branch (Galveston, TX). Each of the three cohorts
consisted of thousands of embryos from six to eight differ-
ent egg mops. Thus, it is likely that the sample populations
were not the offspring of a few closely related individuals,
but were representative of the natural population at each
collection site.

Commencing at hatching, and at weekly intervals there-
after, live squid were sent via overnight express shipping
from the NRCC to the University of North Carolina. Ani-
mals from each of the following eight age classes were used
in the experiments: newly hatched and 1, 2, 3, 4, 5, 6, and
9 weeks after hatching. These age classes correspond to the
early life history stages denned by Segawa (1987), in which
the squid achieve external adult morphology at a dorsal
mantle length (DML) of about 40 mm (age ~6 week) and
begin to mature sexually at 150 mm DML (age >9 weeks).

Prior to the start of the experiments, the animals were
allowed about 30 min to equilibrate in an 80-1 circular
holding tank. The temperature (23 °C) and salinity (35 ppt)
of the water in the holding tanks matched the temperature
and salinity of the water in which the squid were raised.
Circular water flow in the tank helped keep the squid
swimming parallel to the sides of the tank to prevent injury.
There were never more than seven squid in the holding tank
at one time, and the maximum time an individual spent in
the tank was 4 h.

Tethering

Initially, we attempted to measure mantle kinematics in
free-swimming squid. The small size of the hatchling squid,
combined with their inability to maintain position in flow,
made it difficult to videotape at high magnification and thus
obtain adequate spatial resolution for the kinematic mea-
surements. To allow videotaping at high magnification and
to increase the spatial resolution of the edges of the mantie.
and thus the accuracy of the kinematic measurements, the
squid were tethered.

Individual squid were removed from the holding tank
with a glass beaker and anesthetized lightly in a 1:1 solution
of 7.5% MgCl:: artificial seawater (Messenger et al., 1985).
Anesthesia durations varied with the size of the animal
(longer times for larger animals) but were never longer than
2 min. While anesthetized, the squid were tethered (Fig. 1 ).
A needle (0.3-mm-diameter insect pin for smaller animals
or 0.7-mm-diameter hypodermic needle for larger animals)
was inserted through the brachial web of the squid, anterior
to the brain cartilage and posterior to the buccal mass. The
needle was positioned between these two rigid structures to
prevent it from tearing the soft tissue of the squid. The

156

J. T. THOMPSON AND W. M. KIER

DML

Figure 1. The tethering apparatus. A. acrylic plastic base; GB, glue
bead; N, needle; P. post; W, plastic washer. DML indicates dorsal mantle
length; white arrow points at V> DML.

needle was inserted into a hollow stainless steel post (hy-
podermic tubing) attached to a sheet of acrylic plastic. The
needle fit tightly in the hollow post to prevent movement.
Flat, polyethylene washers on the post and needle were
positioned above and below the head to prevent vertical
movement.

Insertion of the needle through the anesthetized squid was
rapid and required minimal handling of the animal. Individ-
uals of this species become nearly transparent under anes-
thesia, making the buccal mass and the brain cartilage
readily visible. Needle placement was verified after the
experiment by examination of the location of the needle
entrance and exit wounds.

Tethered squid were transferred to the video arena (0.4 m
long by 0.2 m wide by 0.15 m deep) filled with aerated
23 °C artificial seawater and were allowed to recover. Teth-
ered squid remained alive and in apparent good health for
up to several hours, though most squid were tethered for
fewer than 15 min.

Critic/tie of tethering

Although tethering is an invasive technique, there were
several indications that it was not unduly traumatic to the
squid. First, tethered squid behaved similarly to the animals
in the holding tank. Both the tethered and free-swimming
squid spent most of the time hovering using the fins and
low-amplitude jets. Second, unlike squid that are in distress
or startled, more than 90% of the tethered animals did not
eject ink. Third, the chromatophore patterns of tethered
squid did not differ qualitatively from the patterns exhibited
by the free-swimming squid in the holding tank. Finally,
squid that were untethered and returned to the holding tank
swam normally and could survive for several hours. It is not
known how Ions; these animals could have survived, be-

cause all the animals were killed for histological analysis
after the day's experiments were completed.

Tethering did, however, affect two aspects of swimming
behavior. Tethered squid ( 1 ) performed escape jets with
higher frequency and (2) performed more consecutive es-
cape jets than the free-swimming squid in the holding tank.
It is possible that the tethering apparatus may have affected
mantle kinematics by restricting the flow of water out of the
funnel. This is unlikely because the post was between 30%
and 50% of the minimum funnel aperture in hatchlings and
less than 20% of the minimum funnel diameter in the largest
animals studied. In addition, the tethering apparatus did not
contact the funnel during the experiments.

Mantle kinematics

Escape-jet behavior was recorded from above with a
Panasonic AG-450 S-VHS professional video camera. The
camera was adjusted so that the squid filled as much of the
field of view as possible. To maximize the measurement
resolution, the animal was oriented with the long axis of the
mantle vertical in the video field (i.e., perpendicular to the
video scan lines). Though the animals were free to rotate
around the tether during the experiments, most remained
near the original orientation. The frame rate of the camera
(60 video fields per second) was more than 10 times faster
than the observed frequency of the mantle jetting cycle. To
reduce image blur, the high-speed shutter of the camera was
set at 1/1000 s. Illumination was adjusted by means of a
variac to the minimum level necessary to provide good
contrast between the squid and the background.

Videotapes were analyzed using a Panasonic AG-1980P
professional S-VHS videocassette recorder to identify es-
cape-jet sequences suitable for digitizing. Only those se-
quences in which the mantle remained in the same orienta-
tion (i.e.. the mantle remained nearly horizontal and did not
twist relative to the head) were digitized. Individual video
fields were digitized using an Imagenation (Beaverton, OR)
PXC200 frame-grabber card in a microcomputer.

Mantle diameter changes during vigorous escape jets
were measured from digitized video fields using morpho-
metrics software (SigmaScan Pro 4.0, SPSS Science, Chi-
cago, IL). Diameter at \A of the dorsal mantle length (DML)
was measured in each video field prior to the start of and
throughout the duration of an escape jet. The mantle diam-
eter at 'A DML (from dorsal mantle edge. Fig. 1) was
selected because the greatest amplitude mantle movements
occurred at that location in all squid examined. We normal-
ized the data by dividing the mantle diameter measured in
each video field by the resting diameter (=diameter of the
anesthetized squid at ]A DML) of the squid. Normalization
by the resting mantle diameter standardized the analysis of
mantle hyperinflation and mantle contraction data among
the squid and allowed for comparisons between animals of

ONTOGENY OF SQUID MANTLE KINEMATICS

157

different size. More than five escape-jet sequences were
analyzed from each animal. Only the sequences that yielded
the greatest mantle hyperinflation and the greatest mantle
contraction were reported.

For many of the escape-jet sequences, the mantle diam-
eter data were plotted against time. Time was estimated
from the video camera frame rate (approximately 0.017 s
per video field). To correct for differences in animal size,
the diameter change between consecutive video fields was
divided by the resting mantle diameter. The rate of mantle
contraction was determined by dividing the mantle diameter
change between successive video fields by 0.017 s. This
calculation yielded a set of incremental rates of mantle
contraction. The highest incremental rate was reported as
the maximum rate of mantle contraction for that animal.

The frequency of escape jets was calculated by dividing
the number of complete escape-jet cycles (the exhalant plus
the inhalant phases) by the time required to perform the
behavior. Time was estimated from the frame rate of the
video camera as above. Measurements were made only from
video sequences of squid that performed two or more escape
jets in rapid succession. Multiple measurements were made
for each squid, but only the highest calculated escape-jet
frequency was reported.

Morphometrics

The dorsal mantle length of anesthetized squid was mea-
sured to the nearest 0. 1 mm using calipers. We chose dorsal
mantle length as a measure of squid size because it is simple
to measure accurately and it correlates strongly with squid
wet weight (Segawa, 1987).

The volume of the mantle cavity was measured for most
animals after videotaping. Each squid was anesthetized
(Messenger et ul.. 1985) at 20 °C for 15 min to relax the
mantle musculature. The animal was then lifted from the
anesthetic by the arms so that the mantle cavity remained
filled with water. The exterior of the squid was gently
blotted dry and the animal weighed on an electronic balance
to the nearest 0.0001 g. The squid was returned to the water
and then lifted by the tip of the mantle so that water emptied
from the mantle cavity. The mantle was squeezed gently, in
the posterior to anterior direction, to aid draining of the
mantle cavity. The outside of the squid was blotted dry and
the animal weighed again. We calculated the volume of the
mantle cavity by dividing the difference in weight between
the two measurements by the density of seawater at 20 °C
( 1 .024 x 10 kg m- ). This procedure was repeated three to
five times for each squid, and the average mantle cavity
volume was recorded. We normalized the volume measure-
ments by dividing the mantle cavity volume by the wet
weight of the squid.

Mantle radius was measured in all the squid. Resting
mantle diameter was measured from digitized video frames

of the dorsal mantle of anesthetized animals. The mantle
was assumed to be cylindrical, and the diameter was mea-
sured at 'A DML. Mantle radius was then calculated from
the diameter data.

The thickness of the mantle wall was measured in 21
specimens. Anesthetized animals were decapitated and a
transverse slice of the mantle was made at one-third DML.
A digital image of the slice was captured using a dissecting
microscope, and the thickness of the mantle wall at the
ventral midline was measured using morphometrics soft-
ware.

The mantle wall thickness and radius were used to cal-
culate mantle circumferential strain during the escape jet.
The circumferential strain experienced during jet locomo-
tion at the midpoint in the thickness of the mantle wall was
calculated using the following equation from MacGillivray
et ol. (1999):

ec = 1 -[(/?, - V2t,)/(R, - V2t,)]

(1)

e^. is the circumferential strain, /?, is the initial ("resting")
outer radius of the mantle, Rt is the final outer radius of the
contracted mantle, t, is the initial thickness of the mantle
wall, and tf is the thickness of the contracted mantle wall.
The resting outer radius and mantle wall thickness, #, and
tt, were measured from the digital images. The final outer
radius of the contracted mantle, Rf. was measured from the
videotapes, and tf- was then calculated using the following
equation from MacGillivray et al. (1999):

tf=Rf- [Rj - t,(2R,- /,)

(2)

By convention, negative circumferential strain values in-
dicate contraction of the mantle, and positive values indicate
hyperinflation of the mantle.

Statistics

All correlations were made using the Spearman rank
order correlation (Sokal and Rohlf. 1981). This nonpara-
metric statistical test was used because the data for dorsal
mantle length were not distributed normally (Kolmogorov-
Smirnov goodness of fit test, P < 0.01 : Zar, 1996) due to
a sample bias toward smaller squid.

The mantle kinematics data were subdivided into the
life-history stages identified by Segawa (1987). This
scheme separates S. lessoniana into seven size classes based
on morphological and ecological characteristics. The squid
used in the experiments include four of Segawa's (1987)
life-history stages: hatchling (5 mm to 10 mm DML), juve-
nile 1(11 mm to 25 mm DML), juvenile 2 (26 mm to 40
mm DML). and young 2 (60 mm to 100 mm DML). After
subdivision into the appropriate life-history stage, the data
in each stage were compared with a one-way ANOVA.
Pairwise comparisons were made using the Student-New-
man-Keuls method of comparison (Zar. 1996). This analysis

158

J. T. THOMPSON AND W. M. KIER

"•5 £
01 4»

** * *

0 10 20 30 40 50 60 70 80 90

Dorsal Mantle Length (mm)

Figure 2. Ontogenetic change in escape-jet frequency. Each point
represents the average frequency of at least two consecutive escape jets.
Escape-jet frequency was inversely correlated with squid si/e (Spearman
rank order correlation coefficient, -0.75. P < 0.0001, n = 38).

was appropriate because the data in each stage were distrib-
uted normally (Kolmogorov-Smirnov goodness of fit test,
P > 0.4 for each stage; Zar, 1996).

The mantle wall thickness and mantle radius data were
log transformed and regressed against dorsal mantle length
using a least-squares technique (Zar. 1996). Student's r
distribution was used to test the slopes against the null
hypothesis slope of 1.0 (Zar, 1996).

Results

Escape-jet behavior

Tethered specimens of Sepioieutliis le.\.wniana escape
jetted spontaneously upon recovery from the anesthesia and
in response to visual stimuli outside the aquarium. Squid
escape jetted periodically during the experimental trials and
frequently jetted multiple times in succession. The number
of escape jets performed consecutively seemed to vary with
the size and age of the animal, with smaller, younger squid
performing more consecutive escape jets than larger, older
squid. Hatchling-stage squid (5 mm to 10 mm DML) often
jetted five times in rapid succession, paused briefly, and
then repeated the series of five jets two or three additional
times. Such behavior was never observed in squid larger
than about 25 mm DML (the juvenile 2 life history stage of
Segawa, 1987). The escape-jet frequency of smaller,
younger squid was higher than that of older and larger squid
(Fig. 2; correlation coefficient, -0.75, P < 0.0001. ;i =
38). For example, newly hatched squid performed four to
five escape -jet cycles per second, whereas two to three
escape-jet cycles per second were recorded for the largest
squid.

Mantle kinematics

The mantle kinematics during escape-jet behavior varied
both in an individual squid over time and among all the

squid studied. There were two distinct modes of mantle
movement immediately prior to the start of an escape jet. In
one mode, there was little mantle hyperinflation and the
mantle cavity was ventilated, presumably by contraction of
the circumferential musculature (Fig. 3A; see Gosline et ai,
1983). In the other mode, the mantle cavity was ventilated
primarily by mantle hyperinflation, presumably by contrac-
tion of the radial musculature (Fig. 3B: see Gosline et al.,
1983). There was no correlation between squid size and the
mode of mantle kinematics prior to the start of an escape jet.
Many of the squid studied exhibited both modes of mantle
kinematics, but the second mode (hyperinflation) was the
most common.

Regardless of age or size, the escape jet was stereotyped.
At the start, the mantle hyperinflated. filling the mantle
cavity with water (Fig. 4A). Next, the collar flaps closed and
the anteriormost edge of the mantle began to contract (anal-
ogous to a drawstring closing a bag) (Fig. 4B). The con-
traction of the anterior mantle edge was most noticeable in

0.0 0.2 0.4 0.6 0.8 1.0 14

0.0 0.2 0.4 0.6 0.8
Time(s)

Figure 3. Mantle diameter change over time. The horizontal line at 0.0
indicates the resting mantle diameter of the anesthetized squid. The neg-
ative numbers indicate mantle contraction, and positive values denote
mantle hyperinflation. (A) A hatchling stage squid (5.5 mm dorsal mantle
length, DML) that performed four consecutive escape jets (indicated by
asterisks). The arrow indicates a mantle hyperinflation immediately prior to
an escape jet. Note that the low-amplitude mantle movements prior to the
escape jets do not involve substantial mantle hyperinflation. (B) A single
escape jet from a young 2 stage squid «o mm DML). The arrow indicates
the mantle hyperinflation prior to the start of the escape jet. Note that the
lower amplitude mantle movements prior to the escape jet consist almost
entirely of mantle hyperinflation and do not involve substantial mantle
contraction.

ONTOGENY OF SQUID MANTLE KINEMATICS

159

Figure 4. Nonconsecutive digitized video frames from an escape jet by a 3-week-old specimen of Sepiii-
teuthis lessoiiiaiw (25-mm DML). (A) The squid immediately before the start of mantle contraction. The mantle
is fully expanded, and the mantle cavity is full of water. The bright region of reflection on the head (arrow) is
the plastic washer of the tethering apparatus. ( B ) The squid just after the start of the escape jet. The anterior edge
of the mantle (arrow) has contracted, and the remainder of the mantle is just beginning to contract. (C) The
mantle near its maximum contraction for the escape jet. The head is drawn back into the mantle cavity, and the
fins are folded along the body. I D) The end of the exhalant phase and the start of the inhalant phase of the escape-
jet. The tins (barely visible on right side) are unfurling and beginning to undulate. The head is maximally
withdrawn into the mantle cavity, and the anterior edge of the mantle is starting to flare (arrow) away from the
head.

squid larger than 40 mm DML (about 5 to 6 weeks post
hatching). One video field (about 17 ms) after the start of
anterior mantle-edge contraction, two events occurred: ( 1 )
the fins were folded against the ventral side of the mantle,
and (2) the remainder of the mantle began to contract
rapidly, expelling water from the mantle cavity through the
funnel (Figs. 4B, C). Nearly simultaneous with folding of

the tins, the head was drawn back into the mantle cavity,
presumably by the activation of the head retractor muscles.
Maximal head retraction was completed within two video
fields (about 34 ms) and was maintained until the end of the
exhalant phase of the jet (Fig. 4C). At the end of the
exhalant phase of the jet, the fins unfolded and began
undulating immediately. Concurrent with fin unfolding, the

160

J. T. THOMPSON AND W. M. KIER

-\I.1J

S -0.20

t; -0.25 •

+ , + +

, « . , ,

i -0.30 •

++£ + + ' + +

o -0.35 -

*+7+ -f+

* -0.40

+ +

S -0.50

-0 SS .

0 10 20 30 40 50 60 70 80 90

<™

U.JV

0.25 •

, .

0.20 •

-^ + +

0.15 •

+ +

+ T_ * + +

0.10 •
0.05 •

+ +j. 4^ + *
+ * 4+ ' +

0.00 •

+ + + +

-0.20
-0.25
-0.30
-0.35
-0.40
-0.45
-0.50
-0.55
-0.60

0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00

•—

=
=

0 10 20 30 40 50 60 70 80 90
Dorsal Mantle Length (mm)

Figure 5. Ontogenetic changes in mantle contraction and mantle hy-
perinflation during the escape jet. Each point represents the maximum
mantle diameter change (contraction or hyperinflation) for an individual
squid during an escape jet. (A) Mantle contraction and circumferential
strain versus dorsal mantle length. The plot shows a significant decrease in
the maximum mantle contraction of the escape jet during ontogeny (Spear-
man rank order correlation coefficient. 0.70. P < 0.0001. n = 55). (B)
Mantle hyperinflation prior to the start of an escape jet versus dorsal mantle
length. The plot shows a significant ontogenetic decrease in maximum
mantle hyperinflation prior to an escape jet (Spearman rank order correla-
tion coefficient. -0.49. P < 0.0001, n = 49). For A and B. the mantle
contraction and hyperinflation scales differ from the circumferential strain
scales because both mantle contraction and hyperinflation are measures of
the outer circumference of the mantle, whereas circumferential strain is a
measure of changes in mantle wall circumference at the midpoint of the
thickness of the mantle wall, and mantle thickness increases during con-
traction (because the mantle wall is isovolumetric).

anterior margin of the mantle flared outward (Fig. 4D). In
animals smaller than 20 mm DML, this flaring was usually
accompanied by even greater contraction of the mantle in
the anterior '/i of the mantle.

Ontogeny of mantle kinematics

A significant ontogenetic change in the amplitude of
mantle movement during escape-jet behavior was observed
in 5. lessoniana. In smaller and younger animals there was
a greater change in mantle diameter than in larger, older
squid (Fig. 5A: Spearman rank order correlation coefficient,
0.7. P <<C 0.001, n = 55). In newly hatched squid, the
mantle contracted 41% to 49% during the escape jet, but it
contracted by only 25% to 32% in larger animals (Fig. 5A;
see Table 1 for descriptive statistics based on life history
stage). After dividing the data into the life-history stages
described by Segawa ( 1987), the average mantle contraction
during the escape jet of the hatchling stage (5 mm to 10 mm
DML) squid was significantly greater than in any other life
history-stage measured (one-way ANOVA, Student-New-
man-Keuls test, P < 0.05; Table 1). In addition, the
average mantle contraction of squid in the juvenile 1 stage
( 1 1 mm to 25 mm DML) was significantly larger than squid
in the young 2 stage (60 mm to 100 mm DML) (one-way
ANOVA, Student-Newman-Keuls test, P = 0.05; Table 1).

There was a significant ontogenetic decrease in the
amplitude of mantle hyperinflation prior to the start of the
escape jet (Spearman rank order correlation coefficient,
-0.49, P « 0.001. n = 49). In newly hatched S. lessoni-
(inu, the mantle hyperinflated between 15% and 27%, but
it hyperinflated only 1% to 15% in older, larger animals
(Fig. 5B; see Table 1 for descriptive statistics). After
dividing the data into the life-history stages of Segawa
(1987). the average mantle hyperinflation prior to the
start of the escape jet was significantly greater in the
hatchling stage squid than in all the other stages (one-

Table 1

Comparison of mantle kinematics during the escape jet among squid divided into the life-history stages defined bv Segawa (1987)

Life-historv stage

Maximum contraction

Maximum hyperinflation

Maximum contraction rate (lengths/s)

Hatchling
Juvenile 1
Juvenile 2
Young 2

-0.40 ± 0.057(13)
-0.32 ± 0.037(23)*
-0.31 ± 0.040(8)
-0.28 ±0.024(9)*

0.18 ±0.072(14)
O.I I ±0.015(15)
0.086 ± 0.054 (11)
0.088 ± 0.038 (9)

8.6 ± 2.1 (14)
4.8 ± 1.2 (16)
3.8 ± 1.7(10)
3.8 ± 0.55 (9)

Values represent mean maximum mantle contraction, mantle hyperinflation, and mantle contraction rate during the escape jet plus or minus the standard
deviation of the mean. The number of squid in the sample is in parentheses. Maximum values for mantle contraction amplitude, mantle hyperinflation
amplitude, or mantle contraction rate for all squid in a life-history stage were pooled to calculate the mean and the standard deviation. In each column, the
mean value for the hatchling stage squid was significantly different from the mean for the juvenile 1. juvenile 2. and young 2 life-history stages (one-way
ANOVA on ranks. P < 0.05). The asterisks in the Maximum contraction column denote a significant difference in mantle contraction between the juvenile
1 and young 2 life-history stages (one-way ANOVA on ranks. P = 0.05). Other within-column comparisons of mantle kinematics were not significantly
different.

ONTOGENY OF SQUID MANTLE KINEMATICS

161

12 -

.2 •$

'•" C

10 •

•IH-

« s

** S-5

S •

•^

C ^^-

o 2

+**

u •£

on
£ c

6 •

**^* ^

4 •

4-+ +^ j* ^

1 ~

-1

+** +

0 •

10 20 30 40 50 60 70 80

Dorsal Mantle Length (mm)

Figure 6. Ontogenetic change in the maximum rate of mantle contrac-
tion. Each point represents the maximum rate of mantle contraction during
the escape jet for one individual. The maximum mantle contraction rate of
the escape jet decreased significantly during ontogeny (Spearman rank
order correlation coefficient. -0.76, P < 0.0001, n = 49).

way ANOVA. Student-Newman-Keuls test, P < 0.05;
Table 1).

The maximum rate of mantle contraction during the es-
cape jet was highest in newly hatched squid and declined
during ontogeny (Fig. 6; Spearman rank order correlation
coefficient, -0.76, P « 0.001, n = 49). The maximum
rate of mantle contraction varied from 7 to 13 mantle
circumference lengths per second in newly hatched squid
and from 3 to 5 lengths per second in the largest squid (Fig.
6). A one-way ANOVA among the life-history stages (Se-
gawa, 1987) indicated that hatchling stage S. lessoniana had
a significantly greater maximum rate of mantle contraction
during the escape jet than all other life history stages (Stu-
dent-Newman-Keuls test. P < 0.05; Table 1).

Morphometrics

Mass-specific mantle cavity volume decreased during
ontogeny (Fig. 7A). Despite the variation among squid of
similar size, there was a significant negative correlation
between mass-specific mantle cavity volume and dorsal
mantle length (Spearman rank order correlation coefficient,
-0.50. P = 0.002. n = 36).

The thickness of the mantle wall increased during ontog-
eny (Fig. 7B). The slope of the regression, 1.29, was sig-
nificantly greater than 1 (Student's t test, P < 0.01 ).

Mantle radius also increased during ontogeny (Fig. 7C).
The slope of the regression, 0.85, was significantly less than
1 (Student's / test. P < 0.01 ).

Discussion

The escape-jet sequence

The general pattern of the escape jet did not vary with
squid age or size. However, the mantle kinematics during

an escape jet did change as the squid grew. These onto-
genetic changes in escape-jet kinematics may arise from
alterations in the neurophysiology (Gilly el ai. 1991),
muscle physiology (Preuss et «/., 1997), morphology
(Moltschaniwskyj, 1995), or mechanical properties of the
mantle. Furthermore, the changes may have implications

•S §0.8

S ai 0.6

1S°
0.0

f %1

4
A

0 10 20 30 40 50 60 70 80
^ Dorsal Mantle Length (mm)

£ 0.0

I -0.5 ]

3-LS

2.0

LogY= 1.3 LogX- 1.9
r2 = 0.98

0.0

0.5

» 1.5 -

•5
cS i.o

S 0.5

0.0

LogY=0.85 LogX - 0.44
r2 = 0.98

0.0 0.5 1.0 1.5 2.0

Log Dorsal Mantle Length (mm)

Figure 7. Ontogenetic changes in mantle morphometrics. (A) Mass-
specitic mantle cavity volume versus dorsal mantle length. Each point
represents the average of between three and five measurements of mantle
cavity volume for each squid ± the standard error of the mean. Dividing
the average mantle cavity volume for one squid by the wet weight of the
same animal normalized the data for the volume of the mantle cavity.
Mass-specific mantle cavity volume decreased significantly during ontog-
eny (Spearman rank order correlation coefficient, -0.50, P = 0.002, /t =
36). (B) Log mantle wall thickness versus log dorsal mantle length. The
equation tor the least-squares regression and the corrected r2 value are
listed at the upper left. The slope of the regression (1.29) was significantly
greater than 1.0 (Student's r test. P < 0.01 ), indicating a positive allo-
metric relationship between mantle wall thickness and mantle length. (C)
Log mantle radius versus log dorsal mantle length. The equation for the
least-squares regression and the corrected r~ value are listed at the upper
left. The slope of the regression (0.85) was significantly less than 1.0
(Student's / test, P < 0.01 ), indicating a negative allometric relationship
between mantle radius and mantle length.

162

J. T. THOMPSON AND W. M. KIER

for the mechanics of escape-jet locomotion during
growth.

Fatigue

The relative proportions of circumferential muscle fiber
types in the mantle of 5. lessoniana change during ontogeny
(Thompson. 2000: Thompson and Kier. 2001). Newly
hatched individuals of S. lessoniana have a larger propor-
tion of mitochondria-rich circumferential muscle fibers
(analogous to vertebrate red muscle fibers, see Bone et til.,
1981, and Mommsen et til.. 1981) than older, larger squid.
Preuss et al. ( 1997) reported a similar change during growth
in the relative proportion of circumferential muscle fiber
types in the mantle of another loliginid squid, Loligo opal-
escent. Preuss et al. (1997) suggested that the greater pro-
portion of mitochondria-rich circumferential mantle muscle
fibers made the hatchling squid more resistant to fatigue
than older, larger animals. The data from the present study
support their hypothesis. Small, young specimens of S.
lessoniana were able to perform more consecutive escape
jets and seemed to tire much less readily than their larger.
older counterparts. However, motivational differences be-
tween small and large squid may also affect jetting behav-
ior.

Newly hatched squid seem to rely more heavily on fre-
quent jet locomotion than do larger squid (Fields. 1965;
Hoar et al., 1994; Preuss et al., 1997). Two reasons have
been suggested for this tendency. First, the fins of newly
hatched squid are rudimentary relative to the adult fins
(Boletzky. 1974; Okutani, 1987; Hoar et til.. 1994). and it
has been proposed that these diminutive fins may not gen-
erate sufficient thrust for locomotion or hovering (Boletzky,
1987; Hoar et al., 1994). Second, most newly hatched squid
live in a fluid regime that is characterized by an intermediate
Reynolds number (estimated from data in Packard. 1969,
and O'Dor et al., 1986: see Jordan. 1992. and Daniel et ai.
1992. for further discussion of intermediate Re) and in
which the near parity of viscous and inertial forces inhibits
coasting after a jet. Unlike large squid that can perform a
single jet and then coast for a considerable distance, small
squid must jet continuously to locomote. Hence, there may
be an advantage in having a large proportion of the loco-
motor musculature specialized for fatigue resistance, partic-
ularly if jetting is the primary mode of locomotion. The
price for such specialization, however, may be a reduction
in the peak force produced by the mantle musculature
during contraction.

Mantle kinematics

The mantle cavity of a hatchling of S. lessoniana holds a
proportionately greater volume of water than the mantle
cavity of a larger squid (Fig. 7 A). In addition, a larger
proportion of this volume is ejected from a hatchling during

an escape jet (Fig. 5 A). Finally, the maximum rate of mantle
contraction during an escape jet is highest in a newly
hatched squid (Fig. 6). Taken together, these data imply that
mass flux (i.e.. the product of the density of water in the
mantle cavity and the volume rate of water flow out of the
mantle cavity) during the escape jet is proportionately
greater in hatchling than in larger, older individuals of S.
lessoniana.

We used the mantle kinematics and morphometric data to
calculate the relative mass flux during the escape jet in two
life stages of 5. lessoniana: a 5.5-mm-DML hatchling stage
and a 65-mm-DML young 2 stage. We modeled the mantle
as a cylinder with the "resting" wall thickness and radius
calculated from the regressions of the mantle wall thickness
(Fig. 7B) and mantle radius (Fig. 1C) data. To simplify the
calculations, we based them on a transverse slice of the
cylinder at ]A DML. We assumed that both the length of the
cylinder and the volume of the cylinder wall were constant;
thus, the cylinder-wall area of the slice was held constant
during the calculations. The initial mass-specific mantle
cavity volume for each squid was obtained from the data in
Figure 7A. We used the data for average mantle contraction
and the maximum rate of mantle contraction from Table 1 to
calculate the amplitude and rate of changes in mantle radius.
We used equation (2) to calculate the increase in mantle
wall thickness during the simulated jet. Finally, we calcu-
lated the relative mass (i.e., mass of water divided by mass
of squid) of water remaining in the mantle cavity during the
simulated jet at 25-ms intervals.

The calculations predict greater relative mass flux during
the escape jet in the hatchling stage squid than in the young
2 stage (Fig. 8 A). The average mass flux over the duration
of the exhalant phase of the escape jet in a hatchling stage
squid is about 2 times greater than that of an animal in the
young 2 stage (Fig. 8 A).

We calculated the average mass flux over the entire
duration of the exhalant phase of the escape jet for the 55
squid from Figure 5A. We modeled the mantle as a cylinder
of constant length with a mantle cavity volume determined
from Figure 7A. Using the mantle contraction data from
Figure 5A, and calculating changes in the thickness of the
mantle wall during the jet using equation (2), we calculated
the normalized mass of water in the mantle cavity (i.e., mass
of water divided by mass of squid) at the start and at the end
of an escape jet. The change in normalized mass was
divided by the duration of the exhalant phase of the jet to
give the average mass flux. The calculations show an onto-
genetic decrease in the normalized average mass flux of the
escape jet (Fig. 8B). Mass flux is proportionately highest in
hatchling squid and decreases rapidly during growth (Fig.
8B; correlation coefficient. -0.81, P < 0.0001, n = 55).
Does the predicted ontogenetic decline in relative mass
flux imply that the mass-specific thrust produced the escape
jet is highest early in ontogeny? Mass flux constitutes only

ONTOGENY OF SQUID MANTLE KINEMATICS

163

0.5

* 0.4 -I

5 0.3 J

0.1
0.0

— — Hatchling
Young 2

0.00 0.04 0.08 0.12 0.16

Time (sec)

01
:/
cs -

>• T

o
Z

X 2 0

IA 1 S
09 '-J

cs
1.0

0.5

0 10 20 30 40 50 60 70 80 90

Dorsal Mantle Length (mm)

Figure 8. Ontogenetic differences in the mass flux of the escape jet.
l A) Calculation of the relative mass of water remaining in the mantle cavity
versus the time course of the exhalant phase of one escape jet. The dashed
line represents a hatchling stage squid (5.5 mm dorsal mantle length) and
the solid line a young 2 stage animal (65 mm dorsal mantle length). For
each squid, the relative mass of water in the mantle cavity at 0.025-s
intervals during the escape jet was calculated using morphometric and
mantle kinematics data. See the Discussion for more details. The data for
each animal were fitted with a polynomial equation. The equations are. }' =
\HX- - 1.8X + 0.31 (r- = 0.99) for the young 2 squid and )' =
9.3X2 - 4.3.Y + 0.44 ( r: = 0.99 1 for the hatchling squid. The derivative
of each equation yields the average mass flux during the escape jet. The
average mass flux of the hatchling animal is approximately 2 times greater
than that for the larger animal. Note that mass flux is highest early in the
escape jet and diminishes at the end of the escape jet. (B) Calculation of the
average mass flux of the escape jet versus dorsal mantle length. Each point
represents the average mass flux of the exhalant phase of the escape jet
normalized by the wet weight of the squid. Normalized average mass flux
decreased significantly during ontogeny (Spearman rank order correlation
coefficient. -0.81. P < 0.0001. n = 55). See the Discussion for more
details.

a portion of the total jet thrust. Under steady-state condi-
tions, the instantaneous thrust produced during a jet is
proportional to the product of the instantaneous mass flux
and the instantaneous velocity of the water exiting the
funnel (averaged over the funnel aperture; Vogel. 1994).
Because it is likely that unsteady effects are important in jet
locomotion (Anderson and DeMont. 2000). an unsteady
term must also be included in an equation used to calculate
jet thrust (see Anderson and DeMont. 2000).

Under both steady-state and unsteady conditions, the
velocity of water exiting the funnel depends, in large part,
on the diameter of the funnel aperture. The funnel complex
is largest in newly hatched squid and decreases in relative
size during ontogeny (Boletzky. 1974; unpubl. obs. of S.
lessoniana and Loligo pealei). Unfortunately, funnel aper-
ture cannot be determined simply from the size of the funnel
of an anesthetized squid because it is a muscular structure
that changes shape during a single jet (Zuev, 1966; O'Dor,
1988; Anderson and DeMont, 2000). Furthermore, measur-
ing funnel aperture accurately during escape-jet locomotion
in small hatchling and juvenile squid is not currently feasi-
ble. Without data on the scaling of the funnel aperture and
dynamic changes in the aperture during a jet cycle, it is not
possible to make precise predictions about the mass-specific
thrust produced during the escape jet.

Whether jet thrust is generated by means of steady or
unsteady mechanisms, the greater relative mass flux pre-
dicted for hatchling-stage individuals of S. lessoniana could
allow a given thrust to be achieved with a relatively low jet
velocity. This may result in a hatchling stage squid having
a higher propulsion efficiency than an older, larger squid.
Anderson and DeMont (2000) calculated the hydrodynamic
propulsion efficiency (17) of the exhalant phase of the jet
stroke using the following equation of rocket motor propul-
sion efficiency:

7,= (2VVI)HV2+ V;).

(3)

where V is the velocity of flow past the squid and Vj is the
velocity of the jet relative to the squid. According to equa-
tion (3). the highest propulsion efficiency is achieved when
V, approximates V. Because V ' } must be greater than V for
a squid to accelerate, relatively lower jet velocity increases
propulsion efficiency. Anderson and DeMont (2000) em-
phasize, however, that the overall propulsion efficiency of
the jet includes both the efficiency of the jet stroke and the
efficiency of refilling the mantle cavity. A thorough onto-
genetic comparison of total hydrodynamic propulsion effi-
ciency must, therefore, also consider the efficiency of man-
tle cavity refilling.

In the intermediate Reynolds number fluid regime of the
newly hatched and juvenile squid, the generation of jet
thrust may not be represented accurately by existing equa-
tions. Previous theoretical treatments consider jet propul-
sion at high Reynolds numbers. In the absence of a math-
ematical model of jet locomotion at these intermediate
Reynolds numbers, measurements of the actual thrust pro-
duced are required. Therefore, direct measurements of the
thrust produced during an escape jet are needed to under-
stand how the ontogenetic changes in mantle kinematics
affect thrust.

164

J. T. THOMPSON AND W. M. KIER

Skeletal support and mantle kinematics

In many vermiform animals, the arrangement of connec-
tive tissue fibers in the body wall helps to control the shape
of the animal during locomotion and movement (e.g., Harris
and Crofton, 1957; Clark and Cowey. 1958). Similarly, the
ontogenetic changes in mantle kinematics during escape -jet
locomotion may result from ontogenetic alterations in the
organization and the mechanical properties of the skeletal
support system of the squid mantle.

In the mantle, skeletal support for locomotion is provided
by a complex arrangement of fibers of muscle and connec-
tive tissue (Ward and Wainwright, 1972; Bone et at., 198 1 ).
As described earlier, the connective tissue fibers are ar-
ranged in distinct networks: the inner tunic, the outer tunic,
and three distinct systems of intramuscular (IM) fibers —
IM-1. IM-2, and IM-3 (Ward and Wainwright. 1972; Bone
ct ai. 1981; for review, see Gosline and DeMont, 1985).
The fibers in all the IM systems are collagenous (Ward and
Wainwright, 1972; Gosline and Shadwick, 1983a; MacGil-
livray etal.. 1999), and the collagen fibers in IM-1 and IM-2
are hypothesized to antagonize the circumferential muscles
that provide power for locomotion.

The organization of collagen fibers in the outer tunic and
IM fiber networks of the mantle changes dramatically dur-
ing the ontogeny of S. lessoniana (Thompson, 2000;
Thompson and Kier, 2001). The IM-1 collagen fiber angle
(i.e., the angle of the collagen fiber relative to the long axis
of the mantle) is lowest in newly hatched squid and in-
creases exponentially during growth in squid up to 15 mm
DML. In squid larger than about 15 mm DML, IM-1 fiber
angle does not change substantially. IM-2 collagen fiber
angle (i.e., the angle of the collagen fiber relative to the
outer curvature of the mantle) is lowest in hatchlings and
rises exponentially until the squid reach 15 mm DML. In
animals larger than 15 mm DML, IM-2 fiber angle increases
only slightly with size. The correlation between these on-
togenetic alterations in connective tissue organization and
the mantle kinematics measured in this study is striking
(Fig. 9). Maximum mantle contraction (Fig. 5A), maximum
mantle hyperinflation (Fig. 5B), and maximum mantle con-
traction rate (Fig. 6) all change exponentially up to a dorsal
mantle length of about 15 mm.

It is possible that the allometric changes in mantle thick-
ness and mantle radius (Fig. 7B, C) might, at least in part,
underlie the ontogenetic change in IM-1 and IM-2 fiber
angle. We were unable, however, to detect a clear relation-
ship between the scaling of mantle thickness or radius and
IM-1 or IM-2 fiber angle.

Simple mathematical models (Thompson, 2000; Thomp-
son and Kier, 2001 ) of the ontogenetic changes in IM-1 and
IM-2 fiber angle predict significantly greater amplitude of
mantle movements during escape-jet locomotion in newly
hatched squid than in older, larger animals. The models.

IM Fiber Angle (degrees)

O UJ -t- t-n O ~~

r> o o o o c

+ o +

+® • + * *' * +
•*+ 1^** ++A A

A+

* A IM-1 Fiber Angle
•^ • IM-2 Fiber Angle
+ Mantle Contraction

-0.15
-0.20

-0.25
-0.30

-0.40
-0.45
-0.50
-0.55

0 10 20 30 40 50 60 70 80
Dorsal Mantle Length (mm)

Figure 9. Correlation between ontogenetic changes in intramuscular
collagen fiber angle and the mantle kinematics of the escape jet. The
maximum mantle contraction during an escape jet (data from Fig. 5A) is
indicated by the black crosses, and the scale is on the right side of the plot.
The gray triangles indicate the average fiber angle of intramuscular fiber
system I (IM-1) collagen fibers, and the gray circles denote the average
fiber angle of IM-2 collagen fibers. Note the correlation between IM-1 and
IM-2 average fiber angles and the mantle kinematics of the escape jet.

which consider only the fiber angle and probable mechan-
ical properties of the IM collagen fibers (see Gosline and
Shadwick. 1983b), predict that mantle circumference
changes up to —45% are possible in hatchling stage squid,
whereas changes of only -25% to -30% are possible in
squid at the young 2 stage (Thompson. 2000; Thompson
and Kier, 2001). The present study supports the predictions
of the models. Maximum mantle contraction during the
escape jet in hatchlings of S. lessoniana ranged from —41%
to —49% and from —27% to —32% in animals at the young
2 stage. Additional work on the ontogeny of the mechanical
properties of squid mantle collagen is necessary to under-
stand better the relationship among connective tissue orga-
nization, mantle mechanical properties, and mantle func-
tion.

Muscle mechanics

The maximum rate of mantle contraction was signifi-
cantly higher in newly hatched individuals of S. lessoniana
than in the larger animals (Fig. 6; Table 1 ). The shortening
velocity of muscle fibers depends on the load of the muscle,
the length of the thick filaments and sarcomeres, and the rate
of cross-bridge cycling (Schmidt-Nielsen, 1997). The load-
ing of the muscle fibers during jet propulsion is difficult to
measure. It should be possible, however, to measure the
contractile properties and myofilament dimensions of the
circumferential muscle from an ontogenetic series of squid
to examine the possibility of a change in performance of the
muscle during ontogeny.

Comparison of the shortening speed of the circumferen-
tial muscles calculated here for S. lessoniana with previous

ONTOGENY OF SQUID MANTLE KINEMATICS

165

measurements in adults of Allotcntliis sulnilcita and Sepia
officiiiulix are complicated by differences in temperature.
The unloaded shortening speed at 1 1 °C of the circumfer-
ential musculature in A. siibuhita and S. offcincilis was
measured to be 2.0 lengths per second and 1.5 lengths per
second, respectively (Milligan et cil., 1997). In S. lessoni-
i/nn. the maximum rate of mantle contraction at 23 °C
ranged from a high of 13 lengths per second in the hatch-
lings to 4 in the young 2 stage squid.

Although the Qn) for cephalopod muscles has not been
measured, previous work on type 1 and type 2 iliofibularis
muscle fibers from Xeno/nis laevis (Lannergren et nl., 1982)
revealed a (Pm of approximately 2 over this temperature
range. Thus, although the difference in measured shortening
velocity between the young 2 stage of S. lesxoniumi and the
adult of A siihiilata may be due to temperature, it is unlikely
that the much higher velocities measured in the hatchlings
are simply an effect of temperature. In addition, the circum-
ferential muscles are contracting against a load during an
escape jet. and thus the unloaded shortening velocity of
circumferential muscle in S. lessoniana will be higher than
the values reported here.

In conclusion, we have described significant ontogenetic
changes in the mantle kinematics of the escape jet in teth-
ered squid. These kinematic changes are correlated strongly
with alterations in the organization of the connective tissue
fibers in the mantle; furthermore, they may affect the mass
flux of the escape jet. An analysis of the mechanics of
escape-jet locomotion in an ontogenetic series of squid is
needed to better comprehend the implications of growth-
related changes in mantle kinematics. Such an analysis will
help us to understand the functional consequences of onto-
genetic changes in morphology and will provide insight into
the evolution of the form and function of hydrostatic skel-
etons.

Acknowledgments

This research was supported by NSF grants to W.M.K.
(IBN-9728707 and IBN-9219495). Grants and fellowships
to J.T.T. from the Wilson Fund, the American Malacologi-
cal Society, and Sigma Xi helped defray research expenses.
We thank L. Walsh at the NRCC for her expertise in
shipping squid cross-country. We thank J. M. Gosline. S.
Johnsen. J. Taylor. T. Uyeno, and two anonymous review-
ers for constructive comments and suggestions on an earlier
version of the paper.

Literature Cited

Anderson. E. J.. and M. E. DeMont. 2000. The mechanics of locomo-
tion in the squid Loligo pealei: locomotory function and unsteady
hydrodynamics of the jet and intramantle pressure. ./. Exp. Biol. 203:
2851-2863.

Boletzky, S. v. 1974. The "larvae" of Cephalopoda: A review. Tliala.\.\ia
Jugosl. 10(1/2): 45-76.

Boletzky, S. v. 1987. Juvenile behaviour. Pp. 45-60 in Cephalopod Life
Cycles. Vol. II: Comparative Reviews. P. R. Boyle, ed. Academic Press,
New York.

Bone, Q., A. Pulsford, and A. D. Chubb. 1VKI. Squid mantle muscle.
J. Mar. Biol. A.isoc. UK 61: 327-342.

Calder, \V. A. 1984. Size. Functinn. and Life Histon: Harvard Univer-
sity Press, Cambridge.

Chen, D. S., G. v. Dykhuizen, J. Hodge, and VV. F. Gilly. 1996.
Ontogeny of copepod prcdation in juvenile squid (Loligo opalescens).
Biol. Bull. 190: 69-81.

Choe, S. 1966. On the eggs, rearing, habits of the fry. and growth of
some Cephalopoda. Bull. Mar. Sci. 16: 330-348.

Clark, R. B., and J. B. Cowey. 1958. Factors controlling the change of
shape of certain nemertean and turbellarian worms. J. Exp. Biol. 35:
731-748.

Daniel, T. L., C. Jordan, and D. Grunbaum. 1992. Hydromechanics of
swimming. Pp. 1 7 — 19 in Advances in Comparative ami Environmental
Physiology. Vol. II. Mechanics of Animal Locomotion. R. M. Alex-
ander, ed. Springer- Verlag. New York.

Fields, W. G. 1965. The structure, development, food relations, repro-
duction and life history of the squid Loligo opalescens Berry. Fish Bull.
131: 1-108.

Gilly, W. F., B. Hopkins, and G. O. Mackie. 1991. Development of
giant motor axons and neural control of escape responses in squid
embryos and hatchlings. Biol. Bull. 180: 209-220.

Gosline, J. M., and M. E. DeMont. 1985. Jet-propelled swimming in
squids. Sci. Am. 252(1): 96-103.

Gosline, J. M., and R. E. Shadwick. 1983a. The role of elastic energy
storage mechanisms in swimming: an analysis of mantle elasticity in
escape jetting in the squid. Loligo opalescens. Can. J. Zoo/. 61:
1421-1431.

Gosline, J. M., and R. E. Shadwick. 1983b. Molluscan collagen and its
mechanical organization in squid mantle. Pp. 371-398 in The Mollusca.
Vol. I: Metabolic Biochemistry ami Molecular Biomechanics, P. W.
Hochachka. ed. Academic Press, New York.

Gosline, J. M., J. D. Sleeves, A. D. Harman, and M. E. DeMont. 1983.
Patterns of circular and radial mantle muscle activity in respiration and
jetting of the squid Loligo opalescens. J. Exp. Biol. 104: 97-109.

Hanlon, R. T., J. W. Forsythe, and S. v. Boletzky. 1985. Field and
laboratory behavior of "Macrotritopus larvae" reared to Octopus defil-
ippi Verany, 1851 (Mollusca: Cephalopoda). Vie Milieu 35(3/4): 237-
242.

Harris. J. E., and H. D. Crofton. 1957. Structure and function in the
nematodes: internal pressure and cuticular structure in Ascaris. J. Exp.
Biol. 34: 116-130.

Hoar, J. A.. E. Sim, D. M. Webber, and R. K. O'Dor. 1994. The role
of fins in the competition between squid and fish. Pp. 27-43 in
Mechanics ami Physiology of Animal Swimming. L. Maddock, Q.
Bone, and J. M. Rayner. eds. Cambridge University Press, Cambridge.

Jordan, C. E. 1992. A model of rapid-start swimming at intermediate
Reynolds number: undulatory locomotion in the chaetognath Suxittu
elegans. J. Exp. Biol. 163: 119-137.

Kier, W. M. 1996. Muscle development in squid: ultrastructural differ-
entiation of a speciali/ed muscle fiber t\pc ./. Morphol. 229: 271-288.

Lannergren. J., P. I imlblom, and B. Johansson. 1982. Contractile
properties of two varieties of twitch fibres in \cimpiis laevis. Acta
Phvsiol. Scand. 114: 523-535.

Lee, P. G., P. E. Turk, W. T. Yang, and R. T. Hanlon. 1994. Biolog-
ical characteristics and biomedical applications of the squid Sepioteu-
this lessoniana cultured through multiple generations. Biol. Bull. 186:
328-341.

MacGHIivray, P. S., E. J. Anderson, G. M. Wright, and M. E. DeMont.
1999. Structure and mechanics of the squid mantle. J. Exp. Biol. 202:
683-695.

166

J. T. THOMPSON AND W. M. K1ER

Marliave, .1. B. 1980. Neustonic feeding in early larvae of Octopus

</(.//<•/;!/ (Wiilker). Veliger 23(4): 350-351.

Messenger, J. B., M. Nixon, and K. P. Ryan. 1985. Magnesium chlo-
ride as an anaesthetic tor cephalopods. Comp. Btochem. Physiol. 82C:

203-205.
Million, B. J., N. A. Curtin, and Q. Bone. I 997. Contractile properties

of obliquely striated muscle from the mantle of squid (Alloteuthis

sii/n/laru) and cuttlefish (Sepia o/fifinalisl J. E.\p. Biol. 200: 2425-

2436.
Moltschaniwskyj, N. A. 1995. Changes in shape associated with growth

in the loliginid squid Phoiololigo sp.: a morphometric approach. Cwt.

J. Zool. 73: 1335-1343.
Mommsen. T. P., J. Ballantyne, I). Mac-Donald. J. Gosline, and P. W.

Hochachka. 1981. Analogues of red and white muscle in squid

mantle. Proe. Nat. Aead. Sci. USA 78: 3274-3278.
Movnihan, M., and A. K. Rodaniche. 1982. The behavior and natural

history of the Caribbean reef squid Sepioteuthis sepioidea. Fortsehr.

Verhalteijsforsch. Supplement 25. Verlag-Parey. Berlin. 151 pp.
O'Dor. R. K. 1988. The forces acting on swimming squid. 7. Exp. Biol.

137: 421-442.
O'Dor, R. K., P. Helm, and N. Balch. 1985. Can rhyncoteuthions

suspension feed? (Mollusca: Cephalopoda). Vic Milieu 35(3/4): 267-

271.
O'Dor, R. K., E. A. Foy, P. L. Helm, and N. Balch. 1986. The

locomotion and energetics of hatchling squid, lllrx i/leeehrosns. Am.

Malac. Bull. 4(1): 55-60.
Okutani, T. 1987. Juvenile morphology. Pp. 33-44 in Cephalopod Life

Cvcles. Vol. II: Comparative Reviews, P. R. Boyle, ed. Academic Press.

New York.
Packard, A. 1969. Jet propulsion and the giant fibre response of Loligo.

Nature 221: 875- S77.

Preuss, T., Z. N. Lebaric, and W. F. Gilly. 1997. Cost-hatching devel-
opment of circular mantle muscles in the squid Loligo opalcsccns. Biol.

Hull. 192: 375-387.

Schmidt-Nielsen, K. 1997. Animal Physiology. 5th ed. Cambridge Uni-
versity Press. Cambridge.
Segawa, S. 1987. Life history of the oval squid. Sepioteuthis lessoniana

in Kominato and adjacent waters central Honshu. Japan. J. Tokyo Univ.
Fish. 74(2): 67-105.

Shadwick, R. E., and J. M. Gosline. 1985. The role of collagen in the
mechanical design of squid mantle. Pp. 294-304 in Biology of Inver-
tebrate and Lower Vertebrate Collagens, A. Bairati and R. Garrone.
eds. Plenum Press. New York.

Sokal, R. R., and F. J. Rohlf. 1981. Biometry. 2nd ed. W. H. Freeman,
New York.

Stearns, S. C. 1992. The Evolution of Life Histories. Oxford University
Press, Oxford.

Sweeney, M. J., C. F. E. Roper, K. M. Mangold. M. R. Clarke, and S. v.
Boli't/.ky (eds.). 1992. Larval ami Juvenile Cephalopods: A Manual
For Their Identification. Smithsonian Contributions to Zoologv, No.
513.

Thompson, J. T. 2000. The ontogeny of mantle structure and function in
the oval squid. Sepioteuthis lessoniana (Cephalopoda: Loliginidae).
Ph.D. dissertation. University of North Carolina at Chapel Hill.

Thompson, J. T., and W. M. Kier. 2001. Ontogenetic changes in
fibrous connective tissue organization in the oval squid, Sepioteuthis
lessoniana Lesson. 1830. Biol. Bull. 201: 136-153.

\ illanueva, R., C. Nozais, and S. v. Boletzkv. 1995. The planktonic life
of octopuses. Nature 377: 107.

Vogel, S. 1994. Life in Moving Fluids. 2nd ed. Princeton University
Press. Princeton.

Ward, D. V., and S. A. Waimvright. 1972. Locomotory aspects of
squid mantle structure. ./. Zonl. lLond.) 167: 437 — 149.

Werner, E. E. 1988. Size, scaling, and the evolution of complex life
cycles. Pp. 60-81 in Size-Structured Populations. B. Ebenman and L.
Persson, eds. Springer- Verlag, New York.

Zar, J. H. 1996. Bit/statistical Analysis. Prentice Hall. Upper Saddle
River. NJ.

Zuev, G. V. 1966. Characteristic features of the structure of cephalopod
molluscs associated with controlled movements. Ekologo-Morfo-
logicheskie Issledovaniya Nektonnykh Zhivotnykh. Kiev. Special Pub-
lication. (Canadian Fisheries and Marine Services Translation Series
101 1. 1968).

Reference: Hiol. Bull. 201: 167-174. (October 2001)

Hydromineral Regulation in the Hydrothermal Vent
Crab Bythograea thermydron

ANNE-SOPHIE MARTINEZ1, JEAN- YVES TOULLEC2, BRUCE SHILLITO1,
MIREILLE CHARMANTIER-DAURES1, AND GUY CHARMANTIER1 *

1 Lahoratoire d'Ecophysiologie des Invertebres, EA 3009 Adaptation Ecophysiologique an cours de

I'Ontogenese, Universite Montpellier II, PI E. Butaillon, 34095 Monipellier cedex 05, France;

Laboratoire Biogenese des Peptides Isomeres, UMR Physiologic el Physiopathologie, Universite P. et

M. Curie, 7 Quai Saint-Bernard, 75252 Paris cedex 05, France; and Laboratoire de Biologic

Cellulaire et Moleculaire du Developpement, UMR 7622, Croupe Biologie Marine, UPMC.

7 Quai Saint-Bernard, 75252 Paris cedex 05, France

Abstract. This study investigates the salinity tolerance
and the pattern of osmotic and ionic regulation of Bytho-
graea thermydron Williams, 1980, a brachyuran crab en-
demic to the deep-sea hydrothermal vent habitat. Salinities
of 33%cr-35%c were measured in the seawater surrounding
the captured specimens. B. thermydron is a marine steno-
haline osmoconformer, which tolerates salinities ranging
between about 31"/cc and 42% c. The time of osmotic adap-
tation after a sudden decrease in external salinity is about
15-24 h. which is relatively short for a brachyuran crab. In
the range of tolerable salinities, it exhibits an iso-osmotic
regulation, which is not affected by changes in hydrostatic
pressure, and an iso-ionic regulation for Na+ and Cl . The
hemolymph Ca2 + concentration is slightly hyper-regulated,
K+ concentration is slightly hyper-hypo-regulated, and
Mg2 + concentration is strongly hypo-regulated. These findings
probably reflect a high permeability of the teguments to water
and ions. In addition to limited information about salinity
around hydrothermal vents, these results lead to the hypothesis
that B. thermydron lives in a habitat of stable seawater
salinity. The osmoconformity of this species is briefly dis-
cussed in relation to its potential phylogeny.

Introduction

Hydrothermal vents, first discovered in 1977 on the Ga-
lapagos Ridge, are unique deep-sea habitats. They are char-

Received 2 January 2001; accepted 9 June 2001.
* To whom correspondence should be addressed. E-mail: charmantierw
univ-montp2.fr

acterized by variable and extreme conditions of some phys-
icochemical parameters, in particular by high temperature,
high sulfide and metal content, high level of carbon dioxide,
low level of oxygen and low pH (Truchot and Lallier, 1998;
Sarradin et ai, 1998, 1999). To live in this environment,
biological communities associated with the vents have de-
veloped behavioral, physiological, morphological, and re-
productive adaptations such as symbiosis (Fisher, 1990),
physiological and biochemical systems for sulfide detoxifi-
cation (Powell and Somero, 1986; Cosson and Vivier, 1997;
Geret et ai, 1998; Truchet et ai, 1998). behavioral and
molecular responses to high temperature (Dahlhoff et ai,
1991; Dixon et ai, 1992; Segonzac et ai, 1993; Desbru-
yeres et ai, 1998; Fisher. 1998), and specialized sensory
organs to locate hot chimneys (Jinks et ai, 1998).

Among this vent fauna live endemic brachyuran decapod
crustaceans (superfamily: Bythograeoidea Williams, 1980;
family: B\thograeidae Williams, 1980; genera: Bythograea
Williams, 1980: Cyanagraea de Saint Laurent, 1984; Seg-
onzacia Guinot, 1989; Aitstinograea Hessler and Martin,
1989) (Tudge et ai. 1998). They have been found in all the
known hydrothermal vents — Bythograea and Cyanagraea
in the East Pacific, Austinograea in the West Pacific, Seg-
onzacia in the mid- Atlantic (Tunnicliffe et ai, 1998). In-
formation published on these brachyurans includes studies
on their biogeography and evolution (Hessler and Wilson,
1983; Newman, 1985: Tunnicliffe. 1988). reproductive bi-
ology and larval development (Van Dover et ai, 1984.
1985; Epifanio et ai, 1999), and ecology and distribution
(Van Dover, 1995; Guinot and Segonzac, 1997). Probably

167

168

A.-S. MARTINEZ ET AL.

due to the difficulty of getting live specimens, physiological
studies are scarcer and have addressed aspects of respiration
(Lallier et ill., 1998), sulfide detoxification (Vetter el <;/.,
1987), and temperature or pressure effects on the mitochon-
dria, heart rate, or oxygen consumption rate (Mickel and
Childress, 1982a,b; Dahlhoff et «/., 1991) of these crabs.

To our knowledge, no information is available on the hy-
dromineral metabolism of the hydrothermal vent animals and
particularly of the brachyuran crustaceans. Salinity is one of
the main environmental factors exerting a selection pressure on
aquatic organisms, and the successful establishment of a spe-
cies in a given habitat depends on the ability of the organisms
to adapt to. among other factors, the typical level and variations
in salinity (Charmantier, 1998). This major adaptive process is
achieved through different behavioral or physiological mech-
anisms. Osmoregulation is one of the most important of these
mechanisms in some animal groups, including crustaceans. It
has been explored in the adults of numerous crustacean species
(reviews in Mantel and Fanner. 1983: Pequeux. 1995).

The present study has been conducted with one species of
bythograeid crab from hydrothermal vents. Bythograea
t/ienuyilron Williams. 1980. This crab is the most fre-
quently observed [density about 20 individuals per nr (Gui-
not and Segonzac, 1997)] and captured species among
brachyuran crustaceans on the East Pacific sites (Guinot,
1989). It is found predominantly in the warm water
(>20°C) suiTounding mussels and vestimentiferans on
which it feeds, and also at the periphery of the vent areas
where temperature is about 2°C (Grassle, 1986. cited by
Epifanio et a!., 1999). These habitats, influenced by the
spatially and temporally variable input of hydrothermal
fluid, are greatly variable over short time and distance.
Information on their salinity does not exist or is unpub-
lished. It is thus unclear whether the salinity of the water
surrounding the vents is as stable as the deep-sea water
environment or is variable under the influence of the hy-
drothermal fluid. Physiological studies have indicated that
adults of B. thermydron are tolerant of wide variations in
temperature, dissolved oxygen, and hydrogen sulfide
(Mickel and Childress, 1982a,b; Vetter et al.. 1987; Airries
and Childress. 1994). but their ability to tolerate salinity
variation and to osmoregulate is not known. The objectives of
the present study were thus to evaluate the salinity tolerance
and the pattern of osmoregulation of B. thennydron. The
salinity of the natural habitat of the crab was also measured. As
the hemolymph osmolality of crustaceans is mostly established
by inorganic ions (essentially Na+ and Cl ) (Pequeux, 1995),
the ionic regulation of this crab was also studied.

Materials and Methods

Animals

Adults of Bvthograea tliennvilron were collected by the
submarine Nuntilc. using resin watertight containers (about

1 X 0.5 X 0.5 m). on the East Pacific Rise (EPR) on the
13°N and 9°N sites [12°46-50'N, 103°57'W and 9°50'N,
104°17'W (Tunnicliffe et al., 1998)]. at a depth of about
2500 m. during the HOPE 99 mission in May 1999. Only a
small number of crabs were available, which resulted in 3 to
10 individuals for each experimental condition. As this
species seems incapable of long-term survival outside the
high-pressure environment of the deep sea (Mickel and
Childress. 1982a; Airries and Childress. 1994). most of the
crabs were transferred into aquaria with running aerated
Pacific surface seawater as soon as they reached the ship
Atalante, and they were used in the following hours for
experiments conducted on board, at atmospheric pressure, at
a water temperature of 13°C. Some of them were also
exposed to high pressure (see below). Crab cephalothoracic
widths were 6-8 cm. Their molt stages (Drach, 1939) were
not checked, but soft (post-molt) crabs were not used in
experiments.

Ambient salinity

Water samples from the depth of the Riftia pachyptila
ring on the 13°N and 9°N EPR sites were collected in
750-ml titanium syringes manipulated by the Nantile. The
water osmolality in mosm/kg was measured on an automatic
micro-osmometer ( Wescor Varro 5520). The corresponding
values of salinity in parts per thousand were calculated by
interpolation of data according to Weast (1969).

Preparation of media

Dilute media were prepared by adding fresh water to
Pacific surface seawater (1002 ± 2 mosm/kg: approxi-
mately 34.6%p), and high-salinity media were prepared by
adding ocean salts (Wimex, Germany) to seawater. Salini-
ties were expressed as osmolality (in mosm/kg) and salt
concentration (in parts per thousand). The osmolality of the
media was measured with a Wescor Varro 5520 micro-
osmometer. Media with the following osmolalities and cor-
responding salinities were prepared: 740 mosm/kg (25.4%c>),
800 (27.5). 900 (31.0). 1002 (34.6). 1100 (38.2), 1200
(41.9), 1300 (45.7). Experiments were conducted at 13°C in
40-1 aerated aquaria that were kept in the dark except at the
time of sampling, when light was briefly necessary.

Salinity tolerance

The objective of the experiment was to estimate the
survival time of the crabs at different salinities. The crabs
were transferred directly from seawater to the experimental
media. Observations were made and dead individuals were
removed 1, 2. 3, 5. 6. 12. 15, and 24 h after the beginning
of the tests. The absence of body movement after repeated
touches with a probe was considered as a proof of death.

OSMOCONFORMITY IN ft THERMYDRON

169

Hydn mineral regulation

Acclimation lime. To estimate the time necessary for
hemolymph osmolality stabilization following a decrease in
salinity, the crabs were first transferred from seawater ( 1002
mosm/kg), into a 740-mosm/kg medium. Hemolymph sam-
ples were taken from surviving animals after 0, 2.45, 5, and
15 h in the dilute medium. As 75% of the animals were dead
at 15 h and 100% shortly afterward, a second experiment
was conducted in an 800-mosm/kg medium. Survival was 60%
at 12 h and 17% at 24 h. Hemolymph samples were taken from
the surviving crabs after 0, 1.2, 3, 6, 12. 24. and 48 h.

Osmotic regulation. The hemolymph osmolality of some
crabs was measured as soon as they were brought on board.
The crabs were then transferred to the different media, and
their hemolymph osmolality was remeasured after a period
of osmotic stabilization in each medium; the length of this
period was determined from the results on adaptation time.
A similar experiment was conducted under high pressure, at
15°C. The crabs were immersed in an 800-mosm/kg me-
dium, in individual 400-ml containers set in a 19-1 pressur-
ized tank called "Incubateur Pressurise pour 1'Observation
en Culture d'Animaux Marins Profonds" (IPOCAMP)
(Shillito, unpub.). The crabs were subjected for 13 h to a
pressure of 260 bars, which approximates the pressure at the
site of capture. The hemolymph was then sampled and its
osmolality was measured.

For sampling, the crabs were rinsed with deionized water
and dried with absorbent paper. Hemolymph was sampled
with a hypodermic needle mounted on a syringe and in-
serted at the basis of a posterior pereiopod. The osmolality

of a 10-ju.l sample of hemolymph was immediately mea-
sured on the Wescor Varro 5520 micro-osmometer.

Ionic regulation. Hemolymph from the same samples
was quickly diluted to 25% in deionized water, stored in
Eppendorf tubes, and kept at -80°C. After transport to the
Montpellier laboratory in liquid nitrogen, the hemolymph
and media samples were dissolved in deionized water to the
appropriate volume, and their ionic contents were deter-
mined using an amperometric Aminco-Cotlove chloridime-
ter for the titration of Cl~, an Eppendorf flame photometer
for Na+, K+, Ca2 + , and a Varian A A- 1275 atomic absorp-
tion photometer for Mg2 + .

Statistical analysis

Statistical comparisons of experimental data were per-
formed by one-way analysis of variance (ANOVA) (Sokal
and Rohlf, 1981) by using the software StatView 4.02
(Abacus Concept, Inc.).

Results

Ambient sal/nitv

The salinity measured from bottom seawater samples was
996-1007 mosm/kg at the 13°N EPR site, and 950
mosm/kg at the 9°N EPR site.

Salinity tolerance

The survival rates of adults of Bythograea thermydron in
Figure 1 were different according to salinity and decreased

100 -
80 -

,— s

~ 60 -

£ 40-

20 -
0

Bythograea
thermydron

-» — 740 mosm/kg 25.4 %o

-0—800 " 27.5 "

••a- 900 " 31.0 "

-A- 1002 " 34.6 "

-• — 1100 " 38.2 "

-•- 1200 " 41.9 "

-o-- 1300 " 45.7 "

6 12

Time (h)

Figure 1. Bythograea thermydron. Survival rate (in %) at different salinities according to the time of
exposure. Number of crabs per condition at the start of the experiment: 3 to 10.

170

A.-S. MARTINEZ ET AL.

with the time of exposure (Fig. 1). They decreased sharply
to less than 25% within 15-24 h at the highest (1300
mosm/kg) and lowest (740. 800 mosin/kg) salinities. Sur-
vival was higher in seawater (1002 mosm/kg) and in salin-
ities ranging from 900 to 1200 mosm/kg.

Hydromineral regulation

Acclimation time. The time of adaptation after a sudden
change in salinity was evaluated at two low salinities (Fig.
2). In both media, the hemolymph osmolality decreased
sharply within 12 h. After 15 h in the 740-mosm/kg me-
dium, hemolymph osmolality had decreased to 805 mosm/
kg — that is, to about 65 mosm/kg above the medium osmo-
lality. As all crabs had died before 24 h, it was not possible
to determine whether hemolymph osmolality had entirely
stabilized at 15 h. After a transfer to the 800-mosm/kg
medium, the hemolymph osmolality stabilized within 24 h.
Its mean values were respectively 817 and 808 mosm/kg (no
significant difference) after 24 h and 48 h in this medium. In
subsequent experiments, the time of exposure to different
media was based on these results and was kept in general at
15-24 h.

Osmotic regulation. Upon the arrival of the crabs on
board the ship following their transfer from the bottom, their
hemolymph osmolality was 1025 ± 4 mosm/kg (n = 18)
and 984 ± 12 mosm/kg (/i = 29) at the 13°N EPR and 9°N
EPR sites respectively. The ability of the crabs to osmo-
regulate was then evaluated in the range of tolerable salin-
ities between 900 mosm/kg and 1200 mosm/kg. The crabs
osmoconformed in the whole range of tested salinities (Fig.

3 A). The hemolymph osmotic concentration was close to
that of the medium, different from it by only 9 to 22
mosm/kg, 15 mosm/kg on average.

The hemolymph osmolality was also measured in crabs
maintained in the 800-mosm/kg medium, under a pressure
of 260 bars. The mean value of hemolymph osmolality
following this treatment for 13 h was 860 ± 9 mosm/kg
(;/ = 3), not significantly different from the value of 856 ±
6 mosm/kg (/; = 3) in control crabs kept in the same
medium for 13 h under atmospheric pressure.

Ionic regulation. The results concerning hemolymph ion
concentrations in the different media are given in Figure
3B-F. In seawater, Na+ and CP were the main osmoeffec-
tors in hemolymph since they accounted for about 95% of
the total hemolymph osmolality, and this trend was retained
in all media. The hemolymph Cl~ concentration followed
that of the medium in the whole range of tolerable salinities.
It tended to be slightly hypo-regulated in most media (Fig.
3B). Na+ regulation was iso-ionic; hemolymph Na+ con-
centration constantly remained slightly above that of the
medium, by 8 to 23 mEq Na+/F ' (Fig. 3C). K+ was
slightly hypo-regulated (by approximately 2.5 mEq K+/
I"1) in the media in which concentrations were above 10.5
mEq K+/l~' (900 mosm/kg). and it was slightly hyper-
regulated (by approximately 3.5 mEq K+/r') in the lowest
salinity (800 mosm/kg, 9.3 mEq K+/r') (Fig. 3D). Hemo-
lymph Ca2+ concentration was slightly hyper-regulated (by
1 .2 to 3.6 mEq Ca2+/r ' ) at most tested salinities (Fig. 3E).
Hemolymph Mg2 + concentration was strongly hypo-regu-
lated (by about 33 to 57 mEq Mg2+/r') over the entire
ransje of salinities (Fis. 3F).

1100

=*1000

oa
O

—

_£>

900 -

800 -

700

Bythograea thermydron

•800 mosm/kg
- 740 mosm/kg

24
Time (h)

Figure 2. Bythogrueu thermydron. Change in hemolymph osmolality according to the time after rapid
transfer from Pacific surface seawater (1002 ± 2 mosm/kg) to dilute media at 740 mosm/kg and 800 mosm/kg.
Error bars: mean ± SD; ;;: 4 to 6 individuals.

OSMOCONFORMITY IN «. THERMYDRON

171

1300
1200
1100
1 1100
100

O.P.

700 800 100 1000 1100 1200 1300

Medium (mosm/kg)

400 500 600
Medium (mEq/1)

300 400 500 600

Medium (mEq/1)

Ca'

Medium (mEq/1)

Figure 3. Byihnitnu'ci thermydron. Variations in hemolymph osmola-
lit\ (A: O.P.) (osmotic pressure in mosm/kg) and ionic concentrations (B
to F) (in mEq • 1~ ' ) after hemolymph osmolality stabilization (about 15-24
h), in relation to the osmolality or ionic concentration of the medium. Time
of exposure to the different media was 24 h (A) or 15-24 h (B to F). Error
bars: mean ± SD; ;;: 3 to 12 individuals; isoconcentration lines are drawn.

Discussion

Salinity tolerance

The limited number of available animals and lack of time
and space on board the ship prevented long-term tolerance
experiments. Specimens of Bythograen thermydron sur-
vived for 24 h in a narrow range of salinities ranging from
about 3 \7(( to 427cc. These crabs, unable to withstand a great
extent of salinity fluctuations, are thus stenohaline animals.
They share this feature with other species of decapods
whose habitat is most often restricted to seawater, for ex-
ample, the Majidae. the Cancridae. and the Calappidae
(review in Mantel and Farmer, 1983: Pequeux, 1995).

Acclimation time

In B. thermydron, the time required to reach an osmotic
steady-state after a sudden decrease in external salinity was

about 15 to 24 h. This is short for a brachyuran crab, similar
to the 15 h required for osmotic equilibration in osmocon-
formers such as the Majidae Maja sp. and Hyas sp. trans-
ferred to 75% seawater (Prosser and Brown, 1965). Osmotic-
adaptation requires longer times in strongly osmoregulating
species, such as 48 h in osmoregulating crabs (Charmantier,
1998) and up to 96 h in crayfish (Susanto and Charmantier,
2000). The short acclimation time found in B. thennvdron
probably indicates a relatively high exchange of water and
ions between the organism and the external medium and a
high permeability of the body surface in this species; it also
reflects the weak salinity stress applied.

Hydromineral regulation

B. thermydron osmoconformed over the narrow range of
tolerable salinities. When salinity varied, the hemolymph
osmolality tended to follow the external osmolality, with a
slight positive difference of only about 15 mosm/kg. This is
probably due to the colloid osmotic pressure of plasma
proteins. B. thermydron is therefore an osmoconformer like
the Majidae Libinia emarginatu, Pugettia producta (Mantel
and Farmer, 1983), Maja sp. (Potts and Parry, 1963). and
Chionoecetes sp. (Mantel and Farmer, 1983; Hardy et ai,
1994); the Cancridae Cancer antennarius (Jones, 1941;
Gross, 1964) and C. pagurus (Pequeux, 1995); and the
Calappidae Calappa hepatica (Kamemoto and Kato. 1969).
As already noted by different authors (reviewed in Mantel
and Farmer, 1983; Pequeux. 1995), osmoconformity does
not permit survival at salinities widely different from sea-
water, and these osmoconformers are marine stenohaline
species. As in other crustaceans, the osmolality of the he-
molymph of B. thermydron was mostly due to inorganic
ions, essentially Na+ and Cl~ (Pequeux. 1995), which ac-
counted for about 95% of the total hemolymph osmolality.
The regulation of these ions was almost iso-ionic. In these
crabs, hemolymph Cl ~ and Na+ concentrations respectively
represented about 93% and 102% of the same ion concen-
trations in the medium. There is thus a slight excess of Na+
and a slight deficit of Cl~ in the hemolymph compared to
the medium, as is noted in other osmoconformers such as
the Majidae Libinia emarginata (Gilles. 1970), Maja sp.
(Potts and Parry, 1963). and Chionoecetes opilio (Hardy et
ai. 1994); the Cancridae Cancer antennarius (Gross, 1964);
and the Calappidae Calappa hepatica (Spencer et ai, 1979).

In B. thermydron, the hemolymph Ca2+ concentration
was slightly hyper-regulated, as noted by Prosser (1973) in
marine crustaceans. K+ concentration was slightly hyper-
hypo-regulated. However, among crustaceans, K+ is often
found in higher concentration in the hemolymph than in the
medium (Mantel and Farmer, 1983).

In B. thermydron, Mg~+ concentration was strongly
hypo-regulated. The concentration of this ion was about 44%
of that found in the medium, a percentage included in the

172

A.-S. MARTINEZ ET AL.

standard range of hemolymph Mg2 + concentration for
brachyurans, that is, between 20% and 80% of the medium
concentration (Prosser, 1973). As in several species of
crabs, Mg2 + might be excreted through the antennal glands
(Morritt and Spicer, 1998). Other osmoconformers such as
Maja squinado or Hyas sp.. which are relatively "unrespon-
sive" (slow-moving) species, have higher hemolymph
Mg2+ concentration (about 80% of that of seawater) (Rob-
ertson, 1960; Frederich et ai, 2000). B. thermydron exhibits
n hemolymph Mg"+ concentration closer to that of more
"active" crabs such as Carcinus nmenas and Pachygrapsits
mannoratus, in which the ion concentration is below 50%
of that found in the medium (Robertson, 1960; Frederich et
a/., 2000). This fact can be related to the active locomotor
behavior of B. thermydron (Williams. 1980; Guinot. 1988;
Guinot and Segonzac, 1997), which is evident in visual
observations and video monitoring (Jean- Yves Toullec,
pers. obs.) that show the crabs frequently moving on chim-
neys, in and out of the warm areas, and among the vesti-
mentiferans or mussels on which they feed. In addition,
these results show that the crabs had retained a strong ability
to hypo-regulate Mg2 + in their hemolymph after their trans-
fer to the surface and one or two days of exposure to
different media. Thus, their osmoconformity and their Na+
and Cl iso-regulation most probably result from a specific
pattern and not from damage to the integument or serious
stress due to the pressure change associated with bringing
the crabs to the surface.

Exposure to high pressure did not affect the hemolymph
osmolality of B. thermydron exposed to low salinity, when
compared to crabs kept at atmospheric pressure. In these
deep-sea hydrothermal crabs, osmoconformity thus appears
to be unaffected by a change in hydrostatic pressure. This
contrasts with the few tested epibenthic crabs in which
osmotic and ionic regulation may vary in relation to pres-
sure. For instance, short-term exposure (1-3 h) to pressure
of 50-100 bars significantly affected the concentration of
the inorganic ions (Na+, K+, Cl~, Ca2 + , Mg2 + ) in hemo-
lymph of Carcinus maenas (Pequeux and Gilles, 1984). but
changed only the Ca2+ content of the hemolymph in Erio-
cheir sinensis (Sebert et ai, 1997).

Ecological implications

Because B. thermydron is a marine stenohaline osmocon-
former, we may hypothesize that this species occupies a
deep hydrothermal habitat where salinity is stable and close
to that of seawater. This hypothesis has been verified in the
present study through direct measurements of the ambient
salinity. The salinity of the hydrothermal water directly
measured on samples taken on the 9°N and 13°N EPR was
approximately 32.7%c to 34.39rc-34.7%r. These values are
close to the salinity of standard Pacific seawater, 34.62%o
(Ivanoff, 1972). The osmoregulation pattern of B. thermy-

dron confirms that these crabs live in an environment where
salinity is stable and close to 33%o-35%c. In addition, these
values of salinity add to the knowledge of the ambient
parameters of the deep-sea hydrothermal environment
(Mid-Atlantic Ridge or East Pacific Rise) where Ca2+,
Mg2 + , Cl~, PO43~. NO2 , NH/, NO3~, and NO2~ con-
centrations have been measured (Truchot and Lallier, 1998;
Sarradin et al., 1998. 1999).

Phytogeny and osmoregiilatory adaptation

The phylogenetic origin of the Bythograeidae is disputed
(Delamare Deboutteville and Guinot, 1981; Guinot, 1988,
1990). According to Williams ( 1980). Tudge et al. (1998),
and Steinberg et al. (1999), Bythograea thermydron
exhibits some similarities to Potamoidea (Potamidae),
Portunoidea (Portunidae), and Xanthoidea (Goneplacidae;
Xanthidae; Trapeziidae). The marine stenohaline osmocon-
former B. thermydron may thus have derived from families
consisting mainly of crab species that are able to strongly
osmoregulate (Jones, 1941; Shaw, 1959; Robertson, 1960;
Ballard and Abbott, 1969; Kamemoto and Kato, 1969;
Harris and Micallef, 1971; Taylor et al., 1977; Birchard et
al., 1982; Blasco and Forward, 1988; review in Mantel and
Farmer, 1983). During its evolution, B. thermydron would
have lost its ancestor's osmoregulatory ability, which had
become superfluous in an environment where salinity is
stable. A similar, if not identical, pattern of evolution has
been reported in some freshwater caridean shrimps — for
example, Palaemonetes paludosus (Dobkin and Manning.
1964) and P. argentinus (Charmantier and Anger, 1999).
These species, which live in fresh water or in low-salinity
habitats, have lost the useless function of hypo-regulation
usually present in osmoregulatory caridean shrimps and
have retained only the capacity to hyper-regulate.

Acknowledgments

The authors warmly thank Prof. Daniele Guinot, who
provided useful ideas on crab phylogeny and reviewed a
draft of the manuscript. They also thank Dr. F. Lallier, the
chief scientist of the HOPE 99 cruise. Dr. P.-M. Sarradin for
the supply of bottom seawater, and Dr. L. Nonnotte for the
loan of the osmometer used aboard ship.

Literature Cited

Airries, C. N., and J. J. Childress. 1994. Homeoviscous properties
implicated by the interactive effects of pressure and temperature on the
hydrothermal vent crab Bvthograea thermydron. Biol. Bull. 187: 208-
214.

Ballard, B. S., and W. Abbott. 1969. Osmotic accommodation in Cal-
linecte.i sapiilus Rathbun. Comp. Biochem. Physiol. 29: 671-687.

Birchard, G. F., L. Drolet, and L. H. Mantel. 1982. The effect of
reduced salinity on osmoregulation and oxygen consumption in the
lady crab, Ovalipes ocellatus (Herbst). Comp. Biochem. Physiol. 71A:
321-324.

OSMOCONFORMITY IN B. THERMYDRON

173

Blasco, E.. and R. B. Forward, Jr. 1988. Osmoregulation of the xanthid
L-rah, Panopeus herhstii. Camp. Biochem. Phv\iol. 90A: 135-139.

C'harmuntier. G. 1998. Ontogeny of osmoregulution in crustaceans: a
review. Invcrtchi: Reprint. l)e\: 33: 177-190.

Charmantier, G., and K. Anger. 1999. Ontogeny of osmoregulation in
the palaemonid shrimp Palaemonetes argeniinu.\ (Crustacea: Deca-
poda). Mm: Ecnl. Prog. Sei: 181: 125-129.

Cosson, R. P., and J.-P. Vivier. 1997. Interactions of metallic ele-
ments and organisms within hydrothermal vents. Call. Biol. Mar. 38:
43-5(1

DahlholT. E.. J. O'Brien, G. N. Somero, and R. D. Vetter. 1991.
Temperature effects on mitochondria from hydrothermal vent inverte-
brates: evidence for adaptation to elevated and variable habitat tem-
peratures. Physiol. Ziial. 64: 1490-1508.

Delaware Deboutleville, C., and D. Guinot. 1981. Considerations sur
les Bythograeoidea Williams, nouvelle superfamille de crabes de la
dorsale Pacifique Est. Vile Reunion des Carcinologistes de Langue
Fninciiise. Banyuls-sur-Mer. France, 1-6 June 1981 (Abstract).

Desbruyeres, D., P. Chevuldonne, A.-M. Alayse, D. Jollivet, F. H.
l.allier, C. Jouin-Toulmond, F. Zal, P.-M. Sarradin, R. Cosson,
J. C. Caprais, C. Arndt, J. O'Brien, J. Guezennec, S. Hourdez. R.
Riso, F. Gaill, L. Laubier, and A. Toulmond. 1998. Biology and
ecology of the "Pompeii worm" (Alvinella pompejana Desbruyeres and
Laubier). a normal dweller of an extreme deep-sea environment: A
synthesis of current knowledge and recent developments. Deep-Sea
Res. H 45: 383-422.

Dixon, D. R., R. Simpson-\Vhi!e, and L. R. J. Dixon. 1992. Evidence
lor thermal stability of ribosomal DNA sequences in hydrothermal vent
organisms. J. Mar. Biol. Assoc. UK 72: 519-527.

Dobkin, S., and R. S. Manning. 1964. Osmoregulation in two species of
Palaemonetes (Crustacea: Decapoda) from Florida. Bull. Mar. Sci. Gulf
Curihh. 14: 149-157.

Drach. P. 1939. Mue et cycle d'intermue chez les Crustaces Decapodes.
Ann. lust. Oceanogi: 19: 103-391.

Epifanio, C. E., G. Perovich, A. I. Dittel, and S. C. Cary. 1999.
Development and behavior of megalopa larvae and juveniles of the
hydrothermal vent crab Byrhograea thermydron. Mar. Ecol. Prog. Ser.
185: 147-154.

Fisher, C. R. 1990. Chemoautotrophic and methanotrophic symbioses in
marine invertebrates. Crit. Rev. Aquat. Sci. 2: 399-436.

Fisher, C. R. 1998. Temperature and sulfide tolerance of hydrothermal
\em fauna. Call. Biol. Mar. 39: 283-286.

Frederich, M., F. J. Sartoris, W. E. Arntz, and H.-O. Portner. 2000.
Haemolymph Mg:* regulation in decapod crustaceans: physiological
correlates and ecological consequences in polar areas. J. Exp. Biol.
203: 1383-1393.

Geret, F., N. Rousse, R. Riso, P.-M. Sarradin, and R. P. Cosson. 1998.
Metal compartmentalization and metallothionein isotorms in mussels
from the Mid-Atlantic Ridge; preliminary approach to the fluid-organ-
ism relationship. Call. Biol. Mar. 39: 291-293.

Gilles. R. 1970. Osmoregulation in the stenohaline crab "Lihinia emar-
ginala" Leech. Arch. Int. Physiol. Binchim. 78: 91-99.

Grassle, J. F. 1986. The ecology of deep-sea hydrothermal vent
communities. Adv. Mar. Biol. 23: 301-362. (Cited in Epifanio et ai.
1999.)

Gross, \V. J. 1964. Trends in water and salt regulation among aquatic
and amphibious crabs. Biol. Bull. 127: 447 — 166.

(iuinot. D. 1988. Les crabes des sources hydrothermales de la dorsale du
Pacifique oriental (campagne Biocyarise. 1984). Oeeanol. Acta. Spec.
Vol. 8: 109-118.

Guinot. D. 1989. Description de Segon;acia gen. nuv. et remarques sur
Scxonzacia mesatlantica (Williams): campagne HYDROSNAKE 1988
sur la dorsale medio-Atlantique (Crustacea Decapoda Brachyura). Bull.
Mus. Natl. Hist. Nat. 11: 203-231.

(iuinot, D. 1990. Austinograea alayseae sp. nov., Crabe hydrothermal
decnuvert dans le bassin de Lau. Pacifique sud-occidental (Crustacea
Decapoda Brachyura). Bull. Mus. Natl. Hist. Nat. 11: 879-903.

Guinot, D., and M. Segonzac. 1997. Description d'un crabe hydrother-
mal nouveau du genre Bythograea (Crustacea. Decapoda. Brachyura)
et remarques sur les Bythograeidae de la dorsale du Pacifique oriental.
Zoosystema 19: 121-149.

Hardy, D., J. Munro, and J.-D. Dutil. 1994. Temperature and salinity
tolerance of the soft-shell and hard-shell male snow crab. Cluonoa ctcs
npiliii. Aquaciiliure 122: 249-265.

Harris. R. R., and H. Micallef. 1971. Osmotic and ionic regulation in
Potamon cilulis. a fresh water crab from Malta. Comp. Biacliem.
Physiol. 38A: 769-776.

Messier, R. R., and G. D. F. Wilson. 1983. The origin and biogeography
of Malacostracan crustaceans in the deep sea. Pp. 227-254 in Evolu-
tion. Time ami Space: The Emergence of the Biosphere, R. W. Sims.
J. H. Price, and P. E. S. Whalley. eds. Academic Press, London.

Ivanoff, A. 1972. Introduction a I'Ocfanographie. Tome I: Pmprieiex
Physiques et Chimiques des Eaux de Met: Vuibert. Paris. 208 pp.

Jinks, R. N., B.-A. Battelle, E. D. Herzog, L. Kass, G. H. Renninger,
and S. C. Chamberlain. 1998. Sensory adaptations in hydrothermal
vent shrimps from the Mid-Atlantic Ridge. Cah. Biol. Mar. 39: 309-
312.

Jones, L. L. 1941. Osmotic regulation in several crabs of the Pacific
coast of North America. J. Cell. Camp. Physiol. 18: 79-92.

Kamemoto, F. I., and K. N. Kato. 1969. The osmotic and chloride
regulative capacities of five Hawaiian decapod crustaceans. Pac. Sci.
23: 232-237.

Lallier, F. H., L. Camus, F. C'hausson, and J.-P. Truchot. 1998. Struc-
ture and function of hydrothermal vent crustaceans haemocyanin: an
update. Cah. Biol. Mar. 39: 313-316.

Mantel, L. H., and L. L. Farmer. 1983. Osmotic and ionic regulation.
Pp. 53-161 in The Biology ofCnistacea. Vol. 5: Internal Anatonn and
Physiological Regulation. L. H. Mantel, ed. Academic Press. New
York.

Mickel, T. J., and J. J. Childress. 1982a. Effects of pressure and
temperature on the EKG and heart rate of the hydrothermal vent crab
Byrhograea thermydron Brachyura. Biol. Bull. 162: 70-82.

Mickel, T. J., and J. J. Childress. 1982b. Effects of temperature,
pressure and oxygen concentration on the oxygen consumption rate of
the hydrothermal vent crab Bvthograea thermvdron Brachyura.
Physiol. Zoo/. 55: 199-207.

Morritt, D., and J.I. Spicer. 1998. The physiological ecology of talitrid
amphipods: an update. Can. J. Zoo/. 76: 1965-1982.

Newman, W. A. 1985. The abyssal hydrothermal vent invertebrate fau-
na: a glimpse of antiquity? Bull. Biol. Soc. Wash. 6: 231-242.

Pequeux, A. 1995. Osmotic regulation in crustaceans. J. Cnistac. Biol.
15: 1-60.

Pequeux, A., and R. Gilles. 1984. Control of extracellular fluid osmo-
lality in crustaceans. Pp. 18-34 in Osmoregulation in Estuarine and
Marine Animal-,. A. Pequeux. R. Gilles. and L. Bolis, eds. Springer,
Berlin.

Potts, W. T. W., and G. Parry. 1963. Osmotic mid Ionic Regulation in
Animals. Pergamon Press, Oxford. 423 pp.

Powell, M. A., and G. N. Somero. 1986. Adaptations to sulfide by
hydrothermal vent animals: sites and mechanisms of detoxification and
metabolism. Biol. Bull. 171: 274-290.

Prosser, C. L. 1973. Inorganic ions. Pp. 79-1 10 in Comparative Animal
Physiology. C. L. Prosser. ed. Saunders, Philadelphia.

Prosser, C. L., and F. A. Brown, Jr. 1965. Comparative Animal Phys-
iology. W. B. Saunders. London. 688 pp.

Robertson, J. U. 1960. Osmotic and ionic regulation. Pp. 317-339 in
Physiology of Crustacea. Vol. I, T. H Waterman, ed. Academic Press.
New York.

174

A.-S. MARTINEZ ET AL.

Sarradin, P.-M., J.-C. Caprais, P. Briand, F. Gaill, B. Shillito. and I).
Desbruyeres. 1998. Chemical and thermal description of the envi-
ronment of the Genesis hydrothennal vent eommunity (13°N, EPR).
Cull. Biol. Mar. 39: 159-167.

Sarradin, P.-M., J.-C. Caprais, R. Riso, R. Kerouel, and A. Aminol.
1999. Chemical environment of the hydrothermal mussel communi-
ties in the Lucky Strike and Menez Gwen vent fields. Mid Atlantic
Ridge. Call. Biol. Mar. 40: 93-104.

Sebert, P., B. Simon, and A. Pequeux. 1997. Effects of hydrostatic
pressure on energy metabolism and osmoregulation in crab and tish.
Ciimp. Biochem. P/nsiol. 116A: 2X1-290.

Segonzac, M., M. De Saint Laurent, and B. Casanova. 1993.
L'enigme du comportement trophique des crevettes Alvinocarididae
des sites hydrothermaux de la dnrsale medio-atlantique. Cull. Biol
Mar. 34: 535-571.

Shaw, J. 1959. Salt and water balance in the East African fresh water
crab. Potaiiinn niloticus (M. Edw.). J. Ev/>. Bint. 36: 157-176.

Sokal, R. R.. and F. J. Rohlf. 1981. Biometry: The Principles ami
Practice of Statistic!, in Biological Research. W. H. Freeman. San
Francisco. 859 pp.

Spencer, A. M., A. H. Fielding, and F. I. Kamemoto. 1979. The
relationship between gill NaK-ATPase activity and osmoregulatory
capacity in various crabs. Physiol. Zool. 52: 1 — 10.

Sternberg, R. V., N. Cumberlidge, and G. Rodriguez. 1999. On the
marine sister groups of freshwater crabs (Crustacea: Decapoda:
Brachyura). J. Zoo/. Sysl. Evol. Re\. 37: 14-38.

Susanto, G. N., and G. Charmantier. 2000. Ontogeny of osmoregula-
tion in the crayfish Astucus leptodactyliis. Physio/. Biochem. Zool. 73:
169-176.

Taylor, E. \V., P. J. Butler, and A. Al-VVassia. 1977. The effect of a
decrease in salinity on respiration, osmoregulation and activity in the
shore crab. Curcimis inaenas (L.) at different acclimation temperatures.
J. Camp. Physio/. 119: 155-170.

Truchet, M., C. Ballan-Dufrancais, A. Y. Jeantet, J.-P. Lechaire. and

R. Cosson. 1998. Le trophosome de Riftia pachyptila et Tevnia
jerichonana (Vestimentifera): bioaccumulations metalliques et metabo-
lisme du soufre. Cuh. Biol. Mar. 39: 129-141.

Truchot. J.-P., and F. H. Lallier. 1998. High CO, content in hydro-
thermal vent water at the Snake Pit area. Mid-Atlantic Ridge. Call.
Biol. Mar. 39: 153-158.

Tudge. C. C., B. G. M. Jamieson, M. Segonzac, and D. Guinot. 1998.
Spermatozoa! ultrastructure in three species of hydrothermal vent crab,
in the genera Bythograea. Austinograea and Segonzacia (Decapoda,
Brachyura. Bythograeidae). Invertebr. Reprod. Dev. 34: 13-23.

TunniclitTe, V. 1988. Biogeography and evolution of hydrothermal-vent
fauna in the eastern Pacific Ocean. Proc. R. Sac. Lund. B 233: 347-366.

Tunnicliffe. V., A. G. McArthur, and D. McHugh. 1998. A biogeo-
graphical perspective of the deep-sea hydrothermal vent fauna. Adv.
Mar. Biol. 34: 353-442.

Van Dover, C. L. 1995. Ecology of the Mid-Atlantic Ridge hydrother-
mal vents. Pp. 257-294 in Hydrothermal Vents and Processes, L. M.
Parson. C. L. Walker, and D. R. Dixon. eds. Geological Society,
London.

Van Dover, C. L., A. B. Williams, and J. R. Factor. 1984. The first
zoeal stage of a hydrothermal vent crab (Decapoda: Brachyura: By-
thograeidae). Proc. Biol. Soc. Wash. 97: 413-418.

Van Dover, C. L., J. R. Factor, A. B. Williams, and C. J. Berg, Jr. 1985.
Reproductive patterns of decapod crustaceans from hydrothermal
\ents. Bull. Biol. Soc. Wash. 6: 223-228.

Vetter, R. D., M. E. Wells, A. L. Kurtsman, and G. N. Somero. 1987.
Sulfide detoxification by the hydrothermal vent crab Bythograea ther-
mnlroii and other decapod crustaceans. Physiol. Zool. 60: 121-137.

Weast, R. C. 1969. Hamlhook of Chemistry and Phynics. 50'" Ed. The
Chemical Rubber Co.. Cleveland. OH. 2033 pp.

Williams, A. B. 1980. A new crab family from the vicinity of submarine
thermal vents on the Galapagos rift (Crustacea: Decapoda: Brachyura).
Proc. Biol. Soc. Wash. 93: 443-472.

Reference: Bio!. Bull. 201: 175-185. (October 2001)

Escape and Aggregation Responses of Three
Echinoderms to Conspecific Stimuli

A. C. CAMPBELL.* S. COPPARD, C. D'ABREO, AND R. TUDOR-THOMAS

School of Biological Sciences, Queen Man', University of London, Mile End Road, London El 4NS, UK

Abstract. In marine invertebrates, waterborne chemical
stimuli mediate responses including prey detection and
predator avoidance. Alarm and flight, in response to dam-
aged conspecifics, have been reported in echinoderms, but
the nature of the stimuli involved is not known. The re-
sponses of Asterias rubens Linnaeus. Psdmmechinus inili-
aris (Gmelin). and Echinus esculentus Linnaeus to conspe-
cifics were tested in a choice chamber against a control of
clean seawater (no stimulus). All three species showed
statistically significant movement toward water conditioned
by whole animals or homogenate of test epithelium. P.
miliaris and E. esculentus displayed a statistically signifi-
cant avoidance reaction, moving away from conspecirtc
coelomic fluid, gut homogenate. and gonad homogenate. A.
nihcnx was indifferent to conspecific coelomic fluid, pyloric
cecum homogenate, and gonad homogenate but moved
away from cardiac gut homogenate. P. miliaris was indif-
ferent to gametes, but the other two species were signifi-
cantly attracted to them. No species showed preference for
one particular side of the chamber during trials to balance
water flow. These echinoderms can distinguish between
homogenates of conspecific tissues that might be exposed
when a predator damages the test, and those that may
emanate from the exterior surface during normal activities.

Introduction

Predation is a strong selective force, and failing to escape
a predator is much more significant in evolutionary terms
than are other selective forces such as failure to mate or
achieve an optimal energy intake (Lima and Dill, 1940).
Animals use a range of cues to detect predators (e.g.. visual.

Received 23 March 1999: accepted 15 May 2001.
* To whom correspondence should be addressed. E-mail: A. C. Campbell @
qmw.ac.uk

auditory, olfactory, and tactile), but the value of these cues
can be limited in aquatic invertebrates. Turbid inshore wa-
ters, for example, may render visual cues vague (Mackie,
1975). and currents can concentrate or dilute chemical ones
(Weissburg and Zimmer-Faust. 1993). The efficiency of
animals tracking chemical cues is greater in calm flowing
water and less in rough turbulent flows (Weissburg and
Zimmer-Faust, 1994). Movements, scents, or tactile stimuli
can warn of predation risk and can originate from the
predator itself or from injured or killed conspecifics (Snyder
and Snyder, 1970). Such stimuli may trigger escape re-
sponses, while others from intact conspecifics may prompt
individuals to aggregate in dense groups where the risk of
predation to an individual is reduced (Slater, 1985; Zahavi
et ul., 1999). Such aggregations occur in other animals such
as birds, where a high density of individuals within the
aggregation or colony has been shown to be related to
decreased frequency of attack by predators (Kruuk. 1964).
Adult echinoderms are sedentary organisms and are vul-
nerable to a range of predators including mammals, birds,
fish, invertebrates, and other echinoderms (Mortensen,
1943; Moore, 1966; Mayo and Mackie, 1976; Bernstein et
al., 1981). They are, however, able to counter predation by
structural and behavioral means such as the use of spines
and globiferous pedicellariae, which are minute, forcep-like
appendages that can seize and, in some cases, inject venom
into the skin of predators (Campbell, 1983). Such mecha-
nisms reduce the ease with which predators can handle their
prey, and two categories of behavioral adaptations serve to
counter predation in marine invertebrates (Legault and Him-
melman, 1993). These are (1) avoidance adaptations that
limit the potential number of encounters with predators; and
(2) escape adaptations that reduce the risk of predation
when a predator has been detected or encountered. Echino-
derms provide good examples of both categories.

175

176

A. C. CAMPBELL ET AL.

Many echinoderms show avoidance behavior such as
burrowing (sand dollars, heart urchins, and some starfish;
Lawrence, 1987), covering themselves with a layer of shell
and sand (sea urchins; Dayton et a!.. 1977), or sheltering
under rocks and in crevices (Orton, 1914). These habits
limit the potential for encounters with surface-moving pred-
ators. A remarkable avoidance behavior is shown by the
echinoid Strongylocentrotus droebachiensis. In summer, it
avoids a predator, the diurnal-feeding wolffish Anarhichas
lupus, by foraging at night (Bernstein et til., 1981); in
winter, when the wolffish is less active, it forages through-
out 24 hours (Bernstein et ai, 1981 ).

Mauzey et al. (1968) described the escape reactions of
various invertebrate prey species, including other echino-
derms, when they encountered predatory starfishes. One of
the clearest escape behaviors is the flight response from
predators shown by S. droebachiensis. which uses its ven-
tral spines to flee when brought into contact with the starfish
Marthasterias glacialis (Jensen, 1966). This urchin also
flees from water conditioned by crabs or lobsters (Bernstein
ft id., 1981; Mann et al., 1984). Flight responses in the sea
urchin Diadema antillaniin have been initiated by the body
fluids of crushed conspecifics (Snyder and Snyder, 1970;
Parker and Shulman, 1986).

Aggregation behavior has been widely reported in echi-
noderms (Reese, 1966), and in some cases aggregations
appear to be related to grazing, detrital feeding, and suspen-
sion feeding (Sloan and Campbell, 1982). More recently
Levitan et til. ( 1992) showed that aggregation can enhance
fertilization success in spawning echinoids. Allee (1927)
concluded that echinoderm aggregations are the result of a
common response to one or more essential environmental
factors, such as food availability, and that they do not
represent true social groupings. On the other hand, Bern-
stein et al. (1983) believed that aggregation behavior in S.
droebachiensis functions as an escape device, reducing the
risk to individuals because of the sheer numbers present.
These authors considered that predatory crabs would find
the entire aggregation of urchins more difficult to handle
than single individuals.

Although echinoderms have only simple receptor organs,
often made up of a few similar receptor cells without
ganglia (Pentreath and Cobb, 1972), many species are sen-
sitive to touch, chemicals, and light, and some may respond
to pressure changes and vibrations (Campbell, 1983). Var-
ious authors (e.g.. Bullock and Horridge, 1965; Chia, 1969;
Lepper and Moore, 1998) have described the locomotory
and defensive responses of asteroids and echinoids to tactile
and chemical stimuli. In most species, the mechanorecep-
tors and chemoreceptors are located superficially in the test
epithelium, from where they can monitor tactile and chem-
ical stimuli (Campbell, 1973; Lepper, 1998). The tube feet
are also sensitive to touch and chemicals, and are used to

detect food and prey (Sloan and Campbell. 1982). A range
of chemicals, of both low and high molecular weight, ini-
tiate responses in asteroids and echinoids (Sloan and Camp-
bell, 1982). Responses to these stimuli range from local
reflex reactions in the spines, pedicellariae, and tube feet to
fully coordinated responses in which the whole organism
moves toward or away from a stimulus source. Experimen-
tal analyses by Bullock (1965) and Campbell and Laverack
( 1968) showed that both the peripheral basi-epithelial nerve
plexus of the test and the radial nerve cords of the central
nervous systems played a role in mediating these responses.

Many echinoderms possess dermal light receptors, but
these are anatomically simple (Yoshida. 1966) and, unlike
the eyes of insects, molluscs, and vertebrates, do not form
detailed images. Spine movements in response to passing
shadows are, however, well known in sea urchins (Millott
and Takahashi, 1963).

This paper investigates the effects of waterborne stimuli
on three common British species, Asterias rubens Linnaeus.
Psammechinus miliaris (Gmelin), and Echinus esculentus
Linnaeus. We tested the hypothesis that escape and aggre-
gation responses in sea urchins and starfish are triggered by
chemical stimuli emanating from the tissues of conspecific
animals and, further, that these responses differ according to
the source of the stimulus. The three species tested show
broadly similar results. By (1) determining whether these
animals display escape and aggregation responses when
presented with conspecific stimuli and (2) identifying the
body tissue or tissues responsible for producing the effec-
tive chemical signal, these experiments add to our knowl-
edge of chemical ecology and the role of chemical stimuli in
promoting aggregation or avoidance behavior in mobile
animals.

Materials and Methods

Specimens of Asterias rubens, Psammechinus miliaris,
and Echinus esculentus were collected from the shore and
by dredging from the Isle of Great Cumbrae. Scotland. The
animals were transferred to a recirculating seawater system
aquarium at Queen Mary, University of London. The spe-
cies were maintained in separate tanks in a 12-h light: 12-h
dark regime at 1 1 °C and 34 ppt salinity. The animals were
acclimated for 7 days before testing and were fed mussels
ad libitem (for A. rubens) and other epifauna and epiflora
(for the echinoids) brought on small rocks from nearby
shores. The size classes of animals that were used in the
experiments were as follows: A. rubens, R (major radius) =
30-50 mm; P. miliaris, 20-35 mm test diameter; and E.
esculentus, 80-120 mm test diameter.

These animals were tested in a choice chamber (see Fig.
1) based on the design of Mann et al. (1984), which was
chosen because it allowed the test subject to be simulta-

ECHINODERM ESCAPE AND AGGREGATION

177

Seawater supply

Header tank I

^^_

I""! Stimulus 1 ^
tank

1 * LJ

' " ^ 1 Stimulus 1 — I

1 tank 1

LJ 2 |

r Choice chamber t
x L* 1

Water flow

i r

= Flow control valve

Water to waste

X = starting position of test animal
over drainage plate

Figure 1. Diagram of the choice apparatus used to test the responses
of three echinoderms to waterborne stimuli (not to scale). Internal dimen-
sions of components: header tank 300 mm long. 200 mm wide, and 190
mm deep; stimulus tanks each 175 mm long. 1 15 mm wide, and 1 15 mm
deep; choice chamber 400 mm long. 160 mm wide. 140 mm deep, and
drainage plate diameter 40 mm. Seawater supply to header tank set to
overflow constantly; flow tubes to choice chamber set to deliver 0.36 I/mm
to each side. Flow tubing 8 mm internal diameter.

neously stimulated by two unmixed water bodies. This is
impossible in a Y-maze where test animals have to move up
the base arm of the Y in a water body containing two
elements that may be partly mixed together before the test
subjects reach the point of choice. Moreover, some species
may move so quickly that they pass into one arm of the Y
before making a purely chemically cued choice (Bartel and
Davenport, 1956). A significant development in choice ap-
paratus occurred when Pratt (1974) designed a choice cham-
ber to investigate the attraction of prey and the stimulus to
attack in the predatory gastropod Urosalpinx cinerea. His
choice chamber featured two slightly inclined slopes drain-
ing centrally in a narrow rectangular chamber and allowed
the test animal to be stimulated by two unmixed water
bodies at the same time (Pratt, 1974).

In the present work, the header tank, stimulus tanks, and
choice chamber itself were all made from clear acrylic
plastic. The tubing connecting the tanks (see Fig. 1) was
made of flexible plastic with an internal diameter of 8 mm.
The header tank acted as the reservoir and supplied the two
stimulus tanks with running seawater, which could be ad-
justed by flow-control valves (see Fig. 1 ). These two stim-
ulus tanks respectively supplied opposite ends of the choice
chamber via 1 -m-long flexible plastic tubes also fitted with
flow-control valves. Each delivered water to the choice
chamber at a rate of about 0.36 1/min. The choice chamber
itself was a narrow rectangular trough with a perforated

circular drainage plate, 40 mm in diameter, fitted flush to the
center of the tank bottom, for \\astewater outflow. The
chamber measured 400 mm long by 1 60 mm wide by 1 40
mm deep (internal measurements).

The choice chamber was positioned on a wet bench close
to the sea urchin holding tank, so quick transfers of exper-
imental animals were possible. Clean uncirculated water
was used to feed the apparatus during the experiments.
However, beforehand, two dyes — green for the left and red
for the right — were added simultaneously to each stimulus
tank and the flow valves adjusted so that the left and right
water flows met exactly in the center of the drainage plate.
"Threads" of dye reached the drainage plate within 2 min of
the system being set to run. After 5 min there were no
pockets of undyed water in the choice chamber, and a front
of dyes was clearly visible, with no mixing, over the cham-
ber drainage plate.

To run the experiments, the system was set with similar
flow rates of clean seawater entering each side of the choice
chamber, thus reducing any confounding effects of rheo-
taxis. The test animal was placed in the middle of the
drainage plate outlet and allowed to acclimate for 1 min.
Then, over the next 10 min, 20 ml of test stimulus extract
(see below) was introduced directly into the outflow tube of
whichever stimulus tank was in use, at the point the tube left
the tank. This inevitably led to dilution, which can be
estimated over the 10 min of stimulus application as fol-
lows: volume of stimulus, 20 ml: volume of water flowing
from stimulus tank to choice chamber. 3600 ml ( = 0.36 1 X
10 min) = 1 : 180. Each stimulus was given as 20 doses of 1
ml each, delivered at 30-s intervals. This delivery rate of
stimuli helped ensure that the stimulus had equally perme-
ated all parts of the appropriate half of the choice chamber
and that the test animals were exposed to as constant a
stimulus as possible. When whole animals were used for the
stimulus, a pair of starfish or urchins were placed in the
appropriate stimulus tank so that seawater flowed over them
on its way to the choice chamber. The response of the test
animal during this time was observed and recorded. The
following responses were possible:

1. Movement towards the stimulus: the test animal
moved fully off the drainage base plate into the water
body conditioned by the stimulus. The minimum
movement for this to be scored as a response was 50
mm for A. mbens, 30 mm for P. niiliaris. and 60 mm
for E. esciilentus. these distances being such that they
brought the animals at least partly off the drainage
plate and clearly into one water stream. Test animals
achieving less distance than this or not moving off the
drainage plate within 30 min were scored as having no
response. Nearly all of the responding animals moved
the full distance from the center of the drainage plate

178

A. C. CAMPBELL ET AL.

to the stimulus inflow point, a distance of 200 mm
(Fig. 1).

2. Movement away from the stimulus: the animal moved
fully off the drainage base plate into the unconditioned
water body.

3. No response: pail of the animal remained on the
drainage base plate or the animal moved to the sides of
the choice chamber such that at least part of its body
was in line with the base plate.

After each experiment the choice chamber was thor-
oughly rinsed and cleaned of all animal debris. The chamber
was washed through with clean seawater for 5 min between
each test to ensure that all residual stimuli were removed. In
addition, the end of the choice chamber into which the
stimulus was introduced was alternated for each successive
test. This was to eliminate the effects of any inequality of
flow between the two sides of the apparatus that might cause
the test animals to favor one side over the other. No animal
was tested more than once each day. Animals were ran-
domly selected from a pool of 50 individuals kept in sepa-
rate holding tanks according to species. The same animal
could have been selected by chance on successive days; in
that case, it was assumed that its response to a stimulus was
independent between days. We are not aware of any studies
that contradict this assumption for starfish.

Forty different animals were exposed to conspecific stim-
uli in each of eight experiments. Thus data were analyzed
with n = 40. For each stimulus tested, animals were drawn
from the same pool of individuals. The following hypothe-
ses were tested using different experimental stimuli as
shown:

1 . That the two sides of the choice chamber gave similar
flow rates and volumes so that the test animals did not
favor one side over the other. This was tested in
experiments when no stimulus was used in either side
of the chamber. These experiments acted as controls.

2. That intact, whole conspecih'cs are attractive or repel-
lent. Here conspecih'cs were used as the stimulus and
were placed in one stimulus tank. There was no stim-
ulus in the other.

3. That test homogenate. composed of spines and epider-
mal tissues from conspecih'cs (one starfish as well as
one large or three small urchins) was attractive or
repellent to conspecih'cs. This material were scraped
into a 50-ml glass beaker and ground up thoroughly in
25 ml of seawater, using a glass rod. The mixture was
then stirred to suspend all fine material before being
added to one stimulus tank. This was to determine
whether the chemical cues active in (2) above resided
in the epithelium and skeleton of the test and its
appendages. Since movement over hard substrates
abrades urchin spines (Campbell, pers. obs.), naked

calcite can be exposed to seawater naturally, and the
inclusion of calcite in this homogenate is appropriate.

4. That coelomic fluid was attractive or repellent to con-
specifics. A syringe was used to draw off 25 ml of
coelomic fluid from an arbitrary number of starfish or
sea urchins via a small hole in the aboral surface. This
was to determine the attractive or repellent effects of
coelomic fluid that is released from animals broken
open by attacking predators.

5. That gut homogenate was attractive or repellent to
conspecifics. Gut tissue was carefully removed from
halved asteroid (excluding the pyloric ceca) or echi-
noid tests (all the gut) and placed in a 50-ml glass
beaker. The tissue collected from one starfish or large
urchin or from five small urchins was ground up in 25
ml of seawater. using a glass rod. The mixture was
stirred well to suspend cells and fragments and was
tested to determine whether gut tissue, which is ex-
posed during predator attacks, might release chemical
stimuli warning conspecifics of predator behavior.

6. That pyloric cecum homogenate (for A. nibens only)
was attractive or repellent to conspecifics. This was
prepared as for (5) above and for a similar purpose.

7. That gonad extract was attractive or repellent to con-
specifics. Gonad tissue was carefully removed from
one halved test of A. rubens or E. escnlentiix (large
echinoid). or from five tests of P. inilinris (small
echinoid). and ground up in 25 ml of seawater, using
a glass rod. The mixture was stirred well to suspend
cells and fragments. When ripe, gonad tissue may
contribute a major part of the contents of the echino-
derm body cavity and may be released and consumed
when predators break open echinoderm tests (Ormond
etui.. 1973).

8. That gametes were attractive or repellent to conspe-
cifics. Both male and female gametes were extracted
from one large or three small sea urchins by injecting
0.5 ml of 0.5 M K.C1 through the peristomial mem-
brane to initiate spawning: the animals shed their
gametes within a few seconds of injection. The ga-
metes were collected over the 5-15 min spawning
period that followed by inverting the urchin over a
50-ml glass beaker filled with seawater and immersing
the gonopores. Male and female gametes of A. nibens
were extracted using the method of Kantanani ( 1969).
in which 30 g of L-methyladenine was dissolved in 2.5
ml of seawater (0.5 ml per arm). Gametes were shed
60 min after injection, collected in seawater. and
stirred immediately before use to keep them sus-
pended. Gametes were tested to see if they would
stimulate gregarious behavior, which is thought to be
important in increasing fertilization success at spawn-
ing (Reese, 1966; Levitan et a!.. 1992).

ECHINODERM ESCAPE AND AGGREGATION

179

The significance of the collected data was examined in
two ways, using a log-likelihood test. First the numbers of
animals moving toward and away from the stimulus in
question for each experiment were pooled and tested against
those not responding with movements at all. The null hy-
pothesis predicted a ratio of 50:50. This showed whether a
significant number of animals moved in response to the
stimuli as opposed to not moving. Second, the number of
animals moving toward the stimulus in question for each
experiment was compared with the number moving away.
Again, the null hypothesis predicted a result of 50:50.

Results

Overall, about 15% of the test animals responded to
stimuli in the choice chamber within 5 min of the start of
each experiment, and 80% had traversed the full length
of one arm of the chamber within 30 min. In the first set of
experiments (Figs. 2a-c) log-likelihood tests revealed that,
with the exception of Asterias rubens (where there was a
lack of significant response to coelomic fluid) and Echinus
escidentus (where there was a lack of response to stimulus-
free water), all three species displayed significant behavioral
responses to eight experimental treatment stimuli in the
choice chamber (P < 0.001-P < 0.025).

In the second set of experiments (Table 1 and Figs. 3a-c),
starfish and urchins tested when no stimulus was introduced
to the apparatus failed to display a significant preference for
one side or the other of the choice chamber (P > 0.05).
This control experiment showed that the apparatus lacked
any intrinsic bias that might have encouraged test animals to
move more to one side than to the other. It therefore
confirmed that subsequent choices made by test animals, in
response to introduced stimuli, would be meaningful. It also
showed that there was effectively no significant response to
the direction of water flow. All three species of echinoderm
tested were significantly attracted to whole conspecifics
(P < 0.001-P < 0.025) and to the homogenates of their
tests, spines, and epidermal tissues (P < 0.001-P < 0.01 )
(Table 1 and Figs. 3a-c). The sea urchins significantly
avoided water bodies containing coelomic fluid, whereas
the starfish showed no significant response to them (Table 1
and Figs. 3a-c). All animals avoided homogenates of con-
specific gut tissue (Table 1 and Figs. 3a-c). A. rubens did
not respond to homogenate of its pyloric ceca or gonads. but
both sea urchin species were significantly repelled by con-
specific gonad homogenates (P < 0.001 ). P. iniliaris was
not significantly attracted to gametes, whereas both E. es-
citlentus and A. ruhens were (Table 1 and Figs. 3a-c).

Discussion

The results show that Asterias rubens, Psammechinus
miliuris. and Echinus escidentus generally respond to water-

borne stimuli derived from conspecifics (Table 1), being
mainly attracted by whole animals, test homogenate. and
gametes and mainly repelled by coelomic fluid, gut, and
gonad homogenates. These findings agree with a number of
studies which have shown that echinoderms perceive and
react to waterborne chemical stimuli (Dix. 1969; Snyder and
Snyder. 1970: Campbell. 1983; Mann el al, 1984; Parker
and Shulman, 1986). Because these animals have low visual
acuity, tactile and chemical cues must be the chief stimuli
received by their sensory systems (Sloan and Campbell,
1982). A distance-mediated chemosensory system was sus-
pected for prey detection in Asterias forbesi, but could not
be definitively demonstrated (Lepper and Moore, 1995,
1998). However, electron microscopy has revealed concen-
trations of suitable receptors in external epithelium in this
species (Lepper, 1998).

Various workers have demonstrated aggregation of con-
specific echinoderms both in the field and in the laboratory
(McKay. 1945; Reese. 1966: Broom. 1975; Tegner and
Dayton, 1976), and the significant attraction we have de-
scribed is likely to mediate this. Three hypotheses have been
put forward to explain aggregation, namely that echino-
derms can benefit from it ( 1 ) by optimizing feeding, (2) by
better resisting the attacks of predators, and (3) by improv-
ing fertilization success at spawning (Bernstein el al.. 1981;
Moore and Campbell, 1985; Levitan et al., 1992). Aggre-
gations of A. rubens were studied by Moore and Campbell
(1985), who showed that not only were individual starfish
attracted by waterborne scents of conspecifics, but foraging
starfish were more attractive than nonfeeding ones. Aggre-
gation in A. rubens may therefore be a response to optimize
food locations, as was shown by Ormond et al. (1973) for
Acanthaster planci and has been described for other animals
(Zahavi et al. 1999). Aggregating behavior of Strongylo-
centrotits droebachiensis was also investigated by Bernstein
et al. ( 1 981. 1983). who found that this species forms dense
feeding and nonfeeding groups of up to 100 individuals per
square meter.

P. iniliaris, in contrast, is often found in small groups of
between 2 and 10 in the field (Campbell, unpubl.). Its
gregarious behavior could be an adaptation to group defense
of feeding areas, enhancing foraging success by locating
other individuals already feeding (Stone et al.. 1993). How-
ever, the small size of the P. iniliaris aggregations makes
them unlikely to be anti-predator mechanisms of the type
described by Bernstein et al. (1983) for S. droebachiensis.
where predation risk to an individual might be reduced by
putting other conspecifics between it and potential predators
(Hamilton. 1971 ). Such behavior is known for other animals
(Zahavi et al., 1999). Laboratory experiments showed that
S. droebachiensis aggregated in the presence of unspecified
crab and lobster predators that were unable to attack them
effectively because they could not encircle the aggregation

180

A. C. CAMPBELL ET AL.

0 Giving no response

69 Moving towards or away from stimulus

Figure 2a

•5

o
o.

•

Figure 2b

Figure 2c

^5
o

•u

o
o.

•

ECHINODERM ESCAPE AND AGGREGATION 181

Table 1

nl the ohscrvcil responses ijfAsterias rubens. Psammechmus miliaris and Echinus esculemus in cnnspccith stimuli

Stimulus used in
choice chamber

No
stimulus

Whole
animal

Test

Inimngenate

Coelomic
Fluid

Gut
homogenate

Pyloric cecum
homogenate

Gonad
homogenate

Gametes

. \. rnht'iis moving towards

stimulus 18 26

,4. ruhcn.\ moving away from

stimulus 16 5

G test value 0.12 NS 15.58****

P. miliaris moving towards

stimulus 17

P. miliaris mining away

from stimulus 19 9

G test value 0.12 NS 5.62**

E. e.scii/entns moving towards

stimulus 16 24

E. esciilcntus moving away

from stimulus 12

G test value 0.6 NS 9.S6

5
18.74****

9
7.84***

12 30

0.18NS 15.40****

33 31

23.0S**** 16.38****

5 35 35

18.74**** 35.72**** 31.6S

0.12 NS

NA
NA

16
0 NS

30
17.46****

25.66

7
10.76***

0.76NS

11
6.26**

Responses toward stimulus vs. away from stimulus: NS = not significant. P > 0.05; *
•• significant. P < 0.01; **** = significant. P < 0.001; NA = not applicable.

= significant, P < 0.05; *' = significant. P < 0.025; *'

with their claws (Bernstein el ai, 1983). Thus gregarious
behavior lowers the intensity of predation and reduces ur-
chin mortality (Bernstein et ai, 1983) and. apart from
optimizing food locations, these groupings appear to be an
effective anti-predator defense.

Orton (1914) found P. miliaris living in paired associa-
tions of 1 male and 1 female so, alternatively, intraspecific
attraction may be explained by spawning aggregation be-
havior, which is well known for echinoids (Moore, 1966),
and which has been shown to increase fertilization success
(Levitan et al., 1992). All the specimens in this study were
collected in August and maintained at 1 1 °C. P. miluiris is
known to breed at Millport from June to August at temper-
atures of 9 °-l 1 °C (Jensen, 1966; Sukarno et al.. 1979). so
it is likely that the individuals tested in these experiments
would be susceptible to factors that might enhance repro-
ductive success. The reasons for aggregation in E. esculen-
tus are less clear, as there have been fewer investigations of
its social behavior than there have been for A. rubens and P.
miliaris. Aggregations of E. escnlentus have been noted
grazing on algal turf (Forster, 1959), and this species is
known to migrate inshore and aggregate for spawning (Elm-
hirst. 1922; Stott. 1931).

The echinoids in the present study all significantly
avoided water conditioned with conspecific coelomic fluid,
gut homogenate. and gonad homogenate by moving away

from these stimuli (P < 0.00 1 ) (Table 1 ). These escape or
alarm reactions are similar to those of Diadema antillaritin.
which fled from fluid extracts of damaged conspecifics
using its oral spines as supplementary locomotory organs, in
a rapid avoidance reaction (Snyder and Snyder, 1970).
Although Snyder and Snyder (1970) were unable to verify
their field observations by laboratory experiments, our re-
sults are consistent with their findings. S. droebachiensis
also displays an alarm response to water conditioned by
crushed conspecifics and predators (Mann et al.. 1984). and
this characteristic may explain why natural aggregations of
this species decreased in number with increasing abundance
of the predatory wolffish Anarhichas lupus (Bernstein et al..
1981). Presumably, when this fish attacked an urchin, it
released chemicals repellent to other echinoids. Using a
choice chamber similar to the one in the present study,
Mann et al. (1984) showed that 79% of active S. droe-
bachiensis moved away from crushed conspecifics. while
70% moved away from predators (Homarus americanus).
These workers found that when active urchins were exposed
to water conditioned by either coelomic fluid, gut. or gonad
tissue from conspecifics, 80%- 87% of the animals exhib-
ited an alarm response. Mann et al. ( 1984) also showed that
when the escape reaction was calculated as a percentage of
active urchins it was apparently independent or temperature,
whereas the food-seeking reaction was temperature-related.

Figure 2. Number of individuals nl each species responding to conspecific stimuli with G-test results and
significance code for each stimulus type, (a) A.steria.s rithens: (bl Psammechinus miliaris: (c) Echinus escnlentus.
NS = 0 > 0.05 * = P < 0.05 * = P < 0.025 *** = P < 0.01 **** = P < 0.001.

0 Giving no response

Z Moving towards or away from stimulus

Figure 3 a

Figure 3b

Figure 3c

•3
o

•a

o
ex

V-c

o
G.

•3

ECHINODERM ESCAPE AND AGGREGATION

183

Thus they suggested that, during the low temperature of
winter months, urchin behavior is determined more by the
presence of predators than by the distribution of food. The
effects of both heterospecific and conspecific body fluids on
several species of Caribbean echinoids were examined by
Parker and Shulman (1986), who found that the degree of
response to conspeciric extracts depended on the extent of
protection afforded by different microhahitats. Further,
there was a correlation between the distance moved by
alarmed urchins and the distance they moved from shelter
when foraging (Parker and Shulman. 1986). Thus there is a
strong implication of predator avoidance in this particular
aspect of echinoid behavior.

P. iniliaris spends much of its time in protected situa-
tions, such as under rocks (Orton, 1929: Jensen. 1966). and
makes excursions from these areas to forage on algae and
encrusting animals such as sponges, hydroids. bryozoans.
and crustaceans (Hancock. 1957). It is during foraging that
P. iniliaris is most vulnerable to predators and that a reliable
signal of the presence of a feeding predator will be most
advantageous in stimulating the urchin to retreat to safety.
The implication of Parker and Shulman's ( 1986) work may
be that the echinoid alarm response to conspeciric scents has
evolved as an adaptation for escape during foraging periods.
The sensitivity to gonad and gut extracts is particularly
significant because most predators of echinoids have to
break open the test in some way to gain access to the main
nutritive elements, the gonads and other viscera. Because E.
esculentus is a sublittoral species (Orton. 1929), behavioral
and ecological observations of this species in the field are
less extensive and evaluation of its foraging and escape
activities in situ is more difficult, than for P. iniliaris. which
occurs on the shore. Some echinoderm species form aggre-
gations at spawning time and shed their gametes in syn-
chrony (Moore. 1966). Therefore, why should the urchins
be repelled by an extract composed of gonad tissue while
not being repelled by gametes, as indicated by the results of
this investigation? Possibly the chemical stimulus that
caused the urchins to move away from the gonad extract
originated in the germinal epithelium or in the nutritive
phagocytic tissue. The responses of A. rubens to reproduc-
tive tissues showed consistencies with those of the echi-
noids, as starfish were not attracted by gonad homogenate
but were by gametes.

In the case of A. rubens, gut homogenate (cardiac and
pyloric stomach) caused an avoidance response: however,
the starfish differed from the urchins in that coelomic fluid
did not do so. Possibly A. rubens, a known predator and
carrion feeder, sometimes acts as a cannibal. Some speci-

mens of .4. rubens held in the aquaria were seen to eat
pyloric ceca previously isolated from other individuals,
which accords with Jangoux's (1982) note of cannibalism in
this species.

Weissburg and Zimmer-Faust ( 1993) pointed out that the
success of chemically mediated alarm responses in protect-
ing individuals from dangerous situations depends on water
turbulence and mixing, because the aquatic environment, as
a medium for the transmission of chemical signals, is pro-
foundly affected by hydrodynamics. Every care was taken
with our experiments to minimize such disturbances. The
use of the dye tests to obtain the correct balance of water
flow through the choice chamber allowed us to determine a
1 0-min regime of stimulus application that subjected the test
animals to the most precise stimulus conditions we could
obtain. Quantification of the amount of stimulus needed to
elicit a response is a desirable but elaborate extension of the
experimental procedure, and one that needs to be addressed
in future work. Various studies have identified the specific
substances to which echinoderms will respond, showing
that these animals can react to chemicals such as amino
acids, which are present in very low concentrations (Mayo
and Mackie, 1976; Sloan and Campbell, 1982; Mann et al,
1984; Lepper and Moore, 1995).

Snyder and Snyder (1970) noted that vinegar produced a
flight response in Diadema antillaritm that was similar to
the one initiated by crushed conspecifics. Because crushed
heterospecifics had no such effect, they rejected the idea that
the escape response was merely due to a change in the
chemistry of the water passing over the urchins. Parker and
Shulman (1986) found similar results, strengthening the
argument for predator avoidance and escape due to extracts
from conspecifics. Solandt and Campbell (1998) demon-
strated that Caribbean echinoids tested in a choice chamber
showed a distinct range of preferences to six algal species,
which further supports the idea that these responses are
based on choice.

The identification here of clear-cut avoidance and aggre-
gation responses for the two echinoid species shows that
they differentiate between, and react to, distinct chemical
stimuli under aquarium conditions at a distance of only 1 m.
Although the stimulus concentrations are poorly defined,
they lie within plausible concentrations for animals living
closely together in aggregations or social groups, where one
may be seriously damaged by a predator. Our results for A.
rubens indicate that the distinction between aggregative and
repellent effects of various conspecific tissues here is less
well defined than it is for the two echinoids. with aut

Figure 3. Number of individuals of each species responding to conspecific stimuli, (a) Av/f/mv rubens; (b)
PsamiiH'iliiiiii-. miliumi; (c) Echinus

184

A. C. CAMPBELL ET AL.

homogenate being the only stimulus that produced a signif-
icant avoidance response by the starfish.

Acknowledgments

The authors acknowledge the help and encouragement
given by Dr. Maurice Elphick, Dr. Craig Young, and Prof.
Paul Tyler. They are indebted to Dr. Carl Smith for his
advice on the statistical treatment of the data and to two
anonymous referees for their constructive comments.

Literature Cited

Allee. W. C. 1927. Animal aggregations. Q. Rev. Biol. 2: 367-398.

Bartel, A. H.. and D. Davenport. 1956. A technique for the investiga-
tion of chemical responses in aquatic animals. Br. ./. Anim. Behav 4:
117-119.

Bernstein, B. B.. B. E. Williams, and K. H. Mann. 1981. The role of
behavioural responses to predators in modifying urchin's (Strong\lo-
centrotus droebachiensis) destructive grazing and seasonal foraging
patterns. Mar. Biol. 63: 39-49.

Bernstein, B. B., S. C. Schroeter, and K. H. Mann. 1983. Sea urchin
(Strongylocentrotus droebachiensis) aggregating behaviour investi-
gated by a subtidal multifuctoriul experiment. Can. J. Fish. Aqi/at. Sci.
40: 1975-1986.

Broom, D. M. 1975. Aggregation behaviour of the brittle-star Ophiothrix
fragili.s. ./. Mar. Biol. Assoc. UK 55: 191-197.

Bullock, T. H. 1965. Comparative aspects of superficial conduction
systems in echinoids and asteroids. Am. Zool. 5: 545-562.

Bullock, T. H., and G. A. Horridjje. 1965. Structure and Function in
the Nervous Systems of Invertebrate -s. W. H. Freeman. San Francisco

Campbell, A. C. 1973. Observations on the activity of echinoid pedi-
cellariae: 11. Jaw responses of tridentate and ophiocephalous pedicel-
lariae. Mar. Behav. Physiol. 3: 17-34.

Campbell, A. C. 1983. Form and function in pedicellariae. Echinoderm
Stud. 1: 139-167.

Campbell, A. C., and M. S. Laverack. 1968. The responses of pedi-
cellariae from Echinus escu/entits (L.). J. E\p. Mar. Biol. Ecoi 2:
141-214.

Chia, E.-S. 1969. Responses of globiferous pedicellariae to inorganic
salts in three regular echinoids. Ophelia 6: 203-210.

Dayton. P. K., R. J. Rosenthal, L. C. Mahen, and T. Antezena. 1977.
Population structure and foraging biology of the predaceous Chilean
asteroid Mevcnaster ge/atinosns and the escape biology of its prey.
Mar. Biol. 39: 361-370.

Dix, T. 1969. Aggregation in the echinoid Evechinns chloroticns. Pac.
Sci. 23: 123-124.

I linhiisi. R. 1922. Habits of Echinus csculenlus. Nature 110: 667.

Forster, G. R. 1959. The ecology of Echinus esculentus L. Quantitative
distribution and rate of feeding. J. Mar. Biol. Assoc. UK 38: 361-367.

Hamilton, \V. D. 1971. Selection of selfish and altruistic behaviour in
some extreme models. Pp. 58-91 in Smithsonian Annual III. Man and
ISeust: Comparative anil Social Behaviour. .1. F. Eisenberg and W. S.
Dillon, eds. Smithsonian Institution Press. Washington. DC.

Hancock. D. A. 1957. The feeding behaviour of the sea urchin Psam-
iiiciliinin nnliaris (Gmelin) in the laboratory. Proc. Zoo/. Soc. Lond.
129: 255-262.

Jangoux, M. 1982. Food and feeding mechanisms: Asteroidea. Pp. I 17-
159 in Echinoderm Nutrition, M. Jangou\ and J. N. Lawrence, eds.
A. A. Balkema. Rotterdam.

Jensen, M. 1966. Breeding and growth of Psammei liinus miliaris (Gme-
lin). Ophelia 1: 65-78.

Kanatani. H. 1969. Induction of spawning and oocyte maturation by
L-methyladenine in starfishes. Exp. Cell Res. 57: 333-337.

Kruuk, H. 1964. Predators and anti-predator behaviour of the black-
headed gull (Earns ridilnindiis L.). Behaviour, Supplement XI. E. J.
Brill. Leiden.

Lawrence, J. M. 1987. A Functional Biology of Echinoderms. Croom
Helm. Bekenham. UK.

Legault. C.. and J. H. Himmelman. 1993. Relation between escape
behaviour of benthic marine invertebrates and the risk of predation. J.
Exp. Mar. Biol. Ecol. 170: 55-74.

Lepper, D. M. E. 1998. infrastructure and morphology of the epidermis
and tube feet of two echinoderms. P. 46 in Proceedings of the Ninth
International Echinoderm Conference. San Francisco, 5—9 August
1W6. R. Mooi and M. Telford. eds. A. A. Balkema. Rotterdam.

Lepper. D. M. E., and P. A. Moore. 1995. Chemical ecology of the
seastar. Asterias forbesi: the role of chemical signals in foraging
behaviour. Client. Senses 20: 729 [Abstract.]

Lepper, D. M. E., and P. A. Moore. 1998. The role of chemical signals
in the foraging behaviour of the sea star A.iterias forbesi. P. 266 in
I'loceedings "I the Ninth International Echinoderm Conference. San
Franct.sco, 5-y August IW6. R. Mooi and M. Telford. eds. A. A.
Balkema. Rotterdam.

Levitan, D. R., M. A. Sewell, and F.-S. Chia. 1992. How distribution
and abundance influence fertilization success in the sea urchin Strongs -
loccniroiii\ Irancist anus. Ecology 73: 248-254.

Lima. S. L.. and L. M. Dill. 1990. Behavioural decisions made under the
risk of predation: a review and prospectus. Can. J. Zool. 68: 619-640.

Mackie, A. M. 1975. Chemoreception. Pp. 69-105 in Biochemical ami
Biophysical Perspectives in Marine Biology, Vol. 2. D. C. Malins and
J. R. Sargent, eds. Academic Press. London.

Mann, K. H., J. L. C. Wright, B. E. Welsford, and E. Hatfield. 1984.
Responses of the sea urchin Strongylocentrotus droebachiensis to
waterborne stimuli from potential predators and potential food algae. J.
/-.»/>. Mar. Biol. Ecol. 79: 233-244.

Mau/ey, K. P., C. Birkland, and P. K. Dayton. 1968. Feeding behav-
iour of asteroids and escape responses of their prey in the Puget Sound
Region. Ecology 49: 603-619.

Mayo, P., and A. M. Mackie. 1976. Studies of avoidance reactions in
several species of predatory British seastars (Echinodermata: Aster-
oidea). Mar. Biol. 38: 41-49.

McKay, D. C. G. 1945. Notes on the aggregating marine invertebrates of
Hawaii. Ecology 26: 205-207.

Millott, N.. and K. Takahashi. 1963. The shadow reaction of DiaJcina
antillariiin Philippi. IV. Spine movements and their implications.
Plnlos. Trans. R. Soc. Lond. B 246: 437-470.

Moore, H. B. 1966. Ecology of echinoids. Pp. 73-85 in Physiology of
Echinodermata, R. A. Boolootian, ed. Wiley Interscience. New York.

Moore, R. J., and A. C. Campbell. 1985. An investigation into the
behavioural and ecological bases for periodic infestations of Asterias
nihens. P. 596 in Proceedings of the Fifth International Echinoderm
Conference, Galway, 24-29 September 1VR4. B. F. Keegan and

B. D. S. O'Connor, eds. A. A. Balkema. Rotterdam.
Mortensen, T. 1943. ,-1 Monograph of the Echinoidea. Camarodonta 11.

C. A. Reitzel. Copenhagen.

Ormond, R. F. G., A. C. Campbell. S. M. Head. R. J. Moore. P. S.

Rainbow, and A. P. Saunders. 1973. Formation and breakdown of

aggregations of the crown-of-thorns starfish, Acanthaster p/anci (L.).

Nature 246: 167-168.
Orton. J. H. 1914. On the breeding habits of Echinus miliaris with a note

on the feeding huhits of Patella vulgala. J. Mar. Biol. Assoc. UK 10:

254-257.
Orton, J. H. 1929. On the occurrence of Echinus esculentus on the

foreshore in the British Isles. ./. Mar. Biol. Assoc. UK 16: 289-296.

ECHINODERM ESCAPE AND AGGREGATION

185

Parker, D. A., and M. J. Shulman. 1986. Avoiding predation: alarm
responses of sea urchins to simulated predation on conspecitic and
heterospecific sea urchins. Mar. Bio/. 93: 201-208.

Pentreath. V. W., and J. L. S. Cobb. 1972. Neurobiology of Echino-
dermata. Biol. Rev. 47: 363-392.

Pratt. D. M. 1974. Attraction to prey and stimulus attack in the predatory
gastropod Urosalpinx cinerea. Mar. Biol. 27: 37-45.

Reese, E. A. 1966. The complex behavior of echinoderms. Pp. 157-218
in Physiology of Echinodermata, R. A. Boolootian. ed. Wiley Inter-
science. New York.

Slater, P. J. B. 1985. An Introduction ro Ethology. Cambridge University
Press, Cambridge.

Sloan, N. A., and A. C. Campbell. 1982. Perception of food. Pp. 3-23
in Echinoderm Nutrition, M. Jangoux and J. N. Lawrence, eds. A. A.
Balkema, Rotterdam.

Snyder, N., and S. Snyder. 1970. Alarm response of Diadema until/a-
rum. Science 168: 276-278.

Solandt, J.-L., and A. C. Campbell. 1998. Habitat selection in Jamaican
echinoids. Pp. 821-827 in Proceedings of the Ninth International
Echinoderm Conference. San Francisco, 5-9 August 1996. R. Mooi
and M. Telford. eds. A. A. Balkema, Rotterdam.

Stone, R. P., C. E. O'Clair, and T. C. Shirley. 1993. Aggregating
behaviour of ovigerous female king crab Paralithodes camtichaticus.
in Auke Bay, Alaska. Can. J. Fish. Aquat. Sci. 50: 750-758.

Stott, F. C. 1931. The spawning of Echinus esctilentus and some changes
in gonad composition. J. £.v/>. Biol. 8: 133-150.

Sukarno, R., M. Jangoux, and E. Van Impel. 1979. Le cycle repro-
ducteur annuel de Psammechinus miliaris (Gmelin) (Echinoidea) en
Zeelande. Pp. 415-416 in Echinoderms: Past and Present. M.
Jangoux, ed. A. A. Balkema, Rotterdam.

Tegner, M. J., and P. K. Dayton. 1976. Sea urchin recruitment patterns
and implications for commercial fishing. Science 196: 324-326.

Weissburg, M. J., and R. K. Zimmer-Faust. 1993. Life and death in
moving fluids: hydrodynamic effects on chemosensory-mediated pre-
dation. Ecology 74: 1428-1443.

Weissburg, M. J., and R. K. Zimmer-Faust. 1994. Odor plumes and
how blue crabs use them in finding prey. J. Exp. Biol. 197: 349-375.

Yoshida, M. 1966. Photosensitivity. Pp. 435-464 in Physiology of Echi-
nodermata. R. A. Boolootian, ed. Wiley Interscience, New York.

Zahavi, A., A. Zahavi. N. Zahavi-Ely, and M. P. Ely. 1999. The
Handicap Principle: A Missing Piece of Danvin 's Puzzle. Oxford
University Press, New York.

Reference: Biol. Bull. 201: 1X6-142. (October 2(101)

Action Potentials Occur Spontaneously in Squid Giant
Axons with Moderately Alkaline Intracellular pH

JOHN R. CLAY1 * AND ALVIN SHRIER2

1 Laborutor\ of Neurophysiology, National Institute of Neurological Disorders and Stroke,

National Institutes of Health, Bethesda, Mainland 20892: and ~ Department of Physiology,

McGill University, Montreal, Quebec, Canada, H3G1Y6

Abstract. This report demonstrates a novel rinding from
the classic giant axon preparation of the squid. Namely, the
axon can be made to fire autonomously (spontaneously
occurring action potentials) when the intracellulur pH (pH,)
was increased to about 7.7, or higher. (Physiological pH, is
7.3.) The frequency of firing was 33 Hz (T == 5°). No
changes in frequency or in the voltage waveform itself were
observed when pH, was increased from 7.7 up to 8.5. In
other words, the effect has a threshold at a pH; of about 7.7.
A mathematical model that is sufficient to mimic these
results is provided using a modified version of the Clay
(1998) description of the axonal ionic currents.

Introduction

The electrical response of squid giant axons /';; vivo to
environmental stimuli can be characterized as being primar-
ily phasic. For example, the axon fires an action potential
once, and only once, in 15 °C seawater in response to a light
flash, thereby triggering the rapid, jet-propelled escape of
the animal (Otis and Gilly, 1990). A more complicated
behavior of the axon occurs in concert with the parallel
small axon system (Young. 1939) during the delayed jet
escape from chemical stimuli applied at the olfactory organ
(Otis and Gilly. 1990). Under these conditions the giant
axon fires from one to three action potentials (or none at all).
A reduction in temperature to 6 °C, which squid often
encounter in deep waters, produces changes in the role of
the axon in these behaviors (Neumeister et <//.. 2000). For
example, the axon usually fires twice in response to a light
flash at 6 °C (Neumeister ct a/.. 2000). A tonic train of a

Received 30 November 2000: accepted 28 June 2001.
* To whom correspondence should be addressed. E-mail: jrclay@
ninds.nih.gov

relatively large number of action potentials does not appear
to be elicited /;; viva. These results are mirrored by the
response of the axon in vitro to current stimuli applied with
the standard axial wire recording technique. Under these
conditions one. and only one, action potential is elicited
with a rectangular current pulse, regardless of pulse duration
or pulse amplitude (Clay. 1998). Moreover, an action po-
tential is not elicited with a relatively slow depolarizing
current rump (J. R. Clay, unpub. obs.). A rapidly changing
stimulus, such as a current step of sufficiently large ampli-
tude, is required.

Given the above results, we were surprised to observe
tonic firing of the axon when the pH of the perfusate used
during recordings from intracellularly perfused axons was
increased to 7.7, or higher. The normal intracellular pH
(pH,) is 7.3 (Boron and DeWeer, 1976). Under these con-
ditions of slightly elevated pH,, action potentials occurred
spontaneously and repetitively. The activity lasted for as
long as a few hours in some preparations. An ionic mech-
anism underlying this observation is proposed.

Materials and Methods

Experiments were performed on internally perfused squid
giant axons at the Marine Biological Laboratory. Woods
Hole. Massachusetts, using standard axial wire voltage- and
current-clamp techniques described elsewhere (Clay and
Shlesinger. 1983; Clay. 1998). The intracellular perfusate
consisted of 300 mM K glutamate and 400 mM sucrose,
with the pH adjusted to the desired level within the 7.2 to
8.5 range by free glutamic acid. In a few experiments the
intracellular buffer consisted of 400 mM sucrose. 250 mM
KF, and 25 mM K2HPO4 (pH, = 7.6-7.8). The extracellular
solution was either filtered seawater (pH = 7.5) or artificial
seawater consisting of 430 mM NaCl, 10 mM KC1. 50 mM

186

ACTION POTENTIALS WITH ALKALINE pH,

187

MgCK. 10 m/WCaCU and 10 inMTris-HCl (pH 7.2). These
extracellular solutions were used interchangeably given that
similar results were obtained in either condition. The tem-
perature was in the 4-6 :C range; in any single experiment
it was maintained constant to within O.I °C by a Peltier
device located within the experimental chamber. Input re-
sistance measurements were made with rectangular current
pulses applied to axons in extracellular medium containing
tetrodotoxin (TTX. Sigma Chemical Co.) at a h'nal concen-
tration of 1 p.M.

Computer simulations of membrane excitability were car-
ried out as described previously (Clay, 1998). The model is
given by

CdV/dt

/L + /K

= 0,

where V is membrane potential in mV, t is time in ms, C is
the specific membrane capacitance (C -= 1 juP • cm 2), 7stim
is the stimulus current (/j,A • cm"2), and the various ionic
current components are described as follows. The sodium
ion current is given as in Vandenberg and Bezanilla ( 1991 )
with

/Na = SN^oWexpUV- EN,)/ 24) - 1 )/((exp( W24) - 1)

• (1 + 0.4exp(-0.38W24)»

where #Na = 107 mS • cm"", £Na = 64 mV. and P0 is the
probability that any single Na+ channel is in the open state
of the Vandenberg and Bezanilla (1991) kinetic scheme.
The various rate constants in the model (in ms"1) are as
follows:

a = 7.55 exp(0.017(V •- 10)),
b = 5.6 exp(- 0.000 17(V - 10)).
•c = 21.0 exp(0.06(V - 10)).
d = 1.8 exp(-0.02(V - 10)).
/ = 0.56 exp(0.00004( V - 10)).
g = exp(0.00004(V - 10)).
/ = 0.0052 exp(-0.038(V - 10)).
j = 0.009 exp(-0.038(V - 10)).
y = 22.0 exp(0.014(V - 10)).
c = 1.26 exp(-0.048(V •- 10)).

The potassium ion current is given by
IK = ,?K"(V. /)sV(exp(V/24) - Ks(r)/K,)/(exp( V/24) - 1).

where i?K = 62.5 mS -cm"2, d;i( V, O/dt = -(a + /3)«( V,
t) - a. with a = -0.0075(V + 64)/(exp( -0. 1 1 ( V +
64)1 -• 1) and J3 = 0.075 exp(-(V + 62)/20). This
represents a modification of the model of /K in Clay ( 1998)
in which n4 kinetics were used. The a and /3 parameters
have also been modified so as to obtain equivalent (or
better) descriptions of the /K results given by the n4 model
in Clay (1998). The Ks parameter, which corresponds to the
potassium ion concentration in the restricted space just
outside of the axolemma, is given by

dKs/dr = O.OI()4/K - 0.08(KS - 10) - 5(KS - If))/

(1 + (K, - l())/2)\

Further details concerning the description of the /Na and /K
components are given in Clay (1998). The background, or
time-independent current in the model consists of three
terms: the "leak" current, /, ; a persistent, tetrodotoxin-
sensitive sodium ion current /NuP (Rakowski ct <//., 1985):
and an inwardly rectifying potassium ion current. /K lr. The
latter, together with /Nap, confers nonlinearity to the back-
ground current in the -90 to -60 mV range, similar to that
observed experimentally. These components are described
by

/NaP = 4.5(W24)(0.()3 exp(VV24) - 0.43)/((exp(W24) - 1)

X (1 + exp(-(V + 65)/7))h

/K.,r = 0.24(V+82)/(l + 0.05(exp(0.15(\'+82)))); and

where gL is either 0.2 (pH, = 7.3) or 0.03 (pH, 8.5). The
formulation for /Na P represents a best fit, by eye, to unpub-
lished measurements of this component kindly provided by
R. F. Rakowski (Finch University of Health Sciences/The
Chicago Medical School).

The simulations were implemented with a fourth-order
Runga-Kutta iteration routine in FORTRAN with a time
step of 1 jus.

Results

As noted above, the physiological intracellular pH for
squid giant axons is 7.3 ± 0.013 (±SE: Boron and DeWeer.
1976). Under these conditions axons are quiescent with a
resting potential of about —60 mV. An action potential
elicited by a brief current pulse is shown in Figure 1A. A
slight oscillatory rebound following the action potential was
apparent, as indicated by the arrow in Figure 1 A. The effect
of changing the intracellular pH to 8.5 is illustrated in
Figure 1 B and C. A few minutes after the solution change.
the membrane potential became oscillatory (inset. Fig. I B).
Moreover, it exhibited a much greater post-excitatory re-
bound (Fig. IB). Several minutes later, the amplitude of the
spontaneously occurring subthreshold oscillations in-
creased, followed by a train of action potentials that lasted
until the experiment was terminated. The result in Figure 1C
occurred shortly after the change in pH|. Similar recordings
were obtained at later times in these experiments by clamp-
ing the membrane potential at the equilibrium point and
then slowly releasing the clamp. The membrane potential
remained at the equilibrium point for a few seconds and
then began to oscillate spontaneously. The oscillations in-
creased in amplitude until an unending train of action po-
tentials occurred, similar to the result in Figure 1C. Results

J. R. CLAY AND A. SHRIER

5 ms

OmV

although small-amplitude subthreshold oscillations were
still observed (Fig. 2C). Spontaneous activity was reestab-
lished when the initial perfusate (pH, = 8.3) was used (Fig.
2D). These results are consistent with a threshold (all-or-
none phenomenon) for autonomous activity with pHj. The
threshold was in the 7.6 to 7.8 range, as indicated by four
experiments in which pH, was changed in 0.2 increments
from pH, = 7.2 to 8.4.

In three experiments on an unrelated topic, spontaneous
firing was observed upon initiation of intracellular perfusion
with a buffer consisting of 400 mM sucrose, 250 mM KF,

5 ms

-50

50 ms

Figure 1. (A) Action potential from a squid giant axon in control
conditions (pH, = 7.3) elicited by a 1-ms, suprathreshold current pulse.
The arrow indicates a slight oscillatory rebound after the action potential.
The inset illustrates a 100-ms epoch during rest (voltage scale 4X). (B)
Action potential 2 mill after changing to an intracellular buffer with pH =
8.5. The oscillatory rebound was considerably larger, and the membrane
potential oscillated about the resting level (inset; voltage scale 4x>. (C)
Initiation of spontaneous firing of action potentials, about 3 min after the
change to pH, = 8.5. The spontaneous activity in this preparation lasted
4 h, at which point the experiment was terminated.

such as those in Figure 1 were observed in 20 out of 24
axons in which the effect was investigated. The frequency
of firing at T = 5 °C. the temperature at which several of
the experiments were performed, was 32.9 ± 6.1 Hz (n =
8; ±SD).

The pH, effect was reversible, as illustrated in Figure 2.
In this experiment. pH, was initially 8.3. The axon fired
spontaneously (Fig. 2A). The intracellular perfusate was
then switched to one having a pH of 7.7 (Fig. 2B). No clear
effect on the electrical activity was observed 15 min after
the solution change. When the intracellular buffer was
changed to one having pH = 7.4, the activity ceased.

A
pH, = 8.3

B
pH, = 7.7

pH, = 7.4

20ms

Figure 2. Autonomous activity with pH( = 8.3 (A) and pH, = 7.7 (B).
No clear effect was apparent with this change in pH. Spontaneous activity
ceased with pH, = 7.4 (C), although subthreshold oscillations were appar-
ent. (D) Spontaneous activity was re-established with pH, = 8.3. Different
preparation than in Figure 1.

ACTION POTENTIALS WITH ALKALINE pH,

189

10 ms

-58

-70

pH, = 7.2

-93 mV

pH, = 8.5

Figure 3. Effect of pH, on background ("leak") conductance. (A) Membrane potential responses to a 30 |U.A •
cirT: hyperpolarizing current pulse with pH, = 7.2 and 8.5, as described in the text. (B) Schematic description
of the change in the current-voltage relation of the axon which is proposed to explain the results in panel A.

and 25 mM K2HPO4 (pH, = 7.6-7.8). These results dem-
onstrate that the effect was not a function of the buffering
system. We primarily used K glutamate which, as Wanke el
nl. (I980a) noted, is appropriate at a concentration of 45
mM for pH in the 9 to 10.8 range. We found that a solution
containing 300 mM K glutamate could be stably titrated
(with free glutamic acid) down to pH 7.2. which allowed us
to cover the pH range of interest (7.2 to 8.5) with a single
buffer system containing an anion — glutamate — which is
known to be "favorable" for squid giant axons (Adams and
Oxford, 1983; Clay. 1988).

The most logical place, a priori, to look for the ionic
mechanism underlying the pH, effect would seem to be the
classical sodium and potassium ion currents. /Na and 7K,
respectively, that underlie the action potential (Hodgkin and
Huxley, 1952). We looked for an effect of a change in pH,
on /Na in voltage-clamp recordings with pH, in the 7 to 9
range, but we did not observe any clear effect. An irrevers-
ible reduction of /Na inactivation does occur for a pH,
greater than 9.5 (Brodwick and Eaton. 1978). which is
outside the range of pHj we have used. Moreover, blockade
of 7Na in squid axons by intracellular protons has been
observed having pKa values of 4.6 and 5.8 (Wanke et til..
1980b) — an effect of pH, which, again, lies outside the
range we have used. We are not aware of any report in the
literature of an effect of a change in pH, in the 7 to 9 range
on /Na. No such effect was observed in this study.

An increase of pH( in the 7 to 9 range increases the
amplitude of 7K at any given depolarization from a holding
level of -50 mV (Wanke et ui. 1980a). although a similar
effect does not occur with relatively negative holding po-
tentials (-80 or -90 mV; Clay, 1990). This holding poten-

tial dependence is consistent with a rightward shift of the IK
inactivation curve along the voltage axis as pH; is increased
in the 6 to 10 range (Clay, 1990). The effect — essentially an
increase in the number of K+ channels available for acti-
vation during the action potential — cannot account for pHr
induced automaticity (simulations not shown).

A clue to the ionic mechanism for pH,-induced automa-
ticity was provided by input resistance measurements in
axons made quiescent with tetrodotoxin (TTX; 1 juAf — Fig.
3). The preparation illustrated in Figure 3A rested at -58
mV in TTX with pH, = 7.2. A hyperpolarizing current pulse
10 ms in duration produced a hyperpolarizing response
having a time constant. T. of 0.7 ms. In an equivalent circuit
model of the membrane, this result is equal to the product of
the membrane resistance and the membrane capacitance.
The specific membrane capacitance is 1 fiF • cm"2. Con-
sequently, the specific membrane resistivity with T = 0.7 ms
is 0.7 kfl • cm"2, a value that is consistent with the classical
small-impedance measurements in figure 23 of Hodgkin and
Huxley (1952). The corresponding result for pH, = 8.5 is
shown in the bottom panel of Figure 3 A. The change of pHj
from 7.2 to 8.5 produced a slight hyperpolarization of rest
potential by about 1 mV. The response to a current pulse of
the same amplitude as in pHj == 7.2 produced a marked
increase in membrane hyperpolarization with a much slower
response time. Indeed, the response was not yet at the
steady-state level at the end of the 10-ms pulse. Similar
observations were made in four different preparations. This
result is consistent with a reduction of net inward current, as
illustrated schematically in Figure 3B. A hyperpolarizing
current pulse having an amplitude of /h intersects the cur-
rent-voltage relation at a much more negative potential with

190

J. R. CLAY AND A. SHRIEK

liAcm"2

J -10.0

Figure 4. Simulations of pH,-induced excitability. Steady-state current-voltage relations are shown for
pH, = 7.3 and pH, = 8.5. The sole difference in the model for the two conditions is in the leak current
conductance, which is 0.2 and 0.03 mS • cm~- for pH, = 7.3 and 8.5. respectively. The equilibrium potential,
that is. the point where the current-voltage relation crosses the voltage axis, is stable for pH, 7.3, as indicated
by the trajectory in the inset adjacent to this point. (The scales are 5 mV and 2 p,A • cm"2). An action potential
elicited by a suprathreshold current pulse for this condition is shown below the current-voltage relation. (The
scales are 50 mV and I ms.) The equilibrium point for pHj = 8.5 is unstable, as illustrated by the adjacent
trajectory. (Same scales as the current-voltage trajectory for pH, = 7.3.) This trajectory spiraled out to the limit
cycle described by the spontaneous action potentials shown adjacent to the current-voltage relation. Scales are
50 mV and 20 ms.

pH, 8.5 as compared to pH, 7.2 (symbols (•) in Fig. 3B).
This result suggests that a change of pH, might affect the
third component of the Hodgkin and Huxley ( 1 952) model
of the action potential, namely the background, or "leak"
current, /, . In particular, the resistance measurements in
Figure 3 imply that the leak component is reduced by an
increase in pH,. This idea has precedence in the work of
Bevan and Yeats (1991), who reported the activation of a
sustained, nonspecific cation conductance in a subpopula-
tion of rat dorsal root ganglion neurons by extracellular
protons. Alkaline pH would reduce the amplitude of this
conductance.

The background current in squid giant axons also consists
of a small-amplitude tetrodotoxin-sensitive sodium ion cur-
rent (referred to as /N;|P) that is activated at relatively
negative potentials, about —80 mV, and has a peak ampli-
tude at —60 mV (Rakowski et ai, 1985). This component
has a current-voltage relation with a negative slope charac-
ter at subthreshold potentials, whereas 7L — a net inward
current component at subthreshold potentials — has an ap-

proximately linear current-voltage relation. We are propos-
ing that the reduction of 7L with increasing pH, allows the
negative slope character of /NaP to destabilize the equilib-
rium point (rest potential) of the axon at -60 mV, thereby
resulting in autonomous activity. We cannot exclude a pHj
dependence of /Nap. However, the pH, -induced change in
resistance illustrated in Figure 3 is not attributable to
changes in /N.,P. since those experiments were carried out in
the presence of TTX, and as shown below, this resistance
change is sufficient to explain the pH,-induced automaticity.
The mechanism we propose for the result in Figure 3 is
illustrated by the simulations and current-voltage relations
in Figure 4. These results are based on the equations pro-
vided in the Materials and Methods. The steady-state cur-
rent voltage relation of the model for control conditions
(pH, 7.3) for -90 < V < -55 mV is shown in Figure 3,
along with an action potential (AP) elicited by a brief,
suprathreshold current pulse. Only a single AP was elicited
in the model even by relatively long-duration current pulses —
regardless of pulse amplitude — as in the earlier analysis

ACTION POTENTIALS WITH ALKALINE pH,

191

(Clay. 1998). The stability of the model for pH, 7.3 is
illustrated by the current-voltage trajectory in the inset of
Figure 3 immediately below the pH, 7.3 equilibrium point.
In this simulation the membrane potential was abruptly
shifted a few millivolts away from equilibrium conditions.
The current-voltage trajectory subsequently spiraled toward
the resting potential (stable focus). The effect of changing
pH, to 8.5 is also shown in Figure 3. The sole change in the
model was a reduction of the leak current conductance, gL,
from 0.2 to 0.03 mS • cm"2. This change resulted in a
hyperpolarization of the equilibrium point from —57.6 to
—59.3 mV and a change in its stability properties from a
stable to an unstable focus (inset above the voltage axis in
Fig. 3). This trajectory spiraled toward a stable limit cycle.
that is. autonomous firing, as illustrated by the inset to the
left of the current-voltage trajectory, with a frequency of
tiring of 29.8 Hz.

Discussion

Excitability of the squid giant axon preparation //; vitro
has traditionally been increased by reductions in the extra-
cellular Ca2 + concentration (Huxley. 1959; Guttman and
Barnhill. 1970). In preliminary experiments, we occasion-
ally observed autonomous activity with 1A normal Ca2+ (2.5
mM), but the effect was transient and episodic. In all prep-
arations examined, repetitive firing was not observed either
autonomously or with current pulse stimulation 15-20 s
after the change to an external medium that was low in
Ca2*. Moreover, axons became inexcitable within a few
minutes in low Ca2+ seawater. This result is not surprising
given that low Ca2+ external medium is deleterious for
neurons (Horn. 1999). The pH,-induced autonomous activ-
ity we report here is reproducible, robust, and long-lasting.
In one axon. we observed stable repetitive firing for 4 h
(with perfusion both intracellularly and extracellularly to
maintain ionic gradients), at which point the experiment
was terminated. Consequently, this preparation may be an
ideal single-cell neuronal oscillator suitable for investiga-
tions concerning mechanisms of rhythmicity.

The ionic model that we propose for the spontaneous
activity is novel and counterintuitive, in that the effect is
attributable to a reduction of inward current, thereby leading
to a destabilization of the rest potential by the /NaP compo-
nent. The result in Figure 2C illustrating subthreshold os-
cillations that increase in amplitude until the threshold for
an action potential is reached is consistent with this aspect
of the model. The only stable element both in the prepara-
tion and the model is the limit cycle, that is. the trajectory
traversed in current-voltage space by the action potential
(Winfree. 1980).

Repetitive firing in nerve cells is well known in a number
of preparations, such as gastropod neuronal somata (Connor
and Stevens. 1971). The rapidly inactivating potassium ion

current. 7A. is believed to play a major role in the activity
(Connor and Stevens. 1971). The delayed rectifier. /K. in
squid giant axons also inactivates, as originally shown by
Ehrenstein and Gilbert (1966). We think that this kinetic
feature does not play a role in our observations because the
mactivation kinetics are shifted rightward along the voltage
axis by an increase in pH, (Clay, 1990), and the onset of
inactivation at 5 C is too slow to be a factor during the
relatively brief times the membrane potential is at depolar-
ized potentials during the action potentials in the pulse train
(Clay, 1989). Moreover. £K inactivation cannot account for
the destabilization of the resting potential illustrated in
Figure 1C. which we believe is the key feature underlying
our results.

To our knowledge, the effect, reported here, of pH, on
excitability in squid axons has not been previously reported.
A similar effect with pH0 was noted in passing by Bicher
and Ohki (1972) in their work with intracellular pH elec-
trodes. They observed an increase in excitability in the giant
axon, including autonomous firing in some preparations,
after the pH of the extracellular bathing medium was raised
to 9. The change in pH0 caused a few tenths rise in pHj,
which we have shown to be sufficient to induce automatic-
ity. It is tempting to speculate that our observations have
physiological relevance, given that they occur within the
normal range of pH in the ocean (7.5 to 8.4: Sverdrup et ai.
1942), and only slightly above the normal, relatively alka-
line, value of 7.3 in the axon (Boron and DeWeer, 1976).
Moreover, transient rises in pH in squid blood have been
reported in exercising squid (Portner et ai, 1991 ). which,
based on our work, would favor an increase in neuronal
excitability. However, not enough is known about the role
of pH, in squid behavior to make an informed conjecture
about the role of the increased excitability in vivo, if it
indeed occurs.

Acknowledgments

We gratefully acknowledge grant support for this work
from the Canadian Institutes for Health Research (A.S.).

Literature Cited

Adams, I). J., and G. S. Oxford. 1983. Interaction of internal anions

with potassium channels of the squid giant axon. J. Gen. Physiol. 82:

424-448.
Bevan, S., and J. Yeats. 1991. Protons activate a cation conductance in

a suhpopulation of rat dorsal root ganglion neurons. J. Physiol. {Land.}

433: 145-161.
Bicher, H. I., and S. Ohki. 1972. Intracellular pH electrode experiments

on the squid giant axon- Bioi him. Biophys. Acki 255: 9(10-904.
Boron, \V. F., and P. DeXVi'tr. 1976. Intracellular pH transients in squid

giant axons caused by CO,. NH,. and metabolic inhibitors. J. Gen.

P/m/o/. 67: 91-112.
Brodwick, M. S., and D. C. Katon. 1978. Sodium channel inactivation

in squid axon is removed by high internal pH or tyrosine specific

reagents. Science 200: 1494-1496.

192

J. R. CLAY AND A. SHRIER

Clay, J. R. 1988. Lack of effect of internal fluoride ions on potassium

channels in squid giant axons. Biophys. J. 53: 647-648.
Clay, J. R. 1989. Slow inactivation and reactivation of the K* channel

in squid axons. A tail current analysis. Biophys. J. 55: 407-414.
Clay, J. R. 1990. /K inactivation in squid axons is shifted along the

voltage axis by changes in the intracellular pH. Biophys. J. 58: 797-

801.
Clay, J. R. 1998. Excitability of the squid giant axon revisited. / Neu-

rophysiol. 80: 903-913.
Clay, J. R., and M. F. Shlesinger. 1983. Effects ot external cesium and

rubidium on outward potassium currents in squid axons. Biophys. J. 42:

43-53.
Connor, J. A., and C. F. Stevens. 1971. Prediction of repetitive firing

behavior from voltage clamp data on an isolated neuronal somata.

J. Physio/. (Land.) 213: 31-53.
Ehrenstein, G., and D. L. Gilbert. 1966. Slow changes of potassium

permeability in the squid giant axon. Biophvs. J. 6: 553-566.
Guttman, R. S., and R. Barnhill. 1970. Oscillation and repetitive firing

in squid axons. J. Gen. Physiol. 55: 104-118.
Hodgkin, A. L., and A. F. Huxley. 1952. A quantitative description of

membrane conductance and its application to conduction and excitation

in nerve. J. Physiol. (Land.) 117: 500-544.
Horn, R. 1999. The dual role of calcium — Pore blocker and modulator of

gating. Proc. Natl. Acad. Sci. USA 96: 3331-3332.
Huxley, A. F. 1959. Ion movements during nerve activity. Ann. NY Acad.

Sci. 81: 221-246.
Neumeister, H., B. Ripley, T. Preuss, and W. F. Gilly. 2000. Effects of

temperature on escape jetting in the squid Loligo Opalescens. J. Exp.
Biol. 203: 547-557.

Otis, T. S., and W. F. Gilly. 1990. Jet-propelled escape in the squid
Loligo opalescens: concerted control by giant and non-giant motor
axon pathways. Proc. Natl. Acad. Sci. USA 87: 2911-2915.

Portner, H. O., D. M. Webber, R. G. Boutilier. and R. K. O'Dor. 1991.
Acid-base regulation in exercising squid (Illex illecebrosus, Loligo
peaiei}. Am. J. Physiol. 261: R239-R246.

Rakowski, R., P. DeWeer, and D. Gadsby. 1985. Threshold channels
can account for steady-state TTX-sensitive sodium current of squid
axon. Biophys. J. 47: A31.

Sverdrup, H. U., M. W. Johnson, and R. H. Fleming. 1942. Pp.
194-195 in The Oceans, Their Physics. Chemistry, and Genera/ Biol-
ogy. Prentice-Hall. New York.

Vandenberg, C. A., and F. Bezanilla. 1991. A sodium channel gating
model based on single channel, macroscopic ionic, and gating currents
in the squid giant axon. Biophys. J. 60: 1511-1533.

Wanke, E., E. Carbone, and P. L. Testa. 1980a. K+ conductance
modified by a titratable group accessible to protons from the intracel-
lular side of the squid axon membrane. Biophvs. J. 26: 319-324.

Wanke, E., E. Carbone, and P. L. Testa. 1980b. The sodium channel
and intracellular H+ blockage in squid axons. Nature 287: 62-63.

Winfree, A. T. 1980. The Geometry of Biological Time. Springer- Ver-
lag, Berlin.

Young, J. Z. 1939. Fused neurons and synaptic contacts in the giant
nerve fibers of cephalopods. Philos. Trans. R. Soc. Land. B 229:
465-503.

Reference: Biol. Bull. 201: 143-203. (October 2001 1

Molecular Phylogeny of the Model
Annelid Ophryotrocha

THOMAS G. DAHLGREN1 2 *, BERTIL AKESSON2. CHRISTOFFER SCHANDER2'3,
KENNETH M. HALANYCH1, AND PER SUNDBERG2

' Woods Hole Oceano graphic Institution, Biologv Department, MS 33, Woods Hole, Massachusetts

02543, USA; Goteborg Universitv, Department of Zoology, Box 463, 405 90 Goteborg, Sweden; and

^University of Copenhagen, Arctic Station, Box 504, DK-3953 Qec/ertarsuaq, Greenland

Abstract. Annelids of the genus Ophryotrocha are small
opportunistic worms commonly found in polluted and nutrient-
rich habitats such as harbors. Within this small group of about
40 described taxa a large variety of reproductive strategies are
found, ranging from gonochoristic broadcast spawners to se-
quential hermaphroditic brooders. Many of the species have a
short generation time and are easily maintained as laboratory
cultures. Thus they have become a popular system for explor-
ing a variety of biological questions including developmental
genetics, ethology, and sexual selection. Despite considerable
behavioral, reproductive, and karyological studies, a phyloge-
netic framework is lacking because most taxa are morpholog-
ically similar. In this study we use 16S mitochondrial gene
sequence data to infer the phytogeny of Ophryotrocha strains
commonly used in the laboratory. The resulting mtDNA to-
pologies are generally well resolved and support a genetic split
between hermaphroditic and gonochoristic species. Although
the ancestral state could not be unambiguously identified, a
change in reproductive strategy (i.e.. hermaphroditism and
gonochorism ) occurred once within Ophryotrocha. Addition-
ally, we show that sequential hermaphroditism evolved from a
simultaneous hermaphroditic ancestor, and that characters pre-
viously used in phylogenetic reconstruction (i.e.. jaw morphol-
ogy and shape of egg mass) are homoplasic within the group.

Introduction

Marine annelids belonging to the group Ophryotrocha
have been used as a laboratory system for much of this
century (e.g.. Bergh, 1895: Bergmann. 1903: Meek.

Received 15 December 2000; accepted 11 June 20111
*To whom correspondence should be addressed. E-mail: tdahlgren®
whoi.edu

1912; Huth, 1933; Hartmann and Lewinski, 1940; Bacci
and La Greca. 1953; Bacci, 1965; Akesson. 1972; Sella,
1988; Vitturi et ai. 2000), not least because they are easy
to maintain in cultures and have short generation times.
Oplirvotroclni has traditionally been treated as a genus
within the eunicimorph family Dorvilleidae (e.g., Fau-
chald, 1977; Eibye-Jacobson and Kristensen, 1994), but
inclusion within the Dorvilleidae has been challenged
(Orensanz, 1990). Many ecological (e.g., Akesson, 1977;
Berglund. 1991; Cassai and Prevedelli, 1999), ethologi-
cal (e.g., Sella, 1991), developmental (e.g., Akesson,
1967. 1973; Zavarzina and Tzetlin, 1991 ), and toxicolog-
ical (e.g., Akesson, 1970, 1975) studies have been con-
ducted on these worms. Charnov ( 1982) and Gambi et al.
(1997) among others, argue that Ophryotrocha is a near
ideal group for studies of the evolution of sex strategies,
since all known forms (gonochorism, sequential and si-
multaneous hermaphroditism) are represented within a
few closely related species.

Since Oplin'otrocha was first described (Claparede and
Mecznikow, 1869), more than 40 species have been
added to the group, most of which are reported from
shallow, nutrient-rich waters such as harbors (e.g.. La
Greca and Bacci, 1962; Akesson, 1976; Paavo et al..
2000). Recent contributions have also shown a consider-
able diversity in the deep sea (Jumars, 1974; Blake, 1985;
Hilbig and Blake. 1991: Lu and Fauchald. 2000). To the
best of our knowledge, the Appendix lists all the de-
scribed species of Ophryotrocha with their type locality.
Only species from shallow, temperate or tropical waters,
however, have been successfully cultured in the labora-
tory (at present —20 distinct forms). Among the cultured
forms, some of which are yet to be formally described.

193

194

T. G. DAHLGREN ET AL.

most taxa are morphologically identical; but each is
believed to be a distinct species because species crosses
failed to produce viable offspring in breeding experi-
ments (Akesson, 1978, 1984, and unpubl.).

Even though the taxonomical and descriptive morpho-
logical literature is extensive (e.g.. La Greca and Bacci.
1962; Pfannenstiel, 1972, 1975; Josefsson, 1975; Akes-
son, 1978; Oug, 1978. 1990; Blake, 1985: Ockelmann
and Akesson. 1990; Hilbig and Blake, 1991; Lu and
Fauchald, 2000; Paavo et cil., 2000), there has been but a
single study (Pleijel and Hide, 1996) focusing on the
phylogenetic history of a broader selection of Ophryotro-
cha species. That study, based on an analysis of 25
morphological characters and 7 electrophoretic protein
loci scored for 20 Ophrytrocha taxa, described two major
clades and suggested that simultaneous hermaphroditism
is the pleisiomorphic condition for the group as a whole.
Basal branching in Pleijel and Hide's tree (1996. Fig.
IB), however, was poorly supported, as depicted by their
highly collapsed strict consensus tree. One of the clades
found by Pleijel and Eijde (1996) is congruent with the
gonochoristic lahmnicti group previously suggested by
Akesson (1984). On the basis of morphological features,
such as similarities in jaw apparatuses and egg-mass
morphology, Akesson ( 1984) and Ockelmann and Akes-
son ( 1990) also recognized the hermaphroditic "gracilis"
and "hartmanni" groups. These taxa form a grade at the
base of the most parsimonious tree favored by Pleijel and
Hide (1996, Fig. I A), and they consequently suggested
that their similarities are sympleisiomorphic.

Chromosome number and karyological characters were
used in an analysis of nine species by Robotti and collab-
orators ( 1991 ). The resulting topology suggested that forms
with equal number of chromosomes constitute monophy-
letic groups. The data also indicated that the distance be-
tween O. robusta and O. Iwnnniiini (2n = 10) is larger than
the distances between members of the two other groups
(2n = 6 and 2n = 8). Further, on the basis of work by
Colombera and Lazzaretto-Colombera (1978) suggesting
that karyotypes evolved towards reduced chromosome num-
bers, Robotti et til. ( 1 99 1 ) proposed that O. rohiista occupies
an ancestral position. The same conclusion was reached by
Vitturi and collaborators (2000) in a study of karyotypes of
10 Ophryotrocha species.

The evolutionary relationships among 18 Ophryotrocha
forms were investigated using the mitochondria! 16S rDNA
gene. Due to organismal availability, we have focused on
intertidal forms that are easily kept in laboratory cultures
(Akesson, 1970, 1975). The present paper builds on the
work of Pleijel and Hide ( 1996) to reconstruct the phylog-
eny of this model annelid system so that knowledge gained
about Ophryotrocha may be assessed in a comparative
context.

Materials and Methods

Taxa

Eighteen Ophryotrocha cultures representing intertidal
and shallow-water forms were selected for this study (Table
1 ). All of these terminal taxa were readily available because
they are maintained as laboratory cultures by B.A. and
represent the best-studied members of Ophryotrocha. Nine
of the eighteen strains included are not formally described
as species. Six are referred to by the name under which they
will be described, followed by "nom. nud." to indicate their
present status as noniina nuda, i.e., names not available. The
remaining three strains are referred to by the location where
they were originally collected. The informally named taxa
are not to be regarded as descriptions sensn International
Code of Zoological Nomenclature (International Commis-
sion on Zoological Nomenclature. 1985).

Outgroup terminals were chosen on the basis of recent
analyses of eunicimorph and annelid phylogeny (Paxton.
1986; Orensanz, 1990; Eibye-Jacobson and Kristensen,
1994; Rouse and Fauchald, 1997). These taxa include Hya-
linoecia tuhicola (Miiller. 1776), Nothria conchylega (Sars,
1835). Eunice pennuhi (Miiller, 1776), Dorvil/ca alhomacii-
luta Akesson and Rice, 1992, and Dinophihis f>yrociliatn\
Schmidt, 1857. Specimens of H. tuhicola, N. conchylega,
and E. pciunita were collected by epibentic dredge, while
Don-illcti iilhoiiuicultitii and Dinophihis gy roc Hiatus were
acquired from laboratory cultures (BA). Locality details are
provided in Table 1.

Dutii collection

The worms were taken from cultures and placed in 70%
ethanol. Voucher specimens from the same cultures were
fixed in 5% formaldehyde for 1 h and subsequently trans-
ferred to 70% ethanol. The voucher specimens are deposited
in the Zoological Museum, Copenhagen (ZMUC) and des-
ignated the numbers given in Table 1 . DN A was extracted
by either employing a Chelex protocol (Sundberg and
Andersson, 1995) or a standard chloroform/phenol protocol
(Doyle and Dickson, 1987). An approximately 400-bp frag-
ment of the mitochondrial large subunit ribosomal RNA
gene was amplified with the universal primers 16Sar-L
(5'-cgcctgtttatcaaaaacat-3'( and 16Sbr-H (5'-ccggtctgaact-
cagatcacgt-3') according to standard protocols (Palumbi,
1996). The amplification profile was 40 cycles of 95 °C for
30 s, 50 °C for 30 s, 72 °C for 45 s with an initial single
denaturing step at 95 °C for 2 min. and a final single
extension step at 72 °C for 7 min. After spin-column puri-
fication (QiaGen, Inc.), the PCR products were sequenced in
both directions on a PharmaciaBiotech ALF-Express auto-
mated sequencer using the TermoSequenase kit (Amer-
shamPharmacia) and Cy-5 labeled 16Sar-L and !6Sbr-H

OPHRYOTROCHA PHYLOGENY

Table 1

Collection data and (jcnBtiiik accession numbers for Ophryotrocha ra.m examined

195

Taxon

Collection site

Coll.
vear

GenBank
Accession Nr

ZMUK

Voucher Nr

Don-illca alhinnaciilaia* Akesson and Rice, 1992

Tarifa. Spain

1990

AF380115

N/A

Dinnphilii.i gyrocilialus* Schmidt. 1857

Xiamen. China

1995

AF380116

N/A

Hyalinui'cia tubicola (Miiller. 1776)

Koster area, Sweden

1997

AF321416

N/A

Nothria conchylega (Sars. 1835)

Koster area, Sweden

1997

AF321417

N/A

Eunice pennala (Miiller. 1776)

Koster area. Sweden

1997

AF32I41S

N/A

O. attherena Paavo el ai. 2000

Kyrenia. Cyprus

1971

AF321419

ZMUC-POL-1110

O. alborana nom. nud.

Algeciras, Spam

1978

AF32I420

ZMUC-POL-1 111

O. cost/owl Akesson. 1978

Duke. NC. USA

1974

AF32I421

ZMUC-POL- 1112

O. diadema Akesson, 1976

L.A. harbor. USA

1972

AF321422

ZMUC-POL- 1113

O. gracilis Huth. 1934

Helgoland, Germany

1988

AF321423

ZMUC-POL-1 114

O. hartimmni Huth, 1933

Malaga, Spain

1990

AF321424

ZMUC-POL- II 15

O. japonica nom. nud.

L.A. harbor, USA

1989

AF321433

ZMUC-POL- 11 16

O. labronica La Greca and Bacci. 1962

Naples. Italy

1965

AF321425

ZMUC-POL- II 17

O. macrovifera nom. nud.

Cyprus

1972

AF321426

ZMUC-POL-1 118

O. notoglandulata Pfannenstiel. 1972

Misaki, Japan

1961

AF321427

ZMUC-POL- 11 19

O. obscura nom. nud.

Pet store. Sweden

1978

AF321436

ZMUC-POL- 11 20

O. permanni nom. nud.

Indian River. Florida

1991

AF321428

ZMUC-POL- 1121

0. puerilis Claparede and Mecznikow. 1869

Malaga. Spain

1990

AF321429

ZMUC-POL- 11 22

O. robusla nom. nud.

Malaga. Spain

1990

AF32I430

ZMUC-POL- 11 23

O. socialis Ockelmann and Akesson. 1990

Helsingor, Denmark

1986

AF321431

ZMUC-POL- 11 24

Eilat-Hurghada

Red Sea

1996

AF321432

ZMUC-POL- 11 25

Qingdao

Qingdao. China

1995

AF321434

ZMUC-POL- 11 26

Sanya sp. 2

South Hainan. China

1995

AF321435

ZMUC-POL- 11 27

; Sequenced by Arne Nygren

primers in accordance with the manufacturer's protocols.
GenBank accession numbers are given in Table 1.

Analysis

The sequences were aligned with Clustal X (Thompson
et uI.. 1994) and proofread by eye. Regions that could not
be unambiguously aligned were excluded from the anal-
ysis. The alignment is deposited at TreeBase and avail-
able at http://phylogeny.harvard.edu/treebase or from
TD. Neighbor-joining (NJ). parsimony, and maximum-
likelihood (ML) analyses were conducted with the
PAUP*4.0b2 software package (Swofford, 2000).
PAUP* was further used for parameter estimations for
the ML searches. For NJ, a Kishino-Hasegawa (1989)
likelihood test found no significant differences between
trees generated under the Jukes-Cantor. Kimura-2-param-
eter. Tamura-Nei, and general-time-reversible (GTR)
models (see Swofford et al.. 1996. for a brief description
of models). Parsimony trees were inferred from an un-
weighted character matrix (i.e., the transition/transver-
sion ratio was assumed to be 1 with the heuristic search
option using the tree-bisection-reconnection (TBR)
branch-swapping algorithm and 100 random-sequence
addition replicates. To reduce the computation time of
the ML search, the most parsimonious tree was used as

starting tree in the ML heuristic search. The model pa-
rameters were estimated from a likelihood analysis of the
most parsimonious tree and included a nucleotide model
with six substitution types (a GTR model), and among-
sites rate heterogeneity used a gamma distribution with
shape parameter of 0.30. A GTR model was chosen since
it is the most general of the ones mentioned above, all of
which are special cases of a GTR (Swofford et al.. 1996).
Nucleotide frequencies were set to empirical values.
Bootstrap analyses for both ML and parsimony employed
1000 iterations.

The four characters — (1) sex strategy, (2) jaw morphol-
ogy. (3) egg-mass morphology, and (4) diploid number of
chromosomes — were chosen in part on the basis of previous
efforts to estimate Ophryotrocha relationships (e.g.. Akes-
son, 1984; Robotti et al.. 1991: Pleijel and Eide. 1996;
Vitturi et al.. 2000). MacClade 3.06 (Maddison and Mad-
dison. 1992) was used to manipulate the molecular data, and
to map sex strategy and morphological and karyological
character state changes on the mtDNA topology. Character
states were scored following Pleijel and Eide (1996) except
for strains from Eilat-Hurghada, Sanya sp. 2, and Qingdao,
which were obtained from the same cultures as the speci-
mens sampled for sequencing. The diploid number of chro-
mosomes is not known for these three forms.

196

T. G. DAHLGREN ET AL.

Table 2

Paim-ise molecular distances (absolute number of unambigously aligned substitutions above diagonal and Jukes Cantor distances below I; Saitva s/>. 2
and O. obscura nom. nud. are distinguished in three positions located in the excluded regions of the alignment

1.1

-,-,

Dttn'iIIfti tilbuniai ttUirti

Oin/>i'liilii\ ^wi'c ilutru\

032

Enntf c pfitnalfi

025

030

H\iil/»"ccia Inhicola

030

0.28

0.14

Nolritl < cm /jv/ci,'i;

0.33

0.35

019

0.16

t> luiht'rens

030

035

0.33

034

0.37

O. atborana nom. nud

0,33

042

0.36

0.35

0.29

II ,:l\ll<m-i

0.34

035

0.42

039

037

030

032

SI)

O. diudema

0.33

0.43

031

0.31

0.32

0.27

0.18

029

O. gracilis

0.33

1)41

036

033

0.38

016

0.35

0.33

026

O lutrtmtinni

033

030

0.32

0.35

035

016

0.31

036

026

0 15

O lubnmHU

0 37

0.41

0,44

042

0.40

0.31

032

008

0.31

034

035

O mticrm-ifeni nom nud.

0.33

034

0.39

0.38

0.37

0.27

0.29

0 10

0.30

0.14

0.31

(I 111

<> nologlandulata

033

038

0.37

034

0.36

030

0.31

II 1 1

024

033

(135

0 12

O permaiini nom. nud

036

0.37

0.45

039

029

0.32

006

030

0.35

0 35

0.06

0.10

Oil

O. pitfrihf.

0.16

033

0.34

0.15

0 39

024

Oil

029

024

025

026

0 10

029

O, niftusru nom nud.

0.30

0.34

0.31

0.33

0.35

025

029

0.2

0.25

0.29

0 30

020

021

0.18

0 19

026

O. aiualis

0.28

0.2")

0.12

0.32

0.34

009

0.29

0.28

0 17

I) 17

029

025

0.29

029

0.2?

025

Eilal-Hurghada

0.32

0,34

038

036

0.34

0.26

030

0 13

0.30

I) 14

I) 13

0 12

0.09

0 1 3

0.11

dill

1)21

026

O. jttpnnica nom. nud.

II 11

0.36

0.34

034

0.35

0.31

0.32

0.11

1)24

034

0.34

0.12

012

0.03

0.12

029

0.17

0.29

0 12

Qingdao

0.38

0.42

0.46

041

0.41

0.28

1)26

0.31

1)21

1)32

0 12

0.35

0.33

0.30

0.31

032

0.28

0,29

0.31

033

Sanya sp, 2

0.36

0.37

0.40

0.37

029

0.32

0.07

029

035

0.14

0.04

o 10

0 II

0.05

029

0.20

0.29

0 12

0 1 1

035

O. <'/>u lira nom nud

O.lo

0.17

I) 4(1

037

0.37

029

0.32

007

0 29

1)35

0.14

004

0 10

0 1 1

005

0 29

0 20

0 29

0 12

II I 1

035

000

Results

The data set consisted of 23 terminal taxa and 485
nucleotide positions. Of the 282 nucleotide positions that
could be unambiguously aligned. 55.0% (155 positions)
were variable and 45.7% (129 positions) were parsimony
informative. Table 2 shows the Jukes-Cantor distances
(below diagonal) and absolute distances (above diagonal)
for the alignment. Figure 1 A shows the highest scoring likeli-
hood tree recovered (Ln likelihood —2741.47651).
More terminal branches were generally well resolved, but
support for basal nodes was low (see bootstrap tree.
Fig. IB).

In the ML tree, two large clades emerge. The first, for
brevity here called A, is further divided into two clades,
Al and A2. Clade Al consists of O. hartmanni, O.
gracilis, O. adherens, O. socialis, and O. puerilis; A2
comprises O. u/horniui. O. diadema, and "Qingdao." The
second major clade, here called B, includes O. japonica
nom. nud., O. notoglandalata, O. costlowi. O. labronica,
Sanya sp. 2, O. obscuru nom. nud.. O. pennunni, O.
incicroriferu nom. nud., Eilat-Hurghada, and O. robusta.
The parsimony and NJ trees (Fig. 2A) differ from the ML
tree in the placement of the root. Instead of a monophy-
letic A group, as in the ML tree, A constitutes a grade
where the two clades Al' (excluding O. puerilis) and A2,
together with O. puerilis, lead to B (Fig. 2A). The general
topologies are. however, similar in the parsimony, NJ,
and ML trees, and most of the nodes are supported over
the 50% bootstra] level (Fig. IB).

In an attempt to assess support for basal nodes, an addi-
tional analysis was performed on an alignment of ingroup
taxa only (aligning and ML procedures as described above).
Inclusion of more distant taxa in an alignment may reduce
the nucleotide positions that can be unambiguously aligned,
and a more restricted selection of taxa could potentially
increase phylogenetic signal by allowing for a "better"
alignment (Halanych et ul.. 1998). The analysis, of the 18
Ophryotrocha ingroup taxa only, did reveal higher boot-
strap support for internal branches of the tree. This restricted
analysis, however, gave lower support or alternative hy-
potheses for some of the more recent clades. More recent
clades in B are less well resolved than in the original
analyses, but O. japonicu nom. nud. and O. notoglandalata
form a strongly supported monophyletic group (Fig. 2B).

Figure 3 shows an arbitrarily chosen most parsimonious
reconstruction (transformation optimization by ACCTRAN)
for each of the four characters mapped on the ML tree (Fig.
1A). Evidence of transformation polarity is given from
outgroup analysis of mtDNA data, and traced characters are
accordingly not scored for outgroup taxa.

Discussion

Ophryotrocha phylogeny was investigated by employing
ML, NJ, and parsimony analyses of 16S rDNA data. One
alignment of these data included the ingroup and five out-
group taxa and resulted in poor support for basal branching
patterns (Fig. IB). The second data set was limited to the 18
Ophryotrocha terminals (i.e., the ingroup) and produced
better supported and nearly identical topologies under ML

OPHRYOTROCHA PHYLOGENY

197

1 Eunice pennata

Hyalinoecia tubicola

Notria conchylega

• Dorvillea albomaculata
— Dinophilus gyrociliatus

O. hartmanni

— ^— — O. gracilis

O. adherens
• O. socialis
' O. puerilis

O. alborana nom. nud."

O. diadema

0.1 substitutions/site

^^^^-^—^— Qingdao
O.japonica nom. nud.

^— O. costlowi

i— O. labronica
Mi Sanya sp.2
H ' O. obscura nom. nud.
•— O. permanni nom. nud.
I O. macrovifera nom. nud.
I Eilat-Hurghada

O. notoglandulata
O. robusta nom. nud.

66/59

70/52

. Hyalinoecia tubicola

• Notria conchylega

• Eunice pennata

• Dorvillea albomaculata
. Dinophilus gyrociliatus

• O. hartmanni

• O. gracilis

• O. socialis

• O. adherens

• O. alborana nom. nud.

• O. diadema
. Qingdao

• O. puerilis

• O.japonica nom. nud.

• O. notoglandulata

• O. costlowi

• O. labronica

• Sanya sp.2

1 O. obscura nom. nud.

• O. permanni nom. nud.

• O. macrovifera nom. nud.

• Eilat Hutg

• O. robusta nom. nud.

Figure 1. (A) Best maximum likelihood tree. -In = 2741.47651. Highlighted nodes indicate clades
congruent with morphological analysis by Pleijel and Hide (1996). (B) 1000 bootstrap consensus. Numbers in
upright type are maximum likelihood and in italic are parsimony support values.

198

T. G. DAHLGREN ET AL

Hyalinoecia tubicola

Notria conchylega

Eunice pennata

Dorvillea albomaculata

Dinophilus gyrociliatus

O. hartmanni

O. gracilis

O. socialis

O. adherens

O. alborana

O. diadema

Qingdao

O.japonica nom. nud.

O. notoglandulata

O. costlowi

O. labronica

Sanya sp.2

O. obscura nom. nud.

O. permanni

O. macrovifera

Eilat-Hurghada

O. robusta

O. puerilis

78/62

65/76

81/97

95/93

64/66

94/100

87/99

76/97

64/92

53AS6

95/99

• O. hartmanni

• O. gracilis

• O. adherens

• O. socialis

• O. puerilis

• O. alborana nom. nud.

• O. diadema

• Qingdao

• O.japonica nom. nud.
1 O. notoglandulata

• O. costlowi

• O. labronica

• O. macrovifera nom. nud.

• Eilat-Hurghada

• O. permanni nom. nud.

• Sanya sp.2

• O. obscura nom. nud.

• O. robusta nom. nud.

Figure 2. (A) Parsimony and neighbor-joining tree topology. (B) Unrooted bootstrap tree from alignment of
ingroup taxa only and drawn to represent the rooting suggested in the original maximum likelihood analysis.
Bootstrap values for maximum likelihood in upright type and for parsimony in italic type.

OI'HKYO'l'ROCHA PHYLOGENY

199

O. harttnanni
O. gradlis
O. adherens
0. socialis
O. puerilis

O. alborana nom. nud.
O. diadema
Qingdao

O.japonica nom. nud.

O. cost/oil1!

O. labronica

Sanya sp.2

O. obscura nom. nud.

O. permanni nom. nud.

O. macrovifera nom. nud.

Eilat-Hurghada

O. notoglandulata

O. robust a nom. nud.

0. harttnanni
O. gradlis
O. adherent
0. socialis
O. puerilis

O. alborana nom. nud.

O. diadema

Qingdao

O.japonica nom. nud.

0. costloun

O. labronica

Sanya sp. 2

O. obscura nom. nud.

O. permanni nom. nud.

0. macrovifera nom. nud.

Eilat-Hutghada ?

O. notoglandulata

O. robusta nom. nud.

Figure 3. Character transformation hypothesis. Alternative rooting as suggested by parsimony analysis is
indicated by an arrow. Dashed gray line depicts equivocal or unknown taxon state. (A) Sex strategy. Full line =
gonochorism; dashed line = simultaneous hermaphroditism; gray line = sequential hermaphroditism. (B) Jaws.
Full line = P and K-type of jaws; dashed line = presence of P-type only. (C) Shape of egg mass. Full line =
tubular; gray line = fusiform; dashed line = irregular. (D) Diploid number of chromosomes. Full line = 6; gray
line = 8; dashed line = 10.

and parsimony methods (Fig. 2B). Topologies from the
analyses restricted to the ingroup taxa were in overall agree-
ment with the ML and parsimony tree of the first dataset.
The problem with alternative hypotheses for placement of
the root may be caused by a long branch phenomenon (<'.#.,
Hendy and Penny, 1989) and cannot be conclusively re-
solved with the data at hand. The parsimony analysis sug-
gests a placement of the root between the (O. luirtmanni, O.
gradlis, O. adherens, O. socialis) clade and the rest of the
tree (Fig. 2A). while ML indicates a rooting between the
clade of hermaphroditic species and the gonochristic species
clade (Fig. 1 A). However, contrary to parsimony, ML meth-
ods account for branch-length information and should give
a better estimate when the model is accurate (Swofford et
ul.. 1996). Therefore, the discussion below focuses on the
ML tree that included 18 ingroup and 5 outgroup taxa.

Figure 1A shows considerable congruence with Pleijel
and Eide's (1996) results from an analysis employing mor-

phological, sex strategy, and protein data. Clades supported
by both sets of data are indicated with the highlighted nodes
in the mtDNA tree (Fig. 1 A). However, the topologies differ
on the internal branching of Opliryoirocha and, possibly, on
the placement of the root. The mtDNA ML data gives some
support for a deep subdivision in two major clades, but no
such subdivision is suggested by Pleijel and Hide (1996).
Recent fossil evidence further suggests that Ophiyotrocha is
an old lineage (Eriksson and Lindstrom, 2000).

The evolution of sex strategies is a debated subject (e.g.,
Ghiselin, 1969; Charnov. 1982; Maynard Smith. 1982:
Hurst. 1992). Our analyses, based on species that are se-
quential ( 1 ) or simultaneous (7) hermaphrodites, and gono-
chorists (10), suggest that, regardless of the placement of
the root, the change from one strategy to the other has taken
place only once within the group (Fig. 3 A). The ancestral
state is, given the available data, ambiguous. The sequential
hermaphrodite O. puerilis is able to switch sex several times

200

T. G. DAHLGREN ET AL

during life, a feature that is rare among metazoans. Using O.
puerilis as a model, Premoli and Sella ( 1995) discussed the
ecological constraints necessary for an evolution from se-
quential to simultaneous hermaphroditism. Our data instead
suggest that an evolution in the opposite direction, from
simultaneous to sequential hermaphrodites, is more proba-
ble within Ophiyotrocha. Such a scenario is also hinted at
by A. Berglund, who — according to Premoli and Sella
(1995) — commented that O. puerilis is "a modified simul-
taneous hermaphrodite in which a reversible mechanism of
temporal inhibition of one of the two sexual phases has
evolved." The problem of whether gonochorism or her-
maphroditism is the ancestral state was also thoroughly
discussed by Sella and Ramella ( 1999). However, they did
not take up a definite position.

In addition to reproduction, jaw morphology has been
used to understand Ophryotrocha relationships. The hind-
most pair of maxillary plates in Ophryotrocha species can
be of two distinct types. Whereas the P-type has a distal row
of fang-like denticles, the K-type plates are distally smooth
but often of a robust construction (Hartmann and Huth.
1936). Whereas a P-type jaw is found in larvae and juve-
niles of all species, the character state for adult worms is
either P or K (e.g., Ockelmann and Akesson. 1990). The
terminology emanates from the German words "kompliz-
iert" (K-type) and "primitiv" (P-type). Based on reproduc-
tion strategy and jaw morphology, Akesson (1973. 1984)
identified the "labronica," the "hartmanni," and the "gra-
cilis" groups within Ophryotrocha. The "labronica" group
of sibling species consists of gonochorists with the K-type
of jaws; it is well supported by the analysis presented here
and represents clade B in Figure 1A. The "hartmanni" and
"gracilis" groups comprise hermaphroditic species. Species
belonging to the former group are distinguished by, among
other characters, spawning an entirely soft, irregularly
shaped egg mass, and the presence of K-type jaws (Ockel-
mann and Akesson, 1990). In contrast, the "gracilis" group
spawns a fusiform egg mass with a hard protective outer
layer, and carries P-type jaws only. The monophyly of the
two groups of hermaphrodites, "gracilis" and "hartmanni,"
are, however, not validated by the present analysis. On the
contrary, the characters "shape of egg mass" and "type of
jaws" are homoplasious in all our trees, irrespective of the
phylogenetic optimization used (i.e., ML or parsimony) and
the placement of the root (Fig. 3B and C). Therefore, in our
trees O. gracilis is no longer a member of the "gracilis"
group despite the close points of similarity in reproductive
traits and morphology between this species and other mem-
bers of the group (Ockelmann and Akesson, 1990; Pleijel
and Hide, 1996). A more extensive study seems to be
justified.

Cytology and karyology have been extensively studied
for a variety of species. Diploid numbers of chromosomes
are known for 18 species (Akesson, 1984; Robotti et al..

1991; Shaojie and Knowles, 1992) that have 2n = 6, 2n =
8 or 2n = 10. The genome size (measured as picograms of
DNA per cell) in 10 studied forms was discontinuously
distributed between 0.4 pg (8 taxa). 0.8 (1 taxon) and 1.16
pg ( 1 taxon) (Sella et al.. 1993: Soldi et al., 1994; Gambi et
al.. 1997). The apparent discontinuous distribution of ge-
nome size (i.e., ^0.4. «=0.8, or ^1.2) was interpreted as an
indication that large parts of the genome are acquired si-
multaneously (Sella et al., 1993). The increments in genome
size, however, do not correspond to increased numbers of
chromosomes (Gambi et al.. 1997). The position of chro-
mosomal nucleolar organizer regions (NOR) has been char-
acterized and found to be highly polymorphic not only
within the genus but also within most of the species (Sella
et al., 1995; Vitturi et al., 2000). On the basis of inferred
low GC contents and only one pair of NOR carrying chro-
mosomes (studied by fluorescent in situ hybridization), Vit-
turi et al. (2000) suggested that O. robusta, with 2n = 10
and a small genome size (0.4 pg), is pleisiomorphic within
the group. A basal position of O. robusta within the "la-
bronica" group is corroborated by the mtDNA data (Fig. 1).
Unfortunately, since O. rubiista is the only studied species
with this combination of characters, it is impossible to tell if
this is an autapomorphy or a sympeisiomorphy.

To summarize, this study presents the first mtDNA gene
tree for Ophryotrocha species. Examination of the tree,
which provides independent data for evaluating the evolu-
tion of reproductive strategies, leads us to suggest that
hermaphroditism or gonochorism evolved once within stud-
ied Ophryotrocha taxa and that sequential hermaphroditism
evolved from simultaneous hermaphroditism.

Acknowledgments

We are indebted to Arne Nygren for letting us use pre-
viously unpublished sequences of Donillea albomaculata
and Dinophilus gyrociliatus. Comments by Fredrik Pleijel
as well as three anonymous referees helped improve the
manuscript. Barbro Lofnertz is acknowledged for assistance
in the laboratory. Financial support for this study was pro-
vided by Swedish Natural Science Research Council (fiA,
PS). National Science Foundation (DEB-0075618) (KMH),
and Knut och Alice Wallenbergs Stiftelse (PS). This is
contribution number 10442 to WHOI.

Literature Cited

Akesson, B. 1967. On the biology and larval morphology of Ophryotro-
cha puerilis Claparede and Metschnikov (Polychaeta). Ophelia 4:
110-119.

Akesson, B. 1970. Ophiyotrocha labronica as test animal for the study
of marine pollution. Helgol. Wiss. Meeresunters. 20: 293-303.

Akesson, B. 1972. Sex determination in Ophryotrocha labronica
(Polychaeta. Dorvilleidae). Pp. 163-172 in Fifth European Marine
Biology Symposium, B. Battaglia. ed. Piccin Editore. Padova.

Akesson, B. 1973. Reproduction and larval morphology of five

OPHRYOTROCHA PHYLOGENY

201

Onhrvotrocha species (Polychaeta. Dorvilleidae). Zoo/. Scr. 2: 145—
155.

Akesson, B. 1975. Bioassay studies with polychaetes of the genus
Ophryotrocha as test animals. Pp. 121-135 in Sublethal Effects of
To\u Chemicals mi At/until- Animals. .1. H Koeman and J. J. T. W. A.
Strik, eds. Elsevier. Amsterdam.

Akesson, B. 1976. Morphology and life cycle of Ophryotrm ha diadcma.
a new polyehaete species from California. Ophelia 15: 23-35.

Akesson, B. 1977. Crossbreeding and geographic races: experiments
with the polvchaete genus Ophryolrocha. Mikrofauna ties Meeresbo-
Jcns 61: I 1-18.

Akesson, B. 1978. A new Ophryotrocha species of the Labronica group
(Polychaeta, Dorvilleidae) revealed in crossbreeding experiments. Pp.
573-590 in NATO Conference Series (Marine Science), B. Battaglia
and J. Beardmore. eds. Plenum Publishing, New York.

Akesson, B. 1984. Speciation in the genus Ophryotrocha (Polychaeta,
Dorvilleidae). Pp. 299-316 in Polvchaete Reproduction, G. Fischer
and H.-D. Pfannenstiel, eds. Gustav Fischer Verlag, Stuttgart.

Bacci, G. 1965. Sex determination and genetic balance of Ophryotrocha
puerilis. a hermaphrodite polychaete worm. Nature 207: 1208-1209.

Bacci, G., and M. La Greca. 1953. Genetic and morphological evidence
for subspecific differences between Naples and Plymouth populations
of Ophryotrocha puerilis. Nature 171: 1115.

Bergh. R. S. 1895. Neue Untersuchungen Uber Ophryotrocha und uber
Anneliden-larven. Zoologische Centralblatt. Leipzig 2: 257-263.

Berglund, A. 1991. To change or not to change sex: a comparison
between two Ophryotrocha species (Polychaeta). Evol. Ecol. 5: 128-
135.

Bergmann, W. 1903. Untersuchungen uber die Eibildung bei Anneliden
und Cephalopoden. Z. Wiss. Zoo/. 73: 278-301.

Blake, J. A. 1985. Polychaeta from the vicinity of deep-sea geothermal
vents in the eastern Pacific. I. Euphrosinidae, Phyllodocidae, Hesion-
idae. Nereididae, Glyceridae. Dorvilleidae, Orbiniidae, and Maldani-
dae. Bull. Biol. Soc. Wash. 6: 67-101.

Cassai, C., and D. Prevedelli. 1999. Fecundity and reproductive effort
in Ophryotrocha labronica (Polychaeta: Dorvilleidae). Mar. Biol. 133:
489-494.

Charnov. E. L. 1982. Sex Allocation. Princeton University Press. Prince-
ton. NJ.

Claparede. I)., and E. Mecznikow. 1869. Beitrage zur Kenntnis der
Entwickelungsgeschichte der Chaetopoden. Z. Wiss. Zool. 19: 163-
205.

Colombera, D., and I. Lazzaretto-Colombera. 1978. Chromosome
evolution in some marine invertebrates. Pp. 487-525 in Marine Or-
ganisms. Genetics. Ecology and Evolution. B. Battaglia and J. A.
Beardmore, eds. Plenum Press, New York.

Doyle, J. J., and E. Dickson. 1987. Preservation of plant samples for
DNA restriction endonuclease analysis. Taxon 36: 715-722.

Eihye-Jacobsen, D., and R. M. Kristensen. 1994. A new genus and
species of Dorvilleidae (Annelida, Polychaeta) from Bermuda, with a
phylogenetic analysis of Dorvilleidae, Iphitimidae and Dinophilidae.
Zool. Scr. 23: 107-131.

Eriksson, M.. and S. Lindstrom. 2000. Ophryotrocha sp., the first
report of a jawed polychaete from the Cretaceous of Skane, Sweden.
Ada Paleontol. Pol. 45: 311-315.

I .UK I i.i M. K. 1977. The polychaete worms. Definitions and keys to the
orders, families and genera. Nat. Hist. A/MS. Los Angel. Cry. Sci. Ser.
28: l-ixs.

Gambi. M. C.. L. Ramella, G. Sella, P. Protto. and E. Aldieri. 1997.
Variation in genome size in benthic polychaetes: Systematic and eco-
logical relationships. J. Mar. Biol. Assoc. U.K. 77: 1045-1057.

Ghiselin. M. T. 1969. The evolution of hermaphroditism among ani-
mals. Q. Rev. Biol. 44: 189-208.

Halanych, K. M., R. A. Lutz, and R. C". Vrijenhoek. 1998. Evolution-

ary origins and age of vestimentiteran tube-worms. Call. Biol. Mar. 39:
355-358.

Hartmann, M., and \V. Huth. 1936. Untersuchungen uber Geschlechts-
bestimmung und Geschlechtsumwandlung von Ophryotrocha puerilis.
Zool. Jahrb. Abt. A/It;. Zool. Plnsiol. Tiere 56: 389-439.

Hartmann. M., and G. von Lewinski. 1940. Untersuchungen uber die
Geschlechtsbestimrnung und Geschlechtsumwandlung von Ophr\otro-
cha puerilis. 111. Die stotfliche Natur der vermannlichenden Wirkung
"starker" Weibchen ("Eistoffe"). Zool. Jahrb. Abt. Allg. Zool. Physiol.
Tiere 60: 1-12.

Hendy, M. D., and D. Penny. 1989. A framework for the quantitative
study of evolutionary trees. Syst. Zool. 38: 297-309.

Hilbig, B., and J. A. Blake. 1991. Dorvilleidae (Annelida: Polychaeta)
from the U.S. Atlantic slope and rise. Description of two new genera
and 14 new species, with generic revision of Ophryotrocha. Zool. Scr.
20: 147-183.

Hurst, L. D. 1992. Intragenomic conflict as an evolutionary force. Proc.
R. Soc. Land.. Biol. Sci. 248: 135-140.

Huth, \V. 1933. Ophryotrocha-Stix&en. Zur Cytologie der Ophryotro-
chen. Z. Zellforsch. Mikrosk. Anat. Berlin 20: 309-381.

International Commission on Zoological Nomenclature 1985. Inter-
national Code of Zoological Nomenclature. 3rd ed. University of Cal-
ifornia Press, Berkeley. CA.

Josefson, A. B. 1975. Ophryotrocha longidentata sp.n. and Don-illea
emcaeformis (Malmgren) (Polychaeta, Dorvilleidae) from the west
coast of Scandinavia. Zool. Scr. 4: 49-54.

Jumars, P. A. 1974. A generic revision of the Dorvilleidae (Polychaeta).
with six new species from the deep North Pacific. Zool. J. Linn. Soc.
54: 101-135.

La Greca. M., and G. Bacci. 1962. Una nuova specie di Ophryolrocha
delle coste tirreniche. Boll. Zool. 29: 13-23.

Lu, H., and K. Fauchald. 2000. Ophryotrocha lipscombae. a new spe-
cies and a possible connection between ctenognath and labidognath-
prionognath eunicean worms (Polychaeta). Proc. Biol. Soc. Wash. 113:
486-492.

Maddison, W. P., and D. R. Maddison. 1992. MacClade: Analysis of
Phytogeny and Character Evolution. Sinauer Associates, Sunderland,
MA.

Maynard Smith, J. 1982. Evolution and the Theory of Games. Cam-
bridge University Press, Cambridge, United Kingdom.

Meek, C. F. U. 1912. A metrical analysis of chromosome complexes,
showing correlation of evolutionary development and chromatin
thread-width throughout the animal kingdom. Philos. Trans. R. Soc.
Lond. B 203: 1-74.

Ockelmann, K. W., and B. Akesson. 1990. Ophryotrocha socialis n.sp..
a link between two groups of simultaneous hermaphrodites within the
genus (Polychaeta, Dorvilleidae). Ophelia 31: 145-162.

Orensanz, J. M. 1990. The Eunicemorph polychaete annelids from
antarctic and subantarctic seas. Pp. 1-183 in Antarctic Research Series.
L. S. Kornicker. ed. American Geophysical Union. Washington. D.C.

Oug, E. 1978. New and lesser known Dorvilleidae (Annelida.
Polychaeta) from Scandinavian ami northeast American waters. Sarsia
63: 285-303.

Oug, E. 1990. Morphology, reproduction, and development of a new
species of Ophryolrocha (Polychaeta: Dorvilleidae) with strong sexual
dimorphism. Sarsia 75: 191-201.

Paavo, B., J. H. Bailey-Brock, and B. Akesson. 2000. Morphology and
life history of Ophryotrocha adherens sp. nov. (Polychaeta. Dorvillei-
dae). Sarsia 85: 251-264.

202

T. G. DAHLGREN ET AL

Palumbi, S. R. 1996. Nucleic acid II: the polymerase chain reaction. Pp.

205-247 in Molecular Systematic*. D. M. Hillis. G. Moritz, and B. K.

Mable, eds. Sinauer Associates, Sunderland, MA.
Paxton, H. 1986. Generic revision and relationships of the family

Onuphidae (Annelida: Polychaeta). Rec. Aiisr. Mus. 38: 1-74.
Pfannenstiel, H.-D. 1972. Erne neue Ophryotrocha-\n (Polychaeta. Eu-

nicidae) aus Japan. Helgol. Wiss. Meeresunters. 23: 1 17-124.
Pfannenstiel, H.-D. 1975. Ophryotrocha natans n. sp. (Polychaeta, Dor-

villeidae): ein Simultanzwitter mil acht mannlichen Segmenten aus

dem Golf von Aqaba. Zoo/. An:. 195: 1-7.
Pleijel, F., and R. Eide. 1996. The phylogeny of Ophrvotrocha (Dor-

villeidae: Eunicida: Polychaeta). J. Nat. Hist. 30: 647-659.
Premoli, M. C., and G. Sella. 1995. Sex economy in henthic

polychaetes. Ethol. Ecol. En>l. 7: 27-48.

Robotti. C. A., L. Ramella, P. Cervella, and G. Sella. 1991. Chromo-
some analysis of nine species of Ophryotrocha (Polychaeta: Dorvillei-

dae). Pp. 625-632 in Systematics, Biology and Morphology of World

Polychaeta. Ophelia Suppl. 5.
Rouse, G. W., and K. Fauchald. 1997. Cladistics and polychaetes. Zoo/.

SIT. 26: 139-204.
Sella, G. 1988. Reciprocation, reproductive success, and safeguards

against cheating in a hermaphroditic polychaete worm. Ophryotrochu

diadema Akesson. 1976. Bio/. Bull. 175: 212-217.
Sella, G. 1991. Evolution of biparental care in the hermaphroditic

polychaete worm Opltrvotmcha diudema. Evolution 45: 63-68.
Sella, G., and L. Ramella. 1999. Sexual conflict and mating systems in

the dorvilleid genus Ophryotrocha and the dinophilid genus Dinophi-

lus. Hydrohiolagia 402: 203-213.
Sella, G., C. A. Redi, L. Ramella, R. Soldi, and M. C. Premoli. 1993.

Genome size and karyotype length in some interstitial polychaete

species of the genus Ophryotrocha (Dorvilleidae). Genome 36: 652-
657.

Sella, G., R. Vilturi, L. Ramella, and M. S. Colomba. 1995. Chromo-
somal nucleolar organizer region (NOR) phenotypes in nine species of
the genus Ophryotrocha (Polychaeta: Dorvilleidae). Mar. Bio/. 124:
425-433.

Shaojie, D., and J. F. Knowles. 1992. Chromosomes of the polychaete
Ophn-otrticha diadema. Ophelia 36: 195-201.

Soldi, R., L. Ramella, M. C. Gambi, P. Sordino, and G. Sella. 1994.
Genome size in polychaetes: relationship with body length and life
habit. Mem. Mus. Nut. Paris 162: 129-135.

Sundberg, P., and S. Andersson. 1995. Random amplified polymorphic
DNA (RAPD) and intraspeciric variation in Oerstedia dorsalis (Hop-
lonemertea, Nemertea). J. Mar Bio/. Assoc. U.K. 75: 483-490.

Swofford, D. L. 2000. PAUP* (Pln'logenetic Analysis Using Par-
simony}. Sinauer Associates. Sunderland, MA.

Swofford, D. L., G. J. Olsen, P. J. Waddell, and D. M. Hillis. 1996.
Phylogenetic inference. Pp. 407-514 in Molecular Systematics, D. M.
Hillis. C. Moritz. and B. K. Mable, eds. Sinauer Associates. Sunder-
land, MA.

Thompson, J. D., D. G. Higgins, and T. J. Gibson. 1994. Clustal W:
improving the sensitivity of progressive multiple sequence alignment
through sequence weighting, position-specific gap penalties and weight
matrix choice. Nucleic Acids Res. 22: 4673-4680.

Vitturi, R., L. Ramella, M. S. Colomba, V. Caputo, and G. Sella. 2000.
NOR regions of polychaete worms of the genus Ophryotrochu studied
by chromosome banding techniques and FISH. J. Heredity 91: 18-23.

Zavarzina, E. G., and A. B. Tzetlin. 1991. Breeding and larval morphol-
ogy of Ophi-yotrncha dimorphica Zavarzina & Tzetlin (Polychaeta: Dor-
villeidae). Ophelia 5: 41 1-420.

Appendix

Checklist of described Ophryotrocha species with original localities

O. adherens Paavo, Bailey-Brock & Akesson, 2000. Cyprus
and Hawai, littoral.

O. akessoni Blake, 1985. Galapagos Rift, East Pacific Ba-
sin, deep sea.

O. iitluntii-ti Hilhig & Blake, 1991. NW Atlantic, slope
depths.

O. haccii Parenti, 1961. Roscoff, France, littoral.

O. hi ficlci Hilbig & Blake. 1991. NW Atlantic, slope depths.

O. claparedii Studer, 1878. Kerguelen. littoral.

O. costlowi Akesson, 1978. Beaufort, North Carolina, lit-
toral.

O. cosinetandra Oug. 1990. Northern Norway, littoral.

O. diadema Akesson, 1976. Los Angeles harbor, littoral.

O. dimorphica Zavarzina & Tzetlin, 1986. Peter the Great
Bay, littoral.

O. tlubiti Harmann-Schroder. 1974. North Sea (off Scot-
land). 68 m.

O. gerluchi Hartmann-Schroder, 1974. North Sea (off Den-
mark). 52 m.

O. geryonicola (Esmark, 1874). Skagerack, Kattegat, sub-
littoral.

O. globopalpata Blake & Hilbig, 1990. Juan de Fuca Ridge,
deep sea.

O. gracilis Huth, 1933. Helgoland, Germany, littoral.

O. hadalis Jumars, 1974. Aleutian Trench, deep sea.

O. hunnuiniu Huth, 1933. NE Atlantic, littoral.

O. irinae Tzetlin, 1980. Kandalaksha Bay. White Sea, lit-
toral.

O. kagoshimaensis Miura, 1997. Kagoshima Bay. Japan,
197 m.

O. labulion Hilbig & Blake. 1991. NW Atlantic, slope
depths.

O. labronica La Greca & Bacci, 1962. Naples. Italy, littoral.

O. lipxcombtii' Lu & Fauchald. 2000. NW Atlantic, slope
depths.

<). liitomlis (Levinsen. 1879). Egesminde. Greenland, lit-
toral.

O. lobifera Oug, 1978. West Norway, in mud, 50 m.

O. hm^identata Josefson. 1975. Skagerack, Kattegat. 50-
100 m.

O. miciolckiu' Hilbig & Blake. 1991. NW Atlantic, slope
depths.

OPHKYOI'ROCHA PHYLOGENY

203

O. nmciduia Akesson, 1973. Skagerack, Kattegat. 2? in.

(). nitinilihiilntii Hilhig & Blake. 1991. NW Atlantic, slope
depths.

O. nu'iliti'iTiini'ii Martin, Abello & Cartes. 1991. Mediter-
ranean, parasitic, 600-1800 m.

O. minimi Levi. 1954. Roscoff. France, littoral.

O. mitiinx Ptannenstiel, 1975. Red Sea. littoral.

O. notitilix (Ehlers. 1908). Southern South America, sublit-
toral.

O. notoxltimlnliiui Ptannenstiel, 1972. Japan, littoral.

O. ohtuxti Hilbig & Blake. 1991. NW Atlantic, slope
depths.

O. iMchyxonui Hilbig & Blake, 1991. NW Atlantic, slope
depths.

O. /uiralhulion Hilbig & Blake, 1991. NW Atlantic, slope
depths.

O. platykeplutk' Blake. 1985. Guayamas Basin, hydrother-

mal vents, deep sea.
O. /iin'rilix piierilis Claparede & Mecznikow. 1869. Naples.

Italy, littoral.
O. piierilis siberti (Mclntosh, 1885). Plymouth. England.

littoral.

O. scarlatoi Averincev. 1989. Franz Josefs Land, littoral.
O. schiibraryi Tzetlin, 1980. Marine aquarium in Moscow,

Russia.
O. socialis Ockelmann & Akesson, 1990. Marine aquarium

in Helsingo'r, Denmark.

O. spatula Fournier & Conlan, 1994. Arctic Canada, littoral.
O. vivipara Banse, 1963. San Juan Archipelago, USA.

22 m.
O. wubaolingi Miura. 1997. Kagoshima Bay, Japan.

200 m.

Reference: Bio/. Bull. 201: 204-217. (October 2001)

Variable Mate-Guarding Time and Sperm Allocation

by Male Snow Crabs (Chionoecetes opilio) in Response

to Sexual Competition, and their Impact on the

Mating Success of Females

AMELIE RONDEAU1'* AND BERNARD SAINTE-MARIE2 1

Institut des sciences de la iner de Rimouski (ISMER), Universite du Quebec a Rimouski, 310 allee des

Ursulines, Rimouski, Quebec G5L 3A1. Canada; and 'Division des invertebres et de la biologie

experimentale, Institut Manrice-Lamontagne, Peches et Oceans Canada, 850 route de la Mer,

C.P. 1000, Mont-Joli, Quebec G5H 3Z4. Canada

Abstract. Two laboratory experiments investigated mate
guarding and sperm allocation patterns of adult males with
virgin females of the snow crab, Chionoecetes opilio, in
relation to sex ratio. Although females outnumbered males
in treatments, operational sex ratios were male-biased be-
cause females mature asynchronously and have a limited
period of sexual attractiveness after their maturity molt.
Males guarded females significantly longer as the sex ratio
increased: the mean time per female was 2.9 d in a 23:209
treatment compared to 5.6 d in a 66 :209 treatment. Female
injury and mortality scaled positively to sex ratio. Males
that guarded for the greatest number of days were signifi-
cantly larger, and at experiment's end had significantly
smaller vasa deferentia, suggesting greater sperm expense,
than males that guarded for fewer days. In both experi-
ments, the spermathecal load (SL) — that is, the quantity of
ejaculate stored in a female's spermatheca — was indepen-
dent of molt date, except in the most female-biased treat-
ment, where it was negatively related. The SL increased as
the sex ratio increased, mainly because females accumu-
lated more ejaculates. However, similarly sized males had

Received 28 September 2000; accepted 30 May 2001.

* Present address: Division des poissons marins. Centre des peches du
Golfe. Peches et Oceans Canada, C.P. 5030 Moncton. Nouveau Brunswick
E1C 4B6, Canada.

t To whom correspondence should be addressed. E-mail: Sainte-
MarieB @dfo-mpo.gc.ca

Abbreviations: CW. carapace width; GT, guarding time per female;
ESR. effective sex ratio; OSR, operational sex ratio; SL. spermathecal
load: VDW. vasa deferentia weijiht.

smaller vasa deferentia and passed smaller ejaculates, such
that, at a given sex ratio, the mean SL was 55% less in one
experiment than in the other. Some females extruded
clutches with few or no fertilized eggs, and their median SL
(3-4 mg) was one order of magnitude smaller than that of
females with well-fertilized clutches (31-50 mg). indicating
sperm limitation. Males economized sperm: all females
irrespective of sex ratio were inseminated, but to a varying
extent submaximally; each ejaculate represented less than
2.5% of male sperm reserves; and no male was fully ex-
hausted of sperm. Sperm economy is predicted by sperm
competition theory for species like snow crab in which
polyandry exists, mechanisms of last-male sperm prece-
dence are effective, and the probability that one male fer-
tilizes a female's lifetime production of eggs is small.

Introduction

Intrasexual competition for mates is a fundamental char-
acteristic of sexual reproduction (Trivers, 1972). The inten-
sity of sexual competition depends mostly on the opera-
tional sex ratio (OSR), which is the number of sexually
active males relative to the number of fertilizable females at
a given site and time (Emlen and Oring, 1977). In many
animal species, females care for progeny and are only
briefly and infrequently receptive, giving them a much
smaller potential rate of reproduction than males. This
causes the OSR to be skewed toward males, a tendency that
may be exacerbated if females become receptive asynchro-
nously. As a result, sexual competition is often more intense
among males than among females; to enhance their repro-

204

MATE GUARDING TIME AND SPERM ALLOCATION BY MALE SNOW CRABS

205

ductive success, males may express flexible behaviors, in-
cluding mate guarding and judicious allocation of sperm
drivers, 1972; Ridley, 1983; Clutton-Brock and Parker,
1992).

Precopulatory mate guarding is taxonomically widespread,
albeit particularly common in the Crustacea, and it may serve
to monopolize a female until she is fertilizable (Parker, 1974;
Ridley, 1983). Postcopulatory mate guarding occurs mostly in
species with direct sperm transfer, and it may help to ensure
paternity for the guarding male by preventing rival males from
inseminating the female until she has fertilized her eggs or is
no longer receptive (Parker, 1970; Smith, 1984). In crustacean
species in which female molting and mating are intimately
linked, postcopulatory mate guarding may also shield the post-
molt female (and the male's reproductive investment) from
predators until her shell has hardened enough to offer protec-
tion (Hartnoll. 1969; Wilber, 1989; Jivoff, 1997a). Males may
vary their mate-guarding pattern in relation to competition and
maximize the number of eggs gained during a breeding season
by balancing the time spent guarding mates against the time
spent searching for new mates (Parker. 1974: Christy, 1987).
Theory predicts (Grafen and Ridley, 1983: Yamamura and
Jormalainen, 1996) and observations typically confirm (see
Jormalainen, 1998) that males respond to increasing sex ratio
by guarding females longer.

Judicious sperm-allocation patterns have evolved in part
because sperm, spermatophores, and seminal fluid can be in
limited supply due to low rates or high costs of production
(Dewsbury, 1982; Pitnick and Markow, 1994). Further,
males may enhance their reproductive success if they adjust
sperm expenditure to the perceived risk of sperm competi-
tion, which may vary as a function of sex ratio, potential for
polyandry, or female mating history (Parker et ai, 1997).
Males typically increase sperm expenditure in the presence
of larger females and scale the amount of sperm allocated to
females positively to the sex ratio and the risk of sperm
competition (Gage. 1991; Gage and Barnard, 1996; Wedell
and Cook. 1999).

Changes in male competition intensity and male mating
patterns may influence female mating success. As compe-
tition becomes more intense, the risk of female injury or
death may increase due to male harassment and more fre-
quent takeover attempts (e.g., Rowe et ai, 1994; Vep-
salainen and Savolainen. 1995). Conversely, when compe-
tition is relaxed and postcopulatory guarding is curtailed,
postmolt females are more exposed to predators (Wilber,
1989; Jivoff. 1997b). A severe reduction in sperm allocation
may lead to sperm limitation and loss of fecundity for
females (Pitnick. 1993; MacDiarmid and Butler, 1999).

Although considerable evidence of flexible patterns of
sperm allocation exists for terrestrial and freshwater animals
with direct sperm transfer, very little is known of this
phenomenon in their marine counterparts (Wilber, 1989;
Jivoff. 1997b; Sainte-Marie et al, 1997; MacDiarmid and

Butler. 1999). The present study on the snow crab (Chio-
noecetes opilio; Majidae), a marine brachyuran of the north-
ern hemisphere, documents mate guarding and sperm allo-
cation in relation to sex ratio for adult males with virgin
females. Male mating strategies predictably are quite flex-
ible in snow crab because the intensity of sexual competi-
tion may be highly variable among years as a result of
intrinsic, circa-decadal oscillations of 1-2 orders of magni-
tude in the abundance ratio of adult males to virgin females.
Such oscillations arise from the interaction of multiyear
variations in year-class strength and of sexual dimorphism
in age at maturity, leading to temporally staggered recruit-
ment patterns for adult females and males (Sainte-Marie et
al.. 1996).

The relationships of mate-guarding time and sperm allo-
cation to sexual competition remain undetermined for snow
crab and congeners; however, other aspects of the sexual
interactions of males with virgin females are very well
documented in the genus Chionoecetes. Female snow crabs
reach sexual maturity at a terminal molt, which occurs from
January to April in the northwest Atlantic (see Alunno-
Bruscia and Sainte-Marie, 1998). Males are attracted to
pre-mature females by chemical cues (Bouchard et al.,
1996; Pelletier et al., 1998) and then engage in courtship
and precopulatory mate guarding until the female molts
(Watson, 1972; Donaldson and Adams, 1989). Females
usually extrude a clutch of eggs within 1-5 d of molting,
whether mated or not (Paul and Adams, 1984; Sainte-Marie
and Lovrich, 1994). Both fertilized and unfertilized eggs
attach to the pleopods: those fertilized are incubated for up
to 2 years; those not fertilized are lost within 5-6 months of
attachment (Sainte-Marie, 1993; Sainte-Marie and Carriere,
1995; Moriyasu and Lanteigne. 1998). Adult males have a
very high potential reproductive rate and can mate effec-
tively with several females in rapid succession (Watson,
1972; Adams and Paul, 1983; Sainte-Marie and Lovrich,
1994). Female asynchronous molting and brief postmolt
sexual attractiveness lead to male competition, and adult
males that are smaller, have a softer shell, or are missing
more pereopods may be displaced from females by more
vigorous males (Stevens et al.. 1993; Sainte-Marie et al.,
1999). Intense male competition also favors polyandry, and
female snow crabs during their first breeding period may
accumulate in their spermathecae the ejaculates of up to six
males (Urbani et al., 1998). Multiple (different males) and
repeated (same male) copulations can happen before or
shortly after the first egg clutch is extruded (Sainte-Marie et
al., 1997, 1999). When multiple mating takes place before
oviposition. last-male sperm precedence usually occurs
through a combination of sperm displacement and post-
copulatory mate guarding (Urbani et al., 1998; Sainte-Marie
et al., 2000). The amount of ejaculate stored by females is
independent of mate body size (Adams and Paul, 1983;
Sainte-Marie and Lovrich, 1994) but is positively related to

206

A. RONDEAU AND B. SAINTE-MARIE

number and duration of copulations, which are hypothe-
sized to he influenced by competition intensity (Sainte-
Marie et ai, 1997).

The present laboratory study of mating behavior in snow
crab was guided by three hypotheses: (i) larger males guard
more than smaller males, (ii) mate guarding lasts longer at
higher than at lower sex ratios, and (iii) females store more
sperm as the sex ratio increases. We also measured the
effects that changes in male competition intensity and re-
productive investment have on the mating success of fe-
males. Two mating experiments were conducted to deter-
mine if sperm allocation patterns in relation to competition
were the same whether sex ratio was manipulated by vary-
ing the density of females or of males. Experiments used
female-biased treatments to explore the potential for sperm
limitation, a major concern where snow crab fisheries re-
move only large adult males (Kruse, 1993; Elner and Be-
ninger, 1995).

Materials and Methods

Collection of crabs

Crabs were collected in October of 1996 and 1997 in the
Saint Lawrence Estuary (48°33'N. 68°23'W), eastern Can-
ada. The carapace width (CW) of all crabs and the right
chela height of males were measured to the nearest 0. 1 mm,
using a vernier caliper as described in Sainte-Marie and
Hazel (1992). Only immature females larger than 40 mm
CW were kept, because they are more likely to reach
maturity at their next molt. Males retained were adults of SO
to 1 15 mm CW. at the mid-range of the 40-155 mm CW
distribution for adult males, with shells of intermediate
condition (i.e., hard exoskeleton with light epibiont foul-
ing). This shell condition prevails from about 8 months to 3
years after the male's terminal molt to adulthood (Sainte-
Marie ft ai., 1995) and coincides with peak sexual compet-
itiveness (Sainte-Marie et til.. 1999).

Selected females and males were brought to the Maurice
Lamontagne Institute and placed in separate tanks supplied
with fresh running seawater. Photoperiod was controlled to
reflect the natural light cycle. Crabs were fed excess thawed
shrimp (Pundtiliix horetilis) and Atlantic herring (Chipen
hareiif>iis) on a semi-weekly basis from the time of capture
to the end of the experiments.

Mating experiments

Mating experiments were conducted in 1997 and in 1998.
Crabs were used only in the winter after collection. Thus the
time elapsed between collection and use in experiments was
similar in both years.

The 1997 experiment ran from 31 January to 4 April (64
d) in nine tanks, each with a bottom surface area of 1.14 nr

(390 1). Tanks received fresh running seawater with ranges
of temperature (-0.5 to 1.5°C) and salinity (24.6%o to
30.2%p) over the duration of the experiment that represented
natural conditions for snow crab. Sex ratio was controlled
by varying the number of females for a constant number of
males: treatments had male-to-female ratios of 2:10 (n = 3
replicates). 2:20 (n = 3). and 2:30 (n = 3). Female and male
crabs were allotted to tanks so that their respective size
distributions were as similar as possible among all repli-
cates. Immature females were missing no pereopods at the
time they were placed in tanks. Some males were missing
1-2 walking legs, but all had two chelae. Excess crabs were
held in reserve tanks. Tanks were checked twice daily for
the presence of molting females, and any exuvium was
removed and measured to determine premolt CW. Several
days after a female had molted to maturity and mating
behavior had ceased, the female was identified with a num-
bered plastic tag tied around the coxopodite of a pereopod.
During the experiment, dead females were replaced by a
female of similar CW and same reproductive stage taken
from the reserves. Pre-mature females are called "pubes-
cent." females that have molted to maturity but not yet
extruded eggs are called "nulliparous." and females that
have layed their first clutch of eggs are called "primipa-
rous." Substitute females were also tagged. At the experi-
ment's end, the CW of priiniparous females was measured.

The 1998 experiment ran from 25 January to 30 March
(65 d) in 12 tanks, each with a bottom surface area of 2.23
nr (740 1). Tanks were supplied with fresh running seawater
ranging from -0.3 to 1.8°C and from 24.4%«? to 31.3%o
salinity over the duration of the experiment. Sex ratio was
controlled by varying the number of males for a constant
number of females; male-to-female ratios in treatments
were 2:20 (n = 4 replicates). 4:20 (n = 3). 6:20 (n = 4), and
10:20 (n = 1 ). The 2cJ:209 treatment was common to the
1997 and 1998 experiments. Methods were identical to
1997, except for the following. All males were intact and
each was identified with a water-resistant, numbered label
fixed to the dorsum. All females in one replicate each of the
2cJ:20$ and 6c?:209 treatments were similarly identified
with a label that bore a letter. Labels were large enough to
be read from a distance but did not impede molting or
mating. Each day we determined the number of males that
were guarding females and recorded specific mating asso-
ciations in the two tanks where all crabs were labeled.
Guarding males were those grasping a female or copulating.

The high densities of crabs in our treatments, reaching up
to 28 crabs m~2, are not unrealistic. Majid crabs are noto-
rious for their gregarious behavior during the mating season
(e.i>.. DeGoursey and Auster, 1992), and densities of 100
crabs m"2 have been documented for Chionoecetes bciirdi
(Stevens et til.. 1994).

MATE GUARDING TIME AND SPERM ALLOCATION BY MALE SNOW CRABS

207

Male reproductive effort

The weight of vasa det'erentia ( VDW). which include the
storage areas tor spermatophores and seminal fluid, was
determined at the end of experiments to evaluate sperm
depletion as a potential indicator of male reproductive ef-
fort. Males in the 1997 and 1998 experiments and 16 reserve
males (unmated = controls) of 1998 were killed and were
injected with, and immersed in. 4c/c seawater-diluted forma-
lin. Males were subsequently dissected and their vasa det'-
erentia were removed, blotted, and weighed to the nearest
milligram.

In 1998 we estimated guarding time for individual fe-
males (GT, in days) in each replicate as

GT = S/f6",

where 116 ,- is the number of males that were guarding a
female on the ;'th day of the experiment and /V9,,, is the
number of females having reached maturity at experiment's
end. GT includes both precopulatory and postcopulatory
mate guarding, which could not be dissociated under the
present experimental conditions. In the two replicates where
all crabs were labeled, we determined for each female the
time elapsed between the occurrence of first grasping and
the maturity molt, and the total number of days she was
guarded.

Female mating success

Female mating success was assessed for primiparous
females (excluding primiparous substitutes) at the end of
experiments by measuring percent fertilized eggs per clutch,
clutch weight, and spermathecal load (SL). Injury and death
are also components of female mating success, so the num-
ber of missing pereopods and the percent mortality of nul-
liparous and primiparous females were compiled for each
replicate.

The percentage of fertilized eggs in a clutch was esti-
mated from a sample of eggs taken from random locations
throughout the clutch while the female was alive. Following
Carriere ( 1995), eggs were processed to highlight nuclei for
determination of the proportion of divided (= fertilized)
eggs per sample. Briefly, eggs were fixed for 1 h in a
solution of 97% glucamine-acetate (GA) buffer, 2% forma-
lin, and 1 % Triton, and then rinsed in GA buffer. Eggs were
then stained for 1 h in a solution of 0.5 jxg Hoechst dye per
ml of GA buffer, and were rinsed twice and preserved in GA
buffer at 4 °C. A sample typically contained 200-2400
eggs, and divided and undivided eggs were counted under
epifluorescent microscopy. Because eggs develop slowly at
cold temperatures, their fertilization status cannot be accu-
rately determined before they are 20 d old (Rondeau, 2000),
so we sampled only primiparous females older than 20 d

postmolt. reasonably assuming no delay between molt and
oviposition.

The female was killed after her eggs were sampled. The
remaining clutch was removed, by severing the base of the
inner ramus of each pleopod, and preserved in 99% ethanol.
The weight of the blotted clutch was measured to the nearest
milligram and then was adjusted using correction factors in
Rondeau (2000) to account for enclosed pleopod rami and
for removed eggs. The right spermatheca was extracted
from the female and preserved in V.'i seawater-diluted for-
malin. Subsequently, SL was determined by peeling away
the wall of the spermatheca and weighing its blotted content
to the nearest 0. 1 mg. Total ejaculate stored by a female can
be estimated by doubling the SL because delivery of sperm
is usually balanced between the two spermathecae (Sainte-
Marie and Lovrich. 1994). Females were considered for
analyses of SL only if they survived for 3 days after their
maturity molt, to ensure they had the opportunity to fully
realize their mating potential. After determination of SL, we
estimated the number of sperm stored in some spermathecae
from the 1998 experiment. To provide an even distribution
over the range of SLs for each treatment. 10 spermathecae
from each of the 2<J:209 and 10c?:209 treatments were
selected before sperm were counted. Our method for count-
ing sperm was to homogenize the spermathecal contents,
dilute the homogenate in seawater to a known volume.
enumerate the sperm in replicate hemacytometers, and then
extrapolate the average sperm count to the total volume
(Adams and Paul, 1983; Sainte-Marie and Lovrich, 1994).

Data analysis

For each replicate, the effective sex ratio (ESR) was
calculated as

ESR =

,„ + AM).

where AM is the number of males and /V9,,, is the number
of mature females available to males during the experiment.
Following Emlen and Oring (1977), the operational sex
ratio (OSR) was calculated as

OSR = S[AM/(«,9, + N6)]/D.

where AM is the number of males. ;i,9, is the number of
fertilizable females on the /'„, day. and D is the duration of
the experiment in days. To calculate OSR. we somewhat
arbitrarily used a fertilizable period of 3 d starting at the
maturity molt. Our choice is justified by the fact that when
males are present, female snow crabs usually are insemi-
nated and extrude eggs within 6-24 h of molting (Watson,
1972; Sainte-Marie and Lovrich, 1994) but may continue to
mate for about 2-3 d after oviposition. However, although
males may compete intensively for nulliparous (pre-ovipo-
sition) females, there is little or no male competition for
females after oviposition (Sainte-Marie et ai, 1997. 1999),

208

A. RONDEAU AND B. SAINTE-MARIE
Table 1

Mean ± standard deviation of effective se.\ ratio (ESR) anil of operational sex ratio lOSR, calculated assuming a 3-d fertilizable period for females)
for treatments in the 1997 and 1998 mating experiments with snow crab

1997

199X

Treatment

ESR

OSR

Treatment

ESR

OSR

25:302

0.06

± 0.00

0.69 ± 0.03

2(5:209

0. 1 1 ± 0.03

0.81 ± 0.04

2d:209

0.09

± 0.01

0.78 ± 0.03

4(5:209

0.18 ±0.01

0.89 ± 0.00

2(5:109

0.1X

± 0.03

0.87 ± 0.00

6(5:209

0.29 ± 0.06

0.93 ± 0.02

10(5:209

0.39

0.94

and thus some uncertainty exists as to how the fertilizable
period should be defined for calculation of OSR. Given this
uncertainty, we preferred to use ESR as a basis for com-
parison of male reproductive effort and female mating suc-
cess across treatments. Both ESR and OSR can take on
values ranging from 0 (no male) to 1 (no female), with 0.5
representing a balanced sex ratio.

For univariate analyses, we used the mean, standard
deviation, and t test or analysis of variance (ANOVA) for
description and sample comparisons of normally distributed
and homoscedastic data (raw or transformed). Otherwise,
the median and Mann-Whitney test were used. One-tailed
tests were performed when the mean or median was ex-
pected a priori to be greater in one sample than in the other.
Correlation analysis examined trends between pairs of vari-
ables such as female size, molt date, guarding time, and SL.
Functional relationships between two variables were inves-
tigated by regression analysis. When relating VDW or SL to
ESR, we used the replicate's mean or median rather than
individual values of the dependent variable so that each
replicate weighted the regression equally. Analysis of co-
variance (ANCOVA) was used to compare clutch weight
among treatments with CW as the covariate, and VDW and
SL between experiments with ESR as the covariate. The
assumption of homogeneity of slopes was met if there was
no significant interaction between factor and covariate
(Sokal and Rohlf. 1995).

In primiparous female snow crabs, SL measured shortly
after the mating period is the sum of individual ejaculates
received from single or multiple mates, less the amount of
ejaculate expended at fertilization (Sainte-Marie el ai,
1997). Frequency distributions of log,0 SL for replicated
treatments were graphed and analyzed using the mixture
distribution method of MacDonald and Pitcher ( 1979) in an
attempt to resolve modes representing one or more accu-
mulated ejaculates.

Results

Size of crabs, molting schedule, and operational sex ratio

Immature females ranged from 45.5 to 66.2 mm CW in
the 1997 experiment and from 44.2 to 60.5 mm CW in the

1998 experiment; males were 80.4-1 1 1.7 mm CW in 1997
and 81.0-113.0 mm CW in 1998. The mean CW of imma-
ture females was homogenous among replicates and treat-
ments in each experiment (two-way ANOVA. F8 ,63 =
0.18. P = 0.993 in 1997 and F1U56 = 1.06, P = 0.400 in
1998), but differed (ANOVA. F,'33g = 20.69, P < 0.001)
between 1997 (53.8 mm) and 1998 (51.7 mm). The mean
CW of males was homogenous among replicates and treat-
ments in 1997 (F6-9 = 0.68, P = 0.671) and 1998 (F841 =
0.32. P = 0.953), and was similar (F, 69 = 2.74, P = o!l03)
between 1997 (94.4 mm CW) and 1998 (98.4 mm CW).

The proportion of immature females that molted was 95.6%
in the 1997 experiment and 70.0% in the 1998 experiment, and
all but two moltees achieved maturity. A negative correlation
between premolt CW and molt date in 1997 (/• = -0.15, n =
186, P = 0.037) and 1998 (r = -0.26. n = 168, P < 0.001)
indicated that larger females tended to molt earlier. The first
female molt occurred on day 2 and day 5 of the 1997 and 1998
experiments, respectively, and molting continued until the end
of each experiment. The cumulative number of molts followed
a logistic pattern over time, and about 75% of molts in each
treatment occurred over a period of about 25 d. After fitting
and comparing logistic regressions. Rondeau (2000) found that
50% of total molts in 1997 occurred on day 20.8 ± 1.0 d in the
least female-biased treatment (2(5:10?), significantly sooner
than in treatments with 2<5 :209 (day 26.2 ± 0.7) and 26 :309
(day 26.8 ± 0.6). In 1998, females also molted sooner in the
least female-biased treatment ( 10(5:209, day 33.9 ± 1.9)than
inthe6(3:209,4cJ:209,and2<J:209 treatments (day 43.7 ±
0.7 to 44.3 ± 0.8). ESR and OSR values paralleled male-to-
female treatment ratios, but OSR was always biased toward
males (Table 1). If the fertilizable period is taken to be 1 d
instead of 3 d — to reflect only the usual time between female
molt and oviposition when males compete to inseminate a
female — then depending on treatment, OSR ranged from 0.87
to 0.96 in 1997 and from 0.93 to 0.98 in 1998.

Female pereopod loss and mortality

Most injury or death of females occurred while they were
in the soft postmolt condition. In both experiments, the
proportion of primiparous females that was missing 0, 1-2,

MATE GUARDING TIME AND SPERM ALLOCATION BY MALE SNOW CRABS

Table 2

209

Percentage of primiparous /finale vimr erah.\ missing 0. 1-2. or .1 or more pereopods h\ replicated sex-ratio treatincnl in the 1997 and 1998 matin,
experiments (n = number of primiparous females)

Treatments

1997

1998

Missing pereopods 2(5:309

2iJ:209

25:109

25:209

4d:209

6^:209

0 56.8

33.3

53.6

50.8

38.0

9.6

\-2 40.9

41.0

39.3

36.9

38.0

46.2

a3 2.3

25.7

7.1

12.3

24.0

44.2

n SS

Females were intact at start of experiment. The number of females in each class of missing pereopods was not independent of treatment (G-test of
independence: P < 0.05 for both years). Females from one 2c5:209 replicate of 1997 were excluded because of incomplete information on number of
missing pereopods.

or >3 pereopods differed significantly among treatments
(Table 2). In 1998. the number of missing pereopods in-
creased relative to the number of males and the sex ratio. In
1997, however, the number of males was held constant and
there was no clear pattern between the number of missing
pereopods and the sex ratio or female density. Mean mor-
tality of combined nulliparous and primiparous females
increased with increasing sex ratio. In the 1997 experiment,
mortality reached 6%, 18%, and 20% in the 2c?:309, 2d:
20$, and 2c?:109 treatments, respectively. One replicate in
the 2c?:209 treatment was excluded from computation of
mortality because combined nulliparous and primiparous
mortality was high (50%). Dead females in this peculiar
replicate were shredded, suggesting that one or both males
were particularly aggressive. Mortality in the 1998 experi-
ment was 15%, 20%. and 35% in the 2c3:209. 4<J:209,
and 6cJ:209 treatments, respectively.

Male dominance and mate-guarding patterns

For the 1998 experiment, males were separated a poste-
riori into two groups based on the total number of guarding
days accumulated by each male over the course of the
experiment. In each replicate, the male that guarded for the
greatest number of days was categorized as "dominant" and
the other male (or males) was considered to be "subordi-
nate." Dominant males (mean CW == 102.4 mm, n = 12)
were as expected larger (one-tailed t test, / == 1.74. P -
0.049) than subordinate males (mean CW = 97.3 mm, ;/ =
42). The average number of days that dominant males
guarded in the 1998 experiment did not decline with in-
creasing number of males per treatment (Table 3). More-
over, the decrease in the contribution of dominant males to
total number of guarding days was not proportional to the
increase in number of males per treatment (Table 3).

Table 3

Cumulative miniher of xuardins lUm h\ male,',, and mean ± Mandard deviation of \;itiinli>if> lime fur individual females lOTl. in relation to sex ratio
in the 1998 mating experiment with anmv crah

Contribution of dominant male to sum of

Number of guarding days by male type

guarding days by all males

Treatment

(a) Dominant sum

Dominant

mean

(b) All males sum

Observed

Expected

*P_

Mean

2d:209

22.0 ±

6.1

166

0.53

0.50

0.60.

>0.05

2.9 ±

1.0

4<5:209

17.3 ±

5.9

124

0.42

0.25

16.87.

<0.001

3.1 ±

0.6

6(5:209

23.2 ±

8.0

251

0.37

0.17

59.83.

<0.001

5.6 ±

1.9

105:209

0.21

0.10

8.93.

<0.01

6.5

The sum of guarding days for the dominant male (a) and for all males (b) in all replicates (n) of each treatment are shown. The dominant male, for which
mean ± standard deviation of guarding days by treatment are given, represents the male that guarded for the greatest number of days in each replicate.
Considering only replicated treatments, there was no effect of sex ratio on mean number of guarding days by the dominant male (ANOVA. F2 5 = 0.69,
P = 0.5281 and a significant effect of sex ratio on mean GT (ANOVA. F2X = 6.97, P = 0.018). Observed contribution of dominant males to sum of
guarding days for all males is the quotient of (a) over (b); expected contribution of dominant males is the quotient of one over the number of males per
treatment. The G-test with William's correction (G.,dJ) verified whether observed contribution departed significantly (P = probability) from expected
contribution.

210

A. RONDEAU AND B, SAINTE-MARIE

0.0

0.2

Effective sex ratio

0.3

0.4

Figure 1. Mean vasa deferentia weight of male snow crabs at end of
1997 (•) and 1998 (A) mating experiments in relation to effective sex ratio
by replicate. Encircled values represent replicates of the 2<3:209 treatment
common to both experiments. Note the outlier (effective sex ratio = 0.08.
vasa deferentia weight = 5.06 g) in the 1997 experiment.

As hypothesized, GT increased as the sex ratio (and
number of males) increased in the 1998 experiment (Table
3). This increase was not simply a density-dependent effect
of more males haphazardly guarding any female. Indeed,
based on the two replicates with all crabs labeled, females
were grasped for the first time sooner (one-tailed Mann-
Whitney test, U = 8, P = 0.002) before they molted in the
66:209 treatment (mean 9.4 d, maximum 33 d) than in the
2<3:209 treatment (mean 2.8 d. maximum 17 d). Also,
dominant males guarded for longer continuous periods prior
to the female's molt in the 6d:202 treatment compared to
the 2d:209 treatment (one-tailed Mann-Whitney test. U :
3, P = 0.007). suggesting they mated fewer females as sex
ratio increased. There was no correlation between the num-
ber of days a female was guarded and the date on which she
molted (2(5:20?, ;• = 0.00. n= 18, P = 1.000: 6<?:209,
r == 0.33, n =-- 10, P = 0.353); therefore, females were
guarded as long at the start as at the end of the experiment.

Sperm depletion

Sperm depletion due to mating was suggested by scatter-
plots showing a weak positive trend between VDW and
ESR (Fig. 1 ), indicating that residual vasa deferentia weight
tended to decline with the number of potential mating
opportunities for males. ANCOVA on log|,,-transformed
data showed that the effect of ESR on VDW was significant
(F,J7 = 5.33. P = 0.034) and that year and ESR did not
interact (F, ,7 = 0.12. P = 0.738). The mean VDW ad-
justed to the overall mean ESR differed between years
(F, , = 5.49, P = 0.031 ) and was 35.5% less in 1998 males

(2.45 g) than in 1997 males (3.80 g). Note the outlier in the
1997 experiment (Fig. 1), corresponding to the 2cJ:209
replicate with unusually high female mortality.

Compelling evidence of sperm depletion was seen in the
contrasting patterns of VDW for dominant and subordinate
males in 1998 (Fig. 2). Two-way ANOVA excluding the
sole 10cJ:209 replicate indicated that the mean log,,, VDW
varied with the log,,, ESR (F237 = 5.78. P = 0.007) and
male hierarchy (F, 37 = 7.00, P = 0.012), but there was no
interaction between the two factors (F2 37 == 2.17, P -
0.129). The mean VDW was progressively smaller as sex
ratio decreased and was less in dominant than in subordinate
and control males (Fig. 2), the difference between control
and dominant males increasing from 2% to 66% as the sex
ratio declined.

Spermathecal load

In every treatment, correlation coefficients were positive
between SL and CW and negative between SL and molt
date. However, the only significant coefficient was between
SL and molt date (;• = -0.31, n = 83. P = 0.005) in the
most female-biased treatment (2<5:309).

As hypothesized, there was a significant positive relation-
ship between SL and ESR in both years (Fig. 3), indicating
that females acquired more ejaculate as the sex ratio in-
creased. ANCOVA on log,0-transformed data showed a
highly significant effect of ESR on SL (F, ,7 = 17.80, P =
0.001) but no interaction between year and ESR (F, ,7 =
0.50, P = 0.488). The mean SL adjusted to the overall mean
ESR differed between years (F, ,8 = 36.41, P < 0.001 ) and
was 55.1% less in 1998 (27.8 mg) than in 1997 (61.9 mg).

0.7 1

.SP 05 -

0.4 1

•a o.:

: 20$

43 : 20$

63 : 20$

Control

Treatment

Figure 2. Mean ± standard de\ iation of log-transformed vasa deter-
entia weight of male snow crabs in relation to male mating status (domi-
nant = m subordinate = *) in the 2<3:209. 46:209 and 6c$:209
treatments of the luc)N mating experiment and in comparison to unmated
control males

MATE GUARDING TIME AND SPERM ALLOCATION BY MALE SNOW CRABS

211

90 1

•5 6°-
I

1« 50 -

a 40

.2 30
•o

* 20

0.0

0.1 0.2

Effective sex ratio

0.3

0.4

Figure 3. Median of right spermathecal load (SL) of primipurous
female snow crabs in relation to effective sex ratio (ESR) per replicate in
1997 and 1998 mating experiments. Regressions fitted to log-transformed
data are significant (• 1997: log,0SL = 0.597 • logll,ESR + 2.310. r --
0.80, FIS = 28.25. P = 0.001: A 1998: log,,,SL = 0.425 • logl(,ESR +
1.810, F, ,, = 5.18. r = 0.58. P = 0.046). Encircled values represent the
2cT:20? treatment common to both experiments.

Frequency distributions of log1()SL of females pooled by
experiment produced a similar pattern in 1997 and 1998,
consisting of four modes, of which the second was most
prominent (Fig. 4). The modes are interpreted as represent-
ing spermathecal loads comprising 1. 2, 3, and 4 or more
ejaculates. By differencing mean SL for two sequential
modes to determine the mean size of successive ejaculates,
it was apparent that ejaculates were much smaller in 1998
than in 1997. and that in any given year the size of ejacu-
lates tended to increase with rank of introduction into the
spermatheca. Mixture analysis was applied to log,0SL for
females pooled by treatment to determine the proportions of
females receiving different numbers of ejaculates (Table 4).
Two striking features emerged: in the treatments with the
highest sex ratio of each experiment (2cJ:109 in 1997 and
6cJ:209 in 1998) no female received only one ejaculate; in
contrast, in the intermediate and lower sex ratio treatments
of each experiment, the proportion of females with four or
more ejaculates was null in 1997 or very small in 1998.
Some trends between ejaculate size and sex ratio may be
biologically meaningful (e.g., the inverse relationship be-
tween size of third and fourth ejaculates and sex ratio in
1998) but must be regarded with circumspection given the
small sample sizes.

Regression of sperm counts on SL with intercept forced
to 0 was significant for the 2c?:209 and 10(5:209 treat-
ments (r > 0.89. n = 10. and P < 0.001 for each regres-
sion). Slopes did not differ significantly between the two
treatments (F, 17 = 3.12. P = 0.095). so we pooled the data
and produced a common regression (Fig. 5).

Clutch weight and percent fertilized eggs

Regression using Iog10-transformed data determined a
positive relationship between clutch weight and female
postmolt CW for each treatment of both experiments (r >
0.59 and P < 0.05 for any given regression). In the 1997
experiment, the slopes of size-fecundity relationships were
identical (F2 167 = 0.00, P = 0.999), but the elevations
differed (F-, ]hy == 3.14, P = 0.046) among treatments.
Mean clutch weight adjusted to overall mean CW decreased
with increasing sex ratio, from 6.77 g in the 2d:309
treatment to 6.47 g in the 2cJ:109 treatment. In the 1998
experiment, size-fecundity relationships had similar slopes

(^3,139 = I-01-

= °-192> and elevations (F, I42

= 1.59,

P = 0.194) among treatments.

Percent fertilized eggs per clutch followed a dichotomous
pattern: either at least 95% of the eggs were fertilized ( =

1997

35-

30-

25-

^•H

III

20-

90.5

15-

10-

20.8
-J

•^.

11 283.0

§ O^T~

1 1

3 0.75

1.00

25 1.50 1.75 2.00 2.25 2.50 2.75

°" 35-

1998

— 30-

26.6

25-

20-

III

15-

55.5

-,/-\

10-

2.7

Vi "

5 -

\oL

0-"-|
0.75

1.00

~n — i — i i i i i

1.25 1.50 175 2.00 2.25 2.50 2.75

Log,0 spermathecal load (mg)

Figure 4. Frequency distribution of right spermathecal load for pri-
miparous female snow crabs in the 1997 and 1998 mating experiments.
Four modes were fitted almost perfectly to the distributions in the 1997
(X2 = 9.78. P = 0.913) and 1998 ()C = 2.78, P = 0.999) experiments
using the mixture analysis method of MacDonald and Pitcher (1979).
Modes are interpreted as representing 1 (mode I). 2 (mode II), 3 (mode III).
or 4 or more (mode IV) accumulated ejaculates. The mg-equivalent mean
value of each mode appears below the roman numerals.

212

A. RONDEAU AND B. SAINTE-MARIE

Table 4

Proportion of primiparous female mow crabs fin boldface} attributed to each of 4 modes identified in frequency distributions of spermathecal loads
(see Fig. 4) and mean spermathecal load for each mode (mg. in parentheses) in replicated xe.\ ratio treatments of the 1997 and 1998
mating experiments

Treatments

1997

1998

Modes 26:109

26:209

26:\09

2.5:209

4d:209

6<J:209

I 0.13

0.12

0.00

0.11

0.19

0.00

(17.1)

(20.4)

(-)

(11.5)

(13.9)

(-)

II 0.72

0.74

0.28

0.76

0.41

0.71

(42.3)

(52.5)

(47. S)

(26.5)

(25.8)

(31.2)

III 0.1S

0.14

0.55

0.12

0.35

0.11

(9S.2)

(1 15.4)

(80.2)

(66.6)

(56.3)

(49.9)

IV 0.00

0.00

0.17

0.02

0.05

0.18

(-)

(255.5)

(182.4)

(116.1)

(86.4)

r 7.io

4.20

2.24

17.52

8.61

12.83

P 0.989

0.980

0.945

0.734

0.929

0.884

n 85

Goodness of fit (x2 value and probability. P) is provided for the multiple-mode model that was adjusted by mixture analysis to spermathecal load
frequency distributions for each treatment, following methods of MacDonald and Pitcher ( 1979). The proportion of total females (H) attributed to each mode
was not independent of treatment (G-test of independence on each year. P < 0.05).

well-fertilized clutch) or a large proportion to none of the
eggs were fertilized (= poorly fertilized clutch). Failure to
fertilize most or all eggs apparently resulted from sperm
limitation. Indeed, females with poorly fertilized or well-
fertilized clutches had median SL values of 4.4 mg (range:
3.0-107.0 mg, ;i = 9) or 49.6 mg (4.4-438.3 mg, H = 145)
respectively in 1997 (one-tailed Mann- Whitney test. U =
137, P < 0.001) and of 3.0 mg (0.0-49.8 mg. n= 1 1 ) or

70 -

60 '

50 '

0 20 40 60 80 100 120 140 160 180
Spermathecal load (mg)

Figure 5. Number of sperm cells (Nl in relation to the right sper-
matheca load (SL) of selected primiparous female snow crabs from the
2cJ:209 (T)and 10d:209 (V) treatments of the 1998 mating experiment.
A common regression was fitted to the data (N = 3.788 • KV SL. n = 20.
r = 0.91, P < 0.001).

30.7 mg (10.2-115.1 mg. n = 126) respectively in 1998
(U = 174, P < 0.001).

No clear relationship between fertilization success and
sex ratio was found. The proportion of females with poorly
fertilized clutches was slightly greater in 1998 (7.8%) than
in 1997 (5.8%) and tended to decline with decreasing sex
ratio in each year, but differences between years or among
treatments within a year were not significant (G-test of
independence, P > 0.05 for all analyses). However, the
proportion of females carrying a poorly fertilized clutch
may be underestimated due to the criterion of a postmolt age
of 20 d for examination of eggs, which excludes females
that molted latest and were more likely to have received
small amounts of sperm. This may be especially true of the
2cJ:309 treatment in 1997 due to the negative correlation
between SL and molt date, and of the 1998 experiment
overall due to a tardy molting schedule.

Discussion

Molting asynchrony combined with brief periods of peak
sexual attractiveness for female snow crabs inflated the
operational sex ratio (OSR) in our experiments and led to a
context of male competition even when females far outnum-
bered males (Table 1 ). As expected in such mating systems,
males exhibited flexibility in their allocation of time and
sperm to females. Below, we discuss how mate-guarding
time, sperm depletion, and sperm expenditure varied in
relation to sex ratio and male dominance. We close the

MATE GUARDING TIME AND SPERM ALLOCATION BY MALE SNOW CRABS

213

discussion by considering the impact of variable patterns of
male mating on female mating success.

Duration of mate guarding

Male snow crabs reacted to an increasing sex ratio in the
1998 experiment by guarding females longer, in accordance
with theory (Grafen and Ridley. 1983: Yamamura and Jor-
malainen, 1996) and experimental demonstrations in other
brachyuran and anomuran crabs (Wilber. 1989: Jivoff,
1997a; Jivoff and Mines, 1998; Wada et ai. 1999). Such a
behavior had been inferred for snow crab from in situ
observations that the proportion of premolt females to post-
molt females in mating pairs is greater in years of higher
than of lower sex ratio (Sainte-Marie et ai, 1999).

Due to different assumptions about female choice and
male mate-guarding costs and capability to defend or take
over mates, models of mate guarding in the Crustacea have
predicted that larger males should associate with females for
shorter (e.g., Grafen and Ridley, 1983) or longer (e.g..
Elwood and Dick, 1990) periods of time than smaller males.
In our study, male snow crabs that accumulated the greatest
number of guard days were larger, reflecting in part a size
advantage for the defense of females and the ability to
displace smaller males (Sainte-Marie et ai, 1997). The
nonproportional decline in the contribution of the dominant
male to total guarding days with increasing number of males
(Table 3) suggests that slight male advantages may become
increasingly important as the intensity of competition esca-
lates. Accordingly, the range of sizes and conditions of
males represented in wild mating pairs was narrower when
females were relatively scarce than when they were more
abundant (Sainte-Marie et ai. 1999).

Sperm depletion

There was clear evidence in our experiments that the
sperm reserves of some males were depleted in relation to
the number of mating opportunities, as evidenced at the
replicate level by progressively smaller mean vasa deferen-
tia weight (VDW) with declining sex ratio (Fig. 1 ). More-
over, dominant males were significantly more sperm-de-
pleted than subordinate males (Fig. 2), suggesting that the
former mated more frequently. This occurred even though
the dominant larger males probably had bigger vasa defer-
entia than the subordinate smaller males at the onset of the
experiment, which can be inferred because VDW usually
scales positively to male CW (see Sainte-Marie et ai.
1995). Part of the difference in VDW with male hierarchy
could be due to dominant males charging their vasa defer-
entia more slowly than subordinate males (see Warner et at.,
1995), since energy expenditure and food deprivation may
increase with guarding time (Robinson and Doyle, 1985;
Sparkes et ai. 1996). Furthermore, we posit that the 35%
VDW difference in favor of 1997 over 1998 males reflected

sperm depletion through successive breeding periods. In-
deed, these males were sampled respectively in the autumn
of 1996 and of 1997, 2 and 3 years into a period of intense
recruitment of adult females and of declining abundance of
large adult males which lasted from 1995 to 1998 (DFO,
2000).

Sperm allocation

Three findings converge to indicate that male snow crabs
allocate sperm parsimoniously and partition it among suc-
cessive matings, a behavior termed sperm economy (Pitnick
and Markow, 1994; Shapiro et ai. 1994). First, all primip-
arous females were subrnaximally inseminated, as evident
from the finding that the largest median SL value of 80 mg
in our experiments (Fig. 3 ) was far less than the record mean
value of 256 mg determined for wild primiparous females in
a year of intense male competition (Sainte-Marie. 1993).
Second, in the 1998 experiment even the largest ejaculates
passed to females (72 mg. difference between mean SL for
modes IV and III in Fig. 4) represented just 2.3% of VDW
of control males, and no male fully exhausted his sperm.
Third, spermathecal load (SL) was independent of female
molt date in all but the most female-biased treatment (26":
309). indicating that sperm depletion was not the usual
cause of reduced female sperm reserves at lower sex ratios.

Sperm economy is predicted by sperm competition theory
when females can be polyandrous. mechanisms of last-male
sperm precedence can be effective, and the probability that
one male fertilizes a female's lifetime production of eggs is
small (e.g., Pitnick and Markow. 1994: Parker et ai, 1997).
all of which are attributes of snow crab. The relatively small
size of snow crab ejaculates explains why males can equally
inseminate several females in rapid succession (Sainte-
Marie and Lovrich, 1994). By contrast, the ejaculates of
blue crab (Callinectes sapidus) represent on average 47% of
male gonad volume (Jivoff. 1997b). indicating a sperm-
maximizing strategy that correlates with the typically mo-
nandrous behavior of females, or otherwise ineffective
sperm-precedence mechanisms, and with the generally high
probability that one male fertilizes a female's lifetime pro-
duction of eggs (see Jivoff, 1997a. b). In blue crab, a severe
depletion of sperm reserves occurs after just one mating,
and males cannot equally inseminate even two females in
rapid succession (Jivoff, 1997b; Kendall and Wolcott.
1999).

In snow crab, the coherent pattern of smaller VDW and
SL in the 1998 experiment compared to the 1997 experi-
ment for a given effective sex ratio (ESR) (Figs. 1 and 3)
indicates that males with relatively smaller gonads pass less
ejaculate than males with relatively larger gonads, and this
is further evidence of sperm economy. Furthermore, since
mean SLs for corresponding modes were distinctly smaller
in 1998 compared to 1997 (Fig. 4). but proportions of

214

A. RONDEAU AND B. SAINTE-MAR1E

females with 1, 2, 3, or 4 or more ejaculates were nearly
identical in the common 2<3:209 treatment (Table 4). we
conclude that the 55% difference in SL between the two
years was due mainly to variation in the size of individual
ejaculates.

Superimposed on the pattern of SL set by relative vasa
deferentia size, in each year SL increased with increasing
sex ratio (Fig. 3). This trend resulted from females accu-
mulating more ejaculates of a progressively larger size with
increasing rank of insertion into the spermatheca (Fig. 3.
Table 4), and it occurred whether sex ratio was controlled
by varying the number of females or males. The greater
number of ejaculates reflects some combination of more
frequent repeated matings (this is certain in 1997, because
only two males were used across all treatments) and mul-
tiple matings with growing intensity of male competition.
The importance of repeated mating relative to polyandry in
providing females with larger sperm stores, especially in the
1998 experiment where the number of males was varied
across the treatments, will be resolved by genetic analyses
using hypervariable microsatellite DNA. Furthermore, the
possibility remains that some measure of the variation of SL
in relation to sex ratio was due to males adjusting the size of
individual ejaculates with changing intensity of competi-
tion.

The greater sperm expenditure at higher sex ratios ob-
served in snow crab represents a widespread response of
males to the risk of sperm competition (Gage, 1991; Gage
and Barnard, 1996; Jivoff. 1997b; Wedell and Cook. 1999).
Moreover, the fact that ejaculate size increased with rank of
insertion into the spermatheca (Fig. 4) is consistent with
predictions and observations for other species that males
expend more sperm with previously inseminated females
than with virgin females (Cook and Gage. 1995: Jivoff.
1997a; Parker et al., 1997). Increasing the number or size of
ejaculates may represent a swamping strategy in species
where sperm mixing occurs and all sperm may potentially
access eggs (Pitnick and Markow, 1994). However, sperm
stratification occurs within the spermathecae of snow crab,
and the advantage of introducing a larger ejaculate may be
that it will more effectively displace and isolate any previ-
ously deposited sperm away from the ovary efferent duct
(Sainte-Marie et al.. 2000).

The fact that both guarding time and SL were usually
independent of female molt date strongly suggests that from
the onset of the experiment male snow crabs adopted com-
plementary mate-guarding and sperm-allocation strategies
that remained fixed in time. As proposed by Wada et al.
( 1999) for the hermit crab Pagiirus middendorffii, the rate at
which a male encounters females and other males may
provide information on the sociosexual context — that is, the
potential number of matings to be realized and the degree of
male rivalry — that determines in part the male's mating
strategy. Similarly. Vepsalainen and Savolainen (1995)

demonstrated that past OSR experience could condition
future male mating behavior in the water strider Gerris
lacustris. A mate-guarding and sperm-allocation strategy
that was established early in the breeding season in reflec-
tion of a male's sperm reserves and dominance rank, and of
sociosexual context, would allow the male to maximize the
number of females that he inseminated. Such a strategy may
be maintained even at the expense of reduced fertilization
rate per mating (= sperm limitation) "because it is less
costly to the male than becoming sperm-depleted before
mating opportunities have ceased" (Warner et al.. 1995).

Female mating success and sperm limitation

Increasing male sexual competition had both positive and
negative effects on female mating success. On one hand. SL
increased with increasing sex ratio (Fig. 3); this implies that
females had progressively more sperm in storage, given the
positive relationship between sperm counts and SL in this
study (Fig. 5) as in Sainte-Marie and Lovrich ( 1994). On the
other hand, increasing male competition had adverse effects
on the post-mating condition and survival of females. The
number of missing pereopods per primiparous female (Ta-
ble 2) and the percent mortality of the fragile nulliparous
and primiparous females increased as the number of males
and the sex ratio increased. In the present study, the fre-
quency of injury and mortality may to some extent have
been amplified by confinement in the tanks. However, there
is field evidence that the number of missing pereopods for
primiparous females may vary in relation to the intensity of
male competition (Sainte-Marie et al.. 1999) and that fe-
males are killed by fighting males (Sainte-Marie and Hazel,
1992). Moreover, a negative relationship between female
fecundity and sex ratio was seen in the 1997 experiment
(P < 0.05) and also in the 1998 experiment, although the
trend was not significant in the latter (Rondeau, 2000). This
decline in fecundity is attributed to the loss of recently
extruded, weakly attached eggs during interactions between
post-oviposition females and males, which may occur more
frequently as male bias in sex ratio increases. These nega-
tive effects of male mating activities on female fitness
constitute another example of intersexual conflict (Rowe et
al.. 1994; Vepsalainen and Savolainen, 1995; Jormalainen,
1998).

Snow crab females incubating a poorly fertilized clutch
were apparently sperm-limited, since they had SLs one
order of magnitude smaller than those of females incubating
a well-fertilized clutch. Similarly, using a subjective index
of SL (none, small, or large) on wild female snow crabs,
Carriere ( 1 995 ) found that the proportion of females with
well-fertilized clutches increased significantly with extent
of spermatheca fullness. In our study, however, there were
a few cases where females had a relatively large SL yet a
small or null proportion of fertilized eggs. This apparent

MATE GUARDING TIME AND SPERM ALLOCATION BY MALE SNOW CRABS

inconsistency could arise if a female was mated by another
male, after initially mating and laying eggs with a dominant
male that was particularly sparing of his sperm.

Sperm limitation occurs naturally when males (i) are
too few to inseminate all receptive females, (ii) allocate
their sperm too parsimoniously among females, or (iii) do
not have time to recharge between matings (Pitnick.
1993: Pitnick and Markow, 1994; Warner el til.. 1995;
Jivoff, 1997b; MacDiarmid and Butler. 1999). For snow
crab, the general cause of sperm limitation was probably
sperm economy and in the case of the 309:2cJ treatment,
additionally, perhaps sperm depletion. Small SLs result-
ing from unfavorable mating conditions during the fe-
male's first breeding season may have negative impacts
on her subsequent reproductive activities. Size-fecundity
relationships for multiparous (= repeat spawners) fe-
males (Sainte-Marie. 19931 and the equation relating
sperm counts to SL (Fig. 5) allow estimation of the
minimum doubled SL value required for fertilization of a
second egg clutch using sperm stored over from a previ-
ous breeding period. This value is determined consider-
ing that an average of 70 sperm cells are expended to
fertilize each oocyte (Sainte-Marie and Lovrich, 1994;
Yamasaki el til., 1994). On this basis, 5.2% (1997) and
9.29r ( 1998) of females did not have enough stored sperm
to produce (without re-mating) a second clutch with all
eggs fertilized. These are necessarily conservative esti-
mates because mortality of stored sperm may occur be-
tween ovipositions (Paul, 1984; Sainte-Marie and Sainte-
Marie. 1999). Females with insufficient sperm stores will
produce fewer or no fertilized eggs, or they will re-mate
at the risk of injury or death (Elner and Beninger, 1995).
In closing, our study has shown that the mating strategies
of male snow crabs are quite flexible, which is adaptive to
the widely varying levels of competition intensity and fe-
male availability that characterize this species. Our study
also points to the potential for sperm limitation to occur in
exploited snow crab populations if the removal rates of
large males are too high. Indeed, fishing will depress the sex
ratio and deplete the most competitive component of the
male population. As a result of reduced sexual competition
the surviving large males may be subject to sperm depletion
through extensive mating, which will exacerbate their
sperm-economy behavior. Thus, by contrast to the predom-
inant view in crab fisheries literature that sperm limitation
could arise from the number of males becoming insufficient
to service all females (see Kruse, 1993; Elner and Beninger,
1995), the present study revealing the sperm-economy be-
havior ot male snow crabs suggests an insidious process of
suboptimal insemination. Further research will consider the
implications of sperm economy for the conservation and
management of snow crab.

Acknowledgments

This study is part of a M.Sc. thesis from Institut des
Sciences de la Mer (ISMER) of Universite du Quebec a
Rimouski. and was supported by a Natural Sciences and
Engineering Research Council of Canada (NSERC) grant to
B. Sainte-Marie. We thank F. Hazel, M. Levasseur, and M.
Carpentier for help in the laboratory. Comments by J.-C.
Brethes. E. Mayrand, and two anonymous reviewers im-
proved this paper at various stages of preparation.

Literature Cited

Adams, A. E., and A. J. Paul. 1983. Male parent size, sperm storage and
egg production in the crab Chionoecetes bairdi (Decapoda. Majidae)
Int. J. Invertebr. Rcproit. 6: I81-IS7.

Alunno-Bruscia, M., and B. Sainte-Marie. 1998. Abdomen allometry,
ovary development, and growth of female snow crab Chionoecetes
opilio in the Gulf of Saint Lawrence (Brachyura, Majidae). Can. J.
Fixh. At/nut. Sci. 55: 459-477.

Bouchard, S., B. Sainte-Marie, and J. N. McNeil. 1996. Indirect evi-
dence indicates female semiochemicals release male precopulatory
behaviour in the snow crab, Chionoecetes opilio (Brachyura: Majidae).
Chemoecology 7: 39 — W.

Carriere, C. 1995. Insemination et fecondite chez la femelle du crabe
des neiges Chionoecetes opilio de 1'estuaire maritime du Saint-Laurent.
Master's thesis. Universite du Quebec a Rimouski. Rimouski. Canada.
88 pp.

Christy. J. H. 1987. Competitive mating, mate choice and mating asso-
ciations of brachyuran crabs. Bull. Mar. Sci. 41: 177-191.
Clutton-Brock, T. H.. and G. A. Parker. 1992. Potential reproductive
rates and the operation of sexual selection. Q. Rev. Biol. 4: 437-456.
Cook, P. A., and M. J. G. Gage. 1995. Effects of risk of sperm
competition on the numbers of eupyrene and apyrene sperm ejaculated
by the moth Pl/n/iu interpitnctella (Lepidoptera: Pyralidae). Behav.
Ecol. Sociohiol. 36: 261-268.

DeGoursey, R. E.. and P. J. Auster. 1992. A mating aggregation of the
spider crab (Lihinia i'inari;inara}. J. Northwest Atl. Fish. Sci. 13:
77-82.
Dewsbury, D. A. 1982. Ejaculate cost and male choice. Am. Nat. 119:

601-610.

DFO (Department of Eisheries and Oceans, Canada). 2000. Snow
crab of the Estuary and Northern Gulf of St. Lawrence (areas 13 to 17).
DFO Science Stock Stums Rep. C4-01 (2000). 13 pp.
Donaldson, W. E., and A. E. Adams. 1989. Ethogram of behavior with
emphasis on mating for the Tanner crab Chionoecetes hainli Rathbun.
/ Crustac. Biol. 9: 37-53.

Elner, R. W., and P. G. Beninger. 1995. Multiple reproductive strate-
gies in snow crab. Cliionoecetex opilio: physiological pathways and
behavioral plasticity. J. Ev/>. Mar. Biol. Ecol. 193: 93-112.
Elwood, R. W.. and J. T. A. Dick. 1990. The amorous Gammants: the
relationship between precopula duration and size-assortative mating in
G. piili-x. Aniin. Bclun. 39: 828-833.
Kmlen, S. T., and L. VV. Oring. 1977. Ecology, sexual selection, and the

evolution of mating systems. Science 197: 215-223.
Gage, A. R., and C. J. Barnard. 1996. Male crickets increase sperm
number in relation to competition and female size. Behav. Ecol. So-
ciohiol. 38: 349-353.

Gage, M. J. G. 1991. Risk of sperm competition directly affects ejacu-
late size in the Mediterranean fruit fly. Anim. Behav. 42: 1036-1037.
Grafen, A., and M. Ridley. 1983. A model of mate guarding. ./. Theor.
Biol. 102: 549-567.

216

A. RONDEAU AND B. SAINTE-MARIE

Hartnoll, R. G. 1969. Mating in the Brachyura. Cru.staceana 16: 161-

181.

Jivoff, P. 1997a. The relative roles of predation and sperm competition

on the duration of the post-copulatory association between the sexes in

the blue crab. Callinecte.s sapidiis. Belun: Ecol. Sociobiol. 40: 175-

186.

Jivoff, P. 1997b. Sexual competition among male blue crab. Callinectes

sapidus. Biol. Bull. 193: 368-380.

Jivoff. P., and A. H. Mines. 1998. Effect of female molt stage and sex
ratio on courtship behavior of the blue crab Callinectes sapidus. Mar.
Biol. 131: 533-542.

Jormalainen, V. 1998. Precopulatory mate guarding in crustaceans:
male competitive strategy and intersexual conflict. Q. Rev. Biol. 73:
275-304.

Kendall. M. S., and T. G. Wolcolt. 1999. The influence of male mating
history on male-male competition and female choice in mating asso-
ciations in the blue crab. Callinectes sapidus (Rathbun). J. Exp. Mar.
Biol. Ecol. 239: 23-32.

Kruse, G. H. 1993. Biological perspectives on crab management in
Alaska. Pp. 355-384 in International Symposium on Management
Strategies for Exploited Fish Populations. B. Baxter, ed. Lowell Wake-
field Fisheries Symposium Series. University of Alaska Fairbanks.
Alaska Sea Grant College Program Report 93-02.
MacDiarmid, A. B., and M. J. Butler. 1999. Sperm economy and

limitation in spiny lobsters. Bchav. Ecol. Sociobiol. 46: 14—24.
MacDonald, P. D. M., and T. J. Pitcher. 1979. Age-groups from
size-frequency data: a versatile and efficient method of analysing
distribution mixtures. J. Fish. Res. Board Can. 36: 987-1001.
Moriyasu, M.. and C. Lanteigne. 1998. Embryo development and re-
productive cycle in the snow crab, Chionoecetes opilio (Crustacea:
Majidae), in the southern Gulf of St. Lawrence. Canada. Can. J. Zi/ol.
76: 2040-2048.

Parker, G. A. 1970. Sperm competition and its evolutionary conse-
quences in the insects. Biol. Rev. 45: 525-567.
Parker, G. A. 1974. Courtship persistence and female-guarding as male

time investment strategies. Behaviour 48: 157-184.
Parker, G. A., M. A. Ball, P. Stockley, and M. J. G. Gage. 1997.
Sperm competition games: a prospective analysis of risk assessment.
Proc. R. Soc. Loml. B 264: 1793-1X02.

Paul, A. J. 1984. Mating frequency and viability of stored sperm in the
Tanner crab Chionoecetes bairdi (Decapoda. Majidae). J. Cnistac.
Biol. 4: 375-381.

Paul, A. J., and A. E. Adams. 1984. Breeding and fertile period for
female Chionoecetes bairdi (Decapoda. Majidae). J. Cnistac. Biol. 4:
589-594.

Pelletier, N., A. Fraser, D. Gauthier, M. Laviolette, and M. Moriyasu.
1998. Mise en oeuvre d'une methode pour 1'analyse biochimique du
mecanisme d'accouplement chez le crabe des neiges (Chionoecetes
opilio). Rapp. Tech. Can. Sci. Halicul. Aquat. 2200. 25 pp.
Pitnick, S. 1993. Operational sex ratios and sperm limitation in popula-
tions of Drosophila pachea. Behav. Ecol. Sociobiol. 33: 383-391.
Pitnick, S., and T. A. Markov*. 1994. Male gametic strategies: sperm
size, testes size, and the allocation of ejaculate among successive mates
by the sperm-limited fly Drosophila puclieti and its relatives. Am. Nat.
143: 785-819.

Ridley, M. 1983. The Explanation of Organic Diversity. The Compara-
tive Method and Adaptations for Mating. Clarendon Press. Oxford. 272

PP-
Robinsnn. B. \V.. and R. VV. Doyle. 1985. Trade-off between male

reproduction (amplexus) and growth in the amphipod Gammarus law-

renciunus. Biol. Bull. 168: 482-4XS.
Rondeau, A. 2000. Economic de sperme par les males et succes repro-

ducteur des femelles primipares chez le crabe des neiges. Chionoecetes

opilio, en function du contexte sexuel. Master's thesis, Universite du
Quebec a Rimouski, Rimouski. Canada. 101 pp.

Rowe, L., G. Arnqvist, A. Sih, and J. J. Krupa. 1994. Sexual conflict
and the evolutionary ecology of mating patterns: water striders as a
model system. Trends Ecol. Evol. 9: 289-293.

Sainte-Marie, B. 1993. Reproductive cycle and fecundity of primiparous
and multiparous female snow crab. Chionoecetes opilio, in the north-
west Gulf of Saint Lawrence. Can. J. Fish. Aquat. Sci. 50: 2147-2156.
Sainte-Marie, B., and C. Carriere. 1995. Fertilization of the second
clutch of eggs of snow crab. Chionoecetes opilio, from females mated
once or twice after their molt to maturity. Fish. Bull. (Seattle) 93:
759-764.

Sainte-Marie, B., and F. Hazel. 1992. Moulting and mating of snow
crabs. Chionoecetes opilio (O. Fabricius), in shallow waters of the
northwestern Gulf of Saint Lawrence. Can. J. Fish. Aquat. Sci. 49:
1282-1293.

Sainte-Marie, B., and G. A. Lovrich. 1994. Delivery and storage of
sperm at first mating of female Chionoecetes opilio (Brachyura: Maji-
dae) in relation to size and morphometric maturity of male parent. J.
Cntstac. Biol. 14: 508-521.

Sainte-Marie, G., and B. Sainte-Marie. 1999. Reproductive products in
the adult snow crab (Chionoecetes opilio}. II. Multiple types of sperm
cells and of spermatophores in the spermathecae of mated females.
Can. J. Zool. 77: 451-462.

Sainte-Marie, B., S. Raymond, and J.-C. Brethes. 1995. Growth and
maturation of the benthic stages of male snow crab, Chionoecetes
opilio (Brachyura: Majidae). Can. J. Fish. Aquat. Sci. 52: 903-924.
Sainte-Marie, B., J.-M. Sevigny, B. D. Smith, and G. A. Lovrich. 1996.
Recruitment variability in snow crab (Chionoecetes opilio)'. pattern,
possible causes, and implications for fishery management. Pp. 451-478
in Proceedings of the International Symposium on Biology, Manage-
ment, ami Economics of Crabs from High Latitude Habitats, B. Baxter,
ed. Lowell Wakefield Fisheries Symposium Series. University of
Alaska Fairbanks, Alaska Sea Grant College Program Report 96-02.
Sainte-Marie, B., J.-M. Sevigny, and Y. Gauthier. 1997. Laboratory
behavior of adolescent and adult males of the snow crab (Chionoecetes
opilio} (Brachyura: Majidae) mated noncompetitively and competi-
tively with primiparous females. Can. J. Fish. Aquat. Sci. 54: 239-248.
Sainte-Marie, B., N. Urban!, J.-M. Sevigny, F. Hazel, and U. Kuhnlein.
1999. Multiple choice criteria and the dynamics of assortative mating
during the first breeding season of female snow crab Chionoecetes
opilio (Brachyura. Majidae). Mar. Ecol. Prog. Ser. 181: 141-153.
Sainte-Marie, G., B. Sainte-Marie, and J.-M. Sevigny. 2000. Ejacu-
late-storage patterns and the site of fertilization in female snow crabs
(Chionoecetes opilio: Brachyura. Majidae). Can. J. Zool. 78: 1902-
1917.

Shapiro, D. Y., A. Marconato, and T. Yoshikawa. 1994. Sperm econ-
omy in a coral reef fish, Thalassoma bifasciatum. Ecology 75: 1334-
1 344.
Smith, R. L., ed. 1984. Sperm Competition and the Evolution of Animal

Matin K Systems. Academic Press, New York. 661 pp.
Sokal, R. R., and F. J. Rohlf. 1995. Biometry. 3rd ed. W H. Freeman,

New York. 887 pp.

Sparkes, T. C., D. P. Keogh, and R. A. Pary. 1996. Energetic costs of
mate guarding behavior in male stream dwelling isopods. Gecologia
106: 166-171.

Stevens. B. G., W. E. Donaldson, J. A. Haaga, and J. E. Munk. 1993.
Morphometry and maturity of paired Tanner crabs, Chionoecetes
bairdi. from shallow- and deepwater environments. Can. J. Fish.
Aquat. Sci. 50: 1504-1516.

Stevens, B. G., J. A. Haaga, and W. E. Donaldson. 1994. Aggregative
mating of Tanner crabs, Chionoecetes bairdi. Can. J. Fish. Aquat. Sci.
51: 1273-1280.
Trivers, R. L. 1972. Parental investment and sexual selection. Pp. 136-

MATE GUARDING TIME AND SPERM ALLOCATION BY MALE SNOW CRABS

217

179 in Si-uuil Xclcctiim and the Descent of Man, I87I-I97I, B.
Campbell, ed., Aldine Press. Chicago.

I'rhani. N.. B. Sainte-Marie, J.-M. Sevigny, D. Zadwornv, and V.
Kiihnk'in. 1998. Sperm competition and paternity assurance during
the first breeding period of female snow crab Cluonot'cctc.s opilio
(Brachyura: Majidae). Can. J. Fish. Ai/iuit. Sci. 55: I 1(14-1 I 13.

Vepsalainen. K., and R. Savolainen. 1995. Operational sex ratios and
mating conflicts between the sexes in the water strider Gerris lacustris.
Am. Nat. 146: 869-880.

\\ada. S., K. Tanuka. and S. Goshima. 1999. Precopulatory mate
guarding in the hermit crab Piigunis middendorffii (Brandt) (Decapoda:
Paguridae): effects of population parameters on male guarding dura-
tion. J. £v/>. Mar. Bio/. Ecol. 239: 289-298.

Warner, R. R., D. Y. Shapiro, A. Marcanato, and C. W. Peterstn. 1995.
Sexual conflict: males with the highest mating success convey the
lowest fertilization benefits to females. Proc. K. Soc. Limit. B 262:
1 35- 1 39.

Watson, J. 1972. Mating behavior in the spider crab. Chionoecele.i

<>/>ili<>. J Fish. Res. Board Can. 29: 447-449.
Wedell. N., and P. A. Cook. 1999. Butterflies tailor their ejaculate in

response to sperm competition risk and intensity. Proc. R. Soc. Land.

B 266: 1033-1039.
Wilber, D. H. 1989. The influence of sexual selection and predation on

the mating and postcopulatory guarding behavior of stone crabs (Xan-

thidae, Menippel Behav. Ecol. Sociohiol. 24: 445-451.
Yamamura, N., and V. Jormalainen. 1996. Compromised strategy

resolves intersexual conflict over precopulatory guarding duration.

Evol. Ecol. 10: hdl-ftSO.
Yamasaki, A., S. Fujita, K. tk-hino, and T. Toshima. 1994. Research

into the ecology of snow crabs in the waters off Kyoto Prefecture — IX.

Numbers of sperm in the spermathecas of female snow crabs. Res. Rep.

Kyoto Inst. Ocean Fi.sh. Sci. 17: 19-24 (Can. Trans/. Fish. Aquat. Sci.

5663).

Reference: Biol. Bull. 201: 21S-226. (October 2001)

The Origins of The Grass Foundation

STEVEN J. ZOTTOLI

Department of Biology. Williams College, Williamstown, Massachusetts 01267, and
The Marine Biological Laboratory, Woods Hole, Massachusetts 02543

Introduction

In the fall of 1935, Albert M. Grass and Ellen H. Rob-
inson both came to the Department of Physiology at Har-
vard Medical School (HMS). This entirely fortuitous con-
fluence of their lives led to their marriage, to a commercial
endeavor — the Grass Instrument Company — that would
provide equipment of high quality to neuroscientists and
other physiologists for over half a century, and finally to the
formation of The Grass Foundation, which has benefited the
neuroscience community since 1955.

The Department of Physiology at Harvard — the seedbed
for these accomplishments — had a deep-rooted commit-
ment to providing both financial and moral support to sci-
entists who were at the beginning of their careers. Albert
and Ellen clearly benefited from this commitment, for it
generated interactions and collaborations that led to and
facilitated the success of the Grass Instrument Company and
then the Foundation.

Thus, the origins of The Grass Foundation must be
sought, not only in the conjoined histories and proclivities
of Albert M. and Ellen R. Grass, but also in scientific and
educational developments that took place in the HMS De-
partment of Physiology between 1906 and 1935, well before
Albert and Ellen met there. This essay is an attempt to
dissect those tangled threads; it ends with a discussion of
The Grass Foundation's hallmark program — the Grass Fel-
lowship Program at the Marine Biological Laboratory in
Woods Hole, Massachusetts — and the impact that this pro-
gram has had on neuroscience.

Albert M. Grass and Ellen H. Robinson

Albert Melvin Grass (1910-1992) was born in Quincy.
Massachusetts, on September 3, 1910, to Henry J. Grass and
Bertha (Martin) Grass. After graduating from Quincy High

Received 12 January 2001: accepted 28 August 2(1(11.

School, he funded his college education by working at
Samson Electric Company testing and installing the ampli-
fiers and systems that provided sound for films (Marshall,
1980; Henry, 1992). He was successful, both academically
and financially, and the B.Sc. degree in electrical engineer-
ing was duly awarded by MIT in 1934. Albert remained at
MIT to work on servo-mechanisms used to simulate earth-
quakes in the study of strains acting on model water towers
and other structures (Marshall, 1980).

In May of 1935, Frederic A. Gibbs — a research fellow in
neurology at the HMS Department of Physiology — con-
tracted with Albert Grass to build a 3-channel electroen-
cephalograph (EEG). Albert and his brother Everett worked
in the basement of their father's house and finished the
project early in the fall of 1935 (Marshall, 1980; Grass,
1984).

This accomplishment, completed in only three months,
led to Albert's being hired as a part-time Research Instru-
ment Engineer by the HMS Department of Physiology, and
he remained in that position from 1935 until 1943. As part
of this job, he continued to improve the EEG (Grass Instru-
ment Co., 1971; Grass. 1984) and its applications (Grass
and Gibbs, 1938). In addition, he worked closely with
scientists, tailoring equipment to their needs (e.g., square
wave stimulators and amplifiers; Forbes and Grass, 1937;
Marshall, 1980).

Ellen Harriet Robinson (1914-2001) was born in Taun-
ton, Massachusetts, on March 29, 1914, the daughter of
Laura (Waldron) Robinson and Francis James Robinson.
She graduated from Taunton High School in 1931 and went
to Radcliffe College, receiving her A.B. degree in Biology
in 1935. She continued her education at Harvard University
with Morgan "Kelly" Upton (1898-1984) and received a
Masters degree from Radcliffe College in 1936; her thesis
was entitled "Three Experiments in Audition."

On Upton's recommendation, Ellen decided to immerse
herself in "a broader field of brain function" (Marshall,

218

THE GRASS FOUNDATION

219

1995) and began graduate work in the Department of Phys-
iology of HMS in the fall of 1935 (Marshall. 19SO. 1995).
A special arrangement was made with Radcliffe College so
that she could take a course in "Research" (Marshall. 1995).
and she was supported by a Porter Fellowship in Physiology
(1935-1936) from the American Physiological Society
(Howell and Greene. 1938; Fenn. 1963). Alfred C. Redfield
(1S90-I9S3) was her initial sponsor on a project entitled
"Auditory Action Potentials" (Fenn, 1963; Brobeck ct al.,
1987). hut her work was ultimately conducted under the
sponsorship of Hallowell Davis (Kemp ft al.. 1937; Fenn,
1963). Ellen collaborated with a number of scientists, in-
cluding Arthur James "Bill" Derbyshire (Grass, 1980). Ed-
ward H. Kemp, and Georges Coppee, a Fellow of the C.R.B.
Educational Foundation. Institute de Physiologic, Liege.
Her studies of the responses of the brainstem to auditory
stimulation led to three publications (Kemp and Robinson,
1937: Kemp er al.. 1936. 1937).

Ellen Robinson and Albert Grass met in the fall of 1935
and were married on June 28, 1936. Ellen continued as a
Ph.D. candidate in the Department of Biology at Harvard
University, recording from the auditory cortex of rabbit in
the laboratory of the noted physiological psychologist Karl
Spencer Lashley (1890-1958). Soon, however, she decided
to devote herself "to motherhood and doing whatever I
could to help Albert provide equipment to a growing num-
ber of scientists" (Marshall. 1995).

The Development of Electroencephalography at the

HMS Department of Physiology and the Formation

of the Grass Instrument Company

The report of voltage changes recorded through the cra-
nium of humans by Hans Berger (electroencephalogram.
EEG; Berger, 1929) and his observations of EEC variations
in patients with epilepsy (Berger, 1932; Gibbs et a/., 1936)
were the basis for continued EEG studies of epilepsy in the
United States.

At the time. Stanley Cobb (1887-1968) and William G.
Lennox ( 1884-1960) were experts on epilepsy (?.#., Cobb,
1922; Lennox and Cobb, 1928; Lennox, 1936: see White,
1984, for Cobb's complete bibliography) carrying out their
investigations at Harvard Medical School (Hughes and
Stone, 1990). In 1929, Cobb offered Frederic A. Gibbs
(1903-1992). who had just received his M.D. from Johns
Hopkins, a fellowship in neuropathology to work on epi-
lepsy. Gibbs worked in the Lennox lab where he met Erna
Leonhardt: the couple were married in 1930 and were
collaborators thereafter (Hughes and Stone. 1990).

The Gibbses wanted to record EEGs from epileptic pa-
tients, and Hallowell Davis' engineer. E. Lovett Garceau
(1906-1966). had built amplifiers and a portable EEG that
could be used for such recordings (Garceau and Davis.
1934, 1935: Garceau and Forbes, 1934). Encouraged by

Cobb. the Gibbses approached Davis to be a collaborator,
and Davis enthusiastically endorsed their plan (White.
1984). With the departure of the engineer. Garceau. and the
need for EEG machines with more than one channel,
Frederic Gibbs sought out advice at the Massachusetts In-
stitute of Technology. There he met Albert Grass (Marshall.
1980; Grass, 1984). With funding from the Macy Founda-
tion, he contracted with Albert Grass in May of 1935 to
"build three channels of EEG amplifiers to drive the West-
ern Union Morse Code inkwriting undulator" (Grass, 1984;
Hughes and Stone, 1990).

The Gibbses went off for the summer to attend the
International Congress of Physiologists in Leningrad and
Moscow, and to visit Berger and engineer J. F. Tonnies in
Germany. In August of 1935, rather late in the summer.
Frederic Gibbs mailed a sketch of Tonnies' neuropolygraph
(designed for A. Kornmiiller's animal studies) to Albert
Grass (Grass, 1984; Hughes and Stone, 1990). By that time.
the EEG machine being constructed by Albert and Everett
Grass in Quincy must have been well along, for when the
Gibbses returned in the fall, it was finished, as mentioned
above. This EEG was used by Lennox, Frederic and Erna
Gibbs, and Davis in their pioneering investigations, which
demonstrated the power of the EEG in the diagnosis of
epilepsy (Gibbs et al., 1935. 1936, 1937; Brazier, 1968).

The demand for EEG machines increased markedly dur-
ing the 1940s. To meet this demand, Albert began to man-
ufacture commercial instruments (Marshall, 1987). Thus.
the small business that started in a basement in 1935 con-
tinued as a partnership between Albert and Ellen Grass, and
ultimately developed and grew to become, 10 years later,
the Grass Instrument Company. The success of the Com-
pany was due to the balance between Albert's engineering
skills and Ellen's scientific expertise, which was critical in
the proper design of equipment to meet the needs of neu-
rophysiologists (Fig. 1 ). Instruments were designed for
"convenience, durability and serviceability" (Morison,
1979).

The Grass Instrument Company was never a typical busi-
ness. In the early years, employees and many of the cus-
tomers were warm and loyal friends of Albert and Ellen
(Morison, 1979). and neurophysiological equipment was
loaned to investigators throughout the world, and especially
to Grass Fellows and courses at the Marine Biological
Laboratory (MBL). To Albert and Ellen, the Company was
always meant to contribute "to the development of human
knowledge and the search for basic scientific truth." (Grass
Instrument Co.. 1967). They took great care to ensure that
equipment was being properly used for the benefit of the
patient. Ricardo Miledi recalls, "When I was in Mexico, I
remember on more than one occasion, seeing a letter [from
the Grasses] inquiring about doctors that intended to pur-
chase their most advanced EEGs and other equipment.
Albert and Ellen were very concerned that their equipment

220

S. J. ZOTTOLI

Figure 1. Albert and Ellen Grass in 1955. the year that The Grass
Foundation was formed. This picture was copied from a newspaper article
entitled "Doctors Told of Findings by Quincy 'Brain' Machine" in the
Quincy Patriot Ledger, Saturday, October 8. 1955.

be used wisely for the benefit of the patients and for re-
search, and not merely to extract money from the patients."

Department of Physiology at Harvard Medical School

Albert and Ellen Grass's success was clearly due, in part,
to the support they received from established researchers in
the Department of Physiology at Harvard Medical School.
Walter B. Cannon and Alexander Forbes were especially
critical in this regard.

"Speaking personally now," Ellen once said, "Dr. Can-
non made very many things possible for Albert and for me.
He invested in us at a time when biomedical engineering
was indeed in its infancy, and the role for women in science
practically negligible" (Grass, 1970).

Walter B. Cannon (1871-1945) served as the George
Higginson Professor of Physiology at Harvard Medical
School for 36 years ( 1906-1942). For this entire period, he
was chairman of the Department of Physiology and an
Emeritus for his last three years. He made many important
contributions to our understanding of how the human body
functions: the use of Roentgen rays to investigate gastroin-
testinal motility (Cannon. 1898; Cannon, 1911; Brooks et

til., 1975: Barger, 1981), the effects of emotions on the
functional state of the body (Cannon. 1915; Davenport,
1981), the basis of surgical shock (Cannon, 1923), the
constancy of the internal environment or homeostasis (Can-
non. 1939). autonomic neuro-effector systems (Cannon and
Rosenblueth, 1937), and the effects of denervation on var-
ious tissues (Cannon and Rosenblueth, 1949).

Walter Cannon spent his undergraduate years at Harvard
University and continued on at Harvard Medical School,
where the faculty held that interested medical students
should be encouraged to conduct original research. Thus, in
the first semester of his medical training Cannon and fellow
student Albert Moser were encouraged by Henry P. Bow-
ditch ( 1840-1911) to conduct a study on deglutition (Beni-
son et al., 1987). Later, as a third-year medical student,
Walter Cannon was approached by William Norton Bullard
(1853-1931). a neurologist at Boston City Hospital, who
offered to fund further research (Taylor, 1931 ).

These early research experiences clearly had a profound
influence on the development of Cannon's scientific philos-
ophies.

Every man active in investigation has more problems in mind
that he can work at himself. A part of his service to the world
consists in training others by giving to others these problems
to work at under his direction. These "others" are ordinarily
his students. — young men who have been stimulated by his
example. They are not yet established in life, they require
remuneration until they have done enough work to warrant
their being taken into independent positions. They should
receive during these years of training (which are very likely to
be productive of good results in research) sufficient compen-
sation to afford comfortable support (Benison et ai, 1987).

Cannon is generally considered to have been exemplary
in his scientific conduct and his concern for human welfare
(Cannon. 1945; Forbes, 1945; Morison, 1945: Davis, 1975).
He "saw that the freedom and beneficence of science could
be guaranteed only within the framework of a just society,
national and international" (Grass, 1970), and was commit-
ted to providing promising young scientists, independent of
nationality, the opportunity to participate and contribute to
the advancement of science (Morison, 1945).

One such scientist was Arturo Rosenblueth (1900-
1970)' who came to Harvard from Mexico as a Guggen-
heim Fellow from 1930 to 1932 to work with Cannon. He
quickly became Cannon's "favorite son" and secured a
position in the Department of Physiology. Their collabora-
tions continued for the next 14 years (e.g.. Cannon and
Rosenblueth. 1937, 1949).

Cannon and Rosenblueth mentored several scientists who
would ultimately become founding and early trustees of The

' The names of the founding and early trustees of the Grass Foundation
arc printed in bold type in this section of the paper. E-mail:
Steven. J.Zottoli (3' Williams.edu.

THE GRASS FOUNDATION

221

Table 1

FimnJini> trustees and early trustees of The Crass Foundation

Founding trustees

Alexander Forbes

Albert M. Grass

Ellen R. Grass

Frederic A. Gibbs

William G. Lennox*

Robert S. Morison

Arturo Rosenblueth*

Richard R. Tovvle

Robert A. Zottoli
Early trustees

George H. Acheson, 1961

Donald B. Lindsley. 1958

Fiorindo A. Simeone, 1968

* Although Lennox and Rosenhlueth are not listed as original members
of the Corporation in the Constitution and Bylaws of The Grass Founda-
tion, they are recognized as founding members in the minutes of The Grass
Foundation.

Grass Foundation (Table 1). Alexander Forbes (1882-

1965) as a fourth-year medical student was encouraged by
Cannon to become involved in research. After receiving his
M.D. degree in 1910 from Harvard Medical School, Forbes
studied with C. S. Sherrington (1857-1952) for two years,
and briefly with Lucas in 1912; afterwards, he returned to
Harvard and the Department of Physiology (Fenn, 1969,
Davis. 1970; Eccles, 1970). Forbes added a strong engineer-
ing background to the department and was continuously at
the forefront of technological advances that he applied to
neurophysiological investigations. These included the use
of the vacuum tube amplifier in conjunction with a string
galvanometer to record action currents in nerve and muscle
(Forbes and Thacher. 1920, see also Gasser and Newcomer,
1921; Forbes et at., 1931; Grass, 1984; Frank, 1986; Sey-
farth, 1996), the study of reflex activity (Forbes, 1922;
Davis, 1975; Seyfarth, 1996), and the use of microelec-
trodes for extracellular recording from cortical cells (Ren-
shaw et at.. 1940; Brazier, 1968).

In fact, Forbes' technical and analytical strengths, along
with those of Hallowell Davis (Forbes et ai, 1931), com-
plemented the more integrative approaches of Cannon and
Rosenblueth (Cannon, 1945). Cannon's encouragement of
Forbes as a young medical student could not have affected
a more appreciative and capable individual. Alexander
Forbes quickly adopted the philosophy of encouraging sci-
entists in his own way. He "anonymously supported others
in the department of physiology" (Davis, 1970; Seyfarth.
1996).

One of the many young medical students supported by
Forbes was Hallowell Davis (1896-1992). He received his
B.A. degree in 1918 and the M.D. degree in 1922 from
Harvard, worked for a year at Cambridge University in
England with Edgar D. Adrian, and then returned to Har-

vard in 1923 as an Instructor in the Department of Physi-
ology (Davis, 1991; Galambos, 1998). Some of his studies
at Harvard include the all-or-none nature of the nerve im-
pulse (Davis et ai, 1926), the use of the EEG in the study
of epilepsy (Gibbs et ai. 1935). recordings from single units
in the "auditory nerve" of cats (Galambos and Davis, 1943;
the recordings turned out to be from cell bodies of the
cochlear nucleus. Davis. 1975). and the tolerance of the
human ear to loud sounds (Davis et ai, 1950). Hallowell
Davis would become the sponsor of Ellen Grass' research
and an exponent of EEG recording at Harvard.

Donald B. Lindsley (currently Trustee Emeritus) had
come to Harvard Medical School with a National Research
Council Fellowship to work with Forbes and Davis in 1933.
During this period. Lindsley recorded motor unit responses
(Lindsley. 1934, 1935a) and pioneered the use of the elec-
tromyogram in neuromuscular disorders (e.g., Lindsley,
1935b, 1936: see Lindsley, 1995, for a review).

Arturo Rosenblueth had encouraged George H. Acheson
(1912-2000), a first-year medical student, and Fiorindo A.
Simeone (1908-1990), a third-year medical student, to
consider conducting original research. Both contributed to
the scientific productivity in the department (e.g., Rosen-
blueth and Simeone, 1934, 1938a. b; Acheson, 1938;
Acheson et al., 1936, 1942; Simeone et ai, 1938) and went
on to distinguished medical careers.

Robert Morison (1906-1986) received an undergradu-
ate degree from Harvard in 1930 and the M.D. in 1935. He
was encouraged to pursue research by his mentor, Rosen-
blueth. during his medical school years. "He [Morison] was
a man of great and thoughtful learning but one who, above
all. wanted to understand the meaning of life and the sig-
nificance of science for that fundamental issue. He under-
stood what it was to make a moral vocation of one's
intellectual work, an effort that requires not only reading,
writing, and thinking, but also something else: the living
out, in daily life, of the values and virtues that animate that
work." (Callahan, 1987). Morison collaborated with many
of those present in the Department of Physiology in the
1930s (e.g., Rosenblueth and Morison, 1934; Rosenblueth
et til., 1936) and went on to Rockefeller University and then
Cornell (Eisner et al.. 1986-1987).

The Formation of The Grass Foundation

As the number of requests for financial support of neu-
roscience endeavors grew. Albert and Ellen recognized that
a mechanism must be found to evaluate proposals and make
decisions (Morison. 1979). The Grass Charity Trust was
formed on December 31. 1948, and charitable disburse-
ments were made after June 27, 1951. This Trust donated
most of its assets to The Grass Foundation (The Grass
Foundation minutes, 1958). which was formed in 1955 "to
assist in advancing knowledge principally in the field of

222

S. J. ZOTTOLI

Figure 2. Four ot the ciriginal Trustees of The Grass Foundation. From left to right: Albert Grass. Frederic
Gibbs. Ellen Grass. Robert Morison. and Erna Gibbs (not a Trustee) at the 111 International Congress of
Electroencephalography and Clinical Neurophysiology held in Boston from August 17-21. 1953.

neurophysiology, and including allied fields of medicine
and science" (Article 2 Section 1 of The Grass Foundation
Constitution and Bylaws).

As we have seen, most of the founding and early trustees
of the Foundation were, at some time in the 1930s, members
of the Department of Physiology at Harvard Medical School
(Fig. 2: Table 1 ). This is only fitting, because their commit-
ment to the support of young scientists, their own exemplary
performance at the bench, and their concern for human
welfare (Morison. 1979) reflect the basic principles that
have molded The Grass Foundation. The Foundation cur-
rently supports programs within the Society for Neuro-
science, at the MBL. and at other institutions. The Grass
Fellowship Program at the MBL was one of the first and
most important projects of The Grass Foundation, and it
continues to flourish.

The Association of Albert and Ellen Grass and The

Grass Foundation With the Marine Biological

Laboratory at Woods Hole

Albert and Ellen Grass's affinity for the MBL developed
over many years and is based on several associations. For
example, Alexander Forbes had a natural affection for the
Woods Hole area. He spent summers on Naushon Island,
which is still owned by the Forbes family, and he was a
distinguished investigator at the MBL, publishing research
done there with Catharine Thachcr (Forbes and Thacher,
1925; Forbes, 1933). Albert Grass was undoubtedly at-

tracted to the MBL because, as a center of neurophysiology,
it was regularly visited in the summers by scientists who
were actively involved in the development of new equip-
ment. With this common interest. Albert developed lasting
friendships with several MBL scientists, including Hairy
Grundfest (1904-1983) and Stephen Kuffler (1913-1980).
Finally. Ellen was drawn to the MBL by her passion for the
marine environment and the animals that live there. This
passion was particularly evident in her Grass Instrument
Company Calendars and the "live displays" presented at the
annual meetings of the Society for Neuroscience.

Of the initiatives in support of basic science at the MBL.
the Grass Fellowship Program most closely embodies the
philosophy of the founding trustees who had "a love for the
adventure of new ideas, a priority for assisting young in-
vestigators, and a program focus to direct its resources to the
growth of neurophysiology" (Grass, 1987). Beginning with
two fellows in 1951 under the auspices of the Grass Charity
Trust, more than 400 young neuroscientists have spent
summers at the MBL conducting independent research.

Established in 1959, the Forbes Lectureship is an integral
part of the Grass Fellowship Program. Each summer, the
Trustees of The Grass Foundation bring one of the world's
outstanding neuroscientists to the MBL "to honor the out-
standing achievements of Dr. Alexander Forbes as a pioneer
and major contributor to the field of neurophysiology. who
has always been an inspired teacher of young students."
(The Grass Foundation minutes, December 1 1. 1958). The

THE GRASS FOUNDATION

223

Table 2

Grass Felliw.-. fram

Year

Fellow

1951

1452
1453

1954
1955

1956
1957

1958
1959

1960
L961

Hal C. Becker
Samuel M. Peacock. Jr.
Ellis C. Berkouitz
Donald M. Maynard
Y. Zotterman
Daniel D. Hansen
David D. Potter
Ricardo Miledi
Yutaka Oomura
Joaquin Remolina
William K. Stephenson
Lionel Adelson
Stanley M. Grain
Clarence Hurdiman
Joan Taylor*
Michael V. L. Bennett
John P. Reuben
William H. Rickles. Jr.
Shirley H. Bryant
Raymond J. Lipicky
Charles F. Stevens
Stephen T. Kitai
Leslie B. Reynolds
Bernice Grafstein
Zach W. Hall
Walter Herzog
Robert H. Wurtz

* Her fellowship was carried out at UCLA.

Forbes Lecturer not only presents two lectures as part of the
MBL's Lecture Series, but also shares space with the Fel-
lows in the Grass Laboratory. Forbes inaugurated the series
with a pair of lectures, on "The Growth of Neurophysiol-
ogy" and on "Electrophysiology of Color Vision."

Many Grass Fellows have gone on to become leaders in
neuroscience (Table 2). and many have come back to the
MBL as investigators, course directors, instructors, Forbes
Lecturers, and as directors and associate directors of the
Grass Laboratory.

Ricardo Miledi is an example of a Grass Fellow who has
had a major impact on neuroscience. He was born in Mexico
City in 1927, received his B.Sc. from the Institute Cientifico
y Literario. Chihuahua, in 1945 and his M.D. from the
Universidad Nacional Autonoma de Mexico in 1955. Miledi
and Joaquin Remolina were working with Arturo Rosen-
blueth and Juan Garcia Ramos at the Institute Nacional de
Cardiologia (Garcia Ramos and Miledi, 1953, 1954; Rosen-
blueth et al. 1954) when Albert Grass and Steve Kuffler
came by to visit Rosenblueth. Informed about the Grass
Fellowship Program by the visitors, Miledi and Remolina
came to the MBL as fellows in 1955 (Fig. 3).

There was no Grass Laboratory or formal program at that
time, and fellows generally worked in separate spaces and

did not interact with one another a great deal. Steve Kuffler,
Harry Grundfest ( 1904-1983). and their collaborators were
perennial investigators at the MBL. and both were Forbes
Lecturers, in 1975 and 1979, respectively. They acted as
mentors for Miledi and other Grass Fellows in the early
years of the program (Zottoli, 1990). Albert and Ellen Grass
would also visit the MBL periodically to make sure that the
Grass Fellows had what they needed in the way of equip-
ment, space, and support.

While at the MBL, Miledi worked on lobster and crayfish
stretch receptors and the squid giant axon (Miledi. 1957).
Miledi has gone on to publish over 460 scientific articles,
and he is one of two neuroscientists to be chosen more than
once by the Trustees of The Grass Foundation as the Forbes
Lecturer (the other is Theodore H. Bullock, in 1963 and
1991). Miledi presented some of his and Bernard Katz's
seminal work on neuromuscular transmission at the MBL in
1964 as the sixth Forbes Lecturer. The topic of his two
lectures was "Localization of ACh receptors and cholines-
terase in muscle fibres." He returned as the Forbes Lecturer
in 1990 (Fig. 3) and delivered two lectures on pioneering
work that utilized frog oocytes to study native receptors and
express exogenous messenger RNA (Kusano et al., 1977,
1982; Barnard et al.. 1982). The subject of his two lectures
was "How to study the brain using frog oocytes." Dr. Miledi
has served two terms as a Trustee of The Grass Foundation
(1992-1995; 1997-2000). He is currently a Distinguished
Professor in the Department of Neurobiology and Behavior
at the University of California, Irvine.

The philanthropic largess of Albert and Ellen goes well
beyond the MBL and has benefited the neuroscience com-
munity in many other ways. For example, "individuals
known to be sound investigators, working under budgetary
or foreign exchange difficulties, often found themselves the
recipients of indefinite loans" of neurophysiological equip-
ment through the auspices of the Grass Instrument Com-
pany (Morison, 1979). Ricardo Miledi remembers, "Joaquin
Remolina and I were once asked to make a list of Grass
equipment to be bought for the Department of Physiology at
the Institute of Cardiology in Mexico. We made a big list
that was sent to the Grass Instrument Company and were
looking forward, with great excitement, to the day when the
equipment would arrive. Then, to our great consternation,
there was a big devaluation of the peso and we were asked
to send a new order with a good number of items deleted.
Later, when the shipment arrived we were all extremely
pleased to see that all the items in our original list had
arrived and a few extra ones had been included, as if to
compensate for our transient worries. I wonder if any other
company exists that would do that?"

Albert Grass died in Quincy, Massachusetts, on May 29,
1992, and Ellen died 9 years later on June 14. 2001. also in
Quincy. The Grass Foundation that embodies their ideals
continues to be committed to providing general support for

224

S. J. ZOTTOLI

Figure 3. Albert Grass and Ricardo Miledi in Woods Hole, Massachusetts. The picture on the left was taken
in 1955 at the Marine Biological Laboratory when Miledi was a Grass Fellow. The picture on the right was a
reenactment of the 1955 photograph taken in 1990 at the National Academy of Sciences in Woods Hole when
Miledi was a Forbes Lecturer for a second time in 1990. The 1990 photograph was taken by Steve Zottoli.

"excellent science." The tireless efforts of Albert and Ellen
to monitor the pulse and flow of neuroscience have led to
initiatives in support of the field, especially in helping those
in need or just starting out. The greatest achievements of the
Foundation's initiatives, such as the Grass Fellowship Pro-
gram at the MBL, have resulted from the ability of the
Trustees to listen, hear and respond to needs of scientists in
a rapidly changing discipline.

Acknowledgments

I would like to thank Ernst-August Seyfarth for his help
and support during this project. George Acheson, Ellen
Grass, Hank Grass. Ron Hoy, Don Lindsley, Ricardo
Miledi. John Reuben, and Richmond Woodward provided
important suggestions for the improvement of an earlier
version of this manuscript. I also thank Helena Warburg for
her efforts in providing biographical information. Mrs. Elin
L. Wolfe at the Countway Library of Medicine deserves
special mention for the research she did on Catherine
Thacher. Finally. I would like to acknowledge the help,
enthusiasm, and ;. alienee of Mike Greenberg.

Literature Cited

Acheson. G. H. 1938. The topographical anatomy of the smooth muscle

of the cat's nictitating membrane. Aunt. Rec. 71: 297-311.
Acheson, G. H., A. Rosenblueth, and P. F. Partington. 1936. Some

afferent nerves producing reflex responses of the nictitating membrane.

,4»i. ./. Physiol. 115: 308-316.
Acheson, G. H., K. S. Lee, and R. S. Morison. 1942. A deficiency in the

phrenic respiratory discharges parallel to retrograde degeneration.

J. Neuroplmiol. 5: 269-273.
Barger, A. C. 1981. New technology for a new century: Walter B.

Cannon and the invisible rays. Physiologist 24: 6-14.
Barnard, F. A., R. Miledi, and K. Sumikawa. 1982. Translation of

exogenous messenger RNA coding for nicotinic acetylcholine recep-
tors produces functional receptors in Xcnoniis oocytes. Prat: R. Soc.

Liintl. H 215: 241-246.
Benison, S., A. C. Barger, and E. L. Wolfe. 1987. Walter B. Cannon:

Tin- Life and Times of a Young Scientist. The Belknap Press of Harvard

Univ. Press. Cambridge. MA. 520 pp.
Berger, H. 1929. Uher das Elektroenkephalogram des Menschcn. I.

Arch. Psychitilr. 87: 527-570.
Berger, H. 1932. Uher das Hlcktrenkeplialognmi des Menschen. Arch.

Psychuin: 97: 6-26.
Brazier, M. A. B. 1968. The Elcctriial Activity of the Nervous S\stcin.

Williams and Wilkens. Baltimore. 317 pp.
Brobeck, J. R., O. E. Reynolds, and T. A. Appel, eds. 1987. History of

THE GRASS FOUNDATION

225

American Physiological Society, The First Century 1XX7-I987. The

American Physiological Society, Bethesda, MD.
Brooks. C. McC.. K. Koizumi, and J. O. Pinkston, eds. 1975. The Life

ami Contributions «t Walter Bradford Cannon IX7I-IV45. SUNY

Downstate Medical Center. Brooklyn. 264 pp.
C'allalian. I). 1987. Robert S. Morison, M.D. Hastings Report February,

1987.
Cannon. \\. B. 1898. The movements of the stomach as studied by

means of the Rontgen rays. Am. J. Physiol. 1: 359-382.
C'annon. \V. B. 1911. The Mechanical Factors of Digestion. Edward

Arnold. London. 227 pp.
Cannon, \V. B. 1915. BoJil\ Changes in Pain, Hunger, Fear and Rage.

Appleton, New York. 31 1 pp.

Cannon, \V. B. 1923. Traumatic Shuck. Appleton. New York. 201 pp.
Cannon. \V. B. 1939. The Wisdom of the Body. W. W. Norton, New

York. 333 pp.
C'annon, \V. B. 1945. The H'«v of an Investigator. W. W. Norton, New

York. 229 pp.
Cannon, \V. B.. and A. Rosenhlueth. 1937. Antononiic Nemo-Effector

Systems. Macmillan. New York. 229 pp.
Cannon. \V. B., and A. Rosenblueth. 1949. The Snpersensitivitv of

Denemiled Structures: A Line of Denervation. Macmilkin, New York.

245 pp.
Cobb. S. 1922. A case of epilepsy with a general discussion of the

pathology. Med. Clin. North Am. S: 1403-1420.
Davenport. H. \V. 1981. Signs of anxiety, rage, or distress. Physiologist

24: 1-5.
Davis, H. 197(1. Alexander Forbes. Pp. 64-66 in Dictionary of Scientific

Biography. Scribner, New York.
Davis, H. 1975. The philosophy of science and the way of investigator.

Pp. 186-193 in The Life and Contributions of Walter Bradford Can-
non. C. McC. Brooks. K. Koizumi and J. O. Pinkston. eds. SUNY

Downstate Medical Center, Brooklyn.
Davis, H. 1991. The Professional Memoirs of Hallowell Davis. Centra]

Institute for the Deaf, St. Louis. MO. 112 pp.
Davis, H., A. Forbes, D. Brunswick, and A. McH. Hopkins. 1926.

Studies of nerve impulse. II. The question of decrement. Am. J. Physiol.

76: 44S-471.
Davis, H., C. T. Morgan. J. E. Hawkins, Jr.. R. Galambos, and F. W.

Smith. 1950. Temporary deafness following exposure to loud tones

and noise. Acta Oto-Ltiryngol. (Suppl.l 88: 1-57.

Eccles, J. C. 1970. Alexander Forbes and his achievement in electro-
physiology. Perxpecr. Biol. Med. 13: 388-404.
Eisner. T.. A. M. Srb, and F. A. Long. 1986-1987. Robert Swam

Morison. Cornell University Faculty Memorial Statements.
Fenn, W. O. 1963. History of the American Physiological Society: The

Third Quarter Century. 1937-1962. The American Physiological So-
ciety. Washington. DC. Pp. 44-46.
Fenn. \V. O. 1969. Alexander Forbes. Biog. Mem. Nail. Acad. Sci. 40:

113-141.
Forbes, A. 1922. The interpretation of spinal reflexes in terms of present

knowledge of nerve conduction. Phvsiol. Rev. 2: 361-414.
Forbes, A. 1933. Conditions affecting the response of the avicularia of

Hiigiila. Biol. Bull. 65: 469-479.
Forbes, A. 1945. Walter Brandford Cannon (1871-1945). Pp. 349-354

in Year Book of the American Philosophical Society. Philadelphia.
Forbes, A., and A. M. Grass. 1937. A simple direct-coupled amplifier

for action potentials. J. Physiol. 91: 31-35.
Forbes, A., and C. Thacher. 1920. Amplification of action currents w ith

the electron tube in recording with the string galvanometer. Am. ./.

/VivwW. 52: 409-471.
Forbes, A., and C. Thacher. 1925. Changes in the protoplasm of Nereis

eggs induced by (3-radiation. Am. J. Physiol. 74: 567-578.
Forbes. A., H. Davis, and .1. H. Emerson. 1931. An amplifier, string

galvanometer, and photographic camera designed for the study of
action potentials in nerve. Rev. Sci. In.str. 2: 1-15.

Frank, R. G., Jr. 1986. The Columbian exchange: American physiolo-
gists and neuroscience techniques. Fed. Proc. 45: 2665-2672.

Galambos, R. 1998. Hallowell Davis 1896-1992. Biog. Mem. Nat.
Acad. Sci. 75: 3-23.

Galambos, R., and H. Davis. 1943. The response of single auditory-
nerve fibers to acoustic stimulation. J. Neurophysiol. 6: 39-58.

Garceau, E. L., and H. Davis. 1934. An amplifier, recording system and
stimulation devices tor the study of cerebral action currents. Am. ./.
Physiol. 107: 305-310.

Garceau, E. L., and H. Davis. 1935. An ink-writing electroencephalo-
graph. Arch. Neurol. Psychol. 34: 1292-1294.

Garceau, E. L., and A. Forbes. 1934. A direct-coupled amplifier for
action currents. Rev. Scient. Instr. 5: 10-13.

Garcia Ramos, J., and R. Miledi. 1953. Estudios sobre el Flutter y la
Fibrilacion. IX La Fibrilacion Ventricular. ,4/r/f. Inst. Cardio. Me.\. 22:
805-834.

Garcia Ramos, J. and R. Miledi. 1954. La Fibrilacion Ventricular. Bol.
San. Mil. Mex. 7: 3-24.

Gasser, H. S., and H. S. Newcomer. 1921. Physiological action currents
in the phrenic nerve. An application of the thermionic vacuum tube to
nerve physiology. Am. ./. Phvsiol. 57: 1-26.

Gibbs, F. A., H. Davis, and \V. G. Lennox. 1935. The electro-enceph-
alogram in epilepsy and in conditions of impaired consciousness. Arch.
Neurol. Psychiatry 34: 1 133-1 14S.

Gibbs, F. A.. W. G. Lennox, and E. L. Gibbs. 1936. The electro-
encephalogram in diagnosis and in localization of epileptic seizures.
Arch. Neurol. Psychiatry 36: 1225-1250.

Gibbs, F. A., E. L. Gibbs, and W. G. Lennox. 1937. Epilepsy: a
paroxysmal cerebral dysrhythmia. Bruin 60: 377-388.

Grass, A. M. 1984. The Electroencephalographic Heritage. Grass In-
strument Co.. Quincy. MA. 41 pp.

Grass, A. M., and F. A. Gibbs. 1938. A Fourier transform of the
electroencephalogram. J. Neurop/i\siol. 1: 521-526.

Grass, E. R. 1970. Text of the Introduction to the first Cannon lecture
supported by the Grass Foundation, Cornell University Medical
School. NYC. Tuesday Nov. 17, 1970.

Grass, J. 1980. Arthur James "Bill" Derbyshire. Taped interview with
"Bill" Derbyshire. August. 1980. UCLA Brain Research Institute.
Neuroscience History Resource Project. Oral History Program, Series
CON. Code DER.

Grass, M. 1987. Alexander Forbes. A handout distributed prior to the
annual Forbes Lectures supported by The Grass Foundation at the
Marine Biological Laboratory.

Grass Instrument Co. 1967. This is Grass Instrument Company. Grass
Instrument Company Catalogue. 1967, Quincy, MA.

Grass Instrument Co. 1971. Perspectives in Electrophysio/ogical In-
strumentation. 1971 calendar of the Grass Instrument Co.. Quincy.
MA.

Henry, M. H. 1992. In Memoriam. Albert Melvin Grass. 1910-1992.
J. Clin. Neurophysiol. 9: 419-421.

Howell, VV. H., and C. W. Greene. 1938. History of the American
Physiological Society Semicentennial, 1887-1937. American Physio-
logical Society. Baltimore, MD.

Hughes. J. R., and J. L. Stone. 1990. An interview with Frederic A.
Gibbs, M.D. and Erna L. Gibbs. Clin. Electroencepha/ogr. 21: 175-
182.

Kemp, E. H., and E. H. Robinson. 1937. Electric responses of the brain
stem to bilateral auditory stimulation. Am. ./. Ph\\iol. 120: 316-322.

Kemp, E. H., G. Coppee, and R. Robinson. 1936. Les voies auditives
au niveau de la moelle allongee (Chat). Mise en evidence des synapses
nerveuses. C. R Soc. Biol. 122: 1244-1297.

Kemp. E. H., G. E. Coppee, and E. H. Robinson. 1937. Electric

226

S. J. ZOTTOLI

responses of the brain stem to unilateral auditory stimulation. Am. J.
Phyxial. 12(1: 304-315

Kusano, K., R. Miledi, and J. Stinnakre. 1977. Acetylcholine receptors
in the oocyte membrane. Nature 270(5639): 739-741.

Kusano, K., R. Miledi, and J. Stinnakre. 1982. Cholinergic and cat-
echolaminergic receptors in the Xenopus oocyte membrane. J. Physiol.
328: 143-170.

Lennox, W. G. 1936. The physiological pathogenesis of epilepsy. Bruin
59: 113-121.

Lennox, W. G., and S. Cobb. 1928. Epilepsy, from the standpoint of
physiology and treatment. Medicine 7: 105-290.

Lindsley, D. B. 1934. Inhibition as an accompaniment of the knee jerk.
Am. J. Physiol. 109: 181-191.

Lindsley, D. B. 1935a. Electrical activity of human motor units during
voluntary contraction. Am. J. Physiol. 114: 90-99.

Lindsley, D. B. 1935b. Myographic and electromyographics studies of
myasthenia gravis. Bruin 58: 470-482.

Lindsley, D. B. 1936. Electromyographic studies of neuromuscular dis-
orders. A rch. Nenrol. Psychiatry 36: 128-157.

Lindsley, D. B. 1995. Life and reflections of a psychologist-psycho-
physiologist from a personal and historical perspective. Int. J. Psyc/io-
physiol. 20: 83-141.

Marshall, L. H. 198(1. Taped interview with Albert Melvin Grass. Nov.
11.1 980. UCLA Brain Research Institute, Neuroscience History Re-
source Project, Oral History Program. Series CON, Code GRA.

Marshall, L. H. 1987. Instruments, techniques, and social units in Amer-
ican neurophysiology, 1870-1950. Physiology in the American Con-
text. 1850-1940. American Physiological Society. Washington, DC.

Marshall, L. H. 1995. Taped interview with Ellen R. Grass. Sept. 15.
1995. UCLA Brain Research Institute. Neuroscience History Resource
Project, Oral History Program, Series CON, Code GRA.

Miledi, R. 1957. The strength latency relation of axons. Acta Physio/.
Liitinoiimer. 7: 155-186.

Morison, R. S. 1945. Walter Bradford Cannon in International Affairs.

Walter Bradford Cannon 1871-1945: A Memorial Exercise. Held at

Harvard Medical School, Monday. Nov. 5. 1945.

Morison, R. S. 1979. Albert and Ellen Grass — An Appreciation. Neu-
roscience Newsletter, Sept. 1979, written 8/1/79.
Renshaw, B., A. Forbes, and B. R. Morison. 1940. Activity of isocor-

tex and hippocampus, electrical studies with microelectrodes. J. Neu-

mplnsiol. 3: 74-105.
Rosenblueth, A., and R. S. Morison. 1934. A quantitative study of the

production of sympathin. Am. J. Phvsinl. 109: 209-220.
Rosenblueth, A., and F. A. Simeone. 1934. The interrelations of vagal

and accelerator effects on the cardiac rate. Am. J. Physiol. 110: 42-55.
Rosenblueth. A., and F. A. Simeone. 1938a. The responses of the

superior cervical ganglion to single and repetitive activation. Am. J.

Physiol. 122: 688-707.
Rosenblueth, A., and F. A. Simeone. 1938b. The action of eserine or

prostigmin on the superior cervical ganglion. Am. J. Physio/. 122:

708-721.
Rosenblueth, A., D. B. Lindsley, and R. S. Morison. 1936. A study of

some decurarizing substances. Am. J. Physio/. 115: 53-68.
Rosenblueth, A., J. Garcia Ramos, and R. Miledi. 1954. The propa-
gation of impulses of myelinated axons. J. Cell. Comp. Phvsiol. 43:

347-364.

Seyfarth, E.-A. 1996. Ernst Theodor von Briicke ( 1800-1941 ) and Al-
exander Forbes ( 1882-1965): Chronicle of a transatlantic friendship in

difficult times. Perfect. Bin/. Med. 40: 45-54.
Simeone, F. A., W. B. Cannon, and A. Rosenblueth. 1938. The sen-

sitization of the superior cervical ganglion to nerve impulses by partial

denervation. Am. J. Physio/. 122: 94-100.
Taylor, E. W. 1931. William Norton Billiard. M.D. 1853-1931. Arch.

Neural. Pxyi hiatn 26: 179-183.
White, B. V. 1984. Stanley Cahh: A Builder of the Modem Neuro-

sciences. Francis A. Countway Library of Medicine, Boston. 445 pp.
/ottoli, S. J. 1990. Taped interview with Ricardo Miledi. July, 1990.

Marine Biological Laboratorv. Woods Hole, MA.

Reports of Papers Presented at

THE GENERAL SCIENTIFIC MEETINGS

OF THE MARINE BIOLOGICAL LABORATORY,

Woods Hole, Massachusetts
13 to 14 August 2001

Program Chairs:

WILLIAM ECKBERG, Howard University

ROBERT GOULD, New York State Institute for Basic Research

ROBERT PAUL MALCHOW, University of Illinois at Chicago

IVAN VALIELA, Boston University Marine Program

Each of these reports was reviewed by two members of a special editorial board
drawn from the research community of Woods Hole, Massachusetts.

Reviewers included scientists from

THE MARINE BIOLOGICAL LABORATORY,

THE WOODS HOLE OCEANOGRAPHIC INSTITUTION,

AND THE NATIONAL MARINE FISHERIES SERVICE.

SHORT REPORTS FROM THE 2001 GENERAL SCIENTIFIC MEETINGS
OF THE MARINE BIOLOGICAL LABORATORY

/•'; \ 1 1 i;i i> REPORT

The Editors

Introduction Ui (he featured report, green fluores-
cent protein: enhanced optical signals 1mm native

cnstals 231

Inoue, Shinya, and Makoto Goda

Fluorescence polarization ratio of GFP crystals 231

(.111 BIOLOGY

Knudson. Robert A., Shinya Inoue, and Makoto Goda

Centrifuge polarizing microscope with dual speci-
men chambers and injection ports

Tran, P. T., and Fred Chang

Transmitted light fluorescence microscopy revisited. . . .

Hernandez, R. V., J. M. Garza, M. E. Graves,

J. L. Martinez, Jr., and R. G. LeBaron

The process of reducing CA1 long-term potentiation
h\ the integrin binding peptide, GRGDSP, occurs
within the first few minutes following theta-burst
stimulation

Kuhns, William J., Dario Riisciano, Jane Kaltenbach,

Michael Ho. Max Burger, and Xavier Femandez-Busquets
Up-regiilation of integrins a, j3, in sulfate-starved ma-
rine sponge cells: functional correlates

Brown, Jeremiah R., Kyle R. Simonetta, Leslie A. Sandberg,

Phillip Stafford, and George M. Langford

Recombinant globular tail fragment of mvosin-Y blocks
vesicle transport in squid nerve cell extracts

Wollert, Torsten, Ana S. DePina, Leslie A. Sandberg,

and George M. Langford

Reconstitution of active pseudo-contractile rings anil
myosin-II-mediated vesicle transport in extracts of
clam oocytes

Clay, John R., and Alan M. Kuzirian

A novel, kinesin-rich preparation derived from squid
giant axons

Weidner. Earl

Microsporidian spore sporoplasm clvnactin in Sjnu-
gitea

Conrad, Mara L., R. L. Pardy, and Peter B. Armstrong
Response of the blood cell of the American IK use-
shoe crab. l.inntln\ polyphemus, to a lipopolysaccha-
ride-like molecule from the green alga ('.hlinilln . . . .

236

240

241

24(i

Silver, Robert

LtB4 evokes tin- < alcium signal that initiates nuclear
envelope breakdown through a multi-enzyme net-
work in sand dollar (Ediniann -iiiu.\ jiiinun) cells .... 24.S

DKYLI OTMIMAL BIOLOGY

Crawford. Karen

Ooplasm segregation in the squid embryo, Loligu
/H-nli'ii 251

Burbach, J. Peter H.. Anita J. C. G. M. Hellemons.

Marco Hoekman, Philip Grant, and Harish C. Pant
The stellate ganglion of the squid Loligo pealeii as a
model for neuronal development: expression of a
POU Class VI homeodomain gene, Rpf-1 252

Link, Brian A.

Evidence for directed mitotic cleavage plane reorien-
tations during retinal development within the ze-
brafish ' 254

Smith, Ryan, Emma Kavanagh, Hilary G. Morrison,

and Robert M. Gould

Messenger RXAs located in spins dogfish oligoden-
drocyte processes 255

Hill, Susan D., and Barbara C. Boyer

Phalloidin labeling of developing muscle in embryos

of the polychaete ('.n/nli-lln sp. 1 257

Rice, Aaron N., David S. Portnoy, Ingrid M. Kaatz,

and Phillip S. Lobel

Differentiation of pharyngeal muscles on the basis of
enzyme activities in the cichlid 'rnnnilirlirnniit intenne-
dius . 258

\l I i;nl;liil IK.}

Twig, Gilad, Sung-Kwon Jung, Mark A. Messerli,
Peter J. S. Smith, and Orian S. Shirihai

Real-time detection of reacti\e o\\gen intermediates
from single microglial cells 261

Silver, Robert B., Mahlon E. Kriebel, Bruce Keller,

and George D. Pappas

Porocytosis: Quantal synaptic secretion of nenro-
transmitter .11 llir neuromuscular junction through
arrayed vessicles 263

Chappell, Richard L., and Stephen Redenti

Endogenous zinc as a neuromodulator in vertebrate
retina: evidence from the retinal slice 265

230

REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

Shashar, Nadav, Douglas Borst, Seth A. Ament,
William M. Saidel, Roxanna M. Smolowitz,
and Roger T. Hanlon

Polarization reflecting iridophores in the arms of the
squid Lo/igo pealeii

Chiao, Chuan-Chiii. and Roger T. Hanlon

Cuttlefish cue visually on area — not shape or aspect
ratio — of light objects in the substrate to produce
disruptive body patterns for camouflage

Errigo, M., C. McGuiness, S. Meadors, B. Mittniann,

F. Dodge, and R. Barlow

Visually guided behavior of juvenile horseshoe crabs . . .

Meadors, S., C. McGuiness, F. A. Dodge,

and R. B. Barlow

Growth, visual field, and resolution in the juvenile
1. iniulu.^ lateral eve

Kozlowski, Corinne, Kara Yopak, Rainer Voigt,

and Jelle Atema

An initial studv on the effects of signal intermittency
on the odor plume tracking behavior of the Ameri-
can lobster, Hniti/ini.^ rimmcn>nt\

Hall, Benjamin, and Kerry Delaney

Cholinergic modulation of odor-evoked oscillations
in the frog olfactory bulb

Zottoli, S. J., D. E. W. Arnolds, N. O. Asamoah,

C. Chevez, S. N. Fuller, N. A. Hiza, J. E. Nierman,

and L. A. Taboada

Dye coupling evidence for gap junctions between
supramedullary/ dorsal neurons of the dinner, Tau-
logolabrns arlspersus

Kaatz, Ingrid M., and Phillip S. Lobel

A comparison of sounds recorded from a catfish
(Orinocodoras t'/grinntiinu, Doradidae) in an aquarium
and in the field

Fay, R. R., and P. L. Edds-Walton

Bimodal units in the torus semicircularis units of the
toadfish (O/MCI/NM Inn)

MARICULTURE

Mensinger, Allen F., Katherine A. Stephenson,
Sarah L. Pollema, Hazel E. Richmond, Nichole Price,
and Roger T. Hanlon

Mariculture of the toadfish OpMinn\ tun

267

271

274

278

Rieder, Leila E., and Allen F. Mensinger

Strategies for increasing growth of juvenile toadfish. . . . 283
Chikarmane, Hemant M., Alan M. Kuzirian, Ian Carroll,
and Robbin Dengler

Development of genetically tagged bay scallops for
evaluation of seeding programs 285

ECOLOGY A\D POPULATION B/OJIH.]

Williams, Libby, G. Carl Noblitt IV, and
Robert Buchsbaum

The effects of salt marsh haying on benthic algal
biomass 287

Hinckley, Eve-Lyn S., Christopher Neill, Richard McHomey,

and Ann Lezberg

Dissolved nitrogen dynamics in groundwater under a
coastal Massachusetts forest 288

Hauxwell, Alyson M., Christopher Neill, Ivan Valiela,

and Kevin D. Kroeger

Small-scale heterogeneity of nitrogen concentrations
in groundwater at the seepage face of Edgartown
Great Pond 290

Novak, Melissa, Mark Lever, and Ivan Valiela

Top down vs. bottom-up controls of microphytobenthic
standing crop: role of mud snails and nitrogen supply
in the littoral of Waquoit Bay estuaries 292

Fila, Laurie, Ruth Herrold Carmichael, Andrea Shriver,

and Ivan Valiela

Stable N isotopic signatures in bay scallop tissue,
feces, and pseudofeces in Cape Code estuaries sub-
ject to different N loads 294

Grady, Sara P., Deborah Rutecki, Ruth Carmichael,

and Ivan Valiela

Age structure of the Pleasant Bay population of Crep-
idula fomicata: a possible tool for estimating horse-
shoe crab age 296

Kuzirian, Alan M., Eleanor C. S. Terry,

Deanna L. Bechtel, and Patrick I. James

Hydrogen peroxide: an effective treatment for ballast
water

297

282

ORAL PRESENTATIONS

Published bv title only

300

Reference: Bio/. Bull. 201: 231. (October 2001)

Introduction to the Featured Report
Green Fluorescent Protein: Enhanced Optical Signals

from Native Crystals

"The bioluminescent jellyfish Aequorea emits 'green' light in vivo, whereas the pure photoprotein aequorin extracted
from the same organism emits 'blue' light on addition of Ca~+." Osamu Shimomura made this observation and
identified a green fluorescing molecule in 1962; then reported its purification and characterization in 1974 from 30,000
specimens of the hydrozoan jellyfish. The result was green fluorescent protein (GFP), which emits at about 509 nm
when it is excited by the blue light (about 460 nm) emitted by aequorin (also purified and characterized by Shimomura).
In the jellyfish, this process — called fluorescence resonance energy transfer (FRET) — results in a signal that, because
of its longer wavelength, can penetrate farther through the turbidity of natural seawater to its target, which might be.
for example, planktonic prey.

The molecular details of GFP emerged about 20 years later (1996) from a pair of independent studies. The
laboratories of Roger Tsien and George Phillips, Jr.. showed the protein to be an unusual, very regular,
barrel-shaped molecule, with its walls (a sheet comprising 11 /3-strands) and caps at both ends of the barrel
enclosing and protecting a fluorophore composed of post-translationally modified amino acids.

The gene encoding GFP was cloned by Douglas Prasher and associates in 1992. And shortly thereafter (1994).
Martin Chalfie and his laboratory showed that the protein, with its fluorophore, could be completely expressed in
bacteria, which would (as if they were jellyfish) glow green when excited with blue light. In the same year. Tulle
Hazelrigg demonstrated that a suitable gene construct would express a fusion protein including GFP, and that the
site of expression could be precisely located in the organism (Drosophila in this case), or in a single cell, merely
by illumination with blue light. With that critical finding. GFP was quickly recognized, and widely used in
developmental, cell, neural, and molecular biology, as a reporter of gene expression and a marker for gene product
localization.

Recently. Osamu Shimomura asked Shinya Inoue to produce a photomicrograph of the fluorescence emitted by
the needle-shaped crystals of purified, native GFP. Inoue agreed, but thought to examine, as well, the anisotropic
properties of the crystals. The novel and surprising results of that investigation are set out in the following short
report by Inoue and Makoto Goda. In brief, the fluorescence from excited GFP crystals is polarized, with the
resonance vectors oriented parallel to the long axis of the crystals. Moreover, when the excitation is also polarized,
the fluorescence measured with an analyzer parallel to the crystal is very much higher (by 20-30 times!) than that
measured perpendicular to it.

These observations, combined with structural studies involving X-ray crystallography, should shed more light
on GFP function and help us improve our interpretation of FRET imaging. Moreover they suggest that, in
investigations where dynamic changes in the orientation of GFP-linked motor or contractile proteins are being
followed, the use of polarized light might well increase the sensitivity of the observations.

— The Editors
August 2001

Reference: Bio/. Bull. 201: 231-233. (October 2001)

Fluorescence Polarization of GFP Crystals

Shinya Inoue (Marine Biological Laboratoiy. Woods Hole. Massachusetts 02543) and Makoto Goda1

Green fluorescence protein (GFP), isolated from the jellyfish, shaped crystals, about 3-/j.m to less than I-/J.ITI wide and some 20-
Aequorea. purified by column chromatography. and dialyzed to 100-/j.m long (1). Examined with a polarizing microscope in
against distilled water to remove salts, forms elongated needle- visible light of wavelength greater than 450 nm, these crystals

show a very weak negative birefringence; i.e., their slow axis

i A|ST Tokyo Japan (larger refractive index) lies perpendicular to the long crystal axis.

231

232

REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

Figure 1. Crystals of purified native GFP. (A) Dichmism (anisotropic

absorbance) in visible light with crystals appearing green to light brown
with the long crystal axis oriented "parallel. " and pale blue to white with
the long crystal axis oriented "perpendicular, " to the polarizer. (In fact, a
polarizer and analyzer were used, off-crossed hy about five degrees, to
accentuate the weak visible light dichroism.) (B. C) Polarization-depen-
dent anisotropv of fluorescence excitation seen in the absence of an
analvzer. The polarizer E-vector in Panel B is oriented parallel to the
prominent crystal in the middle of the panel. In Panel C. it is oriented
perpendicular to the length of the Mime tr\\lul. Bar = 30 fj.tn.

They also show a weak, hut distinct, hlue-green dichroism (Fig.
1A).

Illuminated with blue light of less than 450-nm wavelength, the
same crystals show a very bright, green fluorescence when viewed
through a 527 ± 15-nm band-pass filter. Surprisingly, the bright-
ness of the fluorescence varied by a ratio of as much as 6: 1 when
the crystals were illuminated with polarized blue light and ob-

served in the absence of an analyzer (Fig. IB. 1C). The fluores-
cence was greatest when the long axis of the crystal lay parallel to
the transmission direction of the polarizer E-vector. In other
words, the absorption for the exciting light is six times greater with
its E-vector polarized parallel to the length of the crystal axis than
across. A similarly high ratio and orientation dependence was
observed when the crystals were observed with non-polarized
illumination but through an analyzer. In other words, the fluores-
cence emitted by crystals illuminated with non-polarized light is
again some six times greater for polarization parallel to the long
crystal axis. Between parallel polarizer and analyzer, the orienta-
tion-dependent fluorescence ratio becomes as high as 20: 1 to 30: 1 ,
apparently the product of the excitation and emission anisotropies.
These extremely high polarization ratios show that the resonance
vectors of the dichroic fluorophores are oriented parallel to the
long crystal axis, and that there is little loss of energy or alignment
during fluorescence excitation and emission. [Even the nucleotide
bases in oriented strands of B-form DNA show dichroic ratios of
only 4:1 over the wavelength range 240 to 380 nm (2. 3).]

According to detailed X-ray analyses (4, 5), the 1 1 beta sheets
that make up the barrel-shaped exterior of the GFP molecule are
arranged helically around the barrel axis, with the barrel length
somewhat greater than the diameter. The beta sheets lie at an angle
slightly less than 45 degrees to the barrel axis. Thus, in the
negatively birefringent GFP crystals, it is likely that the long axes
of the barrel-shaped GFP molecules lie more or less across the
length of the long crystal axis. In addition. X-ray data show that
the chromophore responsible for the fluorescence lies within the
beta barrel and is tilted approximately 60 degrees to the long axis
of the barrel. The fluorescence polarization that we observe
strongly indicates that the ftuorophore is arranged with its major
absorbing and emission resonance planes (dipoles) oriented paral-
lel to the long axis of the crystal. Combining the data on fluores-
cence polarization and X-ray analysis, we propose that the beta
barrels are regularly packed with the barrel axes tilted some 60
degrees to the length of the crystal, and possibly wound as con-
centric cylinders around the core of the needle-shaped crystal.

The very high degree of anisotropy for excitation and fluores-
cence of GFP, as well as the dramatic elevation of the orientation-
dependent fluorescence polarization ratio by observation between
parallel polars. suggest their potential use as indicators of the
orientation of molecules with which the GFP or related chro-
mophores are tightly bound. Our observations may also prove
important in using GFP and related compounds in the application
of FRET (fluorescence resonance energy transfer) and other mea-
surements of molecular distances and orientations, because the
interpretation of these measurements relies on the knowledge, or
assumptions, of the orientation-dependent polarizability of the
fluorophores. The observations may also be relevant in explaining
the efficient energy transfer between aequorin and GFP in the
light-emitting organ of the jellyfish itself.

We thank Dr. Osamu Shimomura (Marine Biological Labora-
tory. Woods Hole) for discussions and providing the pure native
GFP crystals. Dr. Kensal Van Holde (Oregon State University) for
discussions on an early version of this manuscript, and Dr. Yoshi-
nori Fujiyoshi (Kyoto University) for generous support of this
project.

FHATURF.D REPORT

233

Literature Cited

1 Morise, H.. O. Shimoiiiiira, F. H. Johnson, and J. \\inant. 1974.

Kinthciiinin 13: 265f>-2662.

2. Seeds, \V. K.. and M. H. F. \Vilkins. 1950. F'unitlay /Jnvn.v.v. Chem.
S,,< 9: 417-423.

3. Inoue, S., and H. Sato. 1964. Pp. 209-248 in Molecular Architecture
in Cell P/ivMo/d.ii.v. T. Hayashi. and A. G. Szenl-Gyorgyi. eds. Prentice-
Hall, Englewood Cliffs. NJ.

4 Ormo, M., A. B. Cubitt, K. Kallio, L. A. Gross, R. Y. Tsien, and S. J.
Remington. 1996. Science 213: 1392-1395.

5 Vang, F., L. G. Moss, and G. N. Phillips, Jr. 1996. Nature Biotech-
nut. 14: 124d-125l.

234 REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

Reference: Biol. Hull. 201: 234. (October 21)01)

Centrifuge Polarizing Microscope with Dual Specimen Chambers and Injection Ports

Robert A. Kmulson (Marine Biological Laboratory, Woods Hole, Massachusetts),
Slunva Inoue, and Makoto Goda]

We reported earlier on a centrifuge polarizing microscope (CPM)
that was designed for observing the weak birefringence of organelles
and fine structures in living cells as they became stratified and ori-
ented under centrifugal fields of up to 10.500 times Faith's gravita-
tional field. In this earlier model ( 1 ). one chamber A contained the
specimen under observation, while the contents of the opposed second
chamber B, which acted solely to balance the rotor, could not be
viewed. We have now improved the CPM so that either chamber can
be viewed and selected at the flick of a lever, within the duration ot
a video frame. In the CPM, an electronic timing circuit synchronizes
the firing of the light source laser precisely to the transit of the
specimen under the microscope (freezing the image to less than
0.5-/j.m specimen motion at up to 1 1 .700 rpm, regardless of the speed
of the 16-cm diameter rotor). The timing circuit, in turn, is triggered
by the signal from a photodiode that picks up the light originating
from a stationary diode laser, and reflected by a small mirror (Ml)
mounted on the spinning rotor near its axis. The complexity of the
electronic timing circuit led us to keep the electronic circuit undis-
turbed and instead to devise an optical system for switching between
the display of the two chambers. To this end, we installed a second
timing mirror (M2) on the rotor, exactly opposite the one for chamber
A, but tilted up by a few degrees, rather than oriented parallel to the
rotor axis as is Ml. In front of the photodiode we also placed a
mounted pair of small mirrors on a "beam switcher" that could either
be flipped up out of the way so that the photodiode would capture the
light reflected from Ml. or flipped down into position so that light
reflected from the tilted mirror M2 would be reflected by the mounted
pair of beam-switcher mirrors and enter the photodiode. Thus, de-
pending on the position of the beam switcher, the timing light would
enter the photodiode, reflected either from mirror Ml or M2. The
timing circuit would then trigger the light source laser at precisely (to
within a few nanoseconds) the time point required to display a stable
image of the specimen in chambers A or B. The response time of the
electronic timing circuit and laser firing device turned out to be so
short that no video frames were lost in flipping the beam switcher and
capturing the images from either of the two chambers.

Figure 1 . left panel, shows the recorded image of sea urchin eggs
stratified in a density gradient in chamber A. while the right panel
shows density-standard beads (Nycomed Amersham, Oslo, Norway)
that reveal the gradient of the identically prepared seawater/Percoll
mix in chamber B. As the figure shows, the density of the unfertilized
Arbacia eggs is approximately 1.060. Immediately after fertilization,
the negative birefringence disappears from the membranes stacked in
the upper half of the clear zone of the stratified eggs (Fig. 1 and Ref.
2). Concurrently, the (de-jellied) egg becomes lighter over the next
minute, presumably by influx of water, and starts to float upward in
the gradient until its density is somewhat less than 1 .040.

In addition to being able to instantly switch between images from

AIST. Tokyo. Japan.

Figure 1. Left: Unfertilized Arbacia eggs stratified in chamber A in a
vim iitt-r/Perco/l density gradient. Ri^ht: Four sets of beads of standard
densities stratified concurrently in the gradient: obseircd in chamber B.
Nominal densities of the beads arc. top to bottom: 1.04, 1.06. 1.09, and 1.10.

chambers A and B. we have devised a method for introducing reagents,
sperm, etc.. into either chamber while the specimen and control are
rotating in the CPM. A plastic unit, notched out to pass the timing light
for mirrors Ml and M2, was placed over the support post for these
mirrors. The unit incorporated an "injection port" drilled along the rota-
tional axis of the rotor, which in turn could be connected through a
selection valve, without stopping the rotor, to either of the two thin pieces
of plastic tubing leading to chambers A or B. Thus, we can now observe
fine structural and density changes — e.g., in marine eggs upon activa-
tion— without having to stop the rotor for a period to remove the speci-
men chamber and introduce the activating agents. Video records resulting
from both of these improvements were presented.

We thank Hamamatsu Photonics K.K. and Olympus Optical Co.
for generous support of this project.

Literature Cited

I Inoue, S., R. Knudson, M. Coda. K. Suzuki, C. Nagano, N. Okada,
H. Takahashi, K. Ichie, M. lida, and K. Yamanaka. 2001. ./.
Microscopy 201: 341 -3Sh.

2. Inoue, S., M. Goda, and R. A. Knudson. 2001. ./. Microscopy 201:
357-367.

CELL BIOLOGY

235

Hiol. Hull. 201: 235-236. (October 2001)

Transmitted Light Fluorescence Microscopy Revisited

P. T. Tran and Fred Cluing (Colitinhiu University. Microbiology Department.
701 W. 168"' Street, New York. New York 10032)

From its introduction in 1967 by Ploem ( 1 ), reflected light fluores-
cence microscopy, commonly called "epi-fluorescence." has enjoyed
wide acceptance. Its optical path is relatively simple: full-spectrum
light passing through an excitation tiller is reflected by the dichro-
matic mirror into the objective lens to illuminate the sample; the
excited sample emits fluorescent light, which is re-collected by the
objective lens and passed through the emission filter to the camera.

The recent development of biosensors based on genetically encoded
variants of green fluorescent protein (GFP), coupled with advances in
digital, multi-modes, epi-fluorescence microscopy, has introduced
new powerful tools for observing protein dynamics and protein-
protein interactions at high spatial and temporal resolution within
living cells. However, there are some disadvantages inherent in epi-
fluorescence microscopy: a) mechanical switching of filter cubes to

Reflected
light (Epi)

Transmitted
light (Trans)

CFP

YFP

Q. ,_

c
CO

iiouu r-

3 2OOO-

.^ 1 5OO •

>, 1000'

'35

0 500 •

c
HI

I 1

i •

o J-

cp J2 w CD (n

-§ o a m m

(3 C5 0 "co

» ° S Q

Q- Q "" "§ 2

3 £ Q- £

OQ_ Li_ U

i~ CQ f^ ^

LL C3 n >-.

^^ n

S- ? ^ 0 Q-
Q. re ._ "-

a LL

5 LU 0

UJ > Q. O

<n Q.

•~ HI .-

c m

Figure 1. Intensely and image misalignment comparisons hefrveen reflected light fluorescence niicro\ci>py li'pi) cnul transmitted light fluorescence
microscopy (trans). (A) Schematic diagrams of the optical paths ofepi- and trans-fluorescence: a-excinitinn filler, b-dichromatic mirror, c-objective lens,
d-specimen. e-emission filter, f-condenser lens. (Bl Fluorescent image of a pollen grain. The rectangle C highlights a pollen spike that is used to measure
image misalignment. The dashed circle D represents the area of the pollen grain used to measure fluorescent intensity. (C) Positions of a pollen spike
(edge-enhanced) imaged with epi- and trans-fluorescence. (D) Mean fluorescence intensities of pollen grain measured in1;/; epi- and trans-fluorescence. See
te.\t for further details of microscopy set-up. Bar = W /LI;;;.

236

REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

view different color fluorescence can cause misalignment of images; b)
multi-pass filter cubes can eliminate the misalignment problem, but may
attenuate the emission light; and e) the epi-fluorescent light source cannot
be used in combination with transmitted light techniques such as Phase
or differential interference contrast (DIC) microscopy.

We investigated the feasibility of using transmitted light fluores-
cence microscopy (referred to here as "trans-fluorescence") to over-
come the limitations of epi-fluorescence described above. In trans-
fluorescence, the specimen is excited by light passing through the
condenser lens, and the fluorescent emission is captured by the ob-
jective lens. This mode eliminates the need for dichromatic mirrors.
Trans-fluorescence was pioneered prior to epi-fluorescence but was
largely abandoned due to two major drawbacks. In the past, excitation
light coming from the condenser lens and going through the objective
lens could not be completely blocked by the emission filter, leading to
a high background signal. Furthermore, optical alignment and match-
ing of the condenser lens and the objective lens were often difficult,
leading hi non-optimal fluorescence. However, recent developments
in filter technology and automated microscope controls prompted us
to test whether trans-fluorescence could provide significant improve-
ment over epi-fluorescence in alignment and fluorescent intensity.

Figure 1A shows schematic diagrams for the epi- and trans-fluo-
rescence optical set-ups. We used a Nikon Eclipse 800 upright mi-
croscope with the following attachments. For epi-fluorescence. a first
excitation filter wheel (Sutler Instruments, model Lambda 10-2)
equipped with band pass excitation filters CFP (436/10 nm), GFP
(HQ 525/50 nm). and YFP (535/30 nm) (Chroma Technology) was
placed at the epi portal; a corresponding dichromatic mirror for each
color protein was placed in the filter cube holder, a Plan Apo 100X/
1.4 N.A. oil objective lens was used; and a second filter wheel
equipped with band pass emission filters CFP (470/30 nm), GFP (HQ
470/40 nm). and YFP (500/20 nm) was placed at the camera port. For
trans-fluorescence, minor modifications from the above set-up were
required. The excitation filter wheel with appropriate filters was
placed at the trans portal; a Universal 1.4 N.A. oil condenser lens was
used to illuminate the sample; and all dichromatic mirrors were
removed from the optical path. The same 100- watt mercury air lamp
light source was used for illumination in the epi and trans portal.
Images of a pollen grain (Carolina Biological) were captured by an
Orca-100 cooled CCD digital camera (Hamamatsu Photonics) con-
trolled by the software package OpenLab 3.0 (Improvision). Images
were captured at an exposure time of 100 ms.

To compare differences in signal intensity between epi- and trans-
fluorescence, we imaged a pollen grain with the GFP filter set using

the two modes. Figure 1 B shows the image of the measured pollen
grain. The average intensity values of the pollen grain (over area "D"),
in 5 separate measurements, were: 2056 ± 224 pollen/252 ± 23
background for the Epi GFP filter cube, 2 190 ± 22 1 pollen/257 ± 25
background for the Epi GFP filter wheels/dichromatic mirror set-up.
and 2062 ± 240 poilen/292 ± 39 background for the Trans GFP filter
set-up (Fig. ID). We conclude that the fluorescent signal intensities
obtained with the epi or trans methods are not different, although the
trans method produced slightly higher background intensity. The trans-
fiuorescent technique is a comparable alternative to epi-fluorescence.

To compare image alignment between epi-fluorescence and
trans-fluorescence, we imaged pollen grains sequentially with the
CFP and YFP filter sets, using the epi mode and then using the
trans mode. Switching between CFP and YFP filter cubes and
dichromatic mirrors was done manually, while switching between
CFP and YFP filters was automated via the filter wheels. Use of the
epi mode with manual switches produced a significant lateral shift
of up to 1 p.m between the CFP and YFP images (Fig. 1C). The epi
mode using automated filter switches with CFP-YFP dual-pass
mirror and the trans mode using automated filter switches alone
produced no measurable image shift (Fig. 1C). However, the epi
mode with CFP-YFP dual-pass mirror attenuated the fluorescent
light by 50% compared to the epi mode using filter cube and the
trans mode (Fig. ID). We conclude that the trans mode produces
better image alignment than modes using the manual switching of
filter cubes; it also produces higher fluorescent intensity compared
to the epi mode with a dual-pass mirror.

These studies suggest that transmitted light fluorescence micros-
copy may be an attractive alternative to reflected light fluorescence
microscopy. The advantages of trans-fluorescence include: a) image
alignment is better than modes using mechanical switching of filter
cubes; b) image fluorescence intensity is comparable to the best epi
set-up and is twice as bright as an epi mode using a dual-pass mirror
and filter wheels; and c) fluorescence can be combined with Phase or
DIC techniques using the same light source.

FC would like to thank the Nikon Corporation for the MBL
Nikon fellowship and for providing the microscope equipment
used in this analysis. PTT is indebted to Shinya Inoue, Rudolf
Oldenbourg, and Ted Salmon for their continued guidance.

Literature Cited

Ploem, J. S., and H. J. Tanke. 1987. ImnnlM-tion i<>
Mifi-ini-iifiy. Oxford University Press, Oxford.

Fluorescence

Reference: Biol. Hull. 201: 236-237. (October 2001)

The Process of Reducing CA1 Long-Term Potentiation by the Integrin Peptide, GRGDSP,
Occurs Within the First Few Minutes Following Theta-Burst Stimulation

R. V. Hernandez, J. M. Garza, M. E. Graves, J. L. Marline:., Jr., and R. G. Li-Baron (University of Texas,
Department of Biology ami the Ccijal Neuroseience Research Center, San Antonio, Texas 78249)

Theta-burst stimulation (TBS) induces Schaffer collateral-CAl
synaptic long-term potentiation (LTP; I. 2), an experimental
model of synaptic plasticity believed to reflect physiological pro-

cesses during normal learning and memory. Various adhesion
receptors may play a role in LTP (3), including integrins, trans-
memhrane signaling receptors that link extracellular ligands to the

CELL BIOLOGY

237

£ 140

£ 120

a 100

I 80

I 60

I 40

g, 20

™ 0

aCSF

GRGDSP

aCSF GRADSP GRGDSP

aCSF

GRGDSP

Figure 1. Hixtnxrumx xiimimiri;inx the effects of altered integrin-binding by GRGDSP on CA1 LTP. (A) Bath application of 250 u.M GRGDSP for 15
min. xtartiiiK at 10 mill pre-TBS and concluded at 5 min. xiihsruntHilly reduces LTP (solid bar). (B) When 25(1 fiM GRGDSP (solid bar) or GRADSP (shaded
bar) is applied for the 10 min before TBS delivery, with a 30-60 s pre-TBS wash-out, CA1 LTP is not reduced. (C) After robust CAI LTP is induced, a
-Ill-inin bath application of 500 juM GRGDSP, startini; at 5 min post-TBS, fails to reduce LTP. Open ham are aCSF controls.

actin cytoskeleton (4). A principal recognition signal for some
integrins is the tripeptide Arg-Gly-Asp (ROD), a sequence found
in various extracellular matrix and cell-surface proteins. Indeed,
integrin-binding to endogenous ligand is perturbed by the peptide
Gly-Arg-Gly-Asp-Ser-Pro (GRGDSP) (5). To assess the role of
integrins in LTP, the effect of GRGDSP was tested on the CAI
field excitatory post-synaptic potential (fEPSP) of the rat hip-
pocampus. In previous studies, we found that 250 /nM GRGDSP,
half the concentration reported by others ( 1 ), was sufficient to
significantly reduce LTP (2), even when applied for 15 min during
a period that included 10 min pre- to 5 min post-TBS (unpubl.
data). Also, application of 250 ju.M GRGDSP at 5 or 30 min
post-TBS had no effect on CAI LTP. These results raised ques-
tions about the time frame of integrin binding during the process of
LTP induction and expression. Current experiments, summarized
in graph form in Figure 1 , now suggest that a critical period of
integrin-binding activity necessary for LTP occurs within the first
few minutes following TBS.

Male Sprague-Dawley rats, 30-40 days old, were decapi-
tated and the brains quickly placed in cold artificial cerebral
spinal fluid (aCSF) consisting of the following (in mM): 124
NaCl. 2 KC1. 1.25 NaH2PO4. 26 NaHCO,. 1 MgCK, 2 CaCU
and 10 dextrose. The brains were cut in 500 /urn horizontal
sections, and the hippocampus was dissected away from sur-
rounding cortex. The isolated hippocampal tissue was incubated
at room temperature during a 2-h recovery period and then
placed on an interface perfusion chamber to record the field
potential response. Tissues were bathed in aCSF at a flow rate
of 2.0 ml/min and supervised with 95% O,/5% CO,. All pep-
tides were mixed in aCSF and bath applied through the perfu-
sion system. Pulled-glass electrodes, with an AgCl wire in-
serted and filled with 150 mM NaCl, served as the recording
electrodes. Temperature was maintained at 31-32°C throughout
the experiments. Test pulses were evoked every 20 s using a
concentric bipolar stimulating electrode.

When 250 p,A/ GRGDSP was applied — beginning 10 min pre-
TBS and concluded at 5 min post-TBS (see Fig. 1A, solid bar;
7.0 ± 3.89r. /; = 3) — the percent change from baseline of the
fEPSP slope, measured at 60 min post-TBS, was substantially
reduced when compared with artificial cerebral spinal fluid (aCSF)

controls (Fig. 1A, open bar: 37.8 ± 16.0%, n = 3). These
experiments replicate previous studies (2) and confirm peptide
activity. However, a 10-min application of 250 IJ.M GRGDSP,
with a 30-60 s wash-out immediately before TBS, did not reduce
LTP (see Fig. IB, solid bar; 48.3 ± 17.9%, /; = 3), as compared
with aCSF (Fig. IB, open bar; 28.2 ± 13.2%, n = 6), or 250 ^M
of the inactive peptide, GRADSP (Fig. IB, shaded bar; 43.3 ±
20.9%, n = 3; ANOVA, P > .05). Finally, to determine whether
a decrease in LTP by post-TBS application of GRGDSP may be
concentration-dependent within ranges previously tested (1, 2), a
40-min bath application of 500 ;u,A/ GRGDSP, beginning 5 min
post-TBS, was tested. This concentration also did not decrease
CAI LTP (Fig. 1C. solid bar; 84.7 ± 30.9%, n = 3) when
compared with aCSF controls (Fig. 1C, open bar; 58.0 ± 14.3%,
n = 7; t test, P > .05).

Based on these new data, we conclude that GRGDSP disrupts
LTP within the first few minutes after TBS, and hypothesize that
tetanic stimulation may initiate a process that modifies the avail-
ability of integrin to bind ligand. The integrin-binding peptide,
GRGDSP. is thought to decrease LTP by competing for integrin
binding sites in the extracellular matrix that recognize the ROD
motif; successful binding by the peptide then disrupts normal
integrin function during LTP expression and maintenance. The
data presented here, however, suggest that integrins may not be
available to bind GRGDSP before TBS, but are quickly and briefly
available after TBS.

Supported by a Specialized Neuroscience Research Projects
grant (NINDS NS39409; RGL/JLM) and the Ewing Halsell Foun-
dation (JLM).

Literature Cited

1. Staubli, U., D. Chun, and G. Lynch. 1998. J. Neurosci. 18: 3460-
3469.

2. LeBaron, R. G.. R. V. Hernandez, J. E. Orfila, and J. L. Martinez,
Jr. 1999. Soc. Neurosci. Ahxlr. 25: 1495.

3 Benson, D. L., L. M. Schnapp, L. Shapiro, and G. W. Huntley. 2000.

Trends Cell Bioi 10: 473-4X2.
4. Hynes, R. O. 1992. Cell 69: I 1-25.
5 Pierschbacher, M. D., and E. Ruoslahti. 1984. Nature 309: 30-

33.

238 REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

Reference: Biol. Bull. 201: 238-239. (October 2001)

Up-regulation of Integrins a3 |3, in Sulfate-Starved Marine Sponge Cells: Functional Correlates

William J. Knhiis1, Dario Rusciano2, Jane Kaltenbach3, Michael Ho1,

Max Burger, and Xavier Fernandez-Busquets
(Marine Biological Laboratory, Woods Hole, Massachusetts 02543)

Integrins are a large family of heterodimeric transmembrane
glycoproteins that attach cells to fibronectin and collagen and other
extracellular matrix proteins of the basement membrane. The at-
tachment is by way of recognition sequences — RGD in the case of
fibronectin. The transmembrane and cytoplasmic domains of inte-
grins provide a conduit for outside-in as well as inside-out signal-
ing ( 1 ). Integrins or their domains have been highly conserved over
many millions of years as judged from their presence in sponges
(2). Cell adhesion and motility are regulated by integrins, but the
pathways that modulate this function are unclear.

Our previous studies have shown that the properties of adhesion
and motility are lost when isolated sponge cells in rotation are
deprived of inorganic sulfate (3). We hypothesized that this kind of
stress would be likely to cause membrane alterations and would
therefore be a useful model for studying integrins and integrin-
ligand binding. Herein we describe the effects of sulfate starvation
upon the expression of a, and j3, integrins in Microciona pmlif-
era, a marine sponge.

Microciona cells were prepared from intact sponge as previ-
ously described (3). Aliquots of the cell suspension at a concen-
tration of 2 X 107/ml were placed in 50-ml centrifuge tubes and
spun at low speed in a table model centrifuge at 16°C. The cells
were resuspended in either sulfate-free artificial seawater (i.e., less
than 10 nM SO.f~). or in seawater with a normal sulfate concen-
tration (26 mM; +SO42~ ). Each flask was rotated for 8 h; the cells
were then centrifuged, the supernatant discarded, and the cell
pellets resuspended in fresh ASW, either normal sulfate or sulfate-
free. This cycle was repeated four more times.

SDS-PAGE analyses were carried out with lysates prepared by
Triton X100 extraction of normal and sulfate-deprived sponge-cell
pellets. The proteins separated by gel electrophoresis were probed
with rabbit antibodies prepared against integrins a?, a^, and /3,
following their electro-transfer to nitrocellulose (NC). The NC was
cut into seven lanes to account for reference standards (1 lane),
integrin staining of ( + SGy + ) lysate (3 lanes), and integrin stain-
ing of (-SO42 + ) lysate (3 lanes). The primary antibodies (rabbit
anti-integrins) were applied to their substrates for 1 h and removed,
and the NC strips washed with PBS. The secondary reagent (goat
anti-rabbit horseradish peroxidase (HRP) conjugate) was applied
for 1 h and removed, and the NC was washed again with PBS.
Color was developed with ECL reagent at a dilution of 1 : 1 applied
to the NC strips, which were then autoradiographed.

1 Hospital for Sick Children. 555 University Ave. Toronto. Ontario.
Canada M5G 1X8. (Author for correspondence.)

• Friednch Miescher Institute. Ch 4002, Basel, Switzerland.

3 Mount Holyoke College. Department of Biology, South Hadley, MA
01075.

J Faculty of Pharmacy. University of Barcelona, Barcelona, Spain

For immunohistochemistry, chemically dissociated cells con-
ditioned in normal or sulfate-free ASW were fixed in 10%
formalin ASW and the centrifuged pellets embedded in paraffin
as described (3). Tissues were sectioned at 6 jam and stained
using mouse monoclonal antibodies (MAB) raised against in-
tegrins a 3 and j3,; following a wash with PBS, the sections were
treated with the secondary antibody (goat anti-mouse HRP
conjugate), washed with PBS. and developed with 3', 3' dia-
minobenzidine. The sections were counterstained for 5 min with
Harris' hematoxylin.

The stained Western blots revealed marked differences be-
tween the integrins derived from Microciona cells prepared in
normal ASW and those processed in sulfate-free ASW (Fig. la).
The expression of a, and £!, integrins was considerably greater
in sulfate-free ASW than in normal ASW. The distinction was
particularly clear in the case of a3 integrin. which displayed a
single prominent band at about 65 kDa in the ( — ) lane, whereas
the ( + ) lane displayed a corresponding band at considerably
lower intensity. The differences were maintained in the lanes
stained with anti-integrin /3,. but the 65-kDa bands were much
more intense. The molecular sizes are somewhat less than those
reported in other sponge species (2, 4). The j3, integrin derived
from the sulfate-free lysate also displayed a very dense broad
band at 200-205 kDa. The multiplicity of bands could be
accounted for, either by cross-reactions between the anti-
integrins and non-integrin proteins, or by glycosylation variants
of the primary bands. The anti-integrin as failed to react with
either cell lysate.

Cell sections stained with integrin MABs generally conformed
to the biochemical distinctions. The cells preconditioned in sulfate-
free ASW and stained with anti-integrin a, displayed many large
darkly stained cells along with a considerable amount of stained
matrix (Fig. Ib). In contrast, the normal counterpart showed cells
and matrix with far less staining intensity, and a more prominent
counterstain (Fig. Ic). Similar differences between cells precondi-
tioned in sulfate-free ASW and normal ASW were found in
sections stained with anti-integrin j3,, but the distinctions were
less pronounced than those observed in cells stained with anti-
integrin «3.

This study confirms those of other workers, and it indicates
that sponges — the oldest animal phylum with a multicellular
lineage — already had membrane structures that provide for
controlled reactions between cells, and for cell matrix reactions
(2. 4). The inverse correlation between integrin up-regulation
and sponge cell motility is of interest and has been described in
systems other than the sponge (5). In some cells, the expression
of integrins that bind fibronectin RGD is correlated with re-
duced cellular motility (6). This is important when considering
the possible relationships between Microciona aggregation fac-

CELL BIOLOGY

239

kDa
205-

116-
97-

67-

43-

30-

a 1

« t

' , 'V

tor (MAP), Microcionu cells, and the a3/3, integrins. MAP
proteins that are expressed from cDNA possess ROD binding
sequences (see ref. 7, p. 29,548. MAP p3/p4 form C). which
could potentially ligate MAP to the integrins to initiate a
trans-cellular motility-reduction signal, a strategy distinct from
the carbohydrate-carbohydrate binding thought to mediate ag-
gregation. Under usual conditions, the high sugar content of
MAP might preclude effective integrin-RGD peptide binding,
but this situation might change when cells are exposed to low
levels of environmental sulfate. The availability of pure MAP
ROD peptide sequences and of integrin peptides may allow for
binding correlations between these synthetic compounds and
their natural counterparts in normal and sulfate-stressed sponge.
The ability to manipulate sponges in viva by sulfate reduction
will provide a powerful tool toward a further understanding of
signaling pathways and their relationships to adhesion and
motility.

Literature Cited

1. Hynes, R. O. 1992. Cell 69: 1 1-25.

2. Pancer, Z., M. Kruse. I. Muller. and W. Muller. 1997. Mol. Biol.
Evol. 14: 391-398.

3. Kuhns, W., O. Popescu, M. Burger, and G. Misevic. 1995. J. Cell.
Biochem. 57: 71-89.

4. Wimmer, W., S. Perovic, M. Kruse, H. Schroder, A. Krasko, R.
Batel, and VV. Muller. 1999. Em: J. Biochem. 260: 156-165.

5. Christopher, R., and Jun-IJn Guan. 2000. Int. J. Mol. Med. 2000 5:
575-581.

6. Zhang, Z., A. Morla, K. Vuori, J. Dauer, R. Juliano, and E.
Ruoslahti. 1993. J. Cell Biol. 122: 235-242.

7. Fernandez-Busquets, X., D. Gerosa. D. Hess, and M. Burger. 1998.
/ Biol. Chem. 273: 29.545-29,553.

r w um i

Figure 1. In) Western blots demonstrating a, and JB, integrins. SDS-
PAGE was performed according to the Laemmli buffer systems on gel slahs
of 75 X 100 X 0. 75 mm. at 125 V. with a Bio-Rad Protean II apparatus
at a gel concentration of 10%. The cell lysates were heated in loading
buffer at 95°C for 5 min. The we/Is were charged with 20 pi (100 pig)
protein. Reference standards 1 10 ij.1) were placed in a separate well: their
migration is shown on the left side of the gel. The separated protein', were
then electrotransferred to nitrocellulose and probed with anti-integrin
rabbit polyclonal antibodies from Bioline Diagnostic! srl (Milan. llal\l at
u dilution of 1-500. (b. c) Integrin inununohistochemistry: sections were
prepared as described in the text. Mouse monoclonal antibodies from
BD-Pharmingen were used at a dilution of 1-200. (b) Sulfate-deprived
cells. Most large cells treated with anti-integrin a: show intense orange-
brown staining: widespread moderate staining of the extracellular matrix

is also noted, (c) Cells preconditioned in normal ASW. Large cell staining
is generally less intense: most cells show lighter anti-integrin staining, or
they stain more prominently with the henuitoxylin counterstain. The extra-
cellular matrix is rrrv lightlv stained with anti-integrin a,..

240 REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

Reference: Biol. Bull. 201: 240-241. (October 2001 I

Kecombinant Globular Tail Fragment of Myosin-V Blocks Vesicle Transport

in Squid Nerve Cell Extracts

Jeremiah R. Brown, Kyle R. Simonetta, Leslie A. Sandberg, Phillip Stafford^, and George M. Longford
(Department of Biological Sciences. Dartmouth College, Hanover, New Hampshire 03755)

Myosin-V, a calmodulin-binding myosin motor, mediates the
movement of vesicles on cortical actin filaments in a variety of cell

1 Motorola. Inc. Chicago, IL.

types. This motor has been shown to transport ER and synaptic
vesicles in neurons, melanosomes in melanocytes. and secretory ves-
icles and the vacuole in yeast. Recent evidence ( 1 ) suggests that the
globular tail of myosin- V, which binds to the surface of vesicles (2, 3),

IP-GST-MyoV-Tail

Ah: aPI96 H2 GST

f -

12345

Vesicles

Ab: QLLQ H2

213 -

128 -

85 -

42.6 -
31.2-

Vesicles

Ab:

QLLQ

GST

2I3"

128-

„,.

42.6-
31.2-

128 -

85 -
42.6 -
31.2-

0.25 05 0.5

GST-MyoV-Tail Trials (mg/ml)

0 0.25 0.5

GST-MyoV-Tail Concentration
(mg/ml)

Figure 1. lA) Western him analyses of immunoprecipitation IIP) experiments using the GST-MyoV-tail. Clarified squid optic lobe homogenate was Triton
X- 100 extracted, incubated with the reeombinant tail fragment for 2 h at 4°C, and recovered using a-GST. (Lime 1) Squid my osin-V enriched fraction (S5B) probed
with a-PIVf>. a polvdonal squid m\osin-\ antibo<l\ I tlenotes hand *'l interest). I Lane 2) IP-GST-MyoV-tail probed with a-P!V6. I Lane 3) IP-GST-MyoV-tail
prohed with a-H2, a monoclonal antibods to the squid kinesin. (Lane 4) IP-GST-MyoV-tail probed with a-GST antibody. (Lane 5) Purified GST-MyoV-tail probed
with a-GST. (Bl Western blot analyses nf sucrose vesicle fractions obtained b\ ninniiix clarified squid brain hoinogeiuile on a sucrose density gradient (0.3/0.6/1.5
M gradient; reside fraction kif.cn from ILVII.ft M interface). (Lane 1 ) Vesicle traction probed with a-QLLQ. a polyclonal antibody to the squid imosin-V tail. (Lane
2) Vesicle fraction probed with a-H2. (C) Western blol analvses of reside fraction incubation GST-MvoV-tail. The reside fraction, prepared by resuspending
vesit nlar pellet from clarified squid brain homogenate. was incubated for 2 h at 4 C with the GST-MyoV-tail. The control iru.v incubated only with buffer. The
supernatant and \-esicle pellet were analy:ed by western blol analyses. (Lane It Control supernatant probed with a-QLLQ. (Lane 2) GST-MyoV-tail incubation
supernatant probed with a-QLLQ. (Lane 3) Control pellet probed with a-QLLQ. (Lane 4) GST-myosin-V tail incubation prohed with a-QLLQ. (Lane 5)
GST-MyoV-tail incubation pellet probed with a-GST. (Lane f>) Purified GST-MyoV-tail probed with a-GST. (D) GST-MyoV-tail inhibition experiments.
GST-M\o\'-iail was added to squid giant a\on extracts at rime -era. GST was added to the control. Vesicles moving/fieUI/minute (motile activity) was measured
for ihc control at 15 min. Motile activin was measured for the GST-MvoV-lai/ at 45 min. Each measured for concentrations of 0.25 mg/ml and 0.5 mg/ml. (E)
The monle aclmlv at cadi GST-MyoV-tail concentration compared with the control is plotted us percent f'rl inhibition. Percent inhibition determined by
coiuparinij 15-mimite control time point with the 45-niimite experimental time point

CELL BIOLOGY

241

interacts with the microtubule-based motor, kinesin. to form a "het-
ero-motor" complex on vesicles. The complex of these two motors,
one microtubule-based and the other actin-based, is thought to facil-
itate the movement of vesicular cargo from microtubules to actin
filaments. Based on our studies of vesicle transport by these two
motors in extracts of squid neurons (4). we hypothesize that one of the
functions of the tail-tail interaction is to provide feedback between the
two proteins to allow a seamless transition of vesicles from microtu-
bules to actin filaments.

To study the interactions of the globular tail domain of myo-
sin-V to kinesin and to neuronal vesicles, we used a glutathionc
S-transferase (GST)-tagged globular tail fragment in motility and
vesicle-binding experiments. The plasmid for the recombinant tail
fragment of mouse myosin-V was provided by Dr. Huang ( 1 ). The
plasmid contained the GST-labeled mouse AF6/cno tail-globular-
domain (GST-MyoV-tail [aa!569 to aa!768] without the coiled
medial tail domain). After expression in E. culi, the GST-tagged
fragment was purified on affinity columns and used in experiments
with squid brain extracts and purified vesicles.

The GST-MyoV-tail fragment was identified on blots with a GST
antibody (Fig. 1A, lanes 4 and 5; 1C, lanes 5 and 6). To determine
whether the GST-MyoV-tail fragment binds squid brain kinesin, squid
brain extracts were incubated with GST-MyoV-tail fragments for 2 h
at 4°C. and then the GST-labeled fragment was immunoprecipitated
with the GST antibody. Blots of the proteins isolated by this immu-
noprecipitation (IP) showed a kinesin band when probed with the H2
antibody to squid brain kinesin (Fig. 1A, lane 3), establishing that the
recombinant mouse myosin-V-tail pulled down native squid kinesin.

The bacterially expressed recombinant globular tail domain of
myosin-V was incubated with purified squid brain vesicles to replace
native myosin-V. Vesicles were purified by sucrose density gradient
from clarified homogenates of squid brain. Vesicle fractions were
examined by DIC and fluorescence microscopy after staining with
DIOC6, a green fluorescent membrane dye. Overlay of the DIC and
fluorescent images demonstrated that the particles observed in the
DIC image were membrane structures. The vesicle fractions were
analyzed by SDS-PAGE and western blots, and both myosin-V and
kinesin were present on these vesicles (Fig. IB. lanes 1 and 2). A
similar vesicle fraction was incubated for 2 h at 4°C with the GST-
MyoV-tail fragment. After incubation, blots of the vesicle pellet
showed a band for the GST-tagged fragment, indicating binding of the
tail domain to vesicles (Fig. 1C, lane 5). Blots of the supematants.

after the vesicles were pelleted, showed a higher concentration of
myosin-V in the GST-MyoV-tail incubation than in the control incu-
bation, indicating displacement of native myosin-V from the vesicles
by the recombinant tail (Fig. 1C. lanes 1 and 2).

The recombinant fragment of myosin-V was used in motility as-
says to determine whether it had replaced native myosin-V on axo-
plasmic vesicles and blocked transport. The GST-MyoV-tail fragment
(0.25 and 0.5 mg/ml) was added to axoplasm in the presence of 5 mM
ATP, and the sample was warmed to 24°C. Purified GST was used as
a control. Actin-based vesicle transport was quantified by counting the
number of vesicles moving/field/min (v/f/m, motile activity) at 2 time
points during a 1 -h incubation. Motile activity for the 0.25 mg/ml trial
decreased from 17.5 ± 5.5 to 2.6 ± 1.3 v/f/m, and for the 0.5 mg/ml
trial, from 16.7 ± 1.7 to 1.5 ± 0.5 v/f/m (Fig. ID). Therefore, the
MyoV tail fragment inhibited vesicle transport by 81% and 91%,
respectively, and thereby exhibited a dominant negative effect in these
functional assays (Fig. IE). These data show that the recomhinant
protein blocked the activity of native myosin-V presumably by bind-
ing to vesicles and competing away the native myosin-V motors.

The GST-MyoV-tail fragment pulled down kinesin by immuno-
precipitation from squid brain homogenates, and it therefore exhibited
binding properties of native myosin-V. The GST-MyoV-tail fragment
blocked vesicle transport in extracts of the squid giant axon. These
data show that the headless myosin-V fragment is an effective inhib-
itor of vesicle transport in cell extracts and can be used to determine
the mechanism of motor recruitment to vesicles. These studies support
the hypothesis that tail-tail interactions may be a mechanism for
feedback between myosin-V and kinesin. allowing transition of ves-
icles from microtubules to actin filaments.

Supported by NSF grant MCB9974709 and MBL Josiah Macy
Fellowship.

Literature Cited

1 Huang, J. D., S. T. Brady, B. W. Richards, I). Stenolen, J. H. Resau.
N. G. Copeland, and N. A. Jenkins. 1999. Nature 397: 267 27(1

2. \Vu, X.. B. Bowers, K. Rao, Q. Wei. and ,j. A. Hammer III. 1998.

./. Cell Bin/. 143: 1X99-19 IS.

3. Reek-Peterson, S. L., P. J. Novick, and M. S. Mooseker. 1999. Mol.
Hint. Cell 10: 1001-1017.

4. Tabb, J. S., B. J. Molyneaux, D. L. Cohen, S. A. Kuznetsov, and
G. M. Langford. ./. Cell Sci. 111:3221 -3234

Reference: Binl. Bull. 201: 241-243. (October 2001)

Reconstitution of Active Pseudo-Contractile Rings and Myosin-II-Mediated Vesicle Transport

in Extracts of Clam Oocytes

Torsten Wollert1, Ana S. DePina, Leslie A. Sandberg, and George M. Langford
(Department of Biological Sciences. Dartmouth College, Hanover. New Hampshire 03755)

Cell division requires the construction of a contractile ring of
actin filaments attached to the plasma membrane at the site of
cleavage. Bipolar myosin-II filaments in the contractile ring gen-

Rostock University, Germany.

erate sliding of anti-parallel bundles of actin filaments, thereby
constricting the cell ( 1 ). A recent study of cell division in Dicivo-
.\ic/itiin (2) showed that myosin-II filaments are recruited to the
contractile ring cell by "cortical flow." The underlying mechanism
of cortical flow is not known. We inhibited the motor activity of
myosin-II in cell-free extracts of clam (Spisu/a solidissimd) oo-

242

REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

mg/ml

0.75 mg/mll

1 mg/ml

140 -

E lac-

_t_

's! 100 •

* 8,

c 80 •

'U 30 min

| 60 •

0) 40 •

nr^n i — i

>
0 •

1 II 1 n _

0 0.1 0.25 0.5 0.75 1

Antibody concentration (mg/ml)

too-

90-

? 70-

c 60-

~ SO-

|g 30 mln|

ll 40-

| 30-

20 •

10 •

o •

0.1 0.25 0.5 075 1

Antibody concentration {mg/ml}

Figure 1. Myosin-II-dependent sliding of actin filament bundles ami vesicle transport in clam oocyte extracts. (A) Successive AVEC-DIC images of a
\liding (tain bundle in a pseudo-contractile ri}ig. Tlte bundle contains 6 or more actin filaments, the limit of detection of video microscopy (5). The tip of
the bundle (asterisk} slides inward the lower left as depicted in each successive image captured at 6 s inten-als. Bar 10 /J./H. (B) Antibody inhibition
experiments. Inhibition of vesicle transport increased as ihe myosin-11-specific antibody concentration increased from O.I to 1.0 mg/ml. A non-specific
rabbit polyclonal (squid myosin-V antibody) was added to the control (labeled 0 mg/ml). (C) Vesicle transport was inhibited by 95% at an antibody
concentration of 1.0 mg/ml. (D) Fluorescence images of the 3-D-actin nenvork at 30 min for each concentration of antibody used. Interactive bundles
self-organized into a 3-D nenvork in the control and at ihc low concentrations of antibody. Network formation was blocked at 0.75 and 1.0 mg/ml.

cytes with a function-blocking myosin-II-specific antibody to in-
vestigate the mechanism of movement of myosin-II to the con-
tractile ring.

Cytoplasmic extracts were prepared from mature oocytes ob-
tained from gravid female clams (3, 4). The clarified extracts were
diluted 2-fold (protein concentration about 15 mg/ml) and adjusted
to pH 7.2. Nocodazole (50 jam) was added to the extracts to block
microtubule assembly, and an ATP regenerating system was added
to maintain ATP levels. The final extract was incubated for 45 min
at 18 °C to initiate transition into the meiotic phase of the cell
cycle. Rhodamine-phalloidin (0.5 j±M) was added to stain the
actin filaments, and the myosin II motor activity was monitored by
AVEC-DIC and fluorescence microscopy (5).

Actin bundles detectable by AVEC-DIC microscopy assembled
spontaneously in the meiotic phase extracts and formed interactive
three-dimensional networks or pseudo-contractile rings by a mech-
anism of self-organization (6). Two types of myosin-dependent
movement associated with the actin networks were observed. First,
overlapping bundles of actin filaments in the network were ob-
served to slide along each other. The advancing tips of sliding
bundles were tracked at speeds greater than 0.2 jum/s for more than

25 /urn before disappearing out of the field of view (Fig. 1A). A
similar sliding motion produced by bipolar myosin-II filaments is
generally accepted as the mechanism by which the contractile ring
constricts the cell during cytokinesis. Therefore, these self-orga-
nized actin networks or pseudo-contractile rings exhibited one of
the principal properties ascribed to the contractile ring.

The second type of motor activity observed in these extracts was
the movement of vesicles on actin filaments. In a given video field
that measured 25 /xrrr, more than 100 vesicles could be observed
moving simultaneously at an average speed of 1.0 /j,m/s (Fig. IB,
control). On occasion, ER-like networks moved on the actin fila-
ments, but most of the moving particles were individual vesicles
that were probably derived from ER during homogenizauon.

To demonstrate that both filament sliding and vesicle transport
were dependent on myosin-II, we performed antibody inhibition
experiments with a rabbit-polyclonal antibody raised to myosin-II
from clam oocytes. The Protein-A-purified, myosin-II-specific an-
tibody recognized a single band on immunoblots of the oocyte
extracts. Inhibition of vesicle transport was determined by com-
paring motile activity after antibody addition with motile activity
in controls. Motile activity was measured by counting the number

CELL BIOLOGY

243

til" moving vesicles per video field per minute (v/t'/m). The control
extracts showed high levels of motile activity (122 v/f/m) for
periods of 60 min or more. However, addition of the myosin-II-
specific polyclonal antibody caused a concentration-dependent in-
hibition of motile activity (Fig. IB). Vesicle transport was inhib-
ited by 95% (5 v/f/m) at an antibody concentration of 1 mg/ml
(Fig. 1C). The antibody inhibition experiments provided direct
evidence that vesicle transport was mediated by myosin-II.

At concentrations of 0.75 and 1.0 mg/ml. the myosin-II-specific
antibody inhibited the formation of the 3-D network (Fig. ID). With
fluorescence microscopy, pseudo-contractile rings of actin bundles
could be seen in the control extracts, but they were absent at these two
antibody concentrations (Fig. ID). The antibody did not inhibit the
assembly of actin filaments but blocked the association of filaments
into bundles: therefore, the myosin-II antibody blocked the self-
organising and bundle-sliding activities observed in the control ex-
tracts. The concentration of myosin-II in these extracts was estimated
to be in the range of 0.1-0.2 mg/ml, so an antibody concentration of
1 mg/ml was about 5-fold higher than the concentration of myosin.
The antibody is polyclonal. and only a subset of IgG molecules are
expected to bind at sites that block motor function; thus we judge the
antibody concentration required for inhibition of filament sliding and
vesicle transport to be within the expected range.

The generation of sliding forces between actin filaments is a
well-established activity of bipolar filaments of myosin-II. There-
fore, the observation that actin filaments self-organized and moved
in an anti-parallel fashion in these extracts fits current models of
the contractile ring. Sliding of actin filaments is assumed to occur
in intact contractile rings, but it has not been observed. These
studies provide a direct view of the sliding activity that occurs in
self-organized actin networks that mimic contractile rings. Self-
organized networks in cell-free extracts such as these may be the
only means available to observe myosin-II-mediated sliding of
actin bundles like those in the contractile ring.

The other novel observation in these experiments was the myosin-
II-dependent movement of vesicles. Myosin-II has not previously
been shown to be a vesicle motor. However, the movement of vesicles
to the contractile ring has been documented, and myosin-II is known
to move toward the equator by cortical flow (2). Myosin-II-mediated
vesicle transport on cortical actin filaments may provide a mechanism
by which myosin-II filaments arrive at the contractile ring. Such a
model is not consistent with several published reports. Yumura and
Uyeda (7), for example, demonstrated that myosin-II molecules that
lack ATPase activity are recruited to the equator. In addition, headless
myosin-II localizes to the equator (8, 9). These observations suggest
that myosin filaments are transported as passive passengers to the
actin cortex rather than through their own motor activity. Our studies,
on the other hand, provide some of the first evidence that myosin-II
binds specifically to vesicles and drives vesicle movement. The motor
activity of myosin-II may thus be another mechanism by which
bipolar myosin-II filaments are recruited to the cortex and to the
contractile ring.

Literature Cited

1. Sanger, J. M., and J. VV. Sanger. 2000. Microsc. Res. Tech. 49:
190-201.

2. Yumura, S. 2001. ./. Cell Iliol. 154: 137-145.

3. DePina, A. S., and G. M. Langford. 1999. Microsc. Res. Tech. 47:
93-106.

4. Sandberg, L.. P. Stafford, and G. M. Langford. 2000. Binl. Bull.
199: 202-203.

5. Langford, G. M. 2001. Methods Mol. Binl. 161: 31-43.

6. Surrey, T., F. Nedelec, S. Leihler, and E. Karsenti. 2001. Science
292: 1167-1171.

7. Yumura, S., and T. Q. P. Uyeda. 1997. Mol. Biol. Cell 8: 2089-2099.

8. Zang, J. H., and J. A. Spudich. 1998. Pmc. Nail. ACM!. Sci. USA 95:
13,652-13.657.

9. Naqvi, N. I., K. Eng, K. L. Gould, and M. K. Balasubramanian.
1999. EMBO J. 18: 854-862.

Reference: Hint. Bull. 201: 243-245. (October 2001)

A Novel, Kinesin-Rich Preparation Derived From Squid Giant Axons

John R. Clay* ami Alan M. Ku~irian (Marine Biological Laboratory, Woods Hole, Massachusetts 02543)

Almost 20 years ago, Robert Allen and colleagues (1, 2) ob-
served in squid giant axons a relatively large number of "submi-
croscopic" particles moving with velocities consistent with fast
axonal transport. These observations were made with video-
enhanced contrast-differential interference contrast microscopy
( 1 1. a methodology which had just been developed. The particles
were estimated to be 30-50 nm in diameter, and they were
proposed to be anatomical correlates of small vesicles apparent in
electron micrographs of Hodge and Adelman (3). We recently
published additional evidence in support of this view (4). More-
over, we demonstrated with immunocytochemistry that a small
fraction of these vesicles contain the delayed rectifier K ' channel.

1 National Institute of Neurological Disorders and Stroke. National In-
stitutes of Health. Bethesda. MD 20892.

This channel is also present in the axolemma, where it underlies
the repolarization phase of the nerve impulse ("action potential"; 4,
5). These vesicles appear not to be targeted to the axon terminals
since they do not contain synaptic vesicle proteins and are not
clathrin coated (4. 6). We have developed novel methodology for
isolating them from axoplasm (4). The initial step used in these
procedures is highlighted in this report.

The medial giant axons were dissected from squid provided by
the Marine Resources Center of the Marine Biological Laboratory,
and the axoplasm was extruded using standard techniques (7; Fig.
1 ). A small amount of buffer was added ( 1 ju,l per cm length of
axon) which contained 10 mM Na acetate, 10 mM HEPES (pH
7.2) with \M glucose, so that the osmolarity was 980 mOsm.
Similar results were obtained with a buffer consisting of 440 mM
K glutamate, 5 mM EGTA, and 10 mM HEPES (4). Axoplasm and
buffer were immediately placed in a small, thick-walled polycar-

244

REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

bonate centrifuge tube (0.2 ml fill volume, Beckman Instruments,
Inc.. Palo Alto, CA) and spun in an ultramicrocentrifuge (Sorvall
RC-M120GX) for 5-6 min at 35,000 gav (Fig. 1). A typical
preparation consisted of axoplasm pooled from 30 axons in lots of
six to minimize the time between dissection of the axons and the
centrifugation step. Centrifugation yielded approximately equal
volumes of translucent materials that we refer to as residual
axoplasm and clear supernatant (Fig. 1 ).

A standard Eppendorf laboratory centrifuge having maximal
centrifugal force of 14,000 gav was not sufficient to produce the
result described above. A force about 2 to 3 times larger was
required. Specifically, the result shown was obtained with a force
in the range of 25,000 to 40.000 g.iv. This step appears to be a
measure of the structural integrity of squid axoplasm. The result
illustrated in Figure 1 has not typically been obtained in August,
when squid viability is known to be poor — and, in our experience,
viability of the axoplasm is also poor. At those times the axoplasm
collapses into a small pellet (approximately 1% of the size of the
residual axoplasm illustrated in Fig. 1 ) with a supernatant volume
equal to that of residual axoplasm plus supernatant obtained at
other times of the year. A similar result was obtained when
chaotropic buffer, such as 0.5 M K iodide, was added to the
axoplasm.

The supernatant (Fig. 1) is rich in low-molecular-weight pro-
teins, as determined by SDS-PAGE with Coomassie blue staining
(Fig. 2, lane a). In particular, it contains tubulin and actin (Fig. 2;
arrows 2 and 3, respectively), which was confirmed by immuno-
blots (not shown) with anti-tubulin and anti-actin (Calbiochem, La
.lolla, CA). Neurofilament proteins were not detected in an ininiu-
noblot using the antibodies described by Grant and colleagues (8).
The supernatant also contains heat shock protein {Hsc70: [9];
arrow 1 in Fig. 2). as demonstrated with an immunoblot with
anti-Hsc70 (Stressgen. Victoria. BC. Canada). Of particular note is
the abundance of the microtubule-based motor protein kinesin in
the supernatant. This protein was not detectable with SDS-PAGE
either with Coomassie blue or silver staining, but was readily
apparent by immunoblot (lane b. Fig. 2: anti-kinesin [Chemicon.
Temecula. CA]). We estimate that 20%- 40% of the total kinesin
in axoplasm is contained in the supernatant of our preparation,
based on densitometer tracings of immunoblots of the supernatant
and residual axoplasm. Immunoblots and single-channel record-

axoplasm ) axon

35,000 X gav
6 min

< — supernatant

.< — residual
axoplasm

V J

V-->

Figure 1. Illustration of the cipcnincnlitl procedure used in tin*
study.

2
3

Figure 2. (a.) SDS-PAGE of the supernatant stained with Coomassie
him1. /A/rmr.v /. 2. and 3 correspond to heat shock protein (Hsc70: the
/uirc;- hand i'l the doublet h\ the arrow), tubulin. and actin. respectively.
(b.) Iiiiiiiuni'hlot of the supernatant with anti-kinesin. detected with en-
hanced cheiniluinineseence. The lines on the right are molecular weight
markers corresponding to 98. 64. 50. 36. and 16 kDa. tup to bottom,
respectively.

ings obtained by adding an aliquot of supernatant to one side of an
artificial lipid bilayer (4) demonstrated that the supernatant also
contains K ' channels. In addition to kinesin. all the other proteins
(Hsc70. neurofilaments. actin. and tubulin) except for K+ channels
were detected in residual axoplasm.

We believe that the kinesin in the supernatant is largely, or
perhaps entirely, bound to small vesicles such as those illustrated
in liaure 4 of our previous work (4). Kinesin has also been found
in soluble fractions of in vitro preparations (10). In those studies
the tissue was homogenized, a procedure which we avoided. We
handled the axoplasm gently until the spin at 35,000 g.iv. which we
believe strips small vesicles from microtubules and any other
elements to which they may be attached. The vesicles then float
into the supernatant because of their buoyancy. A centrifugal force
of 35.000 # „ is not nearly sufficient to bring them down into a
pellet (4).

Our technique provides a vesicle preparation (vesicles destined
for the axolemma) that is free of one of the major contaminants of
vesicle preparations: neurofilament proteins. The preparation also
contains a significant amount of actin and tubulin. which we
believe are not associated with the vesicles. A small fraction of the
heat shock protein is bound to the vesicles (unpubl. obs.K which
appears to be a key factor in further purification of the vesicles (4).
This preparation may be of interest to other investigators in the
cellular motility field.

We gratefully acknowledge Phil Grant for his gift of neurofila-
ment protein antibodies.

Literature Cited

I Allen, R. D.. ,). Metuzals, I. Tasaki, S. T. Brady, and S. P. Gilbert.
1982. Science 218: 1127-1129.

2. Brady, S. T., R. J. Lasek, and R. D. Allen. 1982. Science 218:
1129-1131.

3. Hodge, A. J.. and W. J. Adelman. 198(1. J. Ultnistnict. Res. 70:
220-241.

4 Clay. J. R., and A. M. Kuzirian. 2000. J. Neiirobiol. 45: 172-184.

5 Hodgkin, A. I... and A. F. Huxley. 1952. J. Phyxiol. lLond.t. 116:
449-472.

CELL BIOLOGY

245

(i dt> \\aegh. S.. and S. T. Brady. 1989. J. Mcnrosci. Res. 23: 433-
440.

7. Brown, A., and R. J. Lasek. 1990. Pp. 235-302 in .V</"<</ <n Exper-
imental Animals. D. L. Gilbert. W. J. Adelinan. Jr.. and J. M. Arnold,
eds.. Plenum Press. New York.

8. Grant, P., D. Tseng. R. M. Gould, H. Gainer, and H. C. Pant. 1995.

./. Cum/i. Neuroi 356: 311-326.
9 Tsai. M.-V., G. Murfini, G. Szebenyi, and S. T. Brady. 2000. Mo/.

Biol. Cell 11: 2161-2173.
10. Hollenbeck, P. J. 1989. ./. Cell Biol. 108: 2335-2342.

Reference: Biol. Bull. 201: 245-246. (October 2001)

Microsporidian Spore/Sporoplasm Dynactin in Spraguea

Earl Wculner (Biological Sciences, Louisiana State University, Baton Ront-e, Louisiana 70803)

Intracellular protistan parasites have evolved a diversity of ad-
aptations for survival and replication within host cell vacuoles.
Some of these adaptations require specific membrane-inserted or
surface-attached proteins on the vacuolar envelopes. However,
intracellular microsporidian parasites are not in vacuoles: rather,
they locate directly in contact with the host cell cytoplasm. This
position in the host cytoplasm may be partly due to their means of
entry through injection by an invasion tube. Within the microspo-
ridian genus Spraguea. the parasites are confined to neuronal liber
axoplusms in the central nervous system of anglertish. genus
Lophius. The supramedullary neurons are frequently parasitized by
Spraguea. and the colonies locate in the proximal regions of fibers
adjoining the nerve cell bodies. The supramedullary neurons send
fibers to the cutaneous areas in fish ( 1 ). Recent studies of puffer
fish, genus Takifuga, indicate that these fibers innervate the cuta-
neous mucous glands (2); this observation is supported by our
preliminary investigations on anglerfish. The mucous gland do-
mains of anglerfish skin are the most common sites where infective
Spraguea spores activate and discharge their sporoplasms. Mucus
is a major activator of Spraguea spores and is a primary factor in

effecting spore discharge (3). So the sporoplasms are likely to be
introduced directly into the nerve endings surrounding the cutane-
ous glands in anglerfish. However, established infections of Spra-
guea are always found at the proximal end of the supramedullary
nerve. Our laboratory has therefore hypothesized that Spraguea
sporoplasms are equipped with surface proteins that support sporo-
plasm transport up the fiber to the neuronal cell body area. This
study was therefore designed to probe for a dynactin-dynein as-
semblage because this motor is minus-end-directed and can effect
linkups between membrane and microtubules.

To search for dynactin-dynein motor molecules in Spraguea
sporoplasms, purified spores were incubated in 0.1 M HEPES
buffer (pH 7.0) and transferred to glass coverslips. The spores
were activated to discharge their sporoplasms by a method de-
scribed earlier (3). The discharged sporoplasms retain their attach-
ment to the glass during subsequent washing episodes that remove
unfired and discharged spores; the wash solution was 0.1% con-
canavalin A made up in 0.1 M HEPES (pH 7.0).

The isolated sporoplasms were subjected to optical probes and
Western blots using antibodies to dynactin peptides p15°glued_ p50

Figure I . Spraguea spp. sporoplasm with peroxiduse label directly against p1 -"'K'1"1''. Sporoplasms HIT exceedingly permeable and thus devoid of some
internal matrix: however, the sporoplasm surfaces wen . mi \i\ienil\ labeled with external peroxidase probe I A). IB) This image is a higher magnification
of A. (C) The sporoplasm is within an unh'reJ spore and the />cn>.\idiisc label is within the polaroplast farrow). (/)) This is an enlarged image of C. Bar
si ides represent 11.2? fj.ni.

246

REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

Arpl and intermediate dynein chains. The Western blots to Spra-
gitea sporoplasm proteins indicated that, whether whole sporo-
plasm samples were tested, or only the sporoplasm outer mem-
brane isolates, all four of these peptides were present.
Ultrastructural immunolabeling with an antibody-peroxidase probe
for pl5°£|ued showed the label binding uniformly over the sporo-
plasm surface (Fig. 1A, B). These results were also supported by
immunogold labeling and immunofluorescence (not shown). Since
I hypothesized that the dynactin assemblage should be associated
with the membrane within unfired spores, tests were made to
determine where the dynactin is located in such spores. Spores
were prefixed with 1% glutaraldehyde and subsequently partially
disrupted by the shearing action of a glass homogenizer. These
spores were then subjected to the p'-s°Elued antibody probe, and this
was visualized with a second antibody coupled to peroxidase. The
results showed that the label reacts within the polaroplast domain

(Fig. 1C. D). The position of the dynactin within the polaroplast
supports an old idea: that the spore discharges a tube through
which the membrane of the polaroplast everts to form the sac. The
cytoplasm and nucleus are thought to be introduced into the
everted polaroplast-derived sac. That the sporoplasm membrane is
a secondhand membrane derived from the polaroplast organelle is
supported by the absence of cholesterol in the outer envelope (3)
and the absence of lectin-binding molecules (unpubl. obs.).

Literature Cited

1. Funakoshi. K.. T. Abe, and R. Kishida. 1995. J. Camp. Neural.
358: 552-562.

2. Funakoshi, K., T. Kadota, Y. Atobe, M. Nakano, R. Goris, and R.
Kishida. 1998. Neiirasd. Let!. 258: 171-174.

3. Weidner, K., and A. Findley. 1999. Biol. Bull. 197: 270-271.

Reference: Biol. Bull. 201: 246-247. (October 2001)

Response of the Blood Cell of the American Horseshoe Crab, Limulm polyphemus,
to a Lipopolysaccharide-like Molecule from the Green Alga Chlorella

Mara L. Conrad (Department of Biology, Hunter College, New York, New York 10021),

R. L. Pardy1, and Peter B. Armstrong2

The granular amebocyte is the single cell type in the general
circulation of the horseshoe crab. Linniliis polyphemus. and func-
tions as the most important element in the immune system of the
animal. The cytoplasm of the cell is packed with granules contain-
ing multiple immune effector proteins and peptides ( 1 ). Degranu-
lation of the amebocyte, with the concomitant release of this
complex of antimicrobial effectors, can be elicited by specific
secretagogues such as bacterial lipopolysaccharide (LPS) (2). LPS
is an essential component of the cell wall of all gram-negative
bacteria and is an indicator molecule for the presence of these
bacteria. Although LPS had previously been thought to be unique
to gram-negative bacteria, a similar molecule has recently been
found in the eukaryotic green algae, Chlori'lla, strain NC64A,
maintained in bacteria-free culture (3). Algal LPS, like bacterial
LPS. is composed of lipid A. e-myristic acid, and 2-keto-3-deoxy-
D-manno-octulosonic acid (KDO). This material gelates the LJitiu-
Iits amebocyte lysate, a standard test for LPS (3). However, its
biological activities are essentially uncharacterized.

Does algal LPS operate as an agonist of exocytosis of the
granular amebocyte? Algal LPS was prepared as described
previously (3). To evaluate the ability of the granular amebo-
cyte to react to algal LPS. cultured horseshoe crab blood cells
were challenged with algal or bacterial LPS. and the extent of
exocytosis was evaluated by microscopic inspection (2). The
animal was chilled for 2 h at 4 °C and bled (cardiac puncture
with a 20-gauge needle) directly onto microscope coverslips

1 School of Biological Sciences, University of Nebraska, Lincoln. NE
68583.

~ Department of Molecular and Cellular Biology, University of Califor-
nia. Davis. CA 9561h.

submerged in cold LPS-free 3% NaCl. The cells were allowed
to attach to the coverslips for 5 min: then the coverslips were
assembled into perfusion chambers (with the cell-coated slips
supported above glass slides with chips of #l'/2 coverslips) as
described by Armstrong and Rickles (2). The chambers were
perfused with bacterial or algal LPS in 3% NaCl + 10 mM
CaCU, at room temperature, and observed over time either with
a Nikon inverted phase contrast microscope, or a Zeiss phase
contrast microscope equipped with a Nikon Coolpix digital
camera. Control cultures were perfused with LPS-free 3%
NaCl + 10 mM CaCl2. All glassware used was rendered LPS-
free by heating at 180-200 °C for at least 4 h. Alternately,
blood was collected directly into prechilled, virgin 35 x 10 mm
plastic petri dishes containing 2 ml of LPS-free 3% NaCl, 1
drop of blood per dish. After allowing 5 min for the cells to
attach to the dish, the culture was renewed with 2 ml fresh 3%
NaCl. Blood plasma accelerates the exocytosis response in the
absence of LPS (4). and its removal stabilizes the cells from
spontaneous degranulation. Three percent NaCl was replaced
with test solutions containing LPS in 3% NaCl + 10 mMCaCU.
Ca++ is required for cell flattening and a robust reaction to
endotoxin. As before, the extent of exocytosis was determined
by microscopic inspection.

The granular amebocyte responds positively to algal LPS
(degranulation occurs), but the cell requires a concentration
about II) times higher than for bacterial LPS (E. toli Serotype
()127:B8, Sigma Cat # L-3129). Bacterial LPS provokes a
vigorous exocytosis response at 5 /ag/ml, causing the majority
of the cells to degranulate within 20 min of exposure, whereas
the cultured blood cell requires 50 jug/ml of algal LPS to initiate
a similar response within that period (Fig. 1).

CELL BIOLOGY

247

120

[LPS| (ng/ml)

Figure 1. Stimulation of exocytmis of substratum-attached, cultured
Limulus amehoi MC\ h\ a/gal LPS. (At Prior to exposure, the cells are flattened on
the culture surface mid retain their secretory granule*, \\hieh are ohsen'ahle as the
small spherical vesicles that pack the cytoplasm surrounding the nucleus. <Bl
Thim min after exposure to 30 jj.g/ml algal LPS. all of the substrate-attached cells
have degranulated. Tlie flattened cells show large internal lacunae, which are the
sites of fusion of the membranes of the secretory granules with the plasma
membrane. Tile contracted shinv cells in {B} are situated above the focal plane of
the substratum-attached cells. Tliese are full\ granulated hlctod e~ells that have
separated from the culture surface aiul are now migrating on the surface of the
coagiilin clot fonned from materials secreted by the substrate-attached cells. Cells
that have lost their attachment to the solid culture sulntrutum arc refractory to
stimulation by bacterial LPS (2) and, as sliown here, to a/gal LPS. (C) Tlie
response to a/gal LPS occurs more rapidly at higher concentrations of the agonist
and requires about 10 times more LPS than does the response to bacterial LPS.

Limulus amebocytes generate eicosanoid metabolites (5). The
eicosanoids — oxygenated metabolites of the C20 polyunsaturated
fatty acids, 20:3n-6. 2():4n-6. and 20:5n-3 — operate as second
messengers in activating cellular defense reactions to bacterial
infection in insects. Exposure of blood cells to the eicosanoid
biosynthesis inhibitor naproxin. an inhibitor of cyclooxygenase.
abolished the LPS-induced nodulation response by the blood cells
of the beetle, Zopliohux (6). Consistent with a role for the eico-
sanoids in signaling in the amebocyte, naproxin (1 mM) inhibited
the exocytosis response of the cultured Limulus amebocyte both to
algal and to bacterial LPS.

The response of the granular amebocyte to algal LPS is inter-
esting, both for the characterization of a LPS-like molecule from
algae, and for a better understanding of the immune system of the
animal. In this latter regard, algal colonization and subsequent
erosion of the carapace of the horseshoe crab appears to be an
important cause of mortality of the adult animal (7), so it is of
interest to characterize the different ways by which the immune
system can interact with algae. Our documentation of the parallel
activities of algal and bacterial LPS in the induction of the exo-
cytotic response of the granular amebocyte indicates that both
molecules are capable of activating this important pathway of
immunity in Limulus. This is the first demonstration that the
LPS-like agent from an alga mediates LPS-like biological activi-
ties.

We thank Mr. Jim Barkes of Nikon for assistance with micros-
copy and Dr. Norman Wainwright and Ms. Alice Childs for
conducting LAL assays of algal LPS. This research was supported
by NSF Grant MB 26771 (to PBA) and a fellowship from the
Hunter College-Howard Hughes Medical Institute Undergraduate
Education Program (to MLC).

Literature Cited

1. Iwanaga, S., and S.-i. Kawabata. 1998. Front. Biosci. 3: d973-
d984.

2. Armstrong, P. B., and F. R. Rickles. 1982. Exp. Cell Res. 140:
15-21.

3. Royce, C. L., and R. L. Pardy. 1996. J. Endotoxin Res. 3: 437-
444.

4. Armstrong, P. B. 1980. J. Cell Sci. 44: 243-262.

5. MacPherson, J. C., J. G. Pavlovich, and R. S. Jacobs. 1996. Bio-
chint. Biophvs. Acta 1303(2): 127-136.

6. Bedick, J. C., R. L. Pardy, R. W. Howard, and D. W. Stanley. 2000.
J. Insect Physiol. 46: I48I-14S7.

7. Liebovitz, L., and G. A. Lewbart. 1987. Biol. Bull. 173: 430 (ab-
stract).

248 REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

Reference: Biol. Bull. 201: 248-250. (October 2001)

LtB4 Evokes the Calcium Signal That Initiates Nuclear Envelope Breakdown through a Multi-enzyme

Network in Sand Dollar (Echinaracnius parma) Cells

Robert B. Silver (Departments of Radiology, Pharmacology and Physiology, Wayne State University School

of Medicine, Detroit, Michigan; Decision and Information Sciences Division, Argonne National Laboratory,

Argonne, Illinois; and Marine Biological Laboratory. Woods Hole, Massachusetts)

In our ongoing study of cell division control, we have shown
that a calcium signal that is required for mitosis and that precedes
nuclear envelope breakdown (NEB) arises from perinuclear endo-
membrane vesicles which are derived from a subset of the endo-
plasmic reticulum (1-5). Analyses of the calcium signals indicate
that 1 ) this pre-NEB calcium signal occurs in microdomains; 2) the
signals emitted at individual microdomains are not coherent: and
3 ) neither calcium nor its agonist diffuses more than 1 p,m from the
site of agonist production and calcium release (4. 6, 7). Prompted
by those analyses, we posited and subsequently showed that leu-
kotriene B4 (LtB4) (8) evokes the release of calcium in vivo and in
vitro in a stereospecific fashion and functions as an important
intracellular signal (5, 9, 10).

To approach the mechanism by which the calcium signal is
controlled on the perinuclear vesicles, we have shown that 1 ) an
enzyme network, that includes PLA, and the 5-lipoxygenase path-
way enzymes, glutathione reductase, two forms of glutathione
S-transferase (including a presumptive leukotriene C4 synthase),
and enzymes of the oxidative pentose phosphate and glycolytic
pathways are also present on the calcium regulatory endomem-
branes of the prophase MA (11, 12); and 2) that phospholipase C
activity is absent from prophase MA, thereby excluding 1.4.5-
inositol trisphosphate as the agonist that evokes the pre-NEB
calcium signal (5. 12).

In this paper, we report that the perinuclear calcium-independent
PLA, activity occurs on perinuclear vesicles within the minute
before the pre-NEB calcium signal, and thus is present in the right
time and location to be associated with the pre-NEB calcium
signal. We have also found phosphofructokinase (PFK) activity on
the calcium regulatory vesicles isolated from native prophase
mitotic apparatus. Finally, we have developed a model that incor-
porates the measured kinetics of enzymes found on those vesicles.
This model tracks the production of numerous products, including
LtB4, and the emission of the calcium signal from single vesicles.
The timing of events seen in the model is consistent with the
deterministic nature of the pre-NEB calcium signal.

This study was conducted with eggs and cells from embryos in
the first and second cell cycle, and isolated native prophase MA.
Gametes were obtained from mature adult sand dollars (Echinar-
acnius panna) as previously described (3, 4. 13. 14). The temper-
ature of the gametes and embryos was maintained between 11
and !5rC throughout the experimental procedure. Quantitative
direct-pressure microinjection studies were performed as previ-
ously described (3. 4). Imaging of PLA: activity in vivo was
performed with the self-quenching PLA2 substrate PED6 (15) in
the following fashion. Unquenched derivative of PED6 was found
uniformly distributed in the cytoplasm of the injected cell. For
injection with substrate (i.e., 2-10 pi of 1-2 M-g/n-1 PED6). cells

were viewed with the multispectral video light microscope. Fol-
lowing microinjection, they were transferred to a Zeiss 510 con-
focal light microscope system. The cells tolerated this level of
injected PED6 and the inter-microscope transfer procedure. Spe-
cial care was taken to reduce the photonic load on the cells —
thereby reducing the effects of photobleaching and eliminating
phototoxicity — while maintaining high spatial and temporal reso-
lution. Cells were observed in Multi-Track mode of a Zeiss 510
confocal microscope configured for concurrent viewing of fluo-
rescein and rhodamine fluorescent emissions and differential in-
terference contrast of the specimen. With Multi-Track, we were
able to simultaneously assess PLA2 activity in the fluorescein
emission channel, the formation of lipofuscin due to photon-
induced lipid peroxidation of membranes in the rhodamine emis-
sion channel, and cell structure with differential interference con-
trast. Typically, images were recorded at a frequency of 1 image
per min (0.0167 Hz). Images were recorded, processed, and then
archived in TIFF format on CD-ROM.

Assays for phosphofructokinase activity were performed using
modifications of the methods described by Kemp (16, 17). Reac-
tion volumes were kept to 2 ml, activities of aldolase and glycer-
aldehyde-3-phosphate dehydrogenase were 10-fold less than those
used by Kemp (16, 17), temperature was maintained at 11-12 "C,
and spectrophotometry was performed with an HP 8453 diode
array spectrophotometer. Modeling was performed in WinSAAM
using Michaelis-Menten kinetics for each of the enzymes listed as
part of the network. The kinetics values (e.g.. Km and Vmax) were
obtained in this laboratory with endomembrane subfractions pre-
pared from isolated native prophase mitotic apparatus as first
developed in this laboratory (e.g., 12. 13).

//; vivo hydrolysis of PED6, a self-quenching fluorescent phos-
pholipid substrate for PLA,, was monitored by confocal light
microscopy imaging. The observed PLA-, activity was detected on
discrete perinuclear vesicles seven minutes prior to NEB in the
second cell cycle of sand dollar embryos. This activation, which
occurs in the same region as the pre-NEB calcium signal (1,3, 15 ),
just precedes the pre-NEB calcium signal that occurs 6 min before
NEB (4, 6, 7). This activation also occurs during the period when
NEB in these cells is inhibitable by antagonists of PLA; (e.g.,
7,7-dimethyl-5,8-eicosadienoic acid, bromoenol lactone). 5-li-
poxygenase and leukotriene A4 hydrolase (5. 10. 12). Hydrolysis
of PED6, on perinuclear vesicular membranes, was also observed
to precede NEB in the third and fourth cell cycles of those
embryos. Thus, the location of PLA-, activity, which generates the
initial precursor for LtB4. coincides with the pre-NEB calcium
signal (3-6).

Several enzyme activities have been observed on perinuclear
endomembranes of prophase mitotic apparatus (cf. Fig. 1A; 1-4).

CELL BIOLOGY

249

Panel A

Activation Pathways of Metabolic Network

Purines. Pyrimidines
Amino Acids

6-PG 6-PG DH *> Ru-5-P q PP Isom ^ R-S-P

G-6-P

[Transaldolase and Transketolase |
PFK (f> F-1,6-di P<JjAldolaselJ>G-3-P

ATP

Panel B

0 2 4 6 8 10 12

Time (microsecond steps)

Figure 1. I A i A schematic of the metabolic model for regulation of the calcium signal that occurs as u prerequisite for unclear envelope breakdown,
The en~\mcs noted have hccn identified b\ en~\mutic activitv, ami in some cases b\ immunoassays anil sensitivirv of the pre-NEB calcium signal, NEB and
mitosis to pharmacological antagonists (e.g.. 14. /M. tBi The first 21) /n\ ot output of the model for starting conditions that included I mM ATP and I inM
fructose-6-phosphate. This model assumes that for even LtB4 produced. 200 calcium ions are released from a vesicle lumen to the cytosol. Furthermore,
the model assumes no lag lime for mechanisms of LtB_, evoking calcium release from the vesicle lumen. Therefore, it appears that the rise in cytosolic
calcium occurs immediately upon the start of metabolism, and not because of the action of LtBj. Note the rapid rise and leveling of NADPH and later drop
in LtAj levels that are to be expected given their nature as precursor relationships in the production of LtBj as depicted in Fig. I A. above.

To better understand the spatial and temporal relationships among
these enzymes, a mathematical model was developed that permits
assessment of enzyme catalyzed metabolic flow amongst compart-
ments. The model incorporates the Michaelis-Menten kinetics we
have measured in the calcium regulatory1 endomembrane subtrac-
tion of prophase mitotic apparatus tor the phospholipase A2-5-
lipoxygenase pathway, glutathione reductase and glutathione-S-
transt'erase. and the oxidative pentose phosphate pathways. Other

enzyme activities that we have observed in prophase mitotic ap-
paratus (e.g., creatine kinase. myosin. and kinesin) were also
incorporated in the model. The model, at present, assumes that a
single copy of each enzyme in the reaction scheme (Fig. 1A) is
present on a single 100-nm spherical vesicle resident within a cube
0.5 nm on a side — a so-called "Unit Reaction Volume" based on
ultrastructural observations (e.g., 1. 18). The value of 0.5 /urn was
taken from the measured mean free spacing amongst the vesicles

250

REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

in the aster (1, 18). The model also assumes an immediate effect of
agonist (LtB4) upon calcium release from vesicles. The output of
this model (mentioned above) evokes a calcium signal detectable
with high sensitivity aequorin (e.g.. 4-6) within 250 /is of the start
of PLA, activity, assuming a single calcium ion is released per
LtB4 produced. When the model was modified to consider a ratio
of 100-200 calcium ions released per agonist produced, a calcium
signal detectable with the high sensitivity aequorin we use to
visualize the pre-NEB calcium signal (e.g.. 4. 5) is generated in
less than 6 p,s (Fig. IB).

To extend the hypothesis that an integrated metabolic network is
involved in production of LtB4, isolated endomembrane subfrac-
tions from prophase mitotic apparatus were assayed for phospho-
fructokinase (PFK) activity. The calcium regulatory endomem-
brane subtraction of the prophase mitotic apparatus exhibited
abundant PFK activity, but little or no PFK activity was detected
on the other three endomembrane density subfractions. Under
standard reaction conditions that included 1 mM ATP and con-
centrations of fructose-6-phosphate ranging from 0. 1 nM through
1 mM, the Km and Vmax of this PFK activity were determined to
be 1.38 /J.M and 46.0 /iAf/mg protein/min. respectively. This
relationship, of Vmax being less than Km, indicates that ATP
inhibits the PFK activity, which is consistent with behavior of
PFKs isolated from other sources (e.g., 16, 17). Taken together
with our earlier finding of hexose monophosphate isomerase and
aldolase activities on these membranes, it is now apparent that
glycolysis is one of four metabolic pathways that are present and
active on the calcium regulatory endomembranes.

The results presented in this paper reveal that the first enzymatic
step in LtB4 production occurs at the right place and time for the
pre-NEB calcium signal. In addition, we have found that PFK
activity is part of a network of enzymes on perinuclear vesicles
that regulate production of LtB4 as an agonist of the discrete
pre-NEB calcium signal (5, 12). Lastly, results of our model
support the concept that these networked enzymes act rapidly and
are probably clustered on the vesicles. This represents the first
report of the presence of pre-NEB. calcium-independent PLA,
activity on perinuclear endomembranes, the first demonstration of

PFK activity on perinuclear membranes, and the first report of a
model that incorporates the enzyme network that appears to reg-
ulate nuclear envelope breakdown through a calcium signal and
supports the hypothesis that LtB4 is the agonist for pre-NEB
calcium signal and cell activation (5).

Research grant support by NSF (MCB-99082680) and the Bur-
roughs Wellcome Fund (1002768) is gratefully acknowledged.
Thanks are extended to Drs. John R. Hummel. Howard Rasmussen.
and Raoul F. Reiser for their helpful comments and advice during
these studies, and to the reviewers for their helpful suggestions in
the refinement of the manuscript.

Literature Cited

1 Silver, R. B., R. IX Cole, and W. Z. Cande. 1980. Cell 19:
505-516.

2. Silver, R. B. 1986. Proc. Nail. Acad. Sci. USA 83: 4302-4306.

3. Silver, R. B. 1989. De\: Bio/. 131: 11-26

4. Silver, R. B. 1996. Cell Calcium 20: 161-179.
?. Silver, R. B. 1999. FASEB J. 13: S209-S215.

6. Silver, R. B.. A. P. Reeves, B. P. Kelley, and \V. J. Fripp. 1996.
Biol. Bull. 191: 278-279.

7. Llinas, R., M. Sugimori, and R. B. Silver. 1992. Science 256:
677-679.

8. Samuelsson. B. 1983. Science 220: 568-575.

9 Silver, R. B.. J. B. Oblak, G. S. Jeun, J. J. Sung, and T. C. Dutta.
1994. Biol. Bull 187: 242-244.

10. Silver, R. B. 1995. Biol. Bull. 189: 203-204.

11. Silver, R. B., L. A. King, and A. F. Wise. 1998. Biol. Bull. 195:
209-210.

12. Silver, R. B., and N. M. Deming. 1999. Biol. Bull, 197: 268-270.

13. Silver, R. B. 1986. Methods Enzymol. 134: 200-217.

14. Silver, R. B. 1997. Pp. 83.1-20 in Cells: A Laboratory Manna/.
CSHL Press. Cold Spring Harbor. NY.

15. Hendrickson, H. S., E. K. Hendrickson, I. D. Johnson, and S. A.
Farber. 1999. Anal. Biochem. 276: 27-35.

16. Kemp. R. G. 1975a. Methods En-ymol. 42: 67-71.

17. Kemp, R. G. 1975b. Methods En:\moi 42: 71-77.

18. Silver, R. B., M. S. Saft, A. R. Taylor, and R. D. Cole. 1983.
J. Biol. Cheni. 258: 13.287-13,291.

DIM I OI'MI NTAI HlOl.lXil

251

Reference: Bhl. Bull. 201: 251-252. (October 2001)

Ooplasm Segregation in the Squid Embryo, Loligo pealeii

Karen Crawford (Department of Biology, St. Mar\'s College of Maryland, St. Mary's City, Maryland 20686)

After t'ertili/ation. squid egg ooplasm streams toward the animal
pole to create a clear lens-shaped blastodisc cap where meroblastic
cleavage occurs (1). Exposing the embryo to cold (4°C) after
fertilization inhibits blastodisc formation (2). suggesting that mi-
crotubules may be associated with this ordered movement of
cytoplasm (3). Although microtubules have been correlated with
cytoplasmic movements that follow fertilization in amphibians (4).
ascidians (5). and annelids (6), these embryos do not form blasto-
disc caps and. unlike the squid, they undergo complete or holo-
blastic cleavage. Interestingly, ooplasmic segregation, blastodisc
cap formation, and meroblastic cleavage all take place in the
zebrafish embryo, but here microfilaments and not microtubules
have been shown to direct the segregation of ooplasm from the
yolk (7).

To clarify the role of the cytoskeleton during early development
in squid, embryos cultured at 20 °C in petri dishes lined with 0.2%
agarose (Sigma. Type III and filled with Millipore-tiltered seawa-
ter (MFSW), were treated 30 min after HI vitro fertilization (8) with
either 0.01-20 /^g/ml of the microfilament inhibitor cytochalasin D
(Sigma) (3 trials. 50 to 75 embryos per dish) or 0.5-10 /j,g/ml of
the microtubule inhibitor colchicine (Sigma) (6 trials, 50 to 75
embryos per dish). Stock solutions of each inhibitor were prepared
in dimethyl sulfoxide (DMSO) (Sigma), and DMSO (0.1%) was
therefore added to MFSW as a control. Embryos were observed for
at least 4 h after fertilization during blastodisc formation and after
incubation overnight at 17 °C.

Embryos cultured in MFSW (Fig. la) or MFSW and DMSO
(Fig. Id) formed normal 40-50 /xm thick blastodiscs by 4 h and
underwent normal cleavage and early development. Blastodisc cap
formation occurred in all embryos treated with 0.01. 0.05. O.I. 0.5,
1.0, 2.0. 4.0. and 10.0 jag/ml cytochalasin D; however, at all but
the lowest concentration of microfilament inhibitor, the entire
cortical yolk cell membrane appeared disrupted by the presence of
small and large vesicles of cytoplasm (Fig. Ic). This result was
similar to what had previously been reported in squid (9). Normal
yolk cell membranes were observed in 45% of the embryos treated
with 0.05 /ig/ml cytochalasin D, while all yolk membranes ap-
peared similar to the control in the 0.01 fig/ml group. It is impor-
tant to note that embryos from these different treatment groups
failed to undergo normal development. In contrast, colchicine
prevented ooplasm segregation and blastodisc formation in all
embryos cultured at concentrations of 10.0. 7.5. and 5.0 /xg/ml.
although the thin layer of cytoplasm present prior to fertilization
was maintained (Fig. lb). Embryos cultured in 2.5 /ng/ml all
formed thinner. <20 /im. blastodisc caps by 4 h but failed to retain
them after culture overnight. While blastodisc caps (30-35 /im)
formed in the cultures treated with 1.0 and 0.5 /ag/ml colchicine.
after overnight culture these caps were either lost or reduced to
small abnormal shaped discs or sacs of cytoplasm at the animal
pole. Normal cleavage and development were never observed in
any embryos treated with cytoskeletal inhibitors. Normal blasto-

disc cap formation and cleavage occurred in all embryos treated
with DMSO (0.1%) (Fig. Id).

These results suggest that a microtubule-associated mechanism
is responsible for ooplasm segregation and blastodisc formation in
the squid. Although microfilaments do not seem to be required for
ooplasm movement to the blastodisc, the disruption of the cortical
yolk cell membrane by cytochalasin D suggests that they may have
a role in stabilizing the yolk cell membrane or regulating cortical
cytoplasm flow toward the animal cap. In contrast, microtuhules
do not seem to be involved in ooplasm segregation and blastodisc
formation in the zebrafish embryo (10), where microfilaments have
been shown to direct ooplasm flow along streamers of cytoplasm
within the central region of the yolk cell. Interestingly, in embryos
of another fish, medaka. both microtubules and microfilaments
have been shown to be involved in ooplasm segregation. In this
fish, not only did cytochalasin D inhibit blastodisc formation, but
colchicine treatment also resulted in less directed ooplasm move-
ments (11). Thus it seems, in medaka. microfilaments and micro-
tubule networks may function in concert during ooplasm segrega-
tion.

One possible clue to understanding the subtle similarities and
differences observed in embryos of these fish and the squid may be
found when the location of ooplasm within the unfertilized egg and
the route of cytoplasmic flow are considered. In contrast to the
central flow of ooplasm in zebrafish, ooplasm flow in medaka
occurs cortically along meridonal pathways (11). Similarly, in the
squid embryo, where microtubular arrays can be visualized by
antibody labeling, the ooplasm flow to the cortex of the yolk cell
surface is restricted to the outermost cortical layer of cytoplasm
(unpubl. results). With antibodies to /3-tubulin, unfertilized eggs
were observed to possess circular swirls of tubulin staining within
the cortex. Yolk just below this thin layer did not label for
/3-tubulin. After fertilization, these patterns change and /3-tubulin-
rich streams oriented toward the animal pole are formed along the
outer cortex of the embryo. Perhaps the reliance on two supporting
cytoskeletal mechanisms within the cortex to move ooplasm to the
blastodisc. as shown in medaka. is characteristic of eggs that
possess a dense central yolk and cortical ooplasm, and may un-
derly this process in the squid embryo. With this possibility in
mind, it will be important to reexamine these elements in other
embryos where microtubules alone, or in concert with microfila-
ments, have been linked to ooplasm segregation and movement
following fertilization. In addition, further analysis of microtubule
and microfilament arrays during ooplasm segregation and in the
presence of cytoskeletal inhibitors will further extend our under-
standing of the mechanism of blastodisc formation and early
development in the squid embryo.

This work was supported by a Faculty Development Grant from
St. Mary's College of Maryland to Karen Crawford and would not
have been possible without Dr. Robert Baker and his laboratory
group at the Marine Biological Laboratory.

252

REPORTS FROM THE MBI. GENERAL SCIENTIFIC MEETINGS

Figure 1. Colchicine inhibits ooplasm segregation unj hlustoili^c cup formation in the .u/mJ cmhrvo. (a) Normal hlu\itidisc cap fot'mtititm 3 h
post-fertilization, lateral view (65 •' ) Inn arrowhead murk* the micropyle and chorion. an marks the polar bodies, the yolk cell is labeled, and arrows
indicate the blastodisc in each panel.), (b) Embryo treated with 5.0 ng/ml colchicine. 5 li post-fertilization, lateral view (65XJ. Note the inhibition of
ooplasm segregation and blastodisc cap formation, (c) Embryo treated with 11.05 fJ.g/ml cytochala.iin D (4? < ). Although this embryo has formed a normal
b/astodisc. perturbation of the yolk cell membrane i.\ demonstrated by the presence of large vesicles, (d) Control embryos treated with DMSO (0.1%)
undergo blastodisc cap formation iinil cleave normally, <S' h post-fertilization (-/5X I Note the presence <>f < leiivtn'c fui'n>\\ \ in these embryos.

Literature Cited

1. Arnold, J. M. 1968. Dev. Biol. 18: 180-197.

2. Crawford, K. 2000. Biol. Bull. 199: 207-208.

3. Yahara. I., and F. Kakimoto-Sameshima. 1978. Cell 15: 251-259.

4. Houliston, E., and R. P. Elinson. 1991. Development 112: 107-
117.

5. Sawada, T., and G. Shatten. 1989. Dev. Biol. 132: 331 -342.

6. Eckherg, W. R. 1981. Differentiation 19: 55-58.

7. Leung, C. F., S. E. Webb, and A. I,. Miller. 1998. Develop.
Gnwth Differ. 40: 3 1 3-326.

8. Klein, K. C., and L. A. Jaffe. 1984. Biol. Bull. 167: 518.

9. Arnold, J. M., and L. D. Williams-Arnold. 1974. ./. Embrvol. E\p.
Morphol. 31: 1-25.

10. Leung, C. F., S. E. Webb, and A. L. Miller. 2000. Dev. Growth

Differ. 42: 29-40.
I 1 Abraham. V. C., S. Gupta, and R. A. Fluck. 1993. Biol. Bull. 184:

115-124.

Reference: Bid. Bull. 201: 252-254. (October 2001 )

The Stellate Ganglion of the Squid Loligo pealeii as a Model for Neuronal Development:
Expression of a POU Class VI Homeodomain Gene, Rpf-1

J. Peter H. Burhach (Rudolf Magnus Institute of Neurosciences, Utrecht, The Netherlands),
Anita J. C. G. M. Hellemons1, Marco Hoekman1, Philip Grant2, and Harish C. Pant2

A major challenge in developmental neurobiology is to under-
stand how neuronal systems are specified, for example, how trans-
mitter phenotype and connectivity are established during develop-

1 Department of Medical Pharmacology. Rudolf Magnus Institute of
Neurosciences, University Medical Center Utrecht, Utrecht University.
35S4CG Utrecht. The Netherlands.

: Laboratory of Neurochemistry. NINDS, NIH. Bcthesda. MD 20X'P

ment. Molecular cascades of transcription factors and growth
factors direct neuronal specification ( 1 ). How they operate in
terminal differentiation and adult networks is poorly understood.
To complement our research to characterize molecular cascades in
complex neuronal systems such as midbrain dopaminergic and
hypothalamic systems in the mouse (2. 3. 4). we have turned to a
non-mammalian neuronal system that has a functionally and mor-
phologically more homogeneous structure. That structure is the
stellate ganglion of the squid Lt>lif;<> pealeii. It is a compacted

DEVELOPMENTAL BIOLOGY

253

loligo pealeii RPF-1

comparison

Human RPF-1

PQALEILNQHFEKNTHPSGAELTELSENLSYDREVVRV
PQALEILN-HFEKNTHPSG-E+TE++E-L+YDREWRV
PQALEILNAHFEKNTHPSGQEMTEIAEKLNYDREWRV

Loligo pealeii Phox2

comparison

Mouse Phox2a

AQLKELEKAFAETHYPDIYTREEIAMKIDLTEARVQVW

AQLKELE+-FAETHYPDIYTREE+A+KIDLTEARVQVW

AQLKELERVFAETHYPDIYTREELALKIDLTEARVQVW

Figure 1. Partial amino acid sequences ofhomeodomain proteins predicted from cloned PCR fragments obtained from the stellate ganglion of the M/niil
Loligo pealeii. The degenerate PCR firimers were those used in mammalian hram (2. 3): upstream. 5' -GMRSCGMSAVMGSACMMBCTTYAC-3' ;
dun nst ream. ?'-TGGTTYMRVAAY'CGYHGMGCMARRTG-3'. Sequences without primer-coded sequence are shown and compared with the mammalian
honiolo\;i retina-derived POU factor- 1 (RPF-1 1 of man IA I, and Pho.\2a of mouse (B). In the comparison, identical amino acids are shown and substitutions
h\ physicochemically similar amino acids indicated h\ +. The predicted RPF-1 protein sequence of Loligo pealeii shows a 31/38 idcnlit\ and a 35/38
similarity to the human protein. Pho.\ 2 shares a 34/38 identity and 37/38 simi/aritv to mouse Pho\2a.

cluster of neurons that innervates the muscles of the mantle
through the giant nerve fiber system and controls the jet-propelled
escape response of the squid (5). The aim of this study was to
identity homeodomain genes expressed in the stellate ganglion and
to correlate their expression with development of the ganglion.

Working from the concept that transcription factors involved in
terminal neuronal differentiation are still operating in the adult
system, as demonstrated in mammalian brain (2. 3. 4). we cloned
homeodomain transcripts from the dissected stellate ganglion of
the squid Loligo pealeii using RT-PCR with degenerated primers
designed to conserved motifs in paired-like homeodomain genes
(2, 3). Two homeodomain transcripts were identified from 40
cloned PCR fragments (Fig. 1 ). One fragment ( 1 out of 40 clones)
predicted a homeodomain protein that was highly homologous to
a POU class VI homeodomain gene product recently identified in
man (6): retina-derived POU factor- 1 (RPF-1 ). The other fragment
(6 clones out of 40) was highly similar to the paired-like homeo-
domain genes plio.\2u and phox2b, also termed urix/pmx (7). Both
types of homeodomain genes have been implicated in the specifi-
cation of neuronal systems of the mouse. Plwx2 genes are required
for normal development of central and peripheral components of
the autonomous nervous system, while rpf-l has been implicated
in the development of amacrine and retinal ganglion cells (6. 7).
Other clones represented non-homeodomain-containing genes, in-
cluding abundant transcripts like alpha-tubulin. actin. and collagen.

We chose to determine the embryonic expression of the rpf-l
gene further by a whole mount in situ hybridi/ation protocol using
DIG-laheled cRNA (8), since initial experiments indicated specific
labeling for rpf-l. but no signals for phox2. Comparison of anti-
sense and sense probes showed specific expression of the squid
rpf-l gene in dorsal structures in the mantle in stage 27 embryos of
the squid. Comparison to histologically stained sections of squid

embryos (9) indicated that the labeled structures are part of the
stellate ganglion. No other neural or non-neuronal structures were
labeled at this developmental stage. In earlier stages (22 to 25),
results suggested expression in the developing eye. These results
suggest that this rpf-l gene is expressed in the developing and
adult stellate ganglion of the squid.

Genes like rpf-l and others may have a role in developmental
events in the stellate ganglion, such as establishment of connectivity
and giant axon formation, as well as participating in regulation and
maintenance of the adult giant fiber system. If interference with its
expression, for example by introduction of morpholinos ( 10), can be
achieved, the role of the rpf-l gene and other homeodomain genes can
be established and can serve as a starting point to delineate molecular
cascades in developing neurons.

Part of this research was performed at the Marine Biological
Laboratory, Woods Hole, Massachusetts, and supported by an
MBL Fellowship sponsored by the Baxter Postdoctoral Fellowship
Fund, MBL Associates Fund, James A. and Faith Miller Memorial
Fund, and the H. B. Steinbach Fellowship Fund.

Literature Cited

1. Jessell, T. M. 200(1. Nat. Rev. Genet. 1: 20-29.

2. Smidt, M. P.. H. S. van Schaick, C. Lanclot, .1. J. Tremblay, J. J.
Cox, A. A. van der Kleij, G. Wolterink, J. Drouin, and J. P. H.
Burbach. 1997. Proc. Nail. Acad. Sci. USA 94: 13.305-13.310.

3. Smidt. M. P.. C. H. Asbreuk, J. J. Cox, H. Chen, R. L. Johnson,
and J. P. H. Burbach. 2000. Nut. Neurosci. }: 337-341.

4 Burbach. J. P. H., S. M. l.iickmaii. I). Murphy, and H. Gainer.

2001. Physiol. Rev. 81: 1147-1267.

5. Martin, R. 1965. Z. Zellforsch. Mikmsk. Anal. 67: 77-85.
6 Zhou, H., T. Yoshioka, and J. Nathans. 1996. J. Neurosci. 16:

2261-2274.

254

REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

7. Stanke. M., D. Junghans, M. Geissen. C. Goridis, U. Ernsberger.
and H. Rohrer. 1999. Development 126: 4087-4094.

8. Green, C. B., A. J. Durston, and R. Morgan. 2001. Much. Dei.
101: 105-110.

y. Grant, P., D. Tseng. R. M. Gould. H. Gainer, and H. C. Pant. 1995.

J. Cinnp. Neural. 356: 311-326.

10. Ando, H., T. Furuta, R. V. Tsien, and H. Okamoto. 20(11. Nat.
Genet. 28: 317-325.

Reference: Biol. Bull. 201: 254-255. (October 2001)

Evidence for Directed Mitotic Cleavage Plane Reorientations During Retinal Development

within the Zebrafish

Brian A. Link (Marine Biological Laboratory, Woods Hole, Massachusetts 02543)

The vertebrate retina develops from a single layer of elongated
cells — the optic cup neuroepithelium. At the time of optic cup
formation, individual neuroepithelial cells are multipotent and can
give rise to any of the cell types found within the differentiated
retina (1). As the optic cup neuroepithelium proliferates, the rep-
ertoire of cell type fates becomes restricted. Retinal cell specifi-
cation, the commitment to differentiate as one particular cell type,
occurs at or following the final cell division. The underlying
mechanisms of cellular specification that generate the diversity of
retinal cell types are unknown.

In many invertebrate epithelial cell types, as well as the rat
neuroepithelium, the plane of cell division is regulated during
development by rapidly reorienting the metaphase chromosomal
plane relative to the plane formed by the cellular sheet (2. 3 1. With
regard to cellular specification, particular metaphase orientations
often correlate with specific cell fates for each daughter cell.
Underlying this correlation, studies in both Caenorhabditis el-
egans and Drosophila melanogaster have demonstrated that the
orientation of the mitotic cleavage plane can dictate whether
asymmetically distributed mRNAs or proteins are inherited
equally or unequally by the two daughter cells (4). Whether
vertebrate retinal cells regulate their mitotic cleavage plane though
metaphase reorientations is addressed in this study.

To assess metaphase chromosomal plane orientations in a ver-
tebrate retina, newly fertilized zebrafish embryos ( 1-8 cell stage)
were injected with 5 nl of plasmid DNA (0.1 jU-gVl) encoding a
fusion protein of histone H2B and GFP (H2B::GFP) (5). At this
concentration, expression was mosaic. This fusion protein associ-
ates with chromosomes throughout the cell cycle in an inert
manner, thus fluorescently labeling a subset of the embryo's cell
nuclei. At 22 hours post fertilization (hpf). injected embryos were
prepared for time-lapse microscopy. Zebrafish were anesthetized
with MS222 (to inhibit spontaneous movements), treated with 0.2
mM l-phenyl-2-thiourea (to block pigmentation), and embedded
in 1.5% agarose (to immobilize the embryo). Labeled proliferating
retinal neuroepithelial cells were imaged using a 40X water emer-
sion objective on an upright epifluorescent microscope. Z-series.
50-60 fj.m in depth, were collected with a cooled CCD camera at
intervals of 1-2 min over periods of 10-24 h. At 22 hpf. the
retinoblast pool in zebrafish is expanding because all cells of the
optic cup neuroepithelium are proliferative with an 8-10 h cell
cycle (6).

Mitoses were observed at the apical border of the neuroepithe-
lium adjacent to the retinal pigment epithelium (RPE). Only mi-

totic cells unobstructed by other labeled cells were scored. A
proportion (8/86) of these observable mitoses showed cleavage
plane reorientations (Fig. 1). For all cells, the time required to
progress from metaphase (initial chromatin condensation) to cell
division (end of karyokinesis) showed a range of 9 to 16 min with
a mean of 12.8 ±2.1 min. No significant difference in this time
was observed between cells that reoriented their metaphase plate
and cells that did not (12.5 ± 2.7 min r.v. 12.8 ± 2.1 min).
Interestingly, each cell that did rotate spindles shifted its chromo-
somes by 90 ° so that the plane of cell division was perpendicular
to the plane of the neuroepithelial sheet. Cells that did not rotate
metaphase chromosomes also cleaved with the axis of separation
perpendicular to the RPE-neuroepithelial border.

These results demonstrate that the plane of cell division within
a vertebrate retinal neuroepithelium can be rapidly reoriented, and
in a directed fashion. Rotations of the metaphase spindle ensured
that all 86 cell divisions occurred perpendicular to the RPE-
neuroepithelial border. This consistency in final cell division plane
suggests that during proliferative phases of retinal development,
perpendicular cleavages are actively maintained. Although the
significance of retinal metaphase rotations has not been probed in
this study, the relationship of spindle rotations to cell fate decisions
in other systems is intriguing. Furthermore, similar to inverte-
brates, asymmetric distribution of proteins has also been observed
in vertebrates. For example Numb, an intracellular signal-modify-
ing protein, is localized in a polarized fashion for several neu-
ronal precursor cell types including the rat retinal neuroepithe-
lium (7).

The main result of these studies is the demonstration of mitotic
cleavage plane reorientations in a vertebrate retina. More gener-
ally, by observing mitotic behaviors in situ within a living embryo,
cell cycle parameters such as M-phase length or mitotic
spindle behavior can be measured directly for individual cells,
and heterogeneity can be assessed. This has not been possible
with traditional population studies that use cell cycle markers
in tissue sections. This experimental system also provides the
framework to integrate studies of cleavage plane orientation,
asymmetric distribution of mRNA or protein, and cell fate
decisions in a single biological context. Lastly, the genetic ma-
nipulability of zebrafish will enable mechanistic studies for each of
these processes.

This work was funded by generous support from the Grass
Foundation. The author also thanks John Dowling, Scott Fraser,
and Remhard Koster for their generosity and advice.

DEVELOPMENTAL BIOLOGY

255

Figure 1. Time lapse am/lysis of a cell labeled with H2B..GFP V/NMH
metaphase chromosomal plum' reorientation in a mitotic retina/ nenroep-
ithcluil cell. The piinel in the /mi ('/' right models the 90 ° reorientation shift
in axis from I to 7 minutes. The RPE-nenroepithelial border is located in
the upper right corner for each image. Time in minutes is listed in the
bottom left for each image.

Literature Cited

1. Harris, W. A. 1997. CHIT. Opin. Genet. Dev. 7: 651-658.

2. Guo, S.. and K. Kemphues. 1996. CHIT. Opin. Genet. Dev. 6: 408-
415.

3. Adams, R. 1996. J. AViiro.vr;. 16: 7610-7618.

4 Lu, B.. L. Jan, and Y-N. Jan. 2000. Aniw. Rev. Nem-o.tci. 23:
531-556.

5. Koster, R., and S. Fraser. 2001. Ucv. Biol. 233: 329-346.

6. Hu, M., and S. Easter. 1999. Dev. Biol. 207: 309-321

7 Cayouette. M.. A. \\hitmore. G. Jefferv, and M. Raff. 2001.
J. Neurosci. 21: 5643-5651.

Reference: Biol. Bull. 201: 255-256. (October 2001)

Messenger RNAs Located in Spiny Dogfish Oligodendrocyte Processes

Ryan Smith1, Emma Kavanagh2, Hilary G. Morrison*, and Robert M. Gould2
IN. Y. S. Institute for Basic Research in Developmental Disabilities. Stolen Island, New York)

Oligodendrocytes. the myelin-tbrming cells in the CNS. synthe-
size proteins in two distinct locations: the cell body, and in each
process where myelin sheaths form. Morphologically these "outer
tongue" processes are cytoplasmic channels that run along the
outer surface of each myelin sheath. In mammals, myelin basic
protein (MBP). a major constituent of compact myelin, is synthe-
sized in these processes and moves rapidly (within minutes) into
compact myelin. Proteins synthesized in the cell body take roughly
30 minutes to incorporate into compact myelin.

To place MBP in each sheath. Oligodendrocytes synthesize
protein in many (up to 40, [1]) cytoplasmic processes. We recently
identified many other proteins (nearly 100, based on cDNA se-
quences representing mRNAs enriched in myelin) that are synthe-
sized in rat oligodendrocyte processes using a combination of
subcellular fractionaiion and suppression subtractive hybridization

' Marine Models in Biological Research Program. Woods Hole. MA.

2 N. Y. S. Institute for Basic Research in Developmental Disabilities,
Slaten Island. NY.

' Josephine Bay Paul Center for Comparative Molecular Biology and
Evolution. Marine Biological Laboratory. Woods Hole. MA.

(2, 3). To broaden our understanding of the role that local protein
synthesis plays in myelination, we applied the same approach to
identify proteins synthesized in oligodendrocyte processes of an
elasmobranch. the spiny dogfish (St/ntilns acanthias). We already
found (Gould, unpubl.) that MBP was not synthesized in dogfish
oligodendrocyte processes by /';; xitu hybridization, since the
mRNA is retained in the oligodendrocyte soma and not transported
to the cells' processes.

We prepared "driver" and "tester" cDNAs from total homoge-
nate and myelin for the subtractive hybridization experiment.
Briefly, three female spiny dogfish were killed with an overdose of
anesthetic. Their brains were removed and homogenized in a
buffered hyperosmotic sucrose (1.2 M) solution; previously we
had found that mRNAs located in oligodendrocyte processes are
trapped more effectively in myelin vesicles homogenized with
hyperosmotic homogenization solution (3). A portion of the
homogenate was extracted for RNA (represents the entire popula-
tion of RNAs in the dogfish brain and is the source of "driver")
with TRI reagent (Molecular Research Center). Buffer was added
to the remaining homogenate to reduce the osmolarity to 0.85 M
sucrose, the sample was placed in an ultracentrifuge tube, overlaid

256 REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

Table 1

Distribution of cDNAs obtained by xiibtntctive hybridization

A. Distribution of the cDNA sequences

Classiticution

cDNA

Quantity

Species

Size (bp)

Percentage of Total

Related to known cDNAs

28.4%

Mitochondria! genome
Unrelated to known cDNAs

28.4%
43.2%

Total

B. Distribution of known cDNA sequences

Matches

B-catenin
POMC (9)
HspAS (2)
Dihydropyrimidinase-like

Human
Shark
Mouse
Human

450

177-407
583-602
617

371/440(84%)
e.g.. 405/407 (99%)
e.g., 236/283 (83%)
457/617 (74%)

SINE
B-spectrin

Ribosomal protein LI
Ig heavy chain
Ev,\2/Hox (4)

Shark

Human

Zebrafish

Shark

718
227
134
500
514-826

149/163 (91%)
65/78 (83%)
34/37 (91%)
35/38 (91%)
e.g.. 126/155 (81%)

Note. Matches are taken directly from the BLAST search results, except for dihydropyrimidinase-like protein, where an intervening region (384) bases
were included from both the subject and query sequences. Parentheses: more than one hit. Range in Size field corresponds to lowest/highest size of
multiple-hit cDNA.

with 0.25 M sucrose, and centrifuged (100,000 x g for 3.5 h).
Myelin vesicles floating on the 0.85 M sucrose were collected, and
myelin fraction RNA was prepared with TR1 reagent (Molecular
Research Center) and used to prepare "tester." Messenger RNA
was prepared from both homogenate and myelin fraction RNAs
(MicroPoly(A) Purist™ mRNA purification kit, Ambion). Homog-
enate and myelin mRNAs were then converted to "driver" and
"tester" cDNA, and a subtraction product (enriched in cDNAs that
represent mRNAs enriched in myelin) was prepared with PCR-
Select™ cDNA Subtraction Kit (CLONTECH) according to the
manufacturer's protocol. Several products were amplified by PCR.
subcloned into pGEM T Easy vector (Promega), and clones were
taken to prepare plasmids (minipreps). The cDNAs were se-
quenced in the Josephine Bay Paul Center for Comparative Mo-
lecular Biology and Evolution at the Marine Biological Laboratory
in Woods Hole, Massachusetts.

In all, 74 sequences were analyzed (BLAST (N) search of the
GenBank non-redundant database) (Table 1). Unlike rat cDNAs
(prepared in the same fashion), which mainly represented MBP
and myelin-associated oligodendrocytic basic protein (MOBP)
mRNAs ( 1 ), none of the dogfish cDNA represented MBP or
MOBP homologs. As with rat. about half were unrelated to
mRNAs in the current GenBank database, and high portions were
derived from mitochondria! DNA. Only four of the known se-
quences, /3-catenin, proopiomelanocortin (POMC). heat shock pro-
tein A5 (HspA5). and dihydropyrimidinase-like protein, matched
sequences in the GenBank database throughout. The portions of

HspA5 that matched the human sequence were 3'-coding. The
non-coding portion was less conserved. Small portions of five
other cDNAs — SINE, j3-spectrin, ribosomal protein LI, evx2, and
Ig heavy chain — matched known sequence in the GenBank data-
base.

To confirm that these cDNAs represent mRNAs located in
oligodendrocyte processes. Northern blot studies are needed to
show that the mRNAs are enriched in myelin. Complementary in
situ hybridization studies are planned to further locate the mRNAs
to oligodendrocyte processes. In summary, our results suggest that
the population of mRNAs transported to spiny dogfish oligoden-
drocyte processes is large and varied. Comparative studies are
planned to find out if /3-catenin, POMC. Hsp5a and dihydropyri-
midinase-like protein are expressed in rat oligodendrocytes and if
some of the cDNAs identified in rat oligodendrocyte processes are
expressed in spiny dogfish processes.

This study was supported by grants from the National Multiple
Sclerosis Society (RMG) and the National Science Foundation (RS
and RMG).

Literature Cited

1 Peters, A., and C. C. Proskauer. 1969. Anal. Kec. 163: 243.

2 Gould, R. M., C. M. Freund, F. Palmer, and D. L. Feinstein. 20(10.

J. Neiirochein. 74: 1834-1844.

3. Gould, R. M., C. M. Freund. J. Engler, and H. G. Morrison. 2000.
Biol. Bull. 199: 215-217.

DEVELOPMENTAL BIOLOGY

257

Reference: ttiol. Hull. 201: 257-258. (October 2001)

Phalloidin Labeling of Developing Muscle in Embryos of the Polychaete Capitella sp. I

Susan D. Hill (Department of Zoology, Michigan State University,
East Lansing, Michigan 48824) and Barbara C. Boyer1

Capitella sp. I, previously considered part of the Capitella
capitata complex ( 1 ), is a small polychaete annelid that can be
maintained in culture with ease (J.P. Grassle, Institute of Ma-
rine and Coastal Sciences, Rutgers University, pers. comm.i.
Lecithotrophic eggs are deposited in a maternal brood tube and
can be readily harvested at different stages of development.
Larvae emerge from the brood tube after approximately 8 days
as many-segmented metatrochophores, each bearing a pro-
totroch and telotroch, the classic trochophore stage being by-
passed. The free-swimming metatrochophores are non-feeding
and are competent to settle and metamorphose within a few
hours of emergence. Metamorphosis in this species is not mor-
phologically dramatic, but includes a pronounced elongation,
loss of trochal bands and accompanying locomotory changes,
and a transition from non-feeding to feeding. Postmetamorphic
growth involves a general enlargement of the worm and addi-
tion of segments in a growth zone immediately anterior to the
terminal pygidium.

The development of muscle patterns in soft-bodied bilaterian
animals is not well understood, with most recent information
coming from investigations of acoelomate flatworms (2, 3) and
the medicinal leech, a derived annelid (4, 5). Segmentation
between annelids and arthropods has traditionally been consid-
ered to be homologous; however, the recent assignment of
annelids to the Lophotrochozoa and arthropods to the Ecdyso-
zoa, brings this homology into question. A study of muscle
development in a more ancestral annelid would be useful in
furthering our understanding of the ontogenetic and evolution-
ary origins of segmentation, as well as the cellular interactions
involved in muscle patterning and innervation during embryo-
genesis. To this end we are investigating early muscle devel-
opment in Capitella sp. I. Staged embryos were removed from
the brood tube and labeled with rhodamine-phalloidin following
the procedure used by Reiter er cil. (2) to detect actin filaments
in developing muscle of turbellarian flatworms. Specimens
were observed with an Olympus BX60 fluorescence microscope
and imaged using an Olympus Magnifire digital camera (model
S99860).

Muscle development proceeds from anterior to posterior. As
the stomodeum forms, a ventral arc of muscle becomes apparent
in the lower lip. Phalloidin binding continues dorsally until the
mouth is surrounded (Fig. la). Approximately three days after
fertilization, longitudinal muscles of the body wall begin to
form. Initially eight longitudinal muscles appear in the follow-
ing sequence: ( 1 ) four dorsal strands that will reach from the
prostomium to the pygidium begin to develop; (2) two latero-

' Department of Biological Sciences. Union College. Schenectady, NY
12308

ventral muscles form at the lateral edges of the stomodeum
(Fig. la), then come together at the apex of the prostomium
(Fig. Ib); (3) these two lateroventral muscles also grow poste-
riorly, extending ventrally from the stomodeum into the py-
gidium (Fig. Ib); (4) medially, a second pair of midventral
muscles (Fig. Ib) grows posteriorly from the stomodeum. Sub-
sequently two additional lateral muscles form (Fig. Ib). Lon-
gitudinal muscles initially appear as thread-like single strands
which thicken as more strands are added.

After initial differentiation of longitudinal muscles, circular
muscles begin to develop, appearing first in the peristomial
region. One band forms in the lateral stomodeal region, while a
second passes immediately posterior to the stomodeum. Addi-
tional circular muscle bands are added sequentially from ante-
rior to posterior corresponding to the metameric pattern of the
developing larva (Fig. Ib, c). Development of circular muscles
seems to be initiated in the ventrolateral region of the embryo,
between the midventral muscles (MV) and the lateroventral
muscles (LV). The circular muscles appear as complete bands
ventrally before they are seen dorsally. A few closely spaced
circular fibers also become visible in the telotrochal region. At
this time there is a gap between the more anterior circular
muscles and these telotrochal bands.

As development continues, both longitudinal and circular mus-
cles become more massive with the addition of more strands.
Circular muscle formation continues posteriorly (Fig. Ic). filling
the gap between the developing circular muscles and telotrochal
bands. The number of telotrochal bands also increases.

During larval development additional muscles — longitudinal,
intrasegmental. oblique, setal sac fibers, etc. — are added so that the
muscle pattern in newly emerged metatrochophores is very com-
plex (Fig. Id). Since metamorphosis occurs without major struc-
tural changes, these larval muscles form the basis of the adult
musculature.

Mesoderm formation in polychaetes is attributed to two te-
loblasts. derivatives of the 4d blastomere, which reside between
the posterior ectoderm and the lining of the developing gut (6).
Segmentation is believed to occur as successive blocks of
mesoderm are formed. Currently segmentation in polychaetes is
being investigated in a number of laboratories (7, 8. 9) using
several genetic markers. Our results show that the circular
muscles that differentiate in the larval trunk anterior to the
telotroch are iterated sequentially from anterior to posterior. We
suggest that the phalloidin-staining telotrochal bands are the
nascent segmental muscles of the growth zone.

This work was supported in part by the Union College Faculty
Research Fund. The authors gratefully acknowledge the generous
assistance of Dr. William Eckberg in creating the figure.

258

REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

Figure 1. (a) Ventral view of an earlv embryo (240 X 175 fjun) showing the stomodeum fS) with phalloidin labeling of the lower lip and lateroventral
muscles (LV). (b) Ventral view showing the prototroch (Pi, te/otroch (T), paired midvcinnil muscles (MV), lateroventral muscles (LV), and lateral mu.scli s
(L). Circular musculature formation is incomplete, with a gup between the most posterior circular muscle band and the telotroch. (c) Ventrolateral view
showing thickened midventral (MV) and lateroventral (LV) muscles. Circular muscle bunas arc complete to the telotroch. L = lateral muscle, (d)
Metatrochophore showing greutlv increased complexity of the larval musculature.

Literature Cited

1. Grassle, J. P., and J. F. Grassle. 1976. Science 192: 567-569

2. Reiter, D., B. Buyer. P. Ladurner, G. Main, \V. Salvenmnser, and R.
Rieger. 1996. Kon.\'s Arch. Dev. Biol. 205: 410-423.

3. Ladurner. P.. and R. Reiger. 2000. Dev. Biol. 222: 359-375

4. Jellies, J. 1990. Trends Nenrosci. 13: 126-131.

5. Jellies. J.. and \V. B. Kristan. Jr. I9SS. J. Nenrosci. 8: 3317-
3326.

6 Anderson, I). T. 1966. Aeta Zoo/. Bd XLVII: 1-42.

7 Seaver, E. C., and S. D. Hill. 1999. Am. Zool. 39: 77A

X. Werbrock, A. H., D. A. Meiklejohn, A. Sainz, J. H. Iwasa, and R. M.

Savage. 2001. Dev. Biol. 235: 476-4SX.
4. Seaver, E. C., D. A. Paulsnn, S. Q. Irvine, and M. Q. Martindale.

2001. Dev. Biol. 236: 145-204.

Reference: Biol. Bull. 201: 25S-260. (Ocloher 2001)

Differentiation of Pharyngeal Muscles on the Basis of Enzyme Activities
in the Cichlid Tramitichromis intermedius

Aaron N. Rice, David S. Portno\, Ingrid M. Kaat:, and Phillip S. Lohel (Boston University Marine Program,
Marine Biological Laboratory, Woods Hole, Massachusetts 02543)

One of the key morphological features of cichlid fishes is their
highly developed pharyngeal jaw complex used in feeding ( 1 ).
Although many studies focused on the anatomy (2) and function
( 1 1 of pharyngeal muscles, the potential physiological differences
between them have not been examined in detail. The purpose of
this paper is to investigate the capacity for anaerobic activity of the
muscles in the pharyngeal jaw complex, and to assess whether they

are all the same functional type. Finding different types would
suggest that various muscles in the complex may have functions
other than mastication.

Bass et al. (3) demonstrated that fundamentally different types
of muscles can be distinguished by the activity level of energetic
enzymes. Assaying these enzymes in muscles tissues can indicate
whether a muscle functions primarily through aerobic or anaerobic

DEVELOPMENTAL BIOLOGY

259

(A)

LE2

(B)

"c 10-

0 8 -
|

.^11

'•^ 4 -
O
03

3 2

0 -

LE2 LEO LE4 LP PP PH PHCE PHCI RD Ax

Muscle

Figure 1. I A ) The Tramitichromis intermedius pharyngeal muscles. Arrows indicate the direction of movement of lite pharyngeal jaws due to muscular
contraction. Abbreviations are as follows: LE2, levator externus 2: LE.\ levator externus 3; LE4. levator extermis 4: LP. levator posterior; PP. protractor
pectoralis: PH. pharyngohyoideus; PHCE. pharyngocleithralis externus; PHCI, pharyngocleithralis interims: RD. retractor dorsalis: DPI. dorsal
pharyngeal jaw; VPJ, ventral pharyngeal jaw. IB) En:vinatic activities of L-lactate dehydrogenase from different T. intermedius pharyngeal muscles.

pathways. Comparative analysis between muscles can further elu-
cidate the degree of functional specialization that these tissues
have undergone relative to other muscles in the body. This tech-
nique has been employed in a variety of different taxa — for ex-
ample, fishes (4), frogs (5), and bats (6) — to demonstrate func-
tional differences between muscle types.

Captive-bred Tramitichromis intermedius (born in July 2000
from wild-caught parents from Lake Malawi, Africa) were kept in
75-gallon aquaria. Fish were euthanized with MS-222, and the
opercles were removed. Muscles involved in movement of the
dorsal and ventral pharyngeal jaws were removed and weighed:
levator externi 2. 3, 4 (LE2, LE3. LE4). levator posterior (LP),
protractor pectoralis (PP), pharyngohyoideus (PH). pharyngo-

cleithralis externus (PHCE). pharnygocleithralis internus (PHCI).
and retractor dorsalis (RD) (Fig. 1 A). Axial muscle (Ax) from the
tail was also removed and served as a comparison for fast-twitch
muscle. Muscle nomenclature follows Liem (1). Tissues were
homogenized in 1 ml of buffer (7.5 mM Tris and 1 mM EGTA. pH
7.6). and analyzed for activities of L-lactate dehydrogenase (LDH;
E.C. 1.1.1.27). in order to indicate capacity for anaerobic respira-
tion. Using a Perkin-Elmer Lambda 3B UV/Vis spectrophotome-
ter, enzymes were assayed using 50 mM TEA, 5 mM EGTA. 0. 1 5
mM NADH, 0.24 mM pyruvate, pH 7.6, at 340 nm. Enzyme
activities were calculated as micromoles of product per minute per
gram of tissue (4), and differences between muscle groups were
analyzed using a one-way ANOVA.

260

REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

The mean ( ±SE) LDH activities of the muscles were as follows:
LE2: 3.15 ± 0.71. LE3: 4.56 ± 1.19, LE4: 6.57 ± 1.02, LP:
6.20 ± 1.12, PP: 5.87 ± 1.18. PH: 5.82 ± 1.48. PHCE: 3.38 ±
0.73, PHCI: 4.52 ± 0.72, RD: 8.83 ± 2.57, Ax: 7.44 ± 2.09 (Fig.
IB). These results show that the pharyngeal muscles examined
differ significantly in levels of LDH activity (u = 8, P =
0.0152). A post-hoc Fisher's Protected Least Significant Differ-
ence test revealed that significant differences existed between Ax
and PHCE (P = 0.0469). LE2 and RD (P = 0.0143), LE3 and
RD (P = 0.0374), PHCE and RD (P = 0.0355).

The differences in LDH activity in this muscle complex shows
that several muscles examined have different capacities for anaer-
obic activity. Functional muscle types can be differentiated on the
basis of enzyme activities by comparing ratios between activities
of aerobic and anaerobic enzymes. Without determining aerobic
activity, we cannot conclusively demonstrate that the muscles
examined are different functional types. These preliminary data
suggest that more than one muscle type may be present, but
analysis of the aerobic capacity is necessary.

The presence of different muscle types would suggest that the
pharyngeal complex may be performing a dual function. In addi-
tion to mastication, the pharyngeal jaws have also been hypothe-
sized to function in sound production (7). Spectrograms of sounds
produced by cichlids (8) suggest that this behavior involves very

rapid muscle contraction and occlusion of the pharyngeal jaws
mediated by rapid muscle contraction (7). These muscles would
need to be capable of powerful burst activity, as opposed to more
slow-twitch muscles involved in mastication. Observable differ-
ences in enzymatic properties of pharyngeal muscles are further
representative of the complexity of this structure, and perhaps the
result of its dual function.

This project would not have been possible without the valuable
advice of B. D. Sidell, C. R. Bevier, and R. Voigt. Research was
supported by a grant from the Army Research Office (DAAG55-
91-1-0304) to PSL.

Literature Cited

1. Liem, K. F. 1974. Syst. Zool. 22: 425-441.

2. Anker, G. C. 1978. Nelh. J. Zool. 28: 234-271.

? Bass, A., D. Brdiczka, P. Eyer, S. Hofer, and D. Pette. 1969. Eur.
J. Biochcm. 10: I9X-206.

4. West, J. L., J. R. Bailey, V. M. F. Almeida-Val, A. L. Val, B. D.
Sidell, and W. R. Driedzic. 1999. Can. J. Zool. 77: 690-696.

5. Bevier, C. R. 1995. Physiol. Zool. 68: 1118-1 142.

6. Yacoe, M. E., J. W. Cummings, P. Myers, and G. K. Creighlon.
1982. Am. J. Physiol. 242: R189-R194.

7. Lobel, P. S. 2001. J. At/uaric. Aquat. Sci. 9: 89-108.

8. Lobel, P. S. 1998. Em-iron. Biol. Fishes 52: 443-452.

NEUROBIOLOGY

261

Reference: B,oi Bull. 201: 261-262. (October 2001 )

Real-Tirne Detection of Reactive Oxygen Intermediates From Single Microglial Cells

Gilcul Twig1'2, Sung- Kwon Jung"', Mark A. Messerlr. Peter J. S. Smith*, and Orian S. Shirihai*

A growing body of evidence indicates that activation of
microglia (macrophages resident in brain) aggravates the in-
flammatory process and thus can contribute significantly to the
progression of various neurodegenerative diseases ( 1 ). As with
other tissue-specific macrophages, microglia are thought to
exert some of their cytotoxic effects through the production of
reactive-oxygen-intermediates (ROD. For example, /3-amyloid,
an abundant component of amyloid ("senile") plaques, was
shown to induce the production of ROI by cultured microglia
within 1-2 min (2). Any damage caused to surrounding tissue
by microglial cells is mainly dependent upon the magnitude of
the gradient of ROI that is generated on the surface of the cell's
membrane. Therefore, quantification at a high spatial and tem-
poral resolution of ROI distribution in the microenvironment of
an activated microglial cell is important for the assessment of
neurotoxicity.

The enzyme responsible for the generation of ROI in an oxida-
tive burst in microglial cells is NADPH oxidase, which transfers an
electron from a single cytoplasmic NADPH molecule to an oxygen
molecule, producing superoxide anion (O2" ). O7 and its dismuta-
tion product, hydrogen peroxide (H2O2), diffuse uway from the
microglial cells and have the potential to oxidize cellular compo-
nents in neighboring cells, including proteins, lipids. and DNA (3).
However, H2O2 is a much more stable product than O2 and
therefore can be used as a reliable indicator for detection of an
oxidative burst.

The self-referencing technique has the capacity to detect, with
high spatial and temporal resolution, concentration gradients of
specific molecules surrounding single cells (4, 5). In the current
investigation, we used an H2O2-sensitive microprobe as a sensor of
ROI production by microglia cells. To test the feasibility of the
self-referencing technique for the detection of ROI from single
microglial cells, we activated the NADPH oxidase machinery with
phorbol-12-myristate-13-acetate (PMA), a potent activator of this
enzyme (6).

H2O2 microsensors were prepared as described previously for
oxygen sensors, but with slight modification (7). Briefly, 25-^m
diameter platinum (Pt) wires were immersed in an aqueous
solution of 4 M KCN and 1 M NaOH and then etched down to
~2 /xm diameter by the application of square waves (amplitude,
4.0 V; period, 4 ms). The etched Pt wires were inserted into
pulled glass capillaries, insulated with optical adhesive, and
then coated with 10% cellulose acetate. The total tip diameter of
the sensor was about 3 fj.m. For all measurements, the sensor
was polarized at +0.60 V against a Ag/AgCl reference elec-
trode; its sensitivity was 0.85 ±0.12 pA//xM (mean ± SD. n =
4). Although the sensor can potentially detect other ROI beside

1 The Bruce Rappaport Faculty of Medicine. Technion. Israel.

2 Grass Laboratory and ' BioCurrents Research Center, Marine Biologi-
cal Laboratory, Woods Hole. MA.

H2O2 (such as nitric oxide and O-, ), H^O-, was probably the
major component of the concentration gradient, considering
its longer half-life time and the composition of the media
(<100 [J.M L-arginine).

Purified microglia were isolated from rat brains, as described
elsewhere (8), and were plated into 35-mm diameter culture dishes
at a density of 3000 cells/ml to allow a distribution of single cells
(one cell where no other cells can be detected within a range of
-200 /urn).

Figure 1A demonstrates the experimental protocol that we used
to detect H2O2 production in a single microglial ceil. In the
presence of culture medium only, no significant HoO2 efflux was
detected in the close vicinity of the cell (Fig. 1A, trace (a). 5. 15
/urn). However, 10 min after adding PMA (final concentration of
1 30 nM), a measurement from the same location detected an H,O,
efflux of 0.46 pmol/cnr/s. The magnitude of the H2O2 efflux was
inversely related to the distance from cell surface (traces (c-g) in
Fig. 1A) and was nearly zero when the microprobe was moved 40
/j,m from the cell surface (trace (g) in Fig. 1A). Similar results to
those shown in Figure 1 were obtained in 80% of the isolated
microglial cells (12 out of 15) when PMA was added to the
solution; in the remaining 20% no H-.O, was detected. The range
of peak flux was 0.22 pmol/cnr/s (SD ± 0.17) and the average
threshold detection distance was 22.4 /j.m (SD ± 4.2) (n =12).
The average latency for response was 4.3 min and in all the
responding cells the time period of detectable gradient exceeded
30 min.

To ensure that the signal detected by the probe originated from
the production of ROI, catalase, an enzyme that hydrolyzes H,O^.
was added to the bath. Catalase significantly attenuated the H,O2
efflux regardless of the distance from the cell surface (Fig. 1A:
compare (h) with (i) and Fig. IB: compare closed with open
circles).

The present study demonstrates that a self-referencing mi-
crosensor can detect H2O2 changes in the nano-molar range
near the surface of a single microglial cell. Though numerous
assays are available to measure an oxidative burst within mac-
rophages, the self-referencing technique is unique in providing
the ability to measure the microenvironment around a single
cell or cluster of cells in real-time and with high spatial and
temporal resolution. Because microglial cells can enhance neu-
rotoxicity of the surrounding tissue, this assay may be useful for
quantifying the potential contribution of endogenous micro-
glial-induced activators in neurodegenerative diseases.

We thank Paul Malchow, Jeffery Laskin. and Solomon Graf
for critical reading of the manuscript. This study was supported
by the Grass Foundation Fellowship in Neurophysiology
to G. Twig and by NIH grant NCRR P41 RRO1395 to PJS
Smith.

262

< 40

-20

(A)

(a)
5-15

REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

Literature Cited

pMA

2-12

catalase -

C 1 C

<b>

15-25 (f)
^25-35

35-45

7 min

500 1000 1500
Time (seconds)

2000

1. Gonzalez-Scarano, F., and G. Baltuch. 1999. Anmi. Rev. Netirosci.
22: 219-240.

2. Bianca, V. D., S. Dusi, E. Bianchini. I. Dal Pra, and F. Rossi. 1999.
./. Biol. Client. 274: 15.493-15.439.

3. J. S. Weiss. 1989. N. Engl. J. Med. 320: 365-376.

4. Smith, P. J. S., and J. Trimarchi. 2001. Am. J. Phvsioi 280:
Cl-CI I.

5. Smith. P. J. S., K. Hammar, D. M. Porterfield, R. H. Sanger, and
J. R. Trimarchi. 1999. Microsc. Res. Tech. 46: 39S-417.

6 Khanna, R., L. Roy, X. Zhu, and L. C. Schlichter. 2001. Am. J.
Phvsioi. 280: C796-C806.

7 Jung, S.-K., W. Gorski, C. A. Aspinwall, L. M. Kauri, and R. T.
Kennedy. 1999. Aiwl. Client. 71: 3642-3649.

S. Shirihai, O., P. J. S. Smith. K. Hammar, and D. Dagan. 1998. GIUi
23: 339-34K.

0.7

0.6

| 0.5

| 0.4
o

I. 0.3

I 0.2

o" 0.1

0
-0.1

(B)

PMA
PMA+catalase

10 20 30 40
Distance From Cell Surface [|om]

Figure 1. Self-referencing H,(>, iiieii.iiiri-mciit <>t « single microglial
cell. (A) The difference in current mines detected h\ the microelectrode
when self-referencing at different Jistunces from the cell surface. In nil
traces, the excursion (Jistunce of the probe in the self-referencing format)
u-ns III IJLIII. and the inunhcr on the top , if each trace represents the I\\'o
positions of the microsensor in ^m. Note that addition of PMA Hipper solid
horizontal line) induced a significant elevation in H:O, efflux [la) vs.
(c-g)] and that in the presence of catalase (0.1V mx/mD the H-,02 efflux
was abolished [{i) vs. (h)]. Truce (b) is a "hackgrtniihl" measure taken 60
jj.ni awa\ from the cell. (B) The relationships between H,O: efflux and the
average distance from a membrane surface in the presence of PMA: data
from the same cell shown in (A >. Note that application ofcala/ase abolishes
the H:O: efflux.

NEUROBIOLOGY

263

KflVreiice: Hi»l. Hull. 201: 263-264. (October 2IHII i

Porocytosis: Quanta! Synaptic Secretion of Neurotransmitter at the Neuroniuscular Junction

Through Arrayed Vesicles

Robert B. Silver1", Mahlon E. Kriehel , Bruce Keller , and George D.

We have developed a new hypothesis for secretion, particularly
at the neuromuscular junction and CNS synapses. Our interpreta-
tion of secretion — which is consistent with the structural organi-
zation of the neuromuscular junction reported by McMahan and
co-workers ( 1 ) — is based upon the porocytosis hypothesis (2. 3), in
which the postsynaptic quantal response results from presynaptic
neurotransmitter secretion from many docked vesicles, rather than
from a single vesicular exocytotic event (</! 4. 5). In the mecha-
nism we propose, presynaptic vesicles are arrayed at two levels: 1 )
vesicles are anchored to the active zones of the plasma membrane
and juxtaposed to calcium ion-selective channels by proteins
such as SNAREs (6) to make a unit; and 2) these vesicle-ion
channel-SNARE-membrane-containing units are arranged in spa-
tially periodic arrays. We envision that the organization of the
arrayed active zone material at the frog neuromuscular junction
described by Harlow and co-workers ( 1 ). and the array that we
discuss, are one and the same entity. We view this secretory
"organelle" ( 1 ), which we have called the "synaptomere" (3), to be
the unit of secretion, much as the sarcomere is the unit of contrac-
tion. The synaptomere contains a scaffold that would prevent
vesicular fusion into the terminal membrane and would maintain
vesicles in the linear array so that vesicle and terminal unit mem-
branes are in apposition to the receptors on the postsynaptic fold.
This arrangement is extendable to synapses, although the fine level
of organization of the array structure may vary among secretory
systems.

The porocytosis mechanism we propose provides a quantum of
neurotransmitter, but without the need to invoke fusion of a single
vesicle membrane with the (presynaptic) plasma membrane. The
small observed coefficient of variation (<3%) in end plate poten-
tials indicates that there are only about 200 release sites (9), each
of which secretes one quantum per action potential (2. 7. 8. 9). The
200 sites found on a small muscle fiber establish a maximum
quantitative limit of 1 site per micrometer terminal length tor the
number of secretory organelles at the neuromuscular junction (2, 7.
9) and excludes a single vesicle quantum mechanism. Our math-
ematical modeling efforts have shown that release of neurotrans-
mitter via the quantal vesicular fusion mechanism would result in
a coefficient of variation of 14% to 30%. In summary, the notion
that neurotransmitter release is mediated through a "single quan-
tum-single vesicle" mechanism would appear to be precluded
(2, 3).

1 Departments of Radiology. Pharmacology and Physiology, Wayne
State University School of Medicine, Detroit. MI; Decision and Informa-
tion Sciences Division. Argonne National Laboratory. Argonne. IL.

; Marine Biological Laboratory. Woods Hole. MA.

' Department of Neuroscience and Physiology. SUNY Upstate Medical
University. Syracuse. NY 13210.

4 Psychiatric Institute, and Department of Anatomy and Cell Biology.
College of Medicine, University of Illinois, Chicago, IL 60612.

Strong physiological evidence supports the concept that the
repeating components of the synaptomere function as units, each
secreting one packet of transmitter (10). Most importantly, the
ratio of the large to small class of transmitter packets (MEPPs and
sub-MEPPs), and the number of subunits composing the larger
class, is readily changed with many treatments and conditions ( 10),
showing that the two classes share the same sub-unit. Decreasing
extracellular calcium decreases MEPP frequency and decreases the
number of subunits in the MEPP (Fig. 1 ). In normal calcium, there
is a very small percentage of sub-MEPPs, while in reduced cal-
cium concentration most MEPPs are of the sub-MEPP class. A
postsynaptic effect is ruled out because the modal size of the
sub-MEPP has not changed. Thus, these data indicate that the
number of secreting pores in the array is calcium-dependent. The
concept that a single vesicle would release only a portion of its
contents per flicker is supported by other studies. Neher (11)
calculated that a flicker of a pore would secrete about 8% of the
contents of a small vesicle. Rahamimoff and Fernandez (12) pro-
posed that a cationic transmitter could exchange with Na ions
through a fusion pore to generate the stib-MEPP. In the porocytosis
array model, the 200 physiologically described release sites of a
neuromuscular junction defined by Katz and Miledi (13) are the
synaptomeres. and the attractive "organelles" described by Harlow
and co-workers ( 1 ).

What then is the functional significance of the vesicle array of
the neuromuscular junction? A two-tiered hierarchical array of
vesicles is observed at the neuromuscular junction. Calcium acts
within microdomains during a millisecond timeframe to evoke the
release of neurotransmitter from docked vesicles across two bilay-
crs ( 14). We believe that the calcium ions, with a mobility that is
restricted in space and time, establishes a "salt-bridge" among
adjacent lipid molecules, and in doing so, establishes a pore that
spans the lipid bilayers of both the vesicle and plasma membrane.
That pore will be maintained as long as calcium levels are suffi-
ciently high. Upon the reduction of calcium levels (i.e.. within a
millisecond), the liaison of calcium and lipid is disengaged, and
the lipid molecules are freed to rotate and spread readily by
diffusion, thus resulting in closure of the transient pore (16). We
believe that, aside from the mechanism known as "constitutive
secretion" (15). this porocytotic mechanism extends beyond the
synapse, to include nearly if not all cellular secretory processes.

The observed constancy of the amount of neurotransmitter se-
creted with nerve stimulation, attested to by the small value of the
coefficient of variance of EPPs, can only be explained by release
of small amounts of neurotransmitter molecules from arrays of
vesicles at each release site of the neuromuscular junction. Since
the coefficient of variation of the quantal packet is a function of
I/square root of the number of contributing vesicles, and there are
30-50 in an array, a standard amount of secretion is guaranteed by
the array with each action potential. The array notion is so robust

264

REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

35
30
25

I 20

10
5

Control-2 mM calcium

Mepps

Sub-mepps

30 40 50 60

sis, mediated by acetyl-O-transferase. is known to occur on the
cytoplasmic surface of the vesicle membrane (19). In addition, it is
likely that, concurrently, there are mechanisms for docking and
undocking vesicles that are independent of secretion, each process
having its own identifiable rate constants. Small variations in
amounts of neurotransmitter released are readily accommodated
by modulating (e.g., through small changes in calcium dynamics)
the amount released from many vesicles whose diameters are
observed to vary by 3% to 10% (20, 21). Most importantly, the
array concept permits quanta! size to be frequency-dependent.
Thus, to achieve the observed characteristic constancy of "quantal
release" (i.e., MEPP size), the synapse must rely on secretion
through many vesicles within an array of vesicles. The porocytosis
mechanism we have proposed uniquely meets these requirements.
We believe that the porocytosis mechanism extends to secretion in
other non-synaptic systems.

1/4 Calcium Saline

(0.5 mM calcium, 7.5 mM magnesium)

30 -

25 -

I 20-

I 15 H

10 -

5 -

Sub-mepps

0 10 20 30 40 50 60 70

Bin Number (Each Bin = 100 microvolts)

Figure 1. The effect of calcium concentration on MEPP amplitudes
and on the ratio ofsub-MEPPs to MEPPs in skate electrocytes. Top Panel:
Control in normal saline with 2 mM calcium external to a cell stimulated
to generate I MEPP/s. Note the small percentage of siib-MEPPs. Bottom
Panel: The effect of low (1/4 normal) calcium concentration saline (0.5
mM CaC/,. 7.5 mM MgCI:) and stimulation at a rate of 1 MEPP/s. Note
that most MEPPs are of the suh-MEPP class. A postsynaptic effect is ruled
out because the modal size of the sub-MEPP has not changed.

in maintaining a standard packet size, that vesicle contents may
vary from full to empty. Since 60% of the acetylcholine in the
synapse is present in the cytosol (17, 18), transporters on the
vesicle membrane would continuously "fill" vesicles. In addition,
acetylcholine is readily available for transporters because synthe-

Literature Cited

1 Harlow, M. L., D. Ress, A. Stoschek, R. M. Marshall, and U. J.
McMahan. 2001. Nature 409: 479-484.

2. Kriebel, M. E., B. Keller, J. Holsapple. G. Q. Fox, and G. D.
Pappas. 2000. Neuroscientist 6: 422-427.

3. Kriebel, M. E., B. Keller, R. B. Silver, G. Q. Fox, and G. D.
Pappas. 2001. Brain Res. (In press).

4 Heuser, J. E., T. S. Reese, and D. M. Landis. 1981. ./. Neurocytnl.
3: 109-131.

5. Heuser, J. E., and T. S. Reese. 1974. J. Cell Biol. 88: 564-580.

6. Sudhof, T. C. 2000. Neuron 28: 3 1 7-320.

7 del Castillo, J., and B. Katz. 1954. J. Physiol. 124: 560-573.

8. Quastel, D. M. J. 1997. Biopln-sical J. 72: 728-753.

9. Kriebel, M. E., and B. Keller. 1999. Cell Biol. Int. 23: 527-532.

10. Kriebel, M. E. 1988. Pp. 537-566 in Handbook of Experimental
Pharmacology. Springer- Verlag. Berlin.

1 1 . Neher, E. 1993. Nature 363: 497-498.

12. Rahamimoff, R., and J. M. Fernandez. 1997. Neumn 18: 17-27.

13. Katz, B., and R. Miledi. 1979. Pmc. R. Soc. Land. B 205: 369-378.

14. Llinas, R. 1999. The Suiiid Giant Synapse: A Model for Chemical
Transmission. Oxford University Press, New York.

15. Blasquez, M., and K. L. Shennan. 2000. Biochem. Cell Biol. 78:
181-191.

16. Menikh, A., P. G. Nyholm, and J. M. Boggs. 1997. Biochemistry
36: 3438-3447.

17 Zimmerman, H., and C. R. Denston. 1977. Neuroscience 2: 695-
714.

18. Zimmermann, H. 1982. Pp. 241-259 in Neurotransmitter Vesicles.
Academic Press, New York.

19. Eder-Colli, L., and S. Amato. 1985. Neuroscience 15: 577-589.

20. Fox, G. Q. 1996. Cell Tissue Res. 284: 303-316.

21. Fox, G. Q., and M. E. Kriebel. 1994. Brain Res. 660: 1 13-128.

NEUROBIOLOGY

265

Reference: Biol. Bull. 201: 265-267. (October :illll I

Endogenous Zinc as a Neuromodulator in Vertebrate Retina: Evidence From the Retinal Slice

Richaid L. Chappell {Hunter College, CUNY, 695 Park Ave., New York. New York 10021)

iinil Stephen Reclenti1

Studies of the transretinal electroretinogram (ERG) of the skate
(Raja erinacia) eyecup have provided evidence that endogenous
zinc plays a role as a neuromodulator in vertebrate retina ( 1 ). With
GABA receptor activity blocked by 200 /j,/W picrotoxin, superfu-
sinn of the zinc chelating agent histidine (100 /tA/) increased by
about 2-fold the ON (b-wave) and OFF (d-wave) components of
the ERG. In addition, as shown first in the salamander retina (2)
and more recently in mammalian retinas (3, 4), an accumulation of
zinc has been localized to the base of the photoreceptors in skate
(5). These observations support the suggestion that zinc, co-re-
leased with glutamate from photoreceptor terminals, may serve as

' Ph.D. Program in Biology. The Graduate School and University Cen-
ter. CUNY. 365 Fifth Ave., New York, NY 10016.

a neuromodulator in the outer plexiform layer of the vertebrate
retina. By acting on the receptor terminal to reduce calcium entry,
zinc could serve as a feedback signal to modulate transmitter
release (2). If this is the case, one would expect to observe an
effect of histidine application on the conductance of second-order
cells in the retina of the skate.

We have tested this hypothesis by the use of whole-cell, patch-
clamp recordings from horizontal cells in the skate retinal slice
preparation. The slices (-200 /^im thick) from the all-rod retina of
the skate were positioned on a glass slide and visualized using a
fixed-stage microscope equipped with a water-immersion objec-
tive and Nomarski differential interference contrast optics. Whole-
cell patch recordings were obtained under conditions of steady
ambient illumination from horizontal cells of the inner nuclear

1 .

Ringer Histidine
Control (500 MM)

I*:

-^=j

E -2-

(J -4 .

-s-

*,-
-7.

5 msec

-45-

<Rlnger Control

D .

-50-

| <500nM Histidine ON

I ^ Histidine ON
1 •

H -55-

/ \

r^*

0
1 -60-

\ / ^

<Ringer Wash

-65-

\ •

».m • clam

p = -120mV

•--Ringer Wash

0 5 10 15 20

Time (mm)

1.6-

1.4-

1.0-

0.8-

Vclamp=-12°mV

Ringer I Histidinel Ringer
Control I Wash

•a

'a 0.6
^0.4^
0.2
0.0

Figure 1. I A I Light micrograph of a 200-jj.m retinal slice from skate retina. (B) Whole-cell voltage-clamp recording train a skate horizontal cell during
10 ms steps from a holding potential of —40 mV to — 120 mV in control Ringer solution and after 2 min in 500 /j,M histidine. 1C) Fluorescence micrograph
of a skate horizontal cell recorded and stained with Lucifer \ellow in the retinal slice during whole-cell patch-clamp. iD) Time course of horizontal cell
conductance increase upon histidine application during a 100 ms step to a potential ofVc = —120 mV from a —40 mV holding potential represented by
measured voltage-clamp inward currents. After a Ringer wash, the conductance recovers. {E) Horizontal cell currents measured in histidine fn = 6) and
subsequent Ringer wash fn = 5) at Vc = —120 mV normalized to current in control Ringer. Mean ± SEM.

266

REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

layer, located below the base of the photoreceptors (Fig. 1A).
Glass capillary electrodes, pulled to a resistance of 2 to 4 meg-
ohms, were filled with a standard skate internal solution (5) and the
fluorescent dye, Lucifer yellow (0.3%). In addition, cesium chlo-
ride (204 mM) was added to suppress potassium currents. Holding
potentials of —40 mV were used, thus avoiding the transient
outward currents seen in these cells when they are held at more
negative potentials (6). This simplified the analysis of the relation-
ship between membrane conductance and photoreceptor transmit-
ter release. The preparation was superfused with a continuous
flow of skate-modified Ringer solution (5) at approximately 1 ml
per min. This could be rapidly exchanged with a Ringer solution
to which histidine (100 or 500 /j.A/1 had been added. The
higher concentration provided a faster increase in histidine con-
centration in the experimental chamber, but the ratios of the
current increases measured were found to be independent of drug
concentration.

Responses to 10 ms steps in voltage (Fig. IB) were obtained
from horizontal cells like the one shown in the fluorescence
micrograph of Figure 1C. Note the brightly stained bulbous ter-
minals suggestive of the knob-like endings observed in sections of
Golgi-stained skate external horizontal cells (6). To monitor cur-
rents during solution changes, it was convenient to hold the cell at
-40 mV and step the voltage to -120 mV. Applying histidine
(500 |uM) for 2 min produced a 40% increase in the inward current
as compared with that obtained in Ringer (Fig. IB). Using a
different protocol, in which the duration of the — 120 mV step was
0.1 s, the time course of the current changes during 500 fj.M
histidine applications and Ringer washes was measured and plot-
ted (Fig. ID). Each point represents the average of data from 3
successive steps, except for the "Ringer Control" points (square
symbols) where 9 successive steps have been averaged. The initial
increase in inward current observed in histidine approached satu-
ration in less than 3 min. When the solution was returned to Ringer
for a period of 5 min, 81% recovery was observed. A subsequent
histidine application followed by Ringer wash gave comparable
results.

Data similar to that shown in Figure ID were obtained from 6
horizontal cells, normalized to the current measured at Vc = - 120
mV in control Ringer solution, and averaged (Fig. IE). The inward
current increased 42% in histidine at Vt- —120 mV;

when returned to Ringer, the increment in current was reduced
by 72%.

Since the skate horizontal cell has no ligand-gated GABA
receptors (7), the well known effect of zinc on these receptors is
not relevant, as it is for salamander horizontal cells (2). Glutamute
receptors of skate horizontal cells have not been studied, but the
possibility that zinc is acting directly on horizontal cells to reduce
their permeability seems remote. Retinal horizontal cell glutamate
receptors have been identified as AMPA/kainite receptors (8).
although metabotropic glutamate receptors have been reported in
one case (9). AMPA/kainite receptors studied on neurons else-
where in the nervous system are generally enhanced by zinc at low
concentrations (10, 11). Similar observations have been reported
for retinal horizontal cells (12), but most studies have shown no
effect of zinc on these cells (2, 13. 14). with one exception, where
currents were reduced (15). For example, a zinc concentration of
50 \M — while high enough to block glutamate release from

salamander photoreceptors — showed no effect on horizontal cell
responses to applied glutamate (2). Similarly, it is important to
note that, as a chelating agent, histidine. which has a much higher
affinity for Zn2+ than for Ca2 + and is not membrane-permeable,
would be expected to reduce, not increase, the ERG response if it
were acting directly to reduce calcium entry needed for photore-
ceptor transmitter release ( 1 ).

The skate horizontal cell can serve as a glutamate electrode,
monitoring the amount of photoreceptor transmitter released; i.e..
an increase in photoreceptor transmitter release will be reflected in
an increase in horizontal cell conductance. With these consider-
ations in mind, we interpret the increase in membrane conductance
observed in the presence of histidine to represent an increase in
photoreceptor transmitter release. We believe that this effect is due
to the chelation by histidine of endogenous zinc. Thus, in the
presence of histidine. the inhibitory feedback process is sup-
pressed, calcium entry into the receptor terminals is increased, and
transmitter release is enhanced.

This mechanism probably represents an important component of
"neural" adaptation, comprising processes that are distinct from
those governed directly by the bleaching and generation of rho-
dopsin (16, 17). Moreover, it may well provide insight into mech-
anisms of response dynamics, such as the surround enhancement
effects observed with dynamically-modulated spots of light (18.
19), as well as phenomena referred to as suppressive rod-cone
interaction in amphibians (20), cat (21, 22), and man (23).

Supported by NIH Grant EY00777, PSC/CUNY Grants
622450031 and 632130032, as well as by an NIH/RISE (Research
Institute for Scientific Enhancement) GM60665 award to Hunter
College and by NIGMS grants 2T34 GM07823 (MARC) and R25
GM56945. Research Centers in Minority Institutions award RR-
03037 from the National Center for Research Resources of the
National Institutes of Health, which supports the infrastructure of
the Biological Sciences Department at Hunter College, is also
acknowledged. The contents are solely the responsibility of the
authors and do not necessarily represent the official views of the
NCRR/NIH.

Literature Cited

I Rosenstein, F. J., R. W. Miller, and R. L. Chappell. 2001. Imvs-

tig. Op/n/hilmnl. Vis. Sci. 42: S668.
2. Wu, S. M., X. Qiao, J. L. Noebels, and X. L. Yang. 1993. Vision

Ri-s. 33: 261 1-2616.
_V Kaneda, M., M. Mochizuki, K. Aoki, and A. Kaneko. 1997.

J. Gen. Physinl. 110: 741-747.
4 Akagi, T., M. Kaneda, K. Ishii, and T. Hashikawa. 2001. J. Hix-

locliem. CyiochfiiL 49: 87-96.

5. Qian, H., L. Li, R. L. Chappell, and H. Ripps. 1997. J. Neuro-
physiol. 78: 2402-2412.

6. Malchow, R. P., H. Qian, H. Ripps, and J. E. Dowling. 1990.
./. Gen. Physiol. 95: 177-198.

7. Malchow, R. P., and H. Ripps. 1990. Pm: Mir/. Acail. Sci. USA
87: 8945-8949.

8. Wu, S. M., and B. R. Maple. 1998. Vision Res. 38: 1371-1384.

9. Gaflta, A. C., K. S. Vogel, and C. L. Linn. 1999. Neuroscience 90:
1403-1414,

10 Bresink, L, B. Ebert, C. G. Parsons, and E. Mutschler. 1996.
Neuroplmrmacology 35: 503-509.

NEUROBIOLOGY

267

11. Lin, D. D., A. S. Cohen, and D. A. Coulter. 20(11. J. Nfiirophysiol

85: 1185-11%.
1 1 Yang. X.-L.. L. Ping. T. I.u, V. Shen, and M.-H. Han. 2001. Prog.

Brain Res. 131: 277-293.

13. Schmidt, K.-F. 1999. Neurosci. Lett. 262: 109-112.

14. Shen, Y., and X.-L. Yang. 1999. Neumsci. Lett. 259: 177-180.
I?. McMahon, D. G., D.-Q. Shang, L. Ponomareva, and T. Wagner.

2001. Prog. Brain Res. 131: 419-436.
16. Dow ling, J. E., and H. Ripps. 1970. J. Gen. P/iysiol. 69: 57-75.

18. Chappell, R. L.. K.-I. Naka, and M. Sakuranaga. 1985. J. Gen.
Phy.tiol. 86: 423-453.

19. Chappell. R. L. 2001. Prog. Bruin Res. 131: 177-184.

20. Frumkes, T. E., and T. Eysteinsson. 1987. ./. Neurophysiol. 57:
1361-1383.

21. Pflug, R.. R. Nelson, and P. K. Annelt. 1990. J. Neurophysiol. 64:
313-325.

22. Nelson, R.. R. Pflug, and S. M. Baer. 1990. J. Neumphysiol. 64:
326-340.

17. Green, D. G., J. E. Dowling, I. M. Siegel, and H. Ripps. 1975. 23 Frumkes, T. E., G. Lange, N. Denny, and I. Beczkowska. 1992.

J. Gen. Phvsiol. 65: 483-502.

Vis. Neurosci. 8: 83-89.

Reference: Biol. Bull 201: 267-268. (October 2001)

Polarization Reflecting Iridophores in the Arms of the Squid Loligo pealeii

Nadav Shashar (Hebrew University, Internniversity Institute for Marine Sciences, P.O. Box 469. Eilat 88103,
Israel), Douglas T. Borst\ Seth A. Ament\ William M. Saidel2, Roxanna M. Smolowitz1, and Roger T. Hanlon1

Distinct polarization body patterns have been recorded in cephalo-
pods. In cuttlefish (Sepia officinalis) and squid (Loligo pealeii) these
patterns are postulated to constitute a discrete communication channel
that may be "hidden" from some of their predators ( 1 . 2). In squid, the
patterns of polarization are most prominent as long, narrow stripes
along the arms (3). Examination of the skin of L pealeii has now
revealed very localized rows of iridophore cells that are reflecting and
polarizing incident light and thus producing these patterns. Topical
application of acetylcholine (ACh) to the arms of L pealeii induced a
change in the polarization reflection, as in other squid species (4).
Moreover, silver staining and acetylcholinesterase histochemistry
suggest that these iridophores are under direct neural control, unlike
any known cephalopod iridophore.

Reflection and polarization of incident light by squid iri-
dophores is accomplished by layers of intracellular platelets that
are positioned parallel to each other (5). The spectrum (color) of
the reflection can change from red/pink to blue and depends upon
the distance between platelets, the orientation of the platelets, and
the direction of viewing (6. 7). In squid dermis. iridophores have
been found heretofore only beneath the layer of chromatophores
(4). Iridophores are found in many parts of squid skin, but in most
species they are especially abundant on the mantle. Because the
polarization patterns in Loligo pealeii are created within very
localized areas on the arms (Fig. IB), we examined the skin in
those areas and looked for structures that could potentially reflect
light to produce polarization patterns.

For in vitro examination, pieces of fresh skin containing the
polarizing sections were stretched to original size onto a paraffin-
coated petri dish filled with chilled filtered seawater. The tissue
was then examined with a Zeiss SVII dissecting microscope
equipped with a polarization indifferent digital camera, under
depolarized epi-illumination. and with a rotating linear polarizing
filter (Polaroid HN38S) installed in the outgoing light path. Three

1 Marine Resources Center. Marine Biological Laboratory. Woods Hole.
MA 02543.

: Dept. of Biology. Rutgers, the State University of New Jersey. Cam-
den. NJ 08102.

consecutive images were then taken with the filter set at preset
angles (arbitrarily defined as 0°. 45°. and 90°). The images were
then analyzed with custom-made software, and the polarization
characteristics of the reflected light were determined.

For morphology, arms were preserved in 10% formalin in buff-
ered seawater for 3 d and, after washing, they were cut. mounted,
and transferred to 70% ETOH. Samples were then sectioned at
40-200 p.m intervals and stained with Mayer's hematoxylin and
eosin, Masson's trichrome, and silver (Holmes' silver nitrate meth-
od). Sectioning and processing arm tissue for acetylcholinesterase
histochemistry was done according to the method of Mesulam and
Van Hoesen (8). using the acetylthiocholine medium specified by
Geneser-Jensen and Blackstad (9). Images were then observed
with a Zeiss Axioplan microscope equipped with an internal scal-
ing and calibration system.

Strong partially linearly polarized reflection could be identified in
specific lines along the animals' arms (Fig. 1A-D) and was often
associated with physical colors such as blue or pink. Microscopic
examination of skin tissue at these locations revealed the existence of
a new type of iridophore. These reflecting cells were located in very
narrow areas of the skin, 60 ± 26 /urn (/; = 24) underneath the skin
surface, and organized as long stripes. Cell length was 267 ±131 /J.m
(n = 22). and cell width was 14.8 ± 7.2 /xm (H = 26). These long
stripes of iridophores are consistent with the red and highly polarizing
iridophores reported by Mathger and Denton (7), but the squid arm
iridophores are much narrower. Unlike other squid iridophores. which
are found beneath the chromatophore layer, these cells were situated
above the chromatophores (Fig. IE). Platelets [1.8 ± 1.2 /urn wide
and 14.1 ± 6.4 /j,m long (n = 138)] were set in an angle inside the cell
and were organized parallel to each other with a variability of 7.9° ±
4.0° (n = 25). Inter-platelet space was 1.2 ± 0.7 ^m (n = 150).
providing for an average density of 30.8 ± 9.0 platelets per 100 /J.m.

Previous studies have never furnished evidence of innervated
squid iridophores (4). This is surprising considering the speed with
which changes in color — even iridescent color — occur in cepha-
lopods. Hanlon et al. (4) found that iridophores in the squids
Lolligiinciilu hrevis and Loligo p/ei became iridescent when
treated with ACh, but no nerve fibers were found adjacent to or

268

REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

Figure 1. An arm of a squid us MPH; m normal light lA); through a linear polarizer set ul 45 t<- the orientation of maximal polarized reflection (B):
as a black and white image of A, which is presumably what a color-blind, polarization-insensitive predator would see (C): and when % polarization is coded
into saturation (the color image), and orientation of polarization is encoded into hue (the scale) — note that the polarization reflection is very localized into
a specific stripe along the arm (D). Light microscopy of cross sections in squid arms with H&E staining (E) shows iridophores (short arrows) above the
chromatophores (long arrow), which is a novel arrangement. In (F). a DIC image of a silver-stained section which indicates potential nerve fibers (short
arrows) immediately adjacent to or on an iridophore (long arrow). In (G), acetylcholinesterase staining (short arrows) adjacent to an iridophore cell (long
arrow) also indicates potential locations of innervation.

near the iridophore cells. They therefore surmised that ACh would
diffuse to the iridophore cell surfaces, would bind to ACh recep-
tors there and thus would induce an ultrastructural change in the
platelets to produce iridescence.

However, polarization patterns on cephalopods change in just a
second or two, suggesting neural rather than hormonal control.
Topical application of 10 ' M ACh to isolated skin patches
induced polarization reflections. Silver staining revealed nerve
fibers in very close proximity to the iridophores (Fig. IF), sug-
gesting that these cells may be innervated directly. Finally, stain-
ing for acetylcholinesterase revealed specific active areas at the
attachment of potential nerve fibers to the iridophores (Fig. 1C).

Our results present quite a different cellular structure — and
potential control mechanism — in which a polarization pattern is
produced in the arms of squid. The control of these structures, and
the significance of polarization patterns to squids, remain to be
investigated.

We thank Michael Mitchell for sectioning some samples, John
Messenger and William Kier for evaluating some microscopic
tissue samples, and Louis Kerr and Rudi Rottenfusser for micros-

copy assistance. This study was sponsored by NSF grant 1BN
9722805, BSF grant 1999040, and an MBL fellowship to NS.

Literature Cited

I Hanlon, R. T., M. R. Maxwell, N. Shashar, E. R. Loew, and K.-L.

Boyle. 1999. Hiol. Hull. 197: 49-62.
2. Shashar, N., P. S. Rutledge, and T. W. Cronin. 1996. J. Ev/> Biol.

199: 2077-20X4.

1. Shashar, N., and R. T. Hanlon. 1997. Biol. Bull. 193: 207-208.
4 Hanlon, R. T., K. M. Cooper, B. U. Budelmann, and T. C. Pappas.

199(1. Cell Tissue Res. 259: 3-14.
5. Mirow, S. 1972. Z Zellforsch. Mikrosk. Anal. 125: 176-190.

6 Cooper, K. M., R. T. Hanlon, and B. U. Budelmann. 1990. Cell
Tissue Res. 259: 15-24.

7 Mathger, L. M., and E. J. Denton. 2001. J. E.v/>. Biol. 204: 2103-
21 IX.

8. Mesulam, M. M., and G. W. Van Hoesen. 1976. Brain Res. 109:
152-157.

9. Geneser-Jensen, F. A., and T. W. Blackstad. 1971. Z. ZC///..M, h
Mikrosk. Anal. 114: 460-481.

NEUROB1OLOGY

269

Reference: #/<•/. Bull. 201: 269-270. (October 2001)

Cuttlefish Cue Visually on Area — Not Shape or Aspect Ratio — of Light Objects in the Substrate
to Produce Disruptive Body Patterns for Camouflage

Clnitm-Chin Chiao1 and Roger T. Hanlon (Marine Biological Laboraton\
Woods Hole, Massachusetts 02543)

Cephalopods have at least 20 body patterns for camouflage, yet
these can be organized into four categories: uniform, stipple,
mottle, and disruptive ( 1 ). Among them, disruptive coloration is
probably the most striking because it breaks up the animal's body
outline by visual deception (2). Cuttlefish produce (by direct neural
control of chromatophoresl an array of white skin components that
produce a disruptive coloration on their bodies, and this helps them
achieve camouflage as it is defined by Endler (3). "A colour or
pattern is cryptic if it resembles a random sample of the visual
background as perceived by the predator at the time and place at
which the prey is most vulnerable to predation." The so-called
"White square" on the dorsal mantle of cuttlefish represents a
random sample of white background objects (Fig. 1) that are
common in marine habitats, thereby distracting the attention of
visual predators away from the body outline (2). How do cuttlefish
"decide" to switch to disruptive coloration, and what sensory cues
are involved? We developed a non-invasive assay that monitors
motor output (i.e.. the body pattern of the cuttlefish) resulting from
different visual inputs (computer-generated artificial substrates).
Although many aspects of cephalopod vision are known (4), little
is known about the visual features of the substrate that elicit
disruptive coloration. A recent study (5) of young cuttlefish. Sepia
pharaonis. showed that the size, contrast, and number of white
squares on a black background are the main visual features that
cause cuttlefish to switch from general resemblance of the sub-
strate to disruptive coloration. In this study, we examine the shapes
and aspect ratios of white objects on black backgrounds that lead
cuttlefish to show disruptive coloration.

Five young cuttlefish. Sepia pharaonis (8-10 cm mantle length.
10 weeks old), were reared from eggs in the laboratory of the
National Research Center for Cephalopods (University of Texas
Medical Branch. Galveston) and were maintained in the Marine
Resources Center at the Marine Biological Laboratory. Woods
Hole, Massachusetts. Each animal was placed in a running seawa-
ter tank (25 cm X 40 cm x 10 cm) and was restricted by a
four-wall divider (inside covered by black cloth to prevent light
reflection) to an area (20 cm X 26 cm) where various computer-
generated backgrounds (laminated to be waterproof) were pre-
sented as the substrate. Acclimation to the tank was gauged by the
cessation of excessive swimming and hovering movements and by
the chronic expression of a stable body pattern. A digital video
camera was used to record the body patterning of S. pharaonis
over a period of 30 min (i.e., record 2 s for every 1-min interval;

1 Howard Hughes Medical Institute. 50 Blossom Street. Wellman 429,
Massachusetts General Hospital. Boston. MA 021 14.

total 60 s for each cuttlefish on each substrate). Although cuttlefish
cannot perfectly match backgrounds that are completely artificial,
they do show various grades of disruptive patterns based on certain
visual features of these substrates. Thus, it was possible to quantify
the body patterns corresponding to the shapes or areas of the white
objects in the black background. A simple system for grading
patterns was used to assess an animal's responses to different
substrates (see Ref. 5 for details). The assigned grades were: 1 =
uniformly stippled pattern; 2 = indistinct pattern; 3 = disruptive
pattern. Grading was conducted by playing the videotape and
assigning a grade ( 1-3: whole integers only) every 10 s. Since all
tapes were 60 s long, six grades were assigned for each animal on
each substrate. The combined mean values (and overall standard
deviation) of all animals were plotted in Figure 1.

Six different shapes of medium-sized white objects (same area,
1.53 cm2) were tested to determine whether they would elicit
disruptive coloration (Fig. 1). Two control images were also used:
large circle and large square (same area, 13.80 cm2), which are too
large to elicit the White square in the cuttlefish. The generation of
disruptive or uniform skin patterns in the cuttlefish did not depend
on the shape and aspect ratio of white objects (Fig. 1 ). Although
shapes with equal aspect ratio (i.e., circle, hexagon, pentagon,
square, and triangle — all generally similar to the shape of White
square on the mantle) did not affect the display of a disruptive
body pattern, we were surprised that the elongated oval shape also
elicited the White square and disruptive coloration. This indicates
that cuttlefish may integrate the whole area of white objects to
determine the display of disruptive coloration, regardless of the
shapes and aspect ratios of white objects.

Cuttlefish live in much more complex environments than these
computer-generated backgrounds, and the ability to display appro-
priately camouflaged body patterns is critical to survival of this
soft-bodied creature. We are gradually learning how cuttlefish use
certain features of the visual background to decide upon the type
of camouflage that will avoid detection by predators. The sophis-
ticated skin of Cephalopods provides a novel system with which to
study visual perception and decision-making (6). Further studies
should be aimed at exploring these processes on more natural
backgrounds.

We thank Janice Hanley, Bill Mebane, James Carroll. Hazel
Richmond, and Nicole Gilles for help with cuttlefish rearing and
maintenance. CCC is grateful for the support from Richard Mas-
land of Harvard. This paper is dedicated to Ellen Grass (the Grass
Foundation), who staunchly and enthusiastically supported young
scientists studying all aspects of neurobiology and behavior.

270

REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

Figure 1. Two Itir^e control iniut;c\ In. hi on which the ciitllcfisli Icenreredl did nut show its White square. For /ire imut'es <c-x). cuttlefish were
expected to— mid did— show White si/mire disruptive coloninon. On ,mc inui^e (In they were nut expected to elicit White square due to its highly different
nspect ruliii. vet tliev iliJ. fi) A summary «l icsulls train nil ci^ln minxes. Futtcrninx xrujc J is disruptive. The number enclosed in parentheses indicates
the number of cuttlefish tested. Kesulls o/ the first two iimi^cs \\-ere sixniticiintlv ditlerent from the rei>niinint> si.\ iimiKi's fP < 0.00001).

Literature Cited

1. Hanlon, R. T., and J. B. Messenger. 1988. Philns. Trans. R. Soc.
hn,d. R 320: 437-487.

2. Hanlon. R. T.. and J. B. Messenger. 1996. Cephulapud Bc/uivioni
Cambridge University Press. Cambridge.

3. Endler, J. A. 1991. Pp. 169-196 in Behavioural Ecology. An Emhi-
lionurv Approach. ). R. Krebs and N. B. Davies. eds. Blackwell Sci-
entific Publications. Oxford.

4. Messenger. J. B. 1991. Pp. 364-397 in Evolution of the Eye imd

\'ismil Svstem. ). R. Cnmly-Dillon and R. L. Gregory, eds. Macmillan
Press. London.

5. Chiao, C.-C., and R. T. Hanlon. 2110 1. ./. £v/>. Biol. 204: 2119-
2125.

6. Packard, A. 1995. Pp. 331-367 in Cephalopod Neurobiology, N. J.
Abbott. R. Williamson, and L. Maddock, eds. Oxford University Press.
New York.

NEUROB10LOGY

271

Reference: Kixl. Bull. 201: 271-272. (October 2001 )

Visually Guided Behavior of Juvenile Horseshoe Crabs

M. Errigo1, C. McGuiness2, S. Meadors3, B. Mittmann4, F. Dodge^, and R. Barlow

(Marine Biological Laboraton: Woods Hole, Massachusetts 02543)

The horseshoe crab. Liinnlus polyphemus, has long been an
admirable model for vision research. More than 70 years of re-
search on the physiological properties of the Limulus lateral eye
have uncovered fundamental mechanisms of visual function com-
mon to many animals, including humans (1. 2). Less attention has
been given to the role of the lateral eyes in the animal's behavior.
Initial field studies showed that adult males use their eyes to rind
mates, whereas adult females avoid mate-like objects (3). Our
attempts to study these behaviors in the laboratory were not
successful because adults do not exhibit them in captivity (R.
Barlow, pers. obs.). We therefore turned our attention to juvenile
l.imulus and report here an investigation of their visually guided
behavior both in the field and in the laboratory.

We first studied visually guided behavior of juvenile crabs on
tidal flats (0.3-1 m depth) of the North Monomoy Island Wildlife
Refuge, Chatham, Cape Cod. Massachusetts. Because juvenile as
well as adult animals are most active on the submerged flats during
high tides, we restricted our observations to these periods. We
selected 1 -year-old juveniles, born in the spring of 2000 (stages VI
to X; prosoma widths: 16-39 mm). Their compound lateral eyes
contain from 500 to 600 ommatidia, or about half the number of
the adult eye. When a moving juvenile crab was located, we placed
a high-contrast cylindrical object (7.6 cm diameter; 15 cm high) on
the bottom 1 5 to 45 cm in front of the animal. Twenty-three of the
26 animals tested changed direction and avoided the object; the
other 3 stopped and buried themselves. A low-contrast, gray object
of the same size and placement evoked avoidance behavior in 14
of 20 animals. Five animals continued straight and hit the object.
and one stopped and buried itself. Most animals appeared to
respond to objects placed in front of them because they could see
them, with the black object being more visible than the gray one.
However, we could not eliminate the possibility that they detected
a disturbance in the water when the objects were placed in front of
them.

To examine the visually guided behavior of juveniles under
more controlled conditions, we placed them in shallow seawater
troughs (40 cm x 50 cm; 3 cm water depth: 2 cm sand depth)
under ambient diurnal lighting in the Marine Biological Labora-
tory, Woods Hole, Massachusetts. To test the animals, we trans-
ferred 10 of them to a trough of the same size, containing seawater
(3 cm depth) but no sand. The lack of sand prevents them from
burying themselves and enhances their locomotor activity. We
simulated the illumination of an overcast cloudy day by reflecting
light from a white diffusing surface located above the trough. The

1 Boston University Marine Program.

2 Syracuse University.

3 University of South Carolina.

4 Humboldt-Universitat. Berlin.

5 SUNY Upstate Medical University.

level of illumination at the water's surface was 1.0 cd/trr. After
giving the animals time ( — 30 min) to acclimate to the new trough
without sand, we videotaped their behavior in the vicinity of a high
contrast (black) cylindrical object (diameter: 6.5 cm) placed in the
center of the trough. After 5 min, the object was removed for 5 min
or replaced with a transparent object of the same size. This
sequence of 5-min test intervals was repeated for about 1 h, and
then the animals were returned to their sand-filled troughs. We
digitized the video recordings at 2 frames/sec and traced the paths
of animals on a transparent sheet attached to the monitor. Using
NIH Image software, we also measured their "distance of closest
approach" to the objects (4). The distance at which animals began
to turn from the object ("turning distance," see Ret". 4) was difficult
to judge with precision and was therefore not measured. All
animals appeared about equally active. We did not track specific
individuals.

Figure 1 shows that juvenile horseshoe crabs avoided the black
object (a), but not the transparent one (b). Only twice did an animal
contact the black object, whereas animals contacted and circled the
transparent object many times. In the absence of an object, they
moved about through all areas of the trough. For purposes of
demonstration, we show only 30 traced paths for each of the two
conditions containing targets. Each traced path begins and ends
near the sides of the trough because we did not include the
animals' movements along the sides of the trough.

The "distance of closest approach" of an animal to an object
provides a measure of how well they can see (4). This is the
distance from the center of the animal to the center of the object
(4). NIH Image was used to make these measurements. The data in
Figure 1 (a) yield an average distance of closest approach of 1 1.4 ±
3.1 cm (n = 30). The greatest distance was 18.3 cm, suggesting
that the crabs can see the black object at this distance. The
transparent object (Fig. Ib) is apparently invisible to the animals
because they run into it. After contacting the object, they tend to
circle it one or more times before moving away.

This study presents the first evidence that juvenile horseshoe
crabs can see. Their avoidance of high-contrast objects is similar to
that observed for adult females (4). The previous study suggested
that adult females migrating to shallow waters to build nests search
for unoccupied areas and thus avoid dark, female-size objects.
Juveniles may turn away from dark objects because they represent
potential predators. Adult males avoided dark objects only if they
were held overhead (R. Barlow, pers. obs.). Adult males may view
overhead objects as predators in the same way that juveniles view
dark objects on the bottom. We cannot relate the specific behaviors
we observe to the sex of the 1 -year-old juveniles because their sex
is not known. Horseshoe crabs acquire external sexual features
when they reach maturity, at about six years of age. Perhaps all the
juveniles we tested were females and avoided high-contrast objects
as they do in adult life. Or the juveniles we tested could have been

272

REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

Figure 1. (a) Thirty tracings of the paths of 10 juvenile Limulus in a
trough containing a high-contrast black cylinder. This figure illustrates the
avoidance behavior of the juvenile crabs in the presence of a high-contrast
object. The small black spot represents the relative size of a crab, (b) Thirty
tracings of the paths of 10 juvenile Limulus in a trough containing a clear
cvlinder. This figure illustrates the inability of the crabs to detect a
low-contrast object visuallv. Again, the small black spot represents the
relative size of a crab.

a mix of males and females, and not yet at the time in life when
males change their response to visual objects in front of them from
avoidance to attraction.

Supported by grants from the National Science Foundation.
National Eye Institute and the National Institutes of Mental Health.
C. McGuiness and S. Meadors received REU Fellowships from the
National Science Foundation.

Literature Cited

1 . Ratlif'f, F. 1974. Studies on Excitation and Inhibition in the Retina.
The Rockefeller University Press, New York.

2. Barlow, R. B., J. M. Hitt, and F. A. Dodge. 2001. Bio/. Bull. 200:
169-176.

3. Barlow, R. B., L. C. Ireland, and L. Kass. 1982. Nature 296:
65-66.

4 Powers, M. K., R. B. Barlow, and L. Kass. 1991. Visual Neurosci.
7: 179-189.

Reference: Bio/. Bull. 201: 272-274. (October 2001)

Growth, Visual Field, and Resolution in the Juvenile Limulus Lateral Eye

5. Meadors1, C. McGuiness2, F. A. Dodge*, ami R. B. Barlow3
(Marine Biological Laboratory, Woods Hole, Massachusetts 02543)

When a trilobite larval Limulus hatches from an egg, it begins to
forage with the locomotor abilities of an adult but not with the
vision of an adult. Its lateral eyes have fewer than 2% of the
photoreceptors possessed by an adult. Our understanding of the
way an adult horseshoe crab sees its environment is now suffi-
ciently advanced that its visual processes can be mathematically
modeled ( 1 ). Guided by this model, we have set out to examine

1 University of South Carolina.

2 Syracuse University.

3 SUNY Upstate Medical University.

how the lateral eye and visually guided behavior develop. Here we
report on how photoreceptors, or ommatidia, are added during
development, and how the eyes of juvenile crabs sample visual
space.

We collected eggs from nests on tidal flats of Cape Cod,
Massachusetts, from June 4 to 12. 2001 and maintained them
under natural diurnal lighting in Petri dishes in the laboratory. In
3-4 weeks the eggs hatched into trilobite larvae, or Stage I crabs
(3 mm wide), and 4 weeks later the larvae molted into Stage II
crabs (5.5 mm wide). To analyze the ommatidial array at both
stages, we photographed the eyes with a Zeiss Axiocam attached

NEUROB1OLOGY

273

Figure 1. Scale drawings of a lateral eye of a Stage I Limulus before (left) and after molting to Stage II (right). The solid lines denote the borders
of the eyes, anil the ovals indicate the si-e and location of ommatidia. In both drawings, the apex to the left, with the largest ommatidia. is posterior. The
dashed line marks the visible division benveen the older Stage I and newer Stage II regions of the e\e. Scale bar is 50 fun.

to an Axioplan II compound microscope. The array is clearly
distinguishable in some parts of the eye but is partially obscured in
others by retinal pigmentation. We analyzed photographs taken at
various eye orientations to resolve and reconstruct the arrays of
lateral eyes in Stage I and II crabs.

Lateral eyes of trilobite larvae (Stage I) approximate an equi-
lateral triangle (100-120 /im on a side) containing 14 to 17
ommatidia (Fig. 1). We observed a gradient of ommatidial diam-
eters within the array, with the largest ommatidium at the posterior
apex, and the smallest at the base. In the larval eye in Figure 1
(left), the diameters of the largest and smallest ommatidia are 26
p.m and 15 fim. respectively.

When trilobite larvae molt to Stage II crabs, another triangular
array of ommatidia is added to the anterior edge of the lateral eyes
(Fig. 1, right). The eyes of two Stage II crabs yielded counts of 29
and 33 ommatidia. or about twice the number in the eyes of
trilobite larvae. The new triangular array now has an apex pointing
anteriorly, and a partition between Stage I and Stage II ommatidia
is distinguishable (dashed line in Fig. 1 ). Corneal lenses are visible
in Stage II. but not in Stage I, suggesting that trilobite larvae have
no directional vision. Indeed, we do not know whether either Stage
I or II crabs can see.

Juvenile horseshoe crabs (stages VI [16 mm wide] to X [40
mm] ) were also collected on tidal flats of Cape Cod from June 4 to
12, 2001. We maintained them under diurnal lighting in shallow
troughs in the laboratory. They were fed. and their water was
changed weekly. To assess the growth of their lateral eyes, we
placed five scars along the anterior and ventral edges of the cornea
with a sharp metallic needle (diameter 50 ju,m). Using a Zeiss
SV11 stereoscope, we photographed the scarred eyes before and
after each animal molted. We also photographed their molted
shells to supplement the original records of their eyes.

To assess the visual field of the juvenile eye, we adapted the
method of Herzog and Barlow (2). With the high magnification of
a stereomicroscope. we identified the ommatidium whose optic
axis was aligned with that of the microscope. By changing the
orientation of the molt, we measured the optic axes of numerous
ommatidia and determined the visual field of the eye as well as its

resolution in various parts of the visual field. We analyzed the
growth of the lateral eye at various stages and found, as others
have, that the eye adds ommatidia at each molt (3. 4).

When a Stage IX crab (30 mm wide) molted to a Stage X crab
(38 mm), its right lateral eye increased from 1.8 to 2.2 mm along
the anterioposterior axis, adding approximately 90 ommatidia (490
to 580) in agreement with morphometric data of Waterman (3).
The diameter of ommatidia in the medial and posterior regions of
the eye increased from 64 p.m to 78 ju.m. Scars along the anterior
edge shifted posteriorly, revealing the addition of 5 columns of
about 90 small ommatidia (52 /urn in diameter). This result sup-
ports previous observations that the lateral eye grows by adding
new photoreceptors at its anterior edge (4). A similar result was
reported for the dragonfly eye using the same scarring technique
(5). Curiously, the ventral scars moved dorsally by about 5 om-
matidial diameters. This movement is not associated with the
addition of new ommatidia because the number of ommatidia
medial and posterior to the scars was the same in both the molt and
the crab. Apparently the outer scarred cornea of the Stage X crab
had not grown as much as the underlying matrix of lens facets.

A juvenile crab has about the same visual field as an adult, but
samples it differently. This can be demonstrated by locating the
unique "index" ommatidium, which is the ommatidium with its
optic axis horizontal and normal to the body axis of the crab. It is
located near the center of the adult eye, but in a more posterior
position in juveniles. The younger the crab, the more posterior is
the location of the index ommatidium. For example. 22% of
ommatidia in a Stage VIII eye lie posterior to the index ommatid-
ium. whereas 35% do in a Stage XII eye. Consequently, juveniles
sample the anterior region of visual space with a greater proportion
of ommatidia than an adult eye. However, they do so with about
half the horizontal resolution (0.05 cycles/deg. Ref. 2) of an adult
because they possess fewer columns of ommatidia. On the other
hand, juveniles have about the same vertical resolution as an adult
(0.1 cycles/deg above, and 0.2 cycles/deg below the horizon)
because they possess vertical columns with about the same number
of ommatidia (23 to 26) as an adult.

Pulsatile growth of the eye at the anterior edge modifies its view

274

REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

of the world after each molt. That is, ommutidia viewing the most
anterior region of the animal's visual space now sample a more
lateral region of visual space. As a consequence, the retinotopic
map in the brain must undergo comparable rearrangements to
accommodate inputs from new ommatidia sampling visual space
in front of the animal. The retinotopic map has been determined for
adult animals at the first two synaptic layers in the brain (6). but
that of the juvenile remains to be studied. The retinotopic map
must be plastic to accommodate the changing retinal mosaic as the
eye grows.

Supported by grants from the National Science Foundation,
National Eye Institute, and the National Institutes of Mental
Health. C. McGuiness and S. Meadors received REU Fellow-
ships from the National Science Foundation. We thank the

Monomoy National Wildlife Refuge. Morris Island, Chatham,

Massachusetts.

Literature Cited

1. Passaglia, C. L., F. A. Dodge, and R. B. Barlow. 1998. J. Neuro-
/>/iv.v/,)/. 80: 1800-1815.

2. Herzog, E. H.. and R. B. Barlow. 1992. Vis. Neumsci. 9: 571-580.

3. Waterman, T. H. 1954. J. Morphol. 54: 125-158.

4. Marler. J. J., R. B. Barlow, L. Eisele, and L. Kass. 1983. Binl. Bull.
165: 541

5. Sherk, T. E. 1978. J. E.\p. Zool. 203(2): 183-200.

6 Chamberlain, S. C., and R. B. Barlow. 1982. ./. Neurophysiol. 48:
505-520.

Reference: Biol. Bull. 201: 274-276. (October 2001)

An Initial Study on the Effects of Signal Intermittency on the Odor Plume Tracking Behavior

of the American Lobster, Homarus americanus

Corinne Kozlowski\ Kara Yopak1, Rainer Voigt2 (Boston University Marine Program,
Woods Hole, Massachusetts) and Jelle Atema2

Chemical signals are used by organisms for communication and
location of food, mates, and shelters. The spatial and temporal
distribution of these signals is shaped primarily by environmental
conditions Turbulent odor dispersal causes intermittency in the
chemical signal even when the source emits continuously. There-
fore, animals that use chemical cues to localize odor sources must
overcome signal intermittency. The ability of lobsters to track
continuously released odor plumes has been well-described ( 1 ).
Lobsters may be using one or a combination of two possible
mechanisms to locate an odor source: (a) odor-gated rheotaxis.
which would cause the animal to move upstream, using the mean
current for orientation once a chemical signal is detected (2), and
(b) eddy-chemotaxis, which would require an animal to use the
internal chemical and hydrodynamic fine structure of an odor
plume to locate the source (3). The mechanisms lobsters use to
overcome signal intermittency are still unknown. Male moths use
a sequence of upwind surges and horizontal casting to locate a
female releasing pheromone (4). In the presence of an odor source,
tsetse flies perform a series of overshoots followed by 180° turns
until they come within 1 in of the source (5). In more turbulent
odor plumes, blue crabs decrease their locomotor activity and stop
and turn more frequently (6). Here we explore how lobsters track
odor plumes with a controlled increase in intermittency.

Lobsters (Homarus uniericanus). ranging in carapace length
from 77.5 mm to 98.5 mm. were caught locally and kept in
separate holding tanks with running seawater. Twice weekly the
animals were fed about 2 g of squid. As in previous studies, this
small amount was thought to increase their motivation to track an

' Bowling Green State University. Laboratory for Sensory Ecology.
Bowling Green, OH 43403-0212.

2 Boston University Marine Program. Marine Biological Laboratory,
Woods Hole. MA 02543.

odor source, consisting of 100 ml squid rinse/1 seawater. Each
lobster was tested in a flume ( 1 .8 m X 5.5 m X 0.5 m experimental
arena) with a mean flow rate of 4.5 cm/s. Each lobster was
blindfolded; a white dot of nail polish on the carapace served as a
reference point to digitize the track. After a 20-min acclimation
period, each lobster was placed into a shelter 6 m downstream
from a jet source releasing odor at 100 ml/min through a nozzle
with a 2-mm ID (Re = 200). The trial began once the lobster began
exhibiting tracking behavior (antennule flicking and antennae wav-
ing) in the downstream patch field as visualized with dye; it ended
once the animal was less than one body length away from the
source, or after 20 min. Animals that tracked a continuous jet
plume (Fig. 1A) were randomly tested with odor pulses (1 cm in
length) with gaps of about 5-cm (Fig. IB) and 10-cm (Fig. 1C)
between them. Dye visualization showed that interpulse gaps were
maintained for 2 to 3 m from the source; farther downstream, the
pulses merged due to turbulent dispersal in the flume. Fresh
seawater entered into the flume during each trial to minimize odor
accumulation, and the flume was drained and refilled each night.
All trials were videotaped and digitized using the Metamorph®
Imaging System (Version 3.5. Universal Imaging Corporation) for
analysis. Walking speed, heading, and turning angles were then
calculated for each track.

Under all three plume conditions (continuous, pulsed with 5-cm
gaps, and pulsed with 10-cm gaps), heading and turning angles
remained constant with distance from the source for the successful
tracks; 84.5% of heading angles ranged from -40° to 40°, and
86.9% of turning angles ranged from -20° to 20°. Source local-
ization success decreased with increasing gap length: 10 out of 33
lobsters successfully tracked the continuous jet plume (Fig. 1A).
while 7 of these 10 tracked the plume with 5-cm gaps (Fig. IB),
and only 4 tracked the plume with 10-cm gaps (Fig. 1C). Gener-
ally, lobsters emerged slowly from the shelter area (orientation

NEUROBIOLOGY

275

100

*- Q)

(0 C

in ra

O tS

O =5 -50 •

-100

100 200 300 400

Distance from source (cm)

500

-100

100 200 300 400

Distance from source (cm)

500

100

-100

100

200 300 400

Distance from source (cm)

500

Figure 1. Individual tracks under plume conditions with continuous release (A, N = 10) and deliberate gaps of 5 cm (B. N = 7). and 10 cm
(C. N = 4). Solid lines indicate approximate boundaries of the odor plume, as visualized with dve. Odor source is located at (\ = 0 cm, y = 0 cm, z =
9 cm); shelter is located at (\ = 550 on, y = 0 cm).

phase. 3 to 6 m from the source). Under continuous plume condi-
tions (Fig. 1A), walking speed then increased during the subse-
quent tracking phase (1 to 3 m) and decreased again as the animal
approached the odor source (0 to 1 m). During the tracking phase
and final approach, lobsters mostly stayed within the plume bound-
aries (20% of animals spent more than 5 s outside of the suggested
plume area). Overall, the 10 lobsters that successfully tracked
under continuous plume conditions (Fig. 1A) seemed to show a
straighter approach to the source (higher linearity index; 68% had
a linearity index equal to or greater than 0.9) than when tested

under intermittent plume conditions (Fig. IB. C) (57% had a
linearity index equal to or greater than 0.9). In contrast, in
plumes with deliberate gaps (Fig. IB, C), walking speed re-
mained constant with distance from the source and increased
during the final approach; 81.8% of animals walked outside of
the plume boundaries for at least 5 s during the tracking and
final approach phases. Mean walking speed decreased with an
increase in gap length. Most lobsters that did not locate the odor
source in plumes with deliberate gaps walked along the wall or
did not leave the shelter. Two lobsters showed tracking behav-

276

REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

ior downstream from the odor source (2) but then lost the odor
plume upstream in the jet field.

These results suggest that successfully tracking lobsters use
similar walking paths independent of signal intermittency.
Counter-turning or casting behavior as described for moths (4) was
rarely observed. However, tracking success dropped with increas-
ingly intermittent signal conditions. It appears that lobsters require
a minimal signal encounter rate to continue tracking the plume
successfully to the source. The fact that lobsters stayed mostly
within the odor plume boundaries further suggests that they use its
internal fine structure for guidance.

First and second authors are listed alphabetically; both authors
contributed to the experiment equally and in the same manner.

This study was supported by NSF REU Grant (OCE-0097498) to
CK and KY, and ONR Grant (N00014-981-0822) to JA.

Literature Cited

1. Moore, P. A., N. Scholz, and J. Atema. 1991. J. Client. Etol. 17:
1293-1307.

2. Baker. C. F., and J. C. Montgomery. 1999. Polar Biol. 21: 305-
309.

3. Atema,.). 1998. Bioi Bull. 195: 179-180.

4. Vickers, N. J., and T. C. Baker. 1996. J. Com/.. P/iv.vio/. .4 178:
831-847.

5. Bursell, E. 1984. Physial. Entomol. 9: 133-137.

6. Weissburg, M. J., and R. K. Zimmer-Faust. 1994. J. £v/>. Biol.
197: 349-375.

Reference: Bial. Bull. 201: 276-277. (October 2001)

Cholinergic Modulation of Odor-Evoked Oscillations in the Frog Olfactory Bulb

Benjamin Hall (Marine Biological Laboratory, Woods Hole, Massachusetts 02543) and Kerry Delaney1

The vertebrate olfactory bulb (OB) receives sensory information
from peripheral odorant receptors and transmits this information to
other cortical regions. OB output is encoded in the spiking patterns
of the primary OB neurons — the mitral and tufted mitral cells
(MTCs) — which project directly to higher cortical centers. The
activity of the MTCs is determined both by patterns of odorant
receptor activation and by interactions with intrinsic inhibitory
interneurons within the OB. OB output is thus shaped by the two
major classes of interneurons: the periglomerular cells (PCs) and
the granule cells (GCs). GCs make distributed reciprocal dendro-
dendritic synaptic contacts along the secondary dendrites of MTCs
and. via GABA release, provide both feedback and feedforward
inhibition of the primary neurons ( 1 ). In addition, these reciprocal
circuits are thought to be the site of generation of odor-evoked
oscillations in the OB. In vertebrates, including frogs, GC den-
drites receive prominent cholinergic innervation from the basal
forebrain. mediated in the GC layer by muscarinic acetylcholine
(mACh) receptors (2). Although studies have investigated the
effects of ACh modulation in the OB in response to nerve activa-
tion and at the MTC to GC synapse in slice preparations, the effect
of mACh agonists on natural odorant-evoked oscillations in the
OB is unknown (3, 4).

We examined odor-evoked oscillatory responses in the frog
olfactory bulb using an in vitro nose and brain preparation, in
which we can maintain intact the olfactory circuitry from nose to
cortex (5). We bath-applied the mACh agonist oxotremorine and
monitored local field potential (LFP) electrodes placed in the
external plexiform layer of the OB to examine the effect of this
mACh agonist on odor-evoked activity.

Airborne odorants were delivered to the exposed nares (within
^3 mm) by means of electrically controlled pressure pulses (0.5
psi-1.5 psi/50-300 ms) that introduced a pulse of clean charcoal-
filtered air through a saturated odorized volume (amyl acetate) or

1 Simon Eraser University, Burnaby. British Columbia, Canada V5A

IS6.

via addition of an odorized bolus into a continuous clean airstream.
Oxotremorine (sesquifumarate salt) was mixed fresh daily in reg-
ular Ringer's solution and bath-applied at 100 /nM. Bicuculline
was aliquoted in distilled water and diluted ( 1000-fold) in Ringer's
to 10 iiM.

Odor-evoked oscillations in the frog OB, recorded in the exter-
nal plexiform layer, consisted of an initial biphasic component
(0-300 ms) followed by a slow wave envelope (1 to 2 s in
duration) and a superimposed fast oscillation (=7-12 Hz) (Fig.
1A). The parameters of the odor-evoked response are consistent
with our previous observations and similar to observations in turtle
OB (5, 6). The fast oscillation was completely and reversibly
blocked by bath application of the GABAA antagonist bicuculline
(10 /j,A/). demonstrating that GABAA inhibition is required to
maintain these oscillations (data not shown). Analysis of LFP
recordings determined that oxotremorine had two distinct effects
on the OB LFP response. First, it enhanced the initial component
(0-300 ms) of the olfactory response by «=25% (126.6 ± 4.2%;
mean ± SE; n = 16 trials in four animals; P < 0.001 ) (Fig. IB);
second, it increased the power of the frequency spectrum of LFP
recordings between 2 and 10 Hz, by 75% (175.2 ± 8.1%; mean ±
SE; n = 16 trials in four animals) (Fig. 1C).

In conclusion, the in vitro preparation of frog nose and brain is
a system which, by offering access for stimulating electrodes to the
forebrain nuclei from which these fibers arise, permits us to study
the effects of centrifugal ACh release in the OB. Our data here,
showing enhancement of the LFP in the OB, predict that mACh
receptor activation may improve the spatial coherence of OB
activity. We speculate that one function for this mACh modulation
may be to drive synchrony of the OB output necessary for plas-
ticity at the level of lateral (olfactory) cortex.

This project was generously supported by the Grass Foundation
and by the Canadian Institutes of Health Research. BH would like
to thank all of the Grass Fellows and Kim Hoke and Melissa
Vollrath for their comments on the manuscript.

277

5 10 15 20

Frequency (Hz)

Figure 1. Effects of the muscarinic acety/ciioline (mACh) receptor agonist oxotremorine (100 /jM) on the odor-evoked local field potential (LFP) response
in frog olfactory bulb. {A) The characteristic response to odor application at the nose (200 ms-bar) was an initial, typical/v hiphasic. component followed by a
slow wave and superimposed fast (7-12 H~) oscillations. (B) Bath application of the mACh agonist increased the peak amplitude of the initial component (* in
A). (C) Power spectra/ density analysis of the LFP showed increased power in the presence of the mACh agonist benvecn _ and 10 H~. Inset in C shows
representative averages of three single traces in each condition showing the fust oscillation high-pass filtered to 3.5 H- (scale bar as in A).

Literature Cited

1 Shepherd, G. M., and C. A. Greer. 1990. Pp. 133 169 in The

Synaptic Organization of the Brain, G. M. Shepherd, ed.. Oxford

University Press, New York.
2. Crespo, C., J. M. Blasco-Ibanez, J. G. Brinon, J. R. Alonso, M. I.

Dominquez. and F. J. Martinez-Guijarro. 2000. Ear. J. Neurosci.

12(11): 3963-3974.

3. Castillo, P. E., A. Carleton, J. D. Vincent, and P. M. Lledo. 1999.

J. Neurosci. 19(21): 9180-9191.

4. Elaagouby, A., N. Ravel, and R. Gervais. 1991. Neuroscience
45(3): 653-662.

X Delaney, K. R., and B. J. Hall. 1996. J. Neurosci. Methods 68(2):

193-202.
6. Lam, Y. W., L. B. Cohen, M. Wachowiak, and M. R. Zochowski.

2000. J. Neurosci. 20(2): 749-762.

Reference: Biol. Bull. 201: 277-278. (October 2001)

Dye Coupling Evidence for Gap Junctions Between Supramedullary/Dorsal
Neurons of the Gunner, Tautogolabriis adspersus

S. J. Zottoli. D. E. W. Arnolds, N. O. Asamoah, C. Cherez, S. N. Fuller, N. A. Hiza, J. E. Niennan, ami
and L. A. Taboada (Department of Biology, Williams College, Williamstown, Massachusetts 01267)

Many teleost fish have neurons whose somata lie on the surface
of the medulla oblongata (supramedullary neurons), the spinal cord
(dorsal cells), or both. The somata of these neurons in the cunner,
Tautogolabrus adspersus. are arranged in a single, median, longi-
tudinal row, from the posterior end of the fissura rhomboidalis
through the anterior portion of the spinal cord ( 1 ). Comparative
physiological studies in the cunner and other teleost fish have

indicated that these cells make electrotonic connections with one
another (2). To test the hypothesis that gap junctions exist between
these cells, we have injected individual supramedullary/dorsal
cells with Lucifer yellow and looked for dye coupling with neigh-
boring neurons.

Cunner. 7.3-10.5 cm in body length, were initially anesthetized
in 0.03% ethyl-m-aminobenzoate; when respiration ceased, they

278

REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

were transferred to a holding chamber where a 0.015% solution of
the anesthetic was passed through the mouth and over the gills of
the fish. The caudal portion of the medulla oblongata and rostral
spinal cord were exposed, and a microelectrode filled with Lucifer
yellow (Sigma; 5% in distilled water) was inserted into the soma
of only one neuron. Since the neurons exposed in this area of the
central nervous system could either be in a supramedullary posi-
tion or a spinal cord position, we will refer to the cells as su-
pramedullary/dorsal neurons. The dye was iontophoresed ( - 10 nA
of current was pulsed for 200 ms at a rate of 3/s) for about 1 h.
About 1 h after injection the fish were perfused with 10% phos-
phate-buffered formalin. The brain and spinal cord were then
dissected out, dehydrated, and cleared with methyl salicylate. The
whole brain was viewed with a fluorescent microscope.

The somata of supramedullary/dorsal cells are visible on the
surface of the brain with the aid of a dissecting microscope. In
seven fish, a single soma was located and filled with dye. Lucifer
yellow traveled from the filled cell to adjacent ones in three of the
seven fish. One cell rostral and two cells caudal to the filled cell
contained dye in two fish. In the third fish two cells rostral to the
filled cell contained dye (Fig. 1 ). In the four cases where there was
no apparent dye transfer between neurons, the intensity of the fill
appeared similar to that of filled neurons in which dye coupling did
occur. To control for the possibility that extracellular leakage of
the dye might label neurons other than the one being filled, a single
dorsal gill was penetrated with a dye-filled microelectrode in two
fish. The electrode was withdrawn to just outside the cell mem-
brane and Lucifer yellow was iontophoresed extracellularly for
1 h, the fish was perfused, and the brain processed as described
above. No dye was localized to any cell.

The transfer of Lucifer yellow from one supramedullary/dorsal
cell to others provides morphological evidence for the existence of
gap junctions. The lack of dye coupling in four fish does not
necessarily mean that gap junctions do not exist between su-
pramedullary/dorsal cells in these fish. For example, there may be
a wide distribution of sites of electrical coupling, or Lucifer yellow
may not have crossed the gap junctions (3. 4). When dye coupling
occurred, the fall-off in dye concentration from the filled cell to
adjacent neurons was large, so more distal neurons may be equally
well coupled but not contain dye. The cunner has between 35 and
40 supramedullary/dorsal cells. Electrical coupling measurements
will help determine the extent of coupling between this group of

Figure 1. Lucifer \ellow injection of a single supramedullary/dorsal
cell in the dinner. The cell on the far right was iontophoretically filled with
live; after fixation, dehydration, and clearing, the whole brain was viewed
with a fluorescent microscope. The soma of the filled neuron gives rise to
a single process that extends ventrally and bifurcates near the bottom of
the photomicrograph. Two other rostral somata (arrows) contain dye as
well, providing support for the existence of gap junctions benveen these
cells. This is a sagittal view of the brain with dorsal up and rostral to the
left. Calibration bar = U>0 ^m.

neurons. Supramedullary/dorsal cells in the cunner are sensitive to
tactile stimulation (5). Our results predict that neurons that are
electrotonically coupled will fire synchronously with sufficient
tactile stimulation.

This work was supported in part by Howard Hughes Medical
Institute and Essel Foundation grants to Williams College.

Literature Cited

1. Sargent, P. E. 1899. Anat. An;, 15: 212-225.

2. Bennett, M. V. L. 1960. Biol. Bull. 119: 303.

3 Murphy, A. D., R. D. Hadley, and S. B. Kater. 1983. J. Neurosci.
3: 1422-1429.

4 Peinado, A., R. Yuste, and L. C. Katz. 1993. Neuron 10: 1(13 1 14
5. Zottoli, S. J., F. R. Akanki, N. A. Hiza, D. A. Ho-Sang, Jr., M.

Motta, X. Tan, K. M. Watts, and E.-A. Seyfarth. 1999. Biol. Bull.
197: 239-240.

Reference: Biol. Bull 201: 278-280. (October 2001 1

A Comparison of Sounds Recorded From a Catfish (Orinocodoras eigeninanni, Doradidae)

in an Aquarium and in the Field

Ingrid M. Kaatz and Phillip S. Lobel (Boston University Marine Program,
Marine Biological Laboraton; Woods Hole, Massachusetts 02543)

Parvulescu ( 1 ) raised concerns regarding the suitability of a
small glass aquarium for characterizing fish sounds based upon a
theoretical consideration of sound echoes. Four out of eight au-
thors who cited this paper most recently noted that small aquaria
have complex acoustics, and the other four described the aquarium

environment as yielding imprecise and poor quality sound record-
ings. Further advances in the study of sound production and
communication in fishes require studies in controlled laboratory
environments. Recently. Lugli (2) noted that waveforms and sound
spectra were similar for field- and aquarium-recorded goby sounds.

NEUROBIOLOGY

279

Time, ms

Figure 1. Swimbladder disturbance sounds for three different individuals of a doradid catfish, Orinocodoras eigenmanni: (a) waveform of one entire
sound, field recording, (h) expanded waveform of 10 pulses, field recording (c) expanded waveform of 10 pulses, aquarium recording. The time scale differs
benveen the top (a) and bottom tw'o plots (b & c] hv a factor of 10.

Okumuru et al. (3) observed that artificially generated sounds
recorded close to a hydrophone were free from acoustic artifacts.
Whether more complex, natural fish sounds would also be artifact-
free requires testing. We elicited sounds from a catfish and com-
pared field and aquarium recordings which were specifically made
close to the signal source.

We analyzed sounds produced by the swimbladder mechanism
of a catfish in the disturbance context (fish are restrained by a
human hand) underwater. Similar sounds were produced by the
same fish in conflicts over resting sites (4). Many fishes that
produce sounds in intraspecific behavioral contexts also "release"
these sounds when restrained (5). We chose swimbladder sounds
because they are a common mechanism of sound production for
many fishes (6. 7).

We recorded sounds of nine individuals of a wild-caught neo-
tropical catfish, the doradid Orinocodoras eigenmanni. Standard
length ranged from 5.7 to 8.5 cm. Each individually recognizable
fish was recorded twice in both recording environments. Record-
ings were conducted during 10 July-6 August 1992. Fish were
positioned 7.5 cm from a hydrophone and 23 cm under the water
surface. Fish were held with their left side toward and their
swimbladder centered on the midpoint of the hydrophone. In the
field (Jenkins Pond, Falmouth, MA) fish were recorded in a con-
tainment net. The net had a 60-cm diameter and 60-cm maximum
depth. Water depth at the dock field site (Jenkins Pond) was 90 cm
over a sand bottom. Aquarium recordings were conducted in a

10-gallon glass aquarium on a grass lawn near the pond. The
hydrophone was suspended in the center of the water-filled aquar-
ium. Fish were held in the same relative position to the hydrophone
and water surface as in the field. Temperatures for recording dates
in the aquarium and in the field were not different (24.7 ± 0.6
aquarium, 25.2 ± 0.3 field; n = 3). Sounds were recorded using
a tape recorder (SONY Model WM-D6C: frequency response
40-15,000 Hz ± 3 dB). The hydrophone was pressure sensitive
and had a frequency response range of 10 to 3,000 Hz (BioAcous-
tics, see 8 for specifications). The acoustic analysis software SIG-
NAL (Engineering Systems. Belmont. MA) was used to digitize
and analyze sounds (sampling rate 25 kHz). We only analyzed
sounds which had clear pulse structure. Both recording environ-
ments occasionally yielded some sounds with obscured pulse
number and waveform patterns, due to spurious background noise
or fish movements.

Spectrograms of over 800 sounds were evaluated (580 field. 275
aquarium). The catfish produced similar numbers of sounds in both
recording environments. A minimum often sounds were produced
by each individual on each sampling date. The same types of
sounds were produced by individuals in both recording environ-
ments. Sound duration ranged from 30 ms to 2,400 ms.

In order to assess whether sounds were altered in the aquarium
environment compared to the field, we compared waveforms vi-
sually and pulse durations statistically for sounds from both re-
cording environments. Waveforms of sound pulses for field and

280

REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

aquarium showed the same shapes (Fig. 1 ). No artifacts were
noted. Pulse duration was measured for one sound per individual
(n = 9) for seven pulses in the center third of the sound where
pulse peak amplitudes were consistent. Individual pulse durations
ranged from 6 to 7 ms and were not significantly different between
field and aquarium environments (one way ANOVA). For aquar-
ium-recorded sounds, the pulse duration mean was 6.5 (SE 0.07,
;; = 63). For field-recorded sounds, the pulse duration mean was
6.5 (SE0.07, n = 63).

Disturbance context swimbladder sounds of a catfish showed no
differences in pulse waveform or pulse duration when recorded
close to a hydrophone in both field and small aquarium recording
environments. Kastberger (9) observed that for field recordings of
doradid sounds, pulse pattern was unchanged for up to 30 cm.
Many fishes initiate sound production in close proximity to con-
specifics (10. 11). These results suggest that a small aquarium
environment can provide sound recordings that accurately repre-
sent the sounds a fish produces in the field, yielding reliable
acoustic measurements.

The research was supported by the SUNY-ESF Barbara Suss-
man fund and Sigma Xi. Thanks to John Beckerly for providing
aquarium space, and Matt Bohling. Eric Morgan and David Mann

for technical assistance. Supported in part by Army Research
Office Grant DAAG-55-98- 1-0304.

Literature Cited

1. Parvulescu, A. 1967. Pp. 7-14 in Marine Bio-acoustics, vol. 2.
Pergamon Press, Oxford.

2. Lugli, M., G. Pavan, P. Torricelli, and L. Bobio. 1995. Environ.
Bwi Fishes 43: 219-231.

3. Okumura, T., T. Akamatsu, and H. Y. Van. 2001. Bioacoustics
(in press).

4. Kaatz, I. M. 1999. Ph.D. dissertation, SUNY College of Environ-
mental Science and Forestry, Syracuse, NY. Pp. 162-213.

5. Fish, M. P., and W. H. Mowbray. 1970. Pp. 1-207 in Sounds of the
Western North Atlantic Fishes. The Johns Hopkins Press. Baltimore.

6. Schneider, H. 1967. Pp. 135-158 in Marine Bio-Acoustics, vol. 2.
Pergamon Press. New York.

7. Tavolga, W. N. 1971. Pp. 135-205 in Fish Physiology, vol. 5.
Academic Press. New York.

8. Kaatz, I. M., and P. S. Lobel. 1999. Biol. Bull. 197: 241-242.

9. Kastberger, G. 1977. Zool. Jahrb. Physio/. 81: 281-309.

10. Ladich, F. 1997. Mar. Fres/m: Behav. Physiol. 29: 87-108.

11. Myrberg, A. A., Jr. 1981. Pp. 395-424 in Hearing and Sound
Communication in Fishes, Springer- Verlag, New York.

Reference: Biol. Bull. 201: 280-281. (October 2001)

Bimodal Units in the Torus Semicircularis of the Toadfish (Opsanus tan]

R. R. Fay and P. L. Edds-Walton (Family Hearing Institute, Loyola University Chicago,
6525 N. Sheridan Rd., Chicago, Illinois 60626)

We have been investigating aspects of auditory processing and
directional hearing in the toadfish Opsanus tan. We have shown
that the saccule is an auditory endorgan that encodes both fre-
quency and direction of a sound source ( 1 ). This information is
sent via the Vlllth nerve to nuclei in the medulla, in particular, the
descending octaval nucleus ( 1 ). Our previous work on cells in the
descending octaval nucleus in Opsanus tan has revealed that most
are highly directional ( 1 ) and that these directional auditory cells
project to the midbrain. The torus semicircularis (TS) is a sensory
processing site in the midbrain of fishes and amphibians. Nucleus
centralis in the TS receives input from auditory areas in the
medulla, and nucleus ventrolateralis receives input from lateral
line areas in the medulla (2). Here we report some preliminary
results from extracellular recordings of auditory cells in the TS.

Our protocol is described in detail elsewhere ( 1 ). In brief, the
toadfish is anesthetized and immobilized (pancuronium bromide
injection and lidocaine applied topically), and the dorsal surface of
the midbrain is exposed. Following surgery, the fish is placed in a
cylindrical dish filled with fresh seawater and is secured with a
head holder. The water surface in the dish lies just below the
surgical opening in the skull. The dish is part of a three-dimen-
sional shaker table that provides sinusoidal motion of the animal
with the surrounding water along linear pathways to simulate the
particle motion component of underwater sound at appropriate
frequencies (50-300 Hz) and levels, in the horizontal and mid-
sagittal planes at specified angles (0°. 30°, 60°, 90°, 120°, 150° in

each plane). In addition, we tested for external mechanoreceptive
sensitivity (tentatively identified as lateral line) by producing hy-
drodynamic disturbances using puffs of air at the water surface
along the length of the fish in the absence of an auditory stimulus.
Units were classified as responding to hydrodynamic stimuli if the
evoked spike rate was two standard deviations or more above the
mean background rate.

For extracellular recording we used pulled glass electrodes with
tip sizes of 3-5 ^.m and resistances of 3-10 Mil. Our recording
sites in the TS were confirmed in two ways. First, we used
neurobiotin-filled electrodes (4% in 3 M NaCl) to mark the loca-
tion of the first auditory cell analyzed. Second, the location of the
electrode at all recording sites was plotted using the scale on a
three-dimensional micromanipulator (accuracy to 10 /urn). The
neurobiotin was visualized using standard ABC immunohisto-
chemistry (Vector Labs) in 50-/j.m floating sections, which were
then placed on slides, dehydrated, and coverslipped.

We have recorded from 71 units in the TS. Of the cells that
responded to the auditory stimuli, we have found that 33% have
auditory sensitivity only and 67% respond to both auditory and
hydrodynamic stimulation. Units unresponsive to auditory stimuli
but responsive to hydrodynamic stimuli were observed frequently,
but were not analyzed further. Figure 1 illustrates the responses of
two TS units to varying levels of whole-body vibration in three
orthogonal directions and to the hydrodynamic assay for putative
lateral line sensitivity. Some units demonstrate a relatively large

NEUROBIOLOGY

281

0 *•

-10 0 10 20 30

Displacement (dB re: 1 nanometer)

-30 -20 -10 0 10 20 30
Displacement (dB re: 1 nanometer)

Figure 1. Responses of cells in the torus semicircularis of the toadfisti.
Spike rate versus stimulus level (displacement) is shown for two cells (PI.
B5) in response to whole-bod\ linear translator* motion at 100 H; in three
orthogonal axes. Also shown is the spike rate for PI. B5 during hydrody-
namic stimulation ipsilateral (dash-dot line) and contralateral (dash line)
to the left hemisphere of the brain. The hydrodynamic stimulus consisted of
repeated water disturbances along the length of the fish, and the data
plotted are average spike rates for comparison with the other stimuli. PI
has a strong bimodal response. (Spontaneous rate of 0.5 spikes/sfor both
PI and B5.)

response to hydrodynamic stimulation compared with auditory
(e.g., PI in Fig. 1). while others respond little if at all to the
hydrodynamic stimulus (e.g., B5 in Fig. 1 ). Both units of Figure 1
are highly sensitive and directional with respect to the vibrational
axes producing the greatest responses: PI responds best in the
front-back axis with a displacement threshold of 3 dB re: I nm;
and B5 is most responsive to vertical motion with a displacement
threshold of -10 dB re: 1 nm.

The origin of the bimodal sensitivity may be the convergence of
auditory and lateral line inputs to some of the cells in the TS from
nuclei in the medulla, or bimodal sensitivity may result from
connections between nucleus centralis and nucleus ventrolateralis
within the TS. Our preliminary anatomical data indicate extensive
opportunities for interactions among cells in the two nuclei of the
TS. We are currently evaluating the locations of medullary pro-
jection cells that were back-filled with neurobiotin following in-
jection at TS sites with bimodal response characteristics.

Supported by an R01 grant from NIH, NIDCD to R.R.F. and
from an NIH. NIDCD Program Project Grant to the Parmly Hear-
ing Institute.

Literature Cited

1 . Edds-Walton, P. L., R. R. Fay, and S. M. Highstein. 1999. J. Com/).
Neural. 411: 212-238.

2. McCormick, C. A. 1999. Pp. 155-217 in Springer Handbook of
Auditory Research; Comparative Hearing: Fish and Amphibians, A. N.
Popper and R. R. Fay, eds. Springer-Verlag, New York.

282 REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

Reference: Biol. Bull. 201: 282-283. (October 2001)

Mariculture of the Toadfish Opsanus tan

Allen F. Mensinger1 (University of Minnesota-Duluth, Duhtth, Minnesota, 55812), Katherine A. Stephenson2,
Sarah L Pollema2, Hazel E. Richmond^ Nichole Priced and Roger T. Hanlon1

In response to declining stocks of toadfish in local waters around
Cape Cod, Massachusetts, a toadfish mariculture program was initi-
ated in the summer of 1998 ( 1 ); the aims were to provide researchers
at the Marine Biological Laboratory (MBL) with sufficient numbers
(approximately 400 per year) of this valuable biomedical research
model (2, 3) while lessening pressure on native stocks. The goal was

1 Marine Biological Laboratory, Woods Hole, MA 02543.
- Biology Department, 10 University Drive, University of Minnesota-
Duluth, Duluth. MN 55812.

to raise fish to the target size of 25 cm and 500 gm within three years.
In the first year of the program, culture methods were developed, and
the effects of temperature and stocking density on toadfish growth
were monitored; a preliminary report was published in 1999 ( 1 ). We
continued to observe and monitor this captive population through the
summer of 2001. This paper summarizes the growth rates and mor-
talities of these three-year fish.

Briefly, two toadfish nests (with guardian males) were trans-
ported to the Marine Resources Center of the MBL from Waquoit
Bay, Massachusetts in July 1998. Approximately 400 juvenile fish

ocold •warm

Ooo

Jul-98 Oct-98 Feb-99 Jun-99 Oct-99 Feb-00 Jun-00 Oct-00 Feb-01 Jun-01 Oct-01

DATE

100
90
80
70
60
50
40
30
20
10

o cold • warm

o o

-,00

o o

Jul-98 Oct-98 Feb-99 Jun-99 Oct-99 Feb-00 Jun-00 Oct-00 Feb-01 Jun-01 Oct-01

DATE

Figure 1. Standard length (A) and weight (B) of maricultured toadfish plotted as a function of time for ahnir three years. Daw points represent mean
values offish raised at cold (circlet and mirm (square) temperatures (see the text for specific temperatures). Error bars = I SE.

MARICULTURE

283

detached from both nests in August 1998. In October 1998, 100 of
the juvenile fish were selected for mariculture and placed in
shallow fiberglass tanks (130 X 70 X 10 cm). Fish were raised at
two different temperatures, and their growth was monitored. Half
the fish were maintained at "cold" temperatures (15.8° ± 0.4 °C
average weekly temperature), which have proven successful for
maintaining adult toadfish in captivity. The remaining fish were
maintained at "warm" (19.6° ± 0.8 °C) temperatures in an effort
to increase growth rate. Stocking densities ranged from 10 to 40
toadfish m~2. Fish initially were fed live adult Anemia that had
been bathed in a nutritional supplement (Super Selco). After six
months of culture, the diet was switched to chopped pieces of
squid and butterfish. At the conclusion of the first year, the warm-
water fish averaged 6.4 ± 0.1 cm in length and weighed 13.0 ±
0.3 g. and the cold-water fish averaged 4.0 ± 0.5 cm and 1.7 ±
0.0 g ( 1 ). Survival rate was 78%, with many of the mortalities
attributed to the cannibalistic nature of batrachadoids (4).

At the conclusion of year one, the juvenile fish were transferred to
large, fiberglass tanks measuring either 3.7 X 2.4 m or 3.7 x 1.8 m.
The water level in each tank was maintained at 13 cm. Pieces of PVC
pipe (diameter 7 to 10 cm) were provided as shelters for the fish. The
temperature regimes were maintained (warm and cold), but fish from
the different densities were combined after being sorted by size to
prevent cannibalism. The fish were distributed to the tanks at a density
of 1.8 to 2.4 fish m~2. The two- and three-year age classes were
maintained on a prepared diet consisting mainly of chopped squid and
butterfish and were fed three times per week.

During the second year of culture, the average weekly temper-
atures were 19.4 °C (warm) and 16.9 rC (cold). In year three, the
average warm-water temperature was 20.2 °C. Because the cold-
water fish continued to be small, they were switched to the "warm"
water in February of 2001; as a result, the average weekly tem-
perature for these fish was 19.2 °C during year three.

After 24 months of culturing, the warm-water fish averaged
10.6 ± 0.2 cm and weighed 40.5 ± 2.1 g, and the cold-water fish
averaged 6.2 ± 0.7 cm and 9.0 ± 0.6 g. By the end of the third
year, the warm fish had grown to an average standard length of
13.0 ± 0.3 cm (range 9.5 to 15.5 cm) and average weight of 87.7 ±
5.5 g (25 to 136 g). The cold-water population continued to display
slower growth, with the average fish measuring 9.3 ± 0.2 cm (8.4
to 10.8 cm) and 29.2 ± 2.0 g ( 15 to 45 g) in year three (Fig. 1 ).

Survival rates remained high, following the initial 78% rate in year
one. Approximately two-thirds of the original fish remained alive after

24 months, and 60% survived through July 2001 . The size segregation
instituted in the summer of 1999 greatly reduced cannibalism.

Our eventual goal is to eliminate field collection through the
successful spawning and rearing of captive fish. However, the age
of sexual maturity among the Cape Cod population has never been
firmly established. Five females in the warm-water tanks became
gravid in the spring of 2001, and at least one successfully depos-
ited scores of eggs inside a PVC pipe, suggesting that the onset of
sexual maturity for female toadfish is less than three years. Un-
fortunately, for unknown reasons, these eggs failed to develop. The
onset of sexual maturity in the males remains to be determined.

In summary, we have demonstrated that toadfish can be raised in
captivity for at least three years. At the current maximal growth
rate of 5 cm/year, we estimate that the fish will need at least five
to six years to attain the desired size range of 25 to 30 cm. thus
making the project impractical in terms of cost and time.

One of the main impediments to faster growth is the virtual
cessation of growth during the winter (Fig. 1). Previous observa-
tions led us to hypothesize that keeping the fish at temperatures
about 15 °C above ambient during the winter would circumvent
this "hibernation." Because this expectation has proved incorrect,
future attempts will focus on temperature and photoperiod. Pre-
liminary evidence shows that newly detached juvenile toadfish
raised at 26° to 29 °C grow significantly faster than fish raised at
20 °C (5). We also plan to manipulate the photoperiod during the
winter to stimulate year-round growth.

We wish to thank Waquoit Bay National Estuarine Research
Reserve for use of their facilities, and J. Hanley and B. Mebane for
assistance in tank maintenance. Support was provided by the
Marine Models in Biological Research Program, University of
Minnesota Grant in Aid. NASA Life Science Fellowship. MBL
Associates Fellowship and NIH grant DC01837.

Literature Cited

1 . Tang, K. Q., N. N. Price, M. D. O'Neill, A. F. Mensinger, and R. T.
Hanlon. 1999. Bwl. Bull. 197: 247-248.

2. Mensinger, A. F., and S. M. Highstein. 1999. J. Comp. Neurol. 410:
653-676.

3. Mensinger, A. F., D. J. Anderson, C. J. Buchko. M. A. Johnson,
D. C. Martin, P. A. Tresco, R. B. Silver, and S. M. Highstein. 2000.
J. Neurophysiol. 83: 611-615.

4. Mensinger, A. F., and J. F. Case. 1991. Biol. Bull. 181: 181-188.
5 Rieder, L. E., and A. F. Mensinger. 2001. Biol. Bull. 201: 283-

285.

Reference: Biol. Bull. 201: 283-285. (October 2i«)l i

Strategies for Increasing Growth of Juvenile Toadfish

Leila E. Rieder1'2 and Allen F. Mensinger1 (University of Minnesota-Duluth, Duliith, Minnesota 55812)

A toadfish mariculture program was initiated in the summer of
1998 at the Marine Biological Laboratory. Woods Hole. Massa-

' Marine Biological Laboratory. Woods Hole. MA 02543.
2 Columbia High School, East Greenbush, NY 12061.

chusetts. The purpose of this program was to reduce pressure on
the native toadfish population while providing researchers with a
year-round supply of appropriately sized animals. Although the
toadfish have proven to be amenable to year-round cuituring
(survival rates were 60% to 70% during the initial three years
( 1 , 2 1 ), their growth was slower than that of conspecitics inhabiting

284

REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

southern portions of the geographical range (3). Therefore, in an
effort to accelerate growth, the effects of temperature and diet were
investigated in toadfish that had recently detached from their nests.

Four nests with guardian males were transported from Waquoit
Bay. Massachusetts, to the Marine Biological Laboratory in June
of 2001. The fish for this experiment were selected from a single
nest. Fish began to detach from the nest during the first week of
July, and feeding was initiated on July 9. Nine 40-1 glass aquaria
filled with fresh, filtered seawater were used for the study. The
aquaria were placed in a 3 x 3 matrix on a large aquarium stand.
Each contained 3 cm of sand overlying an air-powered undergravel
filter and had an overhead fluorescent light (14/10 L/D cycle).
Water temperature was maintained at 23°, 26°, or 29 °C in each set
of three aquaria. About 30% to 50% of the aquarium water was
changed daily by slowly adding fresh seawater. Fish were ran-
domly selected from the 400+ juveniles that detached from the
nest, and 10 fish were placed in each aquarium. An additional 25
fish were placed in a shallow 1-m2 fiberglass tray and provided
with fresh running seawater at 20 °C.

Fish were maintained on three diets: live, prepared, and pre-
pared plus. Live food consisted mainly of adult Anemia treated
with a nutritional supplement (Super Selco) and supplemented
with live mysids or newly hatched Fundiilus sp. The prepared diet
consisted of small chunks (approximately 3 mm X 4 mm) of squid,
clam, or mussel. The prepared plus diet consisted exclusively of
small pieces of squid supplemented with crushed commercial fish
food (5 mm pellets. Aquatic Eco-Systems). The three tanks at each
temperature were fed one of the diets exclusively. Control fish in
the fiberglass tanks were fed live Anemia treated with Super Selco
supplemented with mysids.

Experimental fish were fed an average of 6 days per week, and
control fish were fed about 4 times per week. For live food, the
daily ration was sufficient that fish would terminate feeding prior
to prey extermination, and live prey was often observed in the
tanks 24 h after feeding. For the prepared diets, the food was
impaled on a copper wire (28 gauge) affixed to a glass rod and was
waved in front of the fish until it was eaten. Fish were fed once per
day, and individual fish were presented with food continuously
until refusal. To ensure that all fish on prepared diets were fed.
individual fish were visually checked for extended abdomens.

Fish were weighed and measured prior to the initiation of
feeding on July 9. Standard length averaged 1.7 ± 0.02 cm; weight
averaged 0.17 ± 0.01 g. There was no significant difference in size
among the experimental and control aquaria ( ANOVA: P = 0.14;
all statistical analysis was performed with GraphPad InStat version
4.10 for Windows 95, GraphPad Software, San Diego, CA). All
fish were weighed and measured again after 3 weeks of feeding.
Figure 1 shows the standard length and weight distribution for the
three temperatures and diets plus the control.

Examination of the standard lengths offish kept at 23 °C did not
reveal any size difference among the three diets (ANOVA: P =
0.13). However, at both 26 °C and 29 °C. fish eating both
prepared food diets were significantly longer than ones maintained
on live food (ANOVA: P < 0.05). When the experimental tanks
were compared with the controls, fish fed the prepared diet at all
three temperatures and fish given the prepared plus diet at 26 and
29°C were significantly larger than controls (ANOVA: P < 0.05 ).

Examination of weight at each temperature revealed that at

23 °C, the fish on both prepared diets were significantly heavier
than those consuming live prey (ANOVA: P < 0.01 ). At both 26
and 29 UC, fish fed both prepared diets were larger than fish on the
live diet (ANOVA: P < 0.001 ). When treatments were compared
against controls, fish maintained on the prepared diet at all three
temperatures and fish fed the prepared plus diet at the two higher
temperatures were significantly larger than controls (ANOVA:
P < 0.05).

The results indicate that toadfish growth can be accelerated
compared to our previous mariculture methods (1,2), by increas-
ing water temperature and by substituting a diet of prepared food
for one of live food. Previous mariculture efforts (1.2) required 90
days for toadfish to attain the size and weight that fish in the
current experiment reached in 21 days. Survival was high (98%)
and the elevated temperatures were not detrimental to fish health.
This is not surprising because the range of the conspecifics extends
to Florida, and, locally, temperatures of 25 to 30 °C are not
uncommon during summer months in shallow Cape Cod estuaries.

The objective of this experiment was to determine new strate-
gies to accelerate juvenile toadfish growth. Therefore, the experi-
ment was designed to compare our new methodology to our
previous culture techniques. Thus, control fish were placed in
shallow fiberglass trays rather than in 40-1 aquaria. Although every
effort was made to make the experimental tanks identical, small

Dlive S prepared E prepared plus

— 25

I 20

Q
&

a 1

23 26

TEMPERATURE ( °C)

Dlive H prepared O prepared plus

— 04
o

3 02

20 23 26 29

TEMPERATURE ( °C)

Figure 1. The bars represent the average standard length (Al and
weight (B) for each diet at a specific temperature for juvenile toadfish.
Asterisks indicate significantly different means compared to controls
(ANOVA: P < 0.05). Error bars = / SE.

MARICULTURE

285

variations in each tank (water chemistry, ambient light, vibration)
were not examined and may have subtly influenced individual fish.
Finally, the energy expenditure (foraging vs. "hand" feeding)
between the live and prepared diets will need to be addressed in the
future.

The fish in our experiment grew fastest when fed the prepared
diets. However, because hand-feeding hundreds to thousands of
juvenile toadfish is not practical, we are attempting to refine the
feeding techniques to reduce or eliminate this time-consuming
step. The 4- and 5-week-old toadfish have begun foraging, indi-
cating that the food presentation may only be needed during the
first month.

We thank the Waquoit Bay National Estuarine Research Re-
serve for use of their facilities. We thank H. Richmond. J. Hanley.
and B. Mebane for help with aquarium set-up. We thank C. Taylor
for education assistance. Funded by NIH grant DC01837.

Literature Cited

1. Tang, K. Q., N. N. Price, M. D. O'Neill. A. F. Mensinger, and R. T.
Hanlon. 1999. Biol. Bull. 197: 247-248.

2. Mensinger, A. F., K. A. Stephenson, S. L. Pollema, H. E. Richmond,
N. Price, and R. T. Hanlon. 2001. Biol. Bull. 201: 282-283.

3. Wilson, C. A., J. M. Dean, and R. Radtke. 1982. J. Exp. Mar. Biol.
Ecol. 62: 251-259.

Reference: Biol. Bull. 201: 285-286. (October 2001)

Development of Genetically Tagged Bay Scallops for Evaluation of Seeding Programs

Hemant M. Chikannane, Alan M. Kuzirian (Marine Biological Laboratory, Woods Hole, Massachusetts),
Ian Carroll1, and Robbin Dengler (Marine Biological Laboratory, Woods Hole, Massachusetts)

The bay scallop Argopecten irradians (Lamarck 1819) is har-
vested commercially and recreationally throughout its range along
the east and Gulf coasts of North America. At its peak, the
Massachusetts harvest exceeded 1200 metric tons with a monetary
value of $11 million (1). On Martha's Vineyard, bay scallop
harvests are an important economic commodity and can represent,
depending upon the size of the catch and price per pound, between
4% and 10% of the island's annual economy (2).

Argopecten irradians irradians. the northern bay scallop, and
the southern subspecies (A. i. concentricus, A. i. amplicostatus)
exhibit extreme natural variability in harvestable stocks from year
to year: but. in general, natural populations have declined over the
past quarter century (1). Harvest data from Cape Cod, Martha's
Vineyard, and Massachusetts as a whole, compiled from 1965 to
1997, reveal that the maximum harvests (in bushels) occurred in
the 1980s, but that scallop harvests have dropped precipitously
since 1985 (3).

The variability and population declines are attributed to preda-
tion and habitat loss, to harmful algal blooms, and to the species'
short life ( 1. 4. 5). The consequent economic pressures have led to
increased emphasis on scallop aquaculture. development of field
grow-out techniques, transplantations, and seeding programs using
hatchery- or field-collected seed (1. 6). However, the notion that
these efforts contribute significantly to population stabilizations is
poorly supported by hard evidence (3. 7).

One of the difficulties in ascertaining the success of seeding or
stock enhancement programs is that seeded animals cannot be
distinguished from the natural population. The colored tags used
for shrimp and fin fish are not useful for scallops. Allozyme
differences were insufficient to discriminate between native and
transplanted animals (8). In contrast, DNA-based molecular mark-
ers are excellent at distinguishing between subpopulations (9) and
also have the great advantage of being neutral, while not generat-
ing artifacts due to predator preferences or survival. In this paper.

1 Brown University. Providence. RI.

we report progress in the development of scallops with molecular
tags — an aid in the evaluation of stock enhancement programs.

We chose to develop RAPD (Random Amplification of Poly-
morphic DNA) genetic markers that can be detected by the poly-
merase chain reaction (PCR) (10). RAPD-PCR has the great ad-
vantage that DNA sequence information is not required for the
development of useful markers.

Representative adult bay scallops were initially collected from
Nantucket Island and, more recently, from Martha's Vineyard.
Mantle tissue was sampled after the valves opened spontaneously.
This procedure does not kill the animal, which can then be kept
alive for mating experiments. After the tissue was thoroughly
rinsed to remove extraneous biological material, DNA was purified
from it. DNA purification procedures. RAPD-PCR protocols, and
electrophoresis conditions have been described previously (11).
Fifteen primers were screened for amplifiability and reproducibil-
ity, and the relative frequency of bands was determined. Results
for two primers are shown in Table 1 . In both cases, the larger size
bands were present at higher relative frequencies when compared
with those of the smaller size bands. Figure 1 shows a represen-

Table 1

Relative marker frequencies in the population, for primers
AGGTCACTGA (10 bands) and GAAGCGCGAT (9 bands)

Band number

AGGTCACTGA

GAAGCGCGAT

1.0

0.9

1.0

0.8

0.3

0.5

0.6

0.5

0.3

0.1

0.4

0.2

0.4

0.2

0.1

286

REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

A B

Figure 1. Representative RAPD-PCR profiles with Primer AGGTCACTGA (Pn/iel A I. ami Primer GAAGCGCGAT (Panel B). Arrows point In a liii>h
frei/tiencv marker in Panel A, and a low frequency marker in Panel B. 'm' is a Lamhdti Hnullll DNA si~e ladder.

tative gel image for the same two primers. Panel A shows an
example of a RAPD marker that occurs at 100% frequency in the
test population. Pane! B shows a RAPD marker that appears at low
frequency in the same population.

The breeding of marked scallops was a challenge. Bay scallops
are simultaneous hermaphrodites, so inbreeding bay scallops car-
rying selected markers should have been relatively easy. Six ani-
mals were spawned by immersion in 1 mM serotonin, or by
temperature shock (12). and self-fertilization was allowed to occur.
Several hundred embryos were obtained in each case and were
cultured in roller bottles. None of them survived to adulthood,
indicating that self-crosses result in little or no survival of the F,
progeny. This inbreeding depression confirms the results of Kar-
ney (pers. comm.) and Stiles et nl. i 13).

To circumvent this problem, we performed bulk matings with
adults carrying distinctive markers. Crosses were done with four
selected groups of 10 to 25 individuals. These have now gone
through two generations, and are being screened. Animals testing
positive for the high frequency marker (arrow. Fig. 1) will be
conditioned and spawned. From the results obtained thus far, bulk
mating of animals carrying selected markers appears to be the best
approach to genetically tagging Argopecten.

Field trials will be carried out by transplanting animals carrying
the selected marker into test areas where the marker is absent or
present at a very low level, and then determining the relative
frequency of the tag in sampled animals. To the best of our
knowledge, this is the first attempt to tag a cultured mollusc
species with molecular markers for the evaluation of seeding
programs. If this proof-of-principle experiment is successful, it can
be extended to commercial aquacultured species such as Plac-
opecten magellanicus. M\a arenaria, and Mercenaria mercenaria.

This work was supported in part by a grant from the MIT/WHOI
Sea Grant Program to A.M.K. and H.M.C. (Project: R/A-34).

H.M.C. and A.M.K. are indebted to Dr. Dale Leavitt of SEMAC
and Rick Karney of the Martha's Vineyard Shellfish Group for
assistance and advice. We thank the Marine Resources Center,
Marine Biological Laboratory, for providing facilities for mainte-
nance of the scallops.

Literature Cited

1. National Marine Fisheries Service. 2001. Annual Commercial
Landing Statistics. [Online]. Available: http://www.st.nmfs.gov/stl/
comnierciul/landings/annuaMandings.html [August 2001).

2. Karney, R. 1991. Pp. 308-312 in An International Compendium of
Scallop Biology and Culture. S. E. Shumway and P. A. Sandit'er, eds.
World Aquacullure Society. Baton Rouge, LA.

3. Macfarlane. S. L. 1999. SouthEastern Massachusetts Aquaculture
Center (SEMAC) Technical Report 99-01: 1-73.

4 Arnold, W. S.. D. C. Marelli, C. P. Bray, and M. M. Harrison.

1998. Mar. Ecol. Prog. Ser. 170: 143-157.
5. Short, F. T., B. W. Ibelings. and C. DenHartog. 1998. Aauat. Bot.

30: 295-304.

h. Tettelbach, S. T. 1991. Pp. 164-175 in An International Compen-
dium of Scallop Biology and Culture. S. E. Shumway and P. A.
Sandifer, eds. World Aquaculture Society. Baton Rouge. LA.
7. Marelli, D. C., and W. S. Arnold. 1998. J. Shellfish Res. 17: 332.
X. Krause, M. K. 1992. J. Shellfish Res. 11: 199.
9 De Wolf, H., T. Backeljau, and R. Verhagen. 1998. Heredity 81:

486-492.
10 Williams, J. G. K., A. R. Kubelik, K. J. Livak, J. A. Rafalski, and

S. V. Tingey. 1990. Nucleic Acids Res. 18: 6531-6535.
1 1. Chikarmane, H. M., A. M. Kuzirian, R. Kozlowski, M. Kuzirian,

and T. Lee. 2000. Biol. Bull. 199: 227-228.
1 2 Ram, J. L.. G. W. Crawford, J. U. Walker, J. J. Mojares, N. Patel,

P. P. Fong, and K. Kyozuka. 1993. J. Ev/>. Zoo/. 265: 587-598.
13. Stiles, S., J. Choromanski, D. Schweitzer, and Q-Z. Xue. 1996. ./.
Shellfish Res. 16: 461.

ECOLOGY AND POPULATION BIOLOGY
Reference: Binl. Bull. 201: 287-28X. (October 2001)

287

The Effects of Salt Marsh Haying on Benthic Algal Biomass

Libbv Williams, (The College of Wooster, Wooster. Ohio), G. Carl Nohlitt /V1, and Robert Buchsbaum2

Salt marsh haying is a traditional activity on East Coast salt
marshes and is still carried out on a large scale (over 400 hectares
regularly) throughout Plum Island Sound, located in northeastern
Massachusetts.

The removal of approximately 90% of the aboveground biomass
of the salt marsh by haying may alter many ecological processes
within the salt marsh ( 1 ). One such process is the production of
benthic algae. Estrada ct a/. ( 1974) found that nutrients and light
are critical controls on the amount of benthic algae present (2).
When a thick grass canopy shades the algae, their growth is limited
not by nutrients but rather by the available light. However, when
little grass canopy is present, benthic algal growth is limited by the
available nutrients. Grazers are also likely to determine the amount
of algal standing crop. In this project we tested the hypothesis that
there should be a marked increase in benthic algal biomass after an
area has been hayed because the algae is no longer limited by the
available light.

We took core samples at three marsh sites, each about 1 to 2
hectares in area. Two of these are regularly subjected to haying,
and one is an unhayed reference area. The reference area (PUH)
has not been hayed for at least 25 years. One hayed site (EPH) was
last hayed two summers ago (1999). At PUH and EPH, six 1-nr
quadrats were placed randomly in two different vegetation zones,
Spartina alterniflora (low marsh) and Spartina patens (high
marsh) sites. Three quadrats in each vegetation zone at each area
were cleared of aboveground vegetation by clipping, and three
were left as undipped reference quadrats. The second hayed site
(HAY) was hayed in June 2001 before sampling began. At HAY,
we established three 1-nr quadrats within Spartina patens zones
that had just been cleared of vegetation by the hayer.

Six sediment cores (3-cm diameter, 1-cm depth) were taken
from each quadrat at day 0, day 7. day 14, and day 30 after clipping

1 Governor Dummer Academy, Bytield. MA.

2 Massachusetts Audubon Society, Wenham, MA.

(or haying). The six sediment cores were then pooled together into
two sets of three cores. At the end of the 30-day sampling period,
the aboveground plant biomass from the quadrats within the hayed
and reference sites was removed to measure the regrowth of the
vegetation during the experimental period. The material from the
sites was then dried and weighed. The benthic chlorophyll was
extracted and measured from the pooled core samples using the
method of Lorenzen (3).

We used HOBO HLI light intensity loggers to determine the
relative amount of light reaching the sediment surface at both the
treatment and reference sites.

We found no difference between the benthic algal chlorophyll in
the area that was hayed two summers ago and the area that has not
been hayed for 25 years. Consequently, we pooled and treated the
two areas as replicates in further analyses. Furthermore, we found
no statistically consistent increase in algal biomass over the 30
days of the experiment in the clipped or hayed treatments. In
addition, there was not a significant difference in benthic algal
biomass between S. allerniflora (low marsh) and 5. patens (high
marsh) zones regardless of whether they were clipped or not
(Fig. 1).

The benthic chlorophyll concentrations in the June 2001 hayed
area were significantly higher than the benthic chlorophyll in the
undipped treatments (ANOVA, F = 3.330, P = 0.039). How-
ever, neither the benthic algal chlorophyll present in the hayed area
nor that in the reference quadrats was significantly different from
that present in the clipped quadrats. The results suggest that haying
on a large scale, but not small-scale removal of the plant canopy,
increases the amount of benthic algae present.

Based on a limited number of light measurements, there is a
direct relationship between the percentage of light reaching the
sediment surface and benthic chlorophyll concentrations within
each quadrat (Fig. 2). In addition, it appears that there is an inverse
relationship between the plant biomass of each quadrat and the
benthic chlorophyll concentration (Fig. 2).

~>s

2
o

£
O

Ic
CO

6000 "
5000
4000 [
3000 <
2000
1000
0
(

5partma alterniflora zone

xj>^=^

— * — reference
D clipped

) 10 20 30
Days after clipping or haying

10 20 30

Day after clipping or haying

Figure 1. Benthic a/t>al bimmixs (ing chlorophyll per nr) in two different vegetation zones after clip/ting or liuying.

288

REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

Benlhic Chlorophyll a (mg/m3

Benthic Chlorophyll a (mg/m3

•

) 500 1000 1500 2000
plant biomass (g/m2)

00 2500 3000 3.500 4.000 4.500
light (log lumens/sq. meter)

Figure 2. Relationships between benthic alga! biomass, incident light to the marsh surface, and dboveground plant biomass.

The benthic algal biomass is distributed very patchily through-
out each site. As a result, even though six core samples were taken
on each sampling day. more samples might be needed to truly
estimate the actual biomass of each quadrat. The tremendous
variability — not only within each area but also within each quad-
rat— confounds the algal chlorophyll measurements.

It is possible that significant algal growth occurred, but the algae
were grazed heavily and thus did not show an actual increase in
biomass. It is also possible that our removal of aboveground plant
biomass resulted in desiccation of the marsh surface. Desiccation
could limit algal growth directly and by preventing the reminer-
alization of nutrients necessary for future algal growth.

Regrowth of marsh plants in the hayed site occurred particularly
rapidly (to more than 70% of the biomass of an 5. patens refer-
ence) over the 30 days of the experiment. Thus any stimulation of

algal growth by increased light due to haying is likely to be short
term.

The Plum Island Estuary LTER and a Research Experience for
Undergraduates NSF fellowship supported this research. Thanks to
Robert H. Garritt and Kris Tholke for guidance on the chlorophyll
analyses and to Charles G. Hopkinson for advice on experimental
design.

Literature Cited

1. Greenbaum, A., and A. Giblin. 2000. Biol. Bull. 199: 225-226.

2. Estrada, M., I. Valiela, and J. M. Teal. 1974. J. Exp. Mar. Biol.
Ecol. 14: 47-56.

3. Lorenzen, C. J. 1967. Limnol. Oceanogr. 12: 343-346.

Reference: Biol. Bull. 201: 2S8-290. (October 2001)

Dissolved Nitrogen Dynamics in Groundwater Under a Coastal Massachusetts Forest

Eve-Lyn S. Hinckley, Christopher Neill, Richard McHorney1, and Ann Lezberg (The Ecosystems Center,
Marine Biological Laboratory, Woods Hole, Massachusetts 02543)

Land uses, such as agriculture and residential development,
have greatly influenced the amount of nitrogen (N) transported
from coastal watersheds to receiving estuaries. This is a concern to
ecologists and management groups in coastal regions such as Cape
Cod. Martha's Vineyard, and Nantucket, where precipitation per-
colates rapidly through sandy glacial sediments in the vadose zone
(unsaturated layer between the soil and aquifer), causing rapid
vertical transport of N to the aquifer (saturated layer) and horizon-
tal movement of N to coastal waters (1. 2). In many forested
watersheds, ammonium (NHj ) and nitrate (NO^) are transported
to receiving estuaries in low amounts relative to dissolved organic
N (DON), which includes organic acids and other compounds (3,
4). However, in human-altered systems, high amounts of inorganic
N. particularly in the form of NO,", are often transported to aquatic
systems, elevating primary production (5). To make management

1 The Nature Conservancy. Plymouth, MA 02360.

decisions for coastal areas with high anthropogenic N inputs, it is
important to study systems in which human influences are minimal
so that background N transformations can be identified.

Our goal in this study was to quantify N concentrations and to
identify N transformations in groundwater moving along a known
flow path in a forested system with a known land-use history,
minimal septic inputs, and no overland flow. We measured the
relative concentrations of dissolved N species (NH^, NO^, and
DON) in throughfall, soil solution in the vadose (unsaturated)
zone, and groundwater from an oak forest on Job's Neck peninsula
in Edgartown, Massachusetts. We also measured N concentrations
at the seepage face of the Edgartown Great Pond estuary which lies
roughly 500-1000 m downgradient in the groundwater flowpath
from the forest.

We collected throughfall. and water from the vadose zone,
aquifer, and seepage face from June 2000 to August 2001 and
analyzed samples for NHj , NO^, and DON concentrations. We
used spatially extensive sampling to capture fine-scale differences

ECOLOGY AND POPULATION BIOLOGY

289

„ 60

b '

F = 7 986. P < 0-001

5 40

-30
g 20

10
0

F = 24.950. P < 0 001

ab •

» j

F = 11.328, P < 0.001

F =8696, P <0001

Figure 1. Mean concentrations of TON, DON, NHj, and NO, in
water inputs to and outputs from the forest. D = Throughfall, D = Vadose
zone, ^ = Aquifer, • = Seepage face. Means represent the average
concentrations of samples taken from June 2000-August 2001. Only sam-
ples for which all three N analyses were completed are included. Through-
fall N = 72, Vadose zone N = 79, Aquifer N = 138, Seepage face N = 62.
Bars are ± 1 SE and letters above bars indicate significant difference to
the P < 0.001 level.

in vegetation and topography: 60 throughfall collection units and
50 zero-tension lysimeters installed at 40-crn depth in a stratified
random pattern throughout the forest, 40 iron piezometers installed
to the water table along the groundwater flow paths, and 34 points
for shallow groundwater discharge sampling at the seepage face of
Edgartown Great Pond. All water samples were filtered with ashed
(2 h at 550 °C) Whatman GF/F filters and frozen in 60-ml poly-
ethylene bottles until analyzed colorimetrically for NH^, NO^,
and TON concentrations (TDN was analyzed by persulfate diges-
tion). DON concentrations were calculated by subtracting NH4 +
NO7 from TDN concentrations of each sample. We used a one-
way analysis of variance and a Tukey's post-hoc test to determine
statistical differences between means (at 0.05 level of signifi-
cance). All statistical analyses were performed using SYSTAT
(SPSS Inc.. 1997. Version 7.0).

TDN increased significantly (P < 0.001) from 31.44 ± 2.71
ILuW in throughfall to 54.08 ± 3.21 /iM in the aquifer (Fig. 1A).
DON was the principal component of dissolved N in the vadose
zone, aquifer, and at the seepage face (Table 1). These data are
consistent with other studies that show dominance of DON in soil
solution and groundwater of forested watersheds (3, 4). DON
increased significantly (P < 0.001) from 9.15 ± 0.76 /mW in
throughfall inputs to 46.63 ± 2.96 /j,M in the aquifer (Fig. IB).
Most DON consists of organic acids and other compounds that
originate in the upper layers of the forest floor and move to
groundwater during periods of heavy precipitation (4. 6). There
was no significant difference between DON concentrations in the
aquifer and at the seepage face, suggesting that further removal or
accumulation of DON may not occur as groundwater moves hor-
izontally to receiving waters.

NH^ decreased significantly (P < 0.001 ) from 11.97 ± 1.48
juM in throughfall to 3.38 ± 0.50 juM in the vadose zone (Fig. 1C).
This suggests that plants or microbes in the rooting zone immo-
bilized NH4 . NH| concentrations were higher in the aquifer and

the seepage face compared with the vadose zone, but these differ-
ences were not significant and suggest that little additional NH^
uptake occurs below the 40-cm depth at which the vadose zone
samples were collected. NH4+ composed about 12% of TDN in the
aquifer, indicating some export of NH^-N could occur as ground-
water moves to the seepage face (Table 1). NH4 movement from
the vadose zone to the aquifer is consistent with data from other
coastal systems with sandy soils on Cape Cod and may be caused
by low soil pH and low soil cation exchange capacity (2). These
characteristics may cause NH_^ to be more mobile in forests with
very coarse-textured soils compared with other upland forests on
finer-textured soils (7. 8).

NO^ decreased significantly (P < 0.001) from 10.33 ± 1.16
IJiM in throughfall to 0.99 ± 0.08 juM in the aquifer (Fig. ID).
NO^" was about 2% of TDN in the aquifer (Table 1 ), indicating
that very little NO^ moves from the plant-rooting zone to the
aquifer. In the aquifer, concentration of NO,~ was also lower than
NH^, which suggests low rates of nitrification along the flowpath
from soil solution to the aquifer. This pattern is consistent with
NH^ and NOJ concentrations measured in soil solution and
groundwater in Cape Cod coastal forests (2, 6).

NOJ increased significantly (P < 0.001 ) from 0.99 ± 0.08 ^M
in the aquifer to 13.79 ± 5.26 pM at the seepage face (Fig. ID).
NO^ concentrations were highly variable but this overall pattern
suggested that NO^ from additional sources was detected at some
locations along the Edgartown Great Pond shoreline. There are
several possible explanations for this result. Long-distance trans-
port of NO J from septic discharges farther inland are possible but,
we feel, unlikely, given the relative hydrological isolation of Job's
Neck, the west-to-east groundwater movement under the forest,
and our measurements of higher NO7 concentrations at the south-
ern (coastal) end of the pond shoreline. It is also possible that
increases in NO^ result from zones of oxidation of NH^ or DON
to NO," within the seepage face, or from inputs of fixed N derived
from the N-fixing shrub Mvrica pensylvanica, which is present at
many places along the pond shoreline.

From these findings, we conclude that: ( 1 ) relatively low NH4
and NOJ and high DON are transported from the forest to the
coastal pond, (2) incomplete retention of NH^ above the aquifer
and comparatively low NO^ concentrations in the aquifer suggest
that nitrification rates are low in forest soils and in the aquifer, and
(3) there is the possibility that in some places the seepage face may
contribute a small amount of NO^ to discharging groundwater
rather than remove it, because of NHJ or DON oxidation or N
inputs derived from N-fixing species. These findings can serve as
a baseline for understanding how N transformations change with
increasing human development and a shift toward a greater pro-
portion of NO^ reaching the seepage face from the coastal aquifer.

Table 1

Percentage of TDN for each N species measured

NH4+

NO^

DON

Throughfall

38.1

32.8

29.1

Vadose zone

8.8

13.7

77.6

Aquifer

12.0

1.8

86.2

Seepage face

10.4

22.3

67.3

290

REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

This research was supported by the Mellon Foundation. We
thank Tom Chase and Mike Dunphy of The Nature Conservancy
and the Kohlberg family for allowing us to work on their property.

Literature Cited

1 Valiela, I., M. Geist, J. McClelland, and G. Tomask) . 2000. Bio-
geochemistry 49: 277-293.
2. Lajtha. K., B. Seel), and I. Valiela. 1995. Biogeochemistry 28: 33-54.

3. Hedin, L., J. Armesto, and A. Johnson. 1995. Ecology 76: 493-509.

4. Quails, R., B. Haines, and W. Swank. 1991. Ecology 72: 254-266.

5. Valiela. I., G. Collins, J. Kremer, K. Lajtha, M. Geist, B. Seely, J.
Brawley, and C. Sham. 1997. Ecnl. Appl. 7: 358-380.

6. Seely, B., K. Lajtha, and G. Salvucci. 1998. Biogenchemistry 42:
326-343.

7. Vitousek, P., and W. Reiners. 1979. Science 204: 469-474.

8. Gorham, E., P. Vitousek, and W. Reiners. 1979. Anmi. Rc\: Ecol.
ami S\st. 10: 53-84.

Reference: Biol. Bull. 201: 290-292. (October 2001 1

Small-Scale Heterogeneity of Nitrogen Concentrations in Groundwater
at the Seepage Face of Edgartown Great Pond

Alyson M. Hauxwell1, Christopher Neill, Ivan Valiela, and Kevin D. Kroeger (Ecosystems Center and Boston
University Marine Program, Marine Biological Laboratory, Woods Hole, Massachusetts 02543)

Groundwater transports nitrogen to receiving estuaries (1, 2),
but the details of nitrogen exchange, transformations, and losses
are insufficiently known (3). We examined small-scale heteroge-
neity of salinity, ammonium (NH4). nitrate (NO,), dissolved or-
ganic nitrogen (DON), and boron in both vertical and horizontal
profiles near the sandy seepage face of Edgartown Great Pond
(Martha's Vineyard. Massachusetts). The water level of this pond
is managed by dredging an outlet to release accumulated ground-
water; from April to June 2001 the pond was open to the sea.
Sampling for this study was done in June 2001. We focused on
nitrogen because of its role in limiting estuarine production (4),
and on boron because it can be used as a tracer of both wastewater
(5) and seawater (6). Relationships among these solutes allow
inquiry as to sources of the materials and identity of some major
processes (5).

To describe in detail the pattern of distributions of the solutes.
we collected groundwater samples along three parallel transects
running perpendicular to shore from 4.5 m upland to 1 .5 m beyond
the shoreline of Edgartown Great Pond (Fig. IA. B, C). Each
transect consisted of seven points, each sampled at about 16 cm. 41
cm, 66 cm, 91 cm, and 1 16 cm below the ground surface. We
collected 125 ml of water from each point using a well point
piezometer and pressure pump. The samples were filtered through
47-mm glass fiber filters to remove particulates. We measured
salinity using a refractometer and ammonium concentrations using
the alkaline phenol method. Nitrate and TON concentrations were
measured on a Lachat autoanalyzer using the Quick Chem method,
and DON was determined by subtracting NH4 and NO, concen-
trations from TDN concentrations for each sample. Ward Labora-
tories (Kearney, NE) determined boron concentrations on a subset
of the samples.

Salinity (Fig. IA) and ammonium (Fig. IB) concentrations in
groundwater increased seaward. In contrast, NO, (Fig. 1C) con-
centrations decreased seaward. Vertical cross-sections of concen-
trations (Fig. ID, E, F) along the top transect shown in Figure 1 (A.
B, C) suggest how groundwater flow interacts with horizontal
transportation to determine the small-scale patterns of concentra-

1 University of Michigan, Ann Arbor. MI 48109.

tion across the seepage face of this estuary (Fig. ID. E, F). Salinity
of groundwater was 0%<- and increased to 17%<^19%r. under the
pond, about half the salinity of the pond (28<?r) (Fig. ID). The
contours suggest that the fresh groundwater flows over the saltier
water, and discharges in a seepage face a few meters wide. Am-
monium concentrations were highest under the pond and at in-
creasing depths, with one high value under land (Fig. IE). NHj
concentrations increase as salinities increase beyond 13%r (Fig.
1G). This increase is not due to NH4 imported from land to the
pond, or from the pond (the pond has a concentration of only 2 ;uM
NH4). Nitrate concentrations were highest landward and decreased
offshore, with a smaller peak seaward (Fig. IF). DON did not
change significantly through each transect (data not shown) and
decreased only slightly with increasing depth.

One possible explanation for the high NH4 associated with salty
water may be that the pond bottom shares the vertical pattern of
high NH4 concentrations characteristic of anoxic coastal sedi-
ments, with upward diffusion of NH4 regenerated within the sed-
iments by decay of buried organic matter (7). This explanation
seems implausible because 1 ) none of the water samples had a
sulfide odor, hence were not anoxic, and 2) it is difficult to explain
the peak in NO, concentrations if we simply had freshwater
continually flowing toward the seepage face. Perhaps a more
plausible idea is that during the open-to-the-sea stage of the year,
seawater intrudes into the pore space in sediments at the seepage
face, and the Na+ displaces NH4 previously adsorbed to particles.
Such a mechanism has been invoked in the displacement of radium
from many shorelines (8). This mechanism also has the advantage
that it will account for the NO, peak shoreward of the NH4 peak:
during the open-pond phase, saltwater may force its way landward,
and nitrification could transform the exchanged NH4 into NO, as
the porewater moves landward. In most other such estuaries tidal
forces may repeat the pattern that occurs once a year in Edgartown
Great Pond and probably hide the local pattern of concentrations.
This pond hence provides a slow-motion view of what probably
occurs twice daily in tidal dominated estuaries.

The high NO, concentrations landward are likely to be associ-
ated with a wastewater source (Fig. 1H); the concentrations of NO,
are too high to be atmospheric nitrogen passing through soil (W.

ECOLOGY AND POPULATION BIOLOGY

291

meters

Vl 1 ifl 1 1 1 III?

0 000

Salinity (%0)

0 001

0 0 0 2/13<618

5 2 1 1 1)1

Ammonium (//M)

4 340

1 0

Land

Pond

Land

Pond

Land

Pond

Salinity (%o)

Pond: 28 %»

Ammonium

Nitrate (A<M)

Pond: 0/yM

r .16 m

.41 m

f .66 m

- .91 m

1-1.16 m

100

60 *

40 -i

20 J

A
A
fl

L&?. ^.JjAft n

(

) 20 40

60 80

Salinity (%

«)

100 -

H 60^

80 -

3. 40 -

3, 60 -

• E

1 4°-

'E
o

| 20 -

20 -

*• JL *

A
•

»* * n

»»»*» •

H> A H U "1

1 2 4

Boron (ppm)

2 4
Boron (ppm)

Figure 1. (A, B. Ct Plan view of transects where each value is an average for all depths at the location showing (A ) salinity (contour line inten'al 5%c);
(B) NHj (contour line intenal 20 fj.M): (C> NO, (contour line inten'al 10 ju.M| Gray area is land and white area is water. (D, E, F): Cross-section of
transect Figure I (A, B. C) showing (D) salinity (contour line inten'al 5%o); (E! NHj (contour line inten'al 20 nM): (F) NO3 (contour line inten'al 10 /j,M).
(G) NH4 (open shapes) and NO, (closed shapes) concentrations versus salinity for freshwater (•). estuarine water (AJ, and pond water (U). Dashed line
is peak NO, value from soil sources (Pabich et al., unpubl. data). (H) Boron concentrations versus NO, concentrations for fresh, estuarine, and pond water.
(I) Boron concentrations versus NH4 concentrations for fresh (FW), estuarine (EW). and pond water (PI

Pabich et al., unpubl. data), and show higher boron concentrations
than would be likely in uncontaminated sediments (Fig. 1H). The
pattern of boron concentrations supports the idea that in freshwater
the high NO, peak derives from some source on land, probably
wastewater. Boron concentrations within a septic plume can have

ranges above 0.2 ppm boron (5). In saltier groundwater, some
other process produces the smaller peak in concentration under the
pond (Fig. 1H) — in our view, nitrification of the displaced NH4.
Concentrations of solutes in groundwater were variable normal
to the shoreline. The presence of small-scale transient NH4 and

292

REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

NO3 fronts that are the result of local processes in the fresh-
seawater mixing zone at the interface can alter ideas as to how we
estimate land-derived N loads, and can provide insight into the
processes that transfer nitrogen between fresh and saltwater. Un-
derstanding this heterogeneity is important for establishing a
meaningful sampling protocol to estimate nitrogen loading to an
estuary and for interpreting the likely sources.

This research was funded by a National Science Foundation-
Research Experience for Undergraduates Grant (OCE-0097498).
Special thanks to Ann Lezberg and Eve Hinckley for help with
chemistry protocols and to Marci Cole and Joanna York for their
endless help in the lab. Also thanks to William Wilcox of the
Martha's Vineyard Commission and the Kohlberg family for use
of their land.

Literature Cited

1. Valiela, I., G. Collins, J. Kremer, K. Lajtha, M. Geist, B. Seely, J.
Brawley, and C. H. Sham. 1997. Ecol. Appl. 7: 358-380.

2. Giblin, A. E., and A. G. Gaines. 1990. Biogeochemistry 10: 309-
328.

3 Portnoy, J. W., B. L. Nowicki, C. T. Roman, and D. W. Urish. 1998.

Water Rex. 34: 3095-3104.

4. Howarth, R. W. 1988. Aniui. Rev. Ecol. Syst. 19: 89-110.
5 Westgate, E. J., K. D. Kroeger, W. J. Pabich, and I. Valiela. 2000.

Bi,,l. Bull. 199: 221-223.

6. Barth, S. R. 2000. Appl. Geocliem. 15: 937-952.

7. Valiela, I. 1995. P. 437 in Marine Ecological Processes. Springer-
Verlag, New York.

8. Moore, W. S. 2000. J. Geophys. Res. 105: 1 17-122.

Reference: Biol. Bull. 201: 292-294. (October 2001)

Top-down vs. Bottom-up Controls of Microphytobenthic Standing Crop: Role of Mud Snails
and Nitrogen Supply in the Littoral of Waquoit Bay Estuaries

Melissa Novak1 , Mark Lever, and Ivan Valiela (Boston University- Marine Program,
Marine Biological Laboratory, Woods Hole, Massachusetts 02543)

Top-down and bottom-up processes are important in the regu-
lation of primary productivity. In shallow estuaries, the mud snail
llyanassa obsolete may exert strong top-down forces on the bio-
mass of microphytobenthos ( 1, 2), and nutrient availability in the
sediments may also affect microphytic biomass (3, 4, 5, 6). We
examined the relative importance of top-down and bottom-up
effects by experimentally manipulating mud snail densities and
porewater nutrient concentrations.

Twenty cages were set up in Sage Lot Pond in Waquoit Bay.
Massachusetts. The cages were placed in two rows of 10. paral-
leling the shoreline on sandy, subtidal sediments, approximately 7
cm below mean low tide level. The cages were constructed from
plastic boxes (23 cm - 23 cm x 8 cm). The bottom was removed
and windows were cut into the lid and sides. The windows allowed
water exchange and minimized artifacts caused by interference
with water flow. To prevent the escape of enclosed snails, each
window was covered with a 4-mm polypropylene mesh. Each cage
was pushed into the sediment to a depth of approximately 5 cm.

To determine the significance of bottom-up effects, 10 cages
were treated using "diffusers" made of polypropylene microcen-
trifuge tubes with holes drilled into their sides, filled with control-
release fertilizer (9.7% NHj , 8.3% NO,". 6% P2O,). In each cage,
nine tubes were evenly spaced and fully pushed into the sediment,
do that the top was 1 cm below the sediment surface. The nutrient
additions were equivalent to 60 g N/m2 and 8.7 g P/m2.

The success of the fertilizer treatment was established by sam-
pling porewater from six randomly selected locations within each
cage at day 0, 7, 14. 28. and 38. All porewater samples were
pooled samples collected from the entire upper 1 cm of sediment

1 University of Rhode Island, Kingston, RI 02881.

using sippers constructed from modified 10-ml polyethylene sy-
ringes. Samples were filtered through 47-mm glass fiber tillers
with 0.7 p,m pore sizes. Concentrations of phosphate were deter-
mined using a spectrophotometer (7), ammonium following a
fluorometric method (8), and nitrate by using a LACHAT auto
analyzer following the QuikChem method.

To assess top-down effects, densities of 0, 20, 50, 100, and 200
snails per cage were randomly assigned to different cages. The
mean ambient density of mud snails in the area at time 0 was 97
snails/cage (.? = 4.14). Each experimental density was applied to
two cages in each of the fertilized and control treatments. Snails
were recounted after each sampling to ensure that densities were
maintained throughout the experimental period.

To measure the response of benthic microphytes to fertilization and
snail density treatments, six core samples were taken from each cage
at each sampling date. Each coring device consisted of a cut-off 10-ml
syringe with a diameter of 0.95 cm and length of 2.5 cm. Chlorophyll
a concentrations were analyzed spectrophotometrically (9).

Both the fertilization and snail density treatments were effective.
Nutrient concentrations in the upper layer of sediments in the
fertilized cages were significantly higher than in control plots (Fig.
la, b. c) (one-tailed / test: phosphate. P < 0.009; ammonium,
P < 0.004; nitrate. P < 0.03). We note that concentrations of
nutrients in estuarine sediments often exceed those found in our
samples (10). but our concentrations are within the range we find
in the upper 1 cm of sandy substrate in Waquoit Bay sub-estuaries
(M. Lever, unpubl. data). The snail counts in the various cages
remained constant over the course of the experiment.

Linear regressions of chlorophyll a concentrations vs. time were
used to calculate rates of change of the microphyte biomass in each
cage. These rates were then plotted against snail density for both

ECOLOGY AND POPULATION BIOLOGY

293

NOs

-^control
-•-fertilized

PO4

-0.6

O control
• fertilized

40 80 120

Snail density (snails/cage)

160

200

df MS

Fertilizer

Density

Pert. X Density

1 0.12 7.0*
4 0.67 40.2**
4 0.01 0.69

Figure 1. (a-c) Time courses of nitrate (at, ammonium (b). and phosphate (c) concentrations in pore water in control and fertilised cages, (d) Change
in chlorophyll a as a function of snail density. Data points are calculated from regressions of change in chlorophyll over time for individual cages, (e)
Two-wav ANOVA of the results of chlorophyll response to the fertilizer and density treatments. * indicates significance at P = 0.05,' ** indicates
significance at P = 0.01.

fertilized and control treatments (Fig. Id). A two-way ANOVA
showed that both snail densities and nutrient concentrations had
significant effects on chlorophyll a concentrations (Fig. le).

In terms of the effect of herbivore density, we interpret the
results of Fig. Id to suggest that 1) lower snail densities allowed
increases in microphytobenthic biomass (note the position of the
dashed horizontal line showing no change), and 1) snail densities
exceeding the ambient of 97 snails per cage reduced microphyte
biomass. These results suggest that mud snails can control abun-
dance of their food, which means, perhaps, that field densities are
poised at a level that does not deplete the food supply.

In terms of the effect of nutrient supply, the addition of nutrients
significantly increased microphytobenthic biomass (Fig. le). In
spite of the trends suggested by Fig. Id. the response of micro-
phyte biomass to fertilization was not significantly larger at lower
snail densities (Fig. le). The results were insufficient to allow us to
discern the possible interaction between grazing pressure and
nutrient supply, in part because there were only two levels of the

nutrients examined, and because even in the fertilized sediments
concentrations were relatively low.

The results of this experiment suggest that both bottom-up and
top-down processes can be potentially important controls of benthic
microphytes in estuarine sediments. Further experiments in which a
broader range of fertilizer loads is applied will help to determine the
relative importance of top-down vs. bottom-up controls.

Jennifer Wolf helped with the fieldwork. This work was sup-
ported by NSF-Research Experience for Undergraduates Grant
OCE-0097498.

Literature Cited

1 Pace, M. L., S. Shimmel, and W. M. Darlev 1979. Estuar. Coast.
Mar. Sci. 9: 121-134.

2. Connor, M. S., J. M. Teal, and I. Valiela. 1982. J. Exp. Mar. Biol.
Ecol. 65: 29-45.

3. Admiraal, W., H. Peletier, and H. Zomer. 1982. Estuar. Coast.
Shelf Sci. 14: 471-487.

294

REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

4 Maclntyre, H. L., R. J. Geider, and D. C. Miller. 1996. Estuaries
19: 186-201.

5. van Raalte, C. D., I. Valiela, and J. M. Teal. 1976. Limnol.
Oceanogr. 21: 862-872.

6. Graneli, E., and K. Sundback. 1985. J. Exp. Mar. Bio/. Ecol. 85:
253-268.

7. Strickland, J. D. H., and T. R. Parsons. 1972. Pp. 49-64 in A

Practical Handbook of Sea Water Analysis, Fisheries Research Board
of Canada, Ottawa.

8. Holmes, R. M., A. Aminot, R. Kerouel, B. A. Hooker, and B. J.
Peterson. 1999. Can. J. Fish. Aauat. Sci. 56: 1801—1808.

9. Lorenzen, C. J. 1967. Limnol. Oceanogr. 12: 343.

10. Valiela, I. 1995. Pp. 59-78 in Marine Ecological Processes,
Springer, New York.

Reference: Bio/. Bull. 201: 294-296. (October 2001)

Stable N Isotopic Signatures in Bay Scallop Tissue, Feces, and Pseudofeces

in Cape Cod Estuaries Subject to Different N Loads

Laurie Fila1, Ruth Herrold Carmichael, Andrea Shrive r, and Ivan Valiela (Boston University
Marine Program, Marine Biological Laboratory, Woods Hole, Massachusetts 02543)

Scallops (Argopecten irradians) feed on participates in estuar-
ies, and their growth and survival may depend on the quality and
quantity of food particles available ( 1 , 2). To a significant degree,
particle supply in shallow estuaries such as those on Cape Cod
depend on rates of land-derived N load (3). Linkages between
estuarine organisms and terrestrial loadings have been studied in
various ways, including stable isotopic techniques. Isotopic frac-
tionation leads to detectable shifts created by microbial transfor-
mations, trophic steps, as well as to differences due to source of the
N (4, 5).

In this paper we apply isotopic analyses and experiments with
introduced scallops to define the rate at which scallops acquire the
signature of the estuary in which they are located; we examine
whether scallop tissues differ from the signatures of pseudofeces
and feces ejected by scallops, and whether differences in N-
loading rates and sources to different estuaries result in corre-
sponding differences in the signature of scallops within the estu-
aries. Finally, we use results of the introduced scallop experiments
to see if differences in 8'5N signature acquisition are related to
differences in growth or survival of the scallops.

We compared the acquisition of 815N signatures by scallops
incubated in two estuaries of Waquoit Bay, Cape Cod, receiving
different N inputs. Childs River (CR) has a loading rate of 601 kg
N ha"1 y~'. Sage Lot Pond (SLP) has a loading rate of 14 kg N
ha"1 y~'. The difference in N load between these estuaries is due
to different levels of urbanization in their watersheds, and the
differences in wastewater contributions to these two estuaries
result in different isotopic signatures in the N entering the estuaries
from land (5, 6). Juvenile scallops (40-50 mm) were obtained
from Taylor Seafood, Fairhaven, Connecticut. In each estuary we
placed four plastic-coated wire cages, each containing 20 scallops.
Cages were secured 10 cm above the sediment surface in 1 m of
water at mean low tide.

To monitor the acquisition of the 5'5N signature in tissue and
ejecta over time, we removed one cage of scallops from each
estuary on days 3, 6. 12, and 24. Animals were immediately placed
in filtered seawater for 24 hours to clear their guts. Feces and
pseudofeces were filtered through pre-ashed, 7-/j,m Whatman

' Mount Holyoke College

GF/F filters. Scallop tissue was dissected from the shell and dried
at 60 °C overnight. Ejecta were acidified to remove carbonates,
and samples not collected on filters were homogenized.

We determined the 815N signatures of potential food sources,
paniculate organic matter in water (POM, or seston) and sedi-
ments. In each estuary, we sampled the water column and sedi-
ments near the cages on days 0, 3, 6, 12, and 24. Water column
samples were processed in the same manner as ejecta. The top 1
cm of sediment was sampled using a 5-cc syringe as a corer. We
combined four sediment cores for each sample. Sediment samples
were acidified and homogenized. All samples were analyzed using
a Europa Scientific Integra mass spectrometer at the University of
California-Davis.

To determine scallop growth over time, length of shells of
animals from each cage were measured with vernier calipers
accurate to 0.1 mm. The number of dead scallops per cage were
counted on each collection day.

The 8'5N values of tissue from scallops grown in each estuary
were initially 9.23% and during the course of the field incubation
approached S15N values of POM in water and sediments, corrected
by an expected trophic fractionation of 3% (4) (Fig. 1A, B). For
example, if scallops in CR were feeding only on sediments, we
extrapolate that the scallops, at the measured rate of change in
tissue signature, would converge on the mean sediment signature
(corrected by a 3% trophic fractionation) in 93 days. Similarly, if
the scallops were feeding on only seston. the convergence would
take place in 60 days. For the scallops in SLP, the convergence
time would be shorter: 47 days and 36 days, respectively.

To examine whether scallops eject fractionated food particles, we
compared the 815N signature of ejecta (feces + pseudofeces) to the
SI5N signature of food supply from each estuary. Lighter 515N sig-
natures for food in SLP corresponded to lighter 815N signatures in
ejecta from the scallops grown in SLP, while heavier S15N food
signatures in CR corresponded to heavier ejecta signatures from the
CR scallops (Fig. 1C). In both estuaries, the 815N signature of ejecta
was equal to or heavier than that of potential food sources (Fig. 1C).
In addition, 8'5N signatures of ejecta were lighter than 815N signa-
tures of tissue in CR [8.75%-9.85% (Fig. 1A)] and SLP [0.07%-
9.23% (Fig. IB)]. The 2%-3% enrichment from food to ejecta agrees
with trophic level fractionation reported in the literature. The relative
similarity between the 8I5N signatures of seston and sediments makes

ECOLOGY AND POPULATION BIOLOGY

295

1 1 n

C/5

7 -

- 5 H

CHILDS RIVER

Pred 515N

Seston

Sediment

60 d
93d

12 18

Day

24 30

SAGE LOT POND

9 i

6 -

03^

3 -

A x

1:1

8 AA

CR SLP

Sediment • O
Seston ^ ^

Growth
(mm)

Day

SLP

Mortality

CR SLP

3 0.5 ± 0.04 0.6 ± 0.1 0 0

6 0.8 ± 0.1 1.0 ± 0.2 0 0

12 9.5 ± 7.5 3.0 ± 0.3 10 0

24 0.8 ± 0.1 5.6 ± 0.7 20 0

01234567

515N food source (%o)

Figure 1. (A. B) b'5N signature of tissue from scallops grown in Childs River (A) and Sage Lot Pond (B) vs. time. CR regression: y = -0.02\ + 9.37,
F = 1. 08 ns. SLP regression: y = ~O.IO\ + 9.01, F = 10.85*. Predicted 8'5/v" signature lines for tissue are derived from mean seston and sediment
signatures, +3% to correct for trophic shift. The lines represent predicted ultimate tissue signatures for scallops assuming exclusive consumption of either
food source. (C) Sj:iN signature of scallop ejecta (feces + pseudofeces) is generally heavier than that of food sources (seston and sediment). (D) Mean
(± std. error) scallop growth (measured as cumulative change in shell length) and mortality over time, in each study estuary. Mean growth was calculated
using a subsampling of the individuals in the cage fn = 10).

it difficult, however, to determine which food source contributed most
to the diet of scallops during this study.

The faster rate at which SLP scallops approached predicted
815N signatures of their food sources (Fig. 1A, B) may be related
to the faster growth of scallops in SLP (Fig. ID). SLP scallops
grew more quickly and achieved greater length than CR scallops
(Fig. ID). Mean growth rates (from incremental growth data) are
0.24 ± 0.03 mm/day for SLP. and 0.14 ± 0.01 mm/day for CR. In
addition, no scallops in SLP died during the study, whereas those
in CR reached 20% mortality by day 24 (Fig. ID). The data
suggest that conditions in CR were less favorable for scallops than
conditions in SLP. This could be related to lower water quality in
CR (7). which could have lowered feeding rate and possibly
altered the rate of internal turnover of nitrogen within the scallop
tissue.

Scallop 615N signatures moved toward the signatures of their
presumed food sources at a rate suggesting they would converge
with trophic-shift-corrected 8I5N food signatures in 1-3 months of

feeding. Material ejected by scallops had heavier 815iN signatures
than potential food signatures but lighter than tissue signatures.
The increased wastewater N load in CR coincided with a slower
convergence of tissue signatures to trophic-shift-corrected food
signatures, lowered growth, and increased mortality.

Thanks to Marci Cole, Gabby Tomasky, Joanna York, and Mar-
shall Otter for technical assistance, and the residents of 71 Childs
River Road for providing site access. This work was supported by
NSF-Research Experience for Undergraduates Grant OCE-0097498
and the Five College Coastal and Marine Sciences Program's partic-
ipation in the Woods Hole Marine Sciences Consortium.

Literature Cited

1. Cahalan, J., S. E. Siddall, and M. W. Luckenback. 1989. J. Exp.
Mar. Biol. Ecol. 129: 45-60.

2. Rheault, R. B., and M. A. Rice. 1996. J. Shellfish Res. 15:

271-283.

296

REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

3. Valiela, I., G. Tomasky, J. Hauxwell, M. L. Cole, J. Cebrian, and
K. D. Kroeger. 2000. Ecol. ,4/>/>/. 10: 1006-1023.

4. Peterson, B., and B. Fry. 1987. ,4»/i». Rev. Ecol. Syst. 18: 293-320.

5. McClelland, J., I. Valiela, and R. Michener. 1997. Limnol. Ocean-
ogr. 42: 930-937.

6. Valiela. I., M. Geist, J. McClelland, and G. Tomasky. 2000. Bii>-
geochemistn- 49: 277-293.

7. Valiela, 1., K. Foreman, M. LaMontagne. D. Hersh, J. Costa, P. Peckol.
B. DeMeo-Anderson, C. D'Avanzo, M. Babione, C.-H. Sham, J. Braw-
ley, and K. Lajtha. 1992. Estuaries 15: 433-457.

Reference: Bio/. Bull. 201: 296-297. (October 2001)

Age Structure of the Pleasant Bay Population of Crepidula fornicata:
A Possible Tool For Estimating Horseshoe Crab Age

Sam P. Gmdy, Deborah Rntecki, Ruth Cannichael, and Ivan Valiela (Boston University Marine Program,
Marine Biological Laboratory, Woods Hole, Massachusetts 02543)

Crepidula fornicata. the common slipper shell, lives on rocks,
horseshoe crabs (Limit/us polyphemus), and other hard surfaces,
often in stacks of one animal atop another. Unlike many other
gastropods, they tend to remain sessile, and as they grow, their
shells contour to the substrate ( 1 ). The association between horse-
shoe crabs and C. fornicata offers the possibility to use the slipper
shell as a tool to determine the ages and average lifespan of
horseshoe crabs (2). Knowing this information would be helpful
for trying to understand horseshoe crab ecology for use in conser-
vation efforts.

It is difficult to directly estimate horseshoe crab age because
horseshoe crabs lack any hard parts that could be sectioned and
analyzed for growth rings. Their chitinous exoskeleton is molted
with decreasing frequency until a theoretical "terminal molt" (3).
There are also a variety of sizes within visually estimated age
classes because growth is very slow or stops in adults (3).

Other methods have been suggested for aging horseshoe
crabs, including qualitative aging criteria and tagging studies.
From the results of tagging studies it has been estimated that
horseshoe crabs live 9 to 12 years before maturity and 5 to 7
years as adults, for a total lifespan of 14 to 19 years (4). These
age spans are consistent with the prediction of Botton and
Ropes (2) based on laboratory work using C. fornicata as a
proxy for horseshoe crab age.

C. fornicata could indicate age of host horseshoe crabs if 1 )
horseshoe crabs have a terminal molt or do not molt often as
adults, 2) C. fornicata remain on the same horseshoe crab, and 3)
C. fornicata age can be determined with some degree of accuracy
(5). It is also assumed that C. fornicata attach to a host horseshoe
crab as soon as the new cuticle hardens.

Botton and Ropes (2) used a regression proposed by Walne ( 1 )
of C. fornicata length to age to quantitatively estimate the ages of
horseshoe crabs. These C. fornicata were used to formulate this
regression without comparison to a local population of horseshoe
crabs, since the C. fornicata data was from England and horseshoe
crabs were not measured at all.

In this study we measured shell length of C. fornicata and
prosomal width of Linniliis polyphemus in Pleasant Bay. Chatham,
Massachusetts. We measured 496 crabs and their corresponding C.
forniciiui, with the number of C. fornicata per crab ranging widely
from 1 to 30, with an average of 4 per crab. From these data we
fitted cohorts of C. fornicata to a size-frequency distribution. We
also related size of C furnictiui to prosomal width of L.

-g- 64
J 48 -

§• 32 -

64 -

ro "E

5.5

n = 1336

8.9 17.6 27.0 37.9 460

Crepidula shell length (mm)

y = 9.4499X - 5.5932
R2 = 0.9981

234
Time (years)

°CO

/ .

n=122

* ss

A %

4 I?

3 6 S

2 o^

144 176 208 240 272 304

Horseshoe crab prosomal width (mm)

Figure 1. I A) Cohorts of Pleasant Bay population of Crepidula
fornicata: 8.9mm (-1.5 y). 17.6 mm (-2.5 y). 27.0 mm (-3.5 y). 37.9
mm (-4.5 y). and Jft.O mm (-5.5 yl. (B) C. fornicata length vs. age:
i:\trtifiuliititin data from Button and Ropes (21. (C) Length of largest C.
fornicata on horseshoe crabs of different prosomal width. Filled circles
(•) represent malex.open circles (O) represent females.

ECOLOGY AND POPULATION BIOLOGY

297

ux to see if C. fomicata could provide a proxy for L
polyphemus size and age.

The analysis of cohorts demonstrated that C. fornicata in Pleas-
ant Bay can he divided into 5 size cohorts (Fig. 1A), with C.
fornictita of approximately 4-6 mm in length appearing to repre-
sent the most recent spatfall. The cohorts differed in abundance,
reflecting different magnitudes of recruitment from year to year.
Growth rates in this study did not decrease with increased size and
age (Fig. IB). This may be due to low numbers of larger (50 mm
+ ) and older C. fomicata. Published data of sizes and ages (1,2)
match those found in this study, and thus confirm the conversion
from size to age of the C. fomicata. The largest C. fornicata found
resident on a horseshoe crab was 58 mm. This size C. fornicata
could be from 8-11 years old (2).

There was no evident relationship between maximum length and
age of C. fomicata and size of the host horseshoe crabs (Fig. 1C).
Male horseshoe crabs were consistently smaller than females, but
in both sexes the length and presumed age of C. fomicata varied
greatly, and was independent of the size of the crab.

It is not possible to establish a strong relationship between true
horseshoe crab length and the length of the C. fomicata upon it. At
most the data of Figure 1C support that a minimum age can be
calculated by adding the maximum C. fornicata length on a given
horseshoe crab to the minimum age of horseshoe crabs at maturity.
Using 9 years as the age at maturity (4), the crabs in this study
were from 12 to 17 years old.

This study was funded by the Woods Hole Marine Science
Consortium and a grant from the Friends of Pleasant Bay.

Literature Cited

1. Walne, P. R. 1956. Fish. Investig. 6: 1-50.

2. Botton, M. L., and J. W. Ropes. 1988. /. Shellfish Res. 7: 407-412.

3. Shuster, C. 1950. Third rept. investigations of methods of improv-
ing the shellfish resources of Massachusetts. WHOl Contr. No. 564:
IS -23.

4. Shuster, C. 1982. Pp. 1-52 in Physiology and Biology of Horseshoe
Crabs. Alan R. Liss, New York.

5. Ropes, J. 1961. Trans. Am. Fish. Soc. 90: 79-80.

Reference: Biol. Bull 201: 297-299. (October 2001)

Hydrogen Peroxide: An Effective Treatment for Ballast Water

Alan M. Ku-irian, Eleanor C. S. Terry, Deanna L. Bechtel, (Marine Biological Laboratory,
Woods Hole. Massachusetts 02543), and Patrick L. James1

Introduced species have been a problem in the marine and
coastal environments for centuries. Historically, many of these
introductions have a strong geophysical component often associ-
ated with natural disasters. However, in more recent times, "man,
the supreme meddler" ( 1 ) has dramatically changed the rate, num-
ber, and geography of exotic species invasions through importa-
tion, transportation, intentional releases related to agriculture or
aquaculture, as well as unintentional escapes. During the last
century, the problem has dramatically accelerated with the advent
of modern high-speed freighters and their methods of ballast water
exchange.

Transport and discharge of biocontaminated ballast water con-
stitutes a major route (29%) by which potentially invasive spe-
cies— from plants and algae to fish, invertebrates, planktonic and
bacterial micro-organisms, and even potential pathogens — are in-
troduced into coastal waters worldwide. It is estimated that 3000
species are transported daily via ballast water (National Research
Council, 2000). The Great Lakes have experienced the introduc-
tion of at least 129 non-indigenous species (2). while the San
Francisco Bay estuary has recorded 234 exotic species with at least
an additional 125 cryptogenic species (3). At the current estimated
rate, a new species is introduced into the ecosystem every 14
weeks (3).

The problem is not confined to the United States but occurs
worldwide. One noteworthy example is the introduction of the
western Atlantic ctenophore. Mnemiopsis leiclyi, into the Black and
Azov Seas in 1987 and 1988, respectively. This invader has been

' Eltron Research, Inc., Boulder. CO.

blamed for a 20-fold decrease in zooplankton biomass, the subse-
quent sharp decline in anchovy and other pelagic fish stocks, and
a marked disruption in these ecosystems (4).

The United Nations International Maritime Organization (IMO),
established in 1991, developed a voluntary ballast water exchange
(BWE) at sea policy that has now become mandatory (5). BWE is
carried out either by draining and refilling the ballast tanks or by
continuous flushing equivalent to three volume exchanges. The
policy is based upon the rationale that coastal organisms will not
survive at sea and vice versa, so BWE is simpler, less costly, and
thus preferable to controls implemented before departure or upon
arrival (i.e., land-based treatments). Unfortunately, BWE is only
90%-95% effective, and the exchange itself can be dangerous in
foul weather or can produce excessive hull stress. Therefore,
alternative ballast water treatments are being sought.

Some current technologies available for ballast water treatment
include filtration, cyclone or hydrotech-drum settling, UV, ultra-
sonics, and heat. Additional secondary treatment methods include
biocides, ozone, electric pulse or pulse plasma, deoxygenation. and
biological. Some of the biocidal methods involve the storage of
dangerous chemicals and cause unacceptably high levels of cor-
rosion (e.g.. hypochlorite). However, hydrogen peroxide, gener-
ated on-site at low (safe) concen'rations, precludes these hazards
and is more cost-effective than the sophisticated and high-energy-
demanding equipment necessary for ozone generation. Neutral
hydrogen peroxide has been effective in a number of studies, but
only at moderately high concentrations (10-50 ppm; [6]), for
planktonic and some small neustonic organisms. Because most
marine organisms and bacteria cannot tolerate pH extremes (7),
hydrogen peroxide combined with elevated pH (alkaline hydrogen

298

REPORTS FROM THE MBL GENERAL SCIENTIFIC MEETINGS

peroxide) has the potential to produce synergistic effects useful for
treating ballast water. Since alkaline peroxide has not been inves-
tigated for this application, and it is the consequence of the
proposed generating process, we undertook a toxicological labo-
ratory study to test the effects of alkaline peroxide on plankton.
This study is designed to complement the development of an
electrolytic cell (based on patented technology [8]) that is capable
of producing alkaline hydrogen peroxide. An upscale design of the
cell has been proposed for use onboard ship to treat ballast water
to reduce the potential introduction of invasive species.

Plankton were collected from the local waters off Woods Hole,
Massachusetts, by the Aquatic Resources Department of the Ma-
rine Biological Laboratory (MBL). Indigenous zooplankters (Ta-
ble 1 ) were used in this study. The faunal composition of the
plankton varied between collections, but the majority were dom-
inated by crustaceans, both planktonic adults and larvae of benthic
species. Particular attention was directed toward the effects of
alkaline peroxide on the ctenophore. Mnemiopsis leidyi, a known
invasive species (see above), which made up over 90% of many
plankton collections.

The following treatment regimes were tested: 1 ) natural seawa-
ter (NSW) at elevated pH (using NaOH. 8.5, 9.0. 9.5. or 10); and
2) NSW at the four elevated pHs with the addition of 1, 3, or 10
ppm (=mg/l) of hydrogen peroxide (3% USP grade). Hydrogen
peroxide was added first, the pH adjusted (with NaOH), and the
solutions allowed to equilibrate for 30 min (the solution at pH 10
with 10 ppm peroxide precipitated and was removed from the
testing matrix). The pH was re-adjusted before the addition of
mixed zooplankton (minimum of 25 animals/condition). Times for
50% (LD50) and 100% mortality (mortality time) were recorded.
Mortality time was defined as the point when all of the plankton
species ceased movement and became unresponsive to tactile
stimuli. To ensure the accuracy of mortality time determinations,
all animals were returned to NSW (via serial dilution) after each
treatment to test for recovery. LDSOs were also calculated and
compared with mortality times; the values averaged 43% of the
100% mortality times. Since 100% mortality was the desired
outcome, the data were reported using that method. Plankton left in

Table 1

List of major taxa of species present in the mixed plankton samples

Phylum Cnidaria
Class Hydrozoa

Pennaria sp. (Medusae)

Phylum Ctenophora
Mnemiopsis leidvi

Phylum Annelida
Class Polychaeta
Platynereis sp.

(epitoke stages and eggs)

Phylum Arthropoda
Class Crustacea

Hi'inants americanus
(advanced larval stages)

Ovalipes ocellatus

(zoea, megalops stages)
Procellanid zoea
Calanoid copepods
Euphausids. spp.
Mysid shrimp, spp.

Phylum Mollusca
Class Bivalvia

Various larval spp.

Phylum Chordata
Class Osteichthyes
Syngnathidae sp.
Various larval spp.

45-

40-

I 30-

™ 15-
O

2 10-

Treatment Regimes (H O ppm/pH)

Figure 1. Combined effects of pH and peroxide on mixed plankton.
Mortality times (i.e.. time to 100% mortality) were recorded when all
sH'iiniiiiiii; activity had ceased and all of the animals were unresponsive to
tactile stimuli. Animals were subsequently place in natural seawater (NSW)
ami ohM'rvt'il for recovery as a test of these end points. No recovery was
observed.

NSW (pH 7.8-8.0) served as controls. There were at least six
replicates for each treatment.

The ctenophore, Mnemiopsis leidyi, was tested with the same
treatment regimes. Because of their size and buoyancy, even when
dead, for accuracy it was necessary to record mortality times when
the compound cilia of the comb rows and the cilia in the digestive
tract both ceased beating. The test animals were placed back into
NSW and observed for signs of recovery. All data were analyzed
statistically using ANOVA or Student's t paired comparisons.

Plankton placed in NSW with elevated pHs all survived for at
least 24 h. and the majority of those in pH 8.5-9.5 were alive for
as long as three days. Only those animals at pH 10 did not survive
beyond 24 h. Mnemiopsis responded similarly.

When solutions containing mixed plankton and alkaline perox-
ide were tested, no significant differences were found between pH
values within each peroxide concentration (ANOVA: F values <
2: P > 0.2) (Fig. 1). However, for each peroxide concentration,
there were significant decreases in mortality times (Student's t
paired comparisons: / > 4.7; P < 0.001). Similar results were
obtained with Mnemiopsis; i.e.. there were no pH effects within
each peroxide concentration (ANOVA: F < 1.8; P > 0.15).
However, increases in peroxide concentrations significantly short-
ened mortality times (Fig. 2). When animals were placed in 10
ppm peroxide, beating of all the comb rows immediately stopped;
and within seconds, the activity of the digestive cilia also ceased.
Therefore, the effects of this concentration were not graphed. The
difference between the means of the two peroxide concentrations
(1 and 3 ppm) was highly significant, with Student's t value / >
//.5 \\-itli P < 0.001.

The results indicate that, up to pH 10. the increased alkalinity
has little toxic effect on either mixed plankton or ctenophores; and
survivorship after 24 h was equal to NSW controls. Hydrogen
peroxide, even at 1 ppm, had a mean ( 100%) mortality time of 30
min for mixed plankton samples; and for the ctenophores. the
times were even shorter ( X3.7). Peroxide at 3 ppm was three times

ECOLOGY AND POPULATION BIOLOGY

299

E E E E

a a. a. a.

~ ~ n «•

Treatment Regimes (H O /pH)

Figure 2. Effects of pH and peroxide concentrations on the cteno-
phore, Mnemiopsis leidyi.

more efficient at disinfection than 1 ppm for plankton and approx-
imately twice as effective for the comb jellies. When 10 ppm
peroxide was used with mixed plankton, the mortality times de-
creased again, this time at twice the rate. The within-treutment
variance was extremely low for both the 3- and 10-ppm peroxide,
and thus the means between the two treatment regimes were highly
significant. For the ctenophore Mnemiopsis, 10-ppm peroxide was
essentially lethal upon contact (<1 mini.

In summary, the data from these tests indicate that NSW at pHs
slightly elevated above that of ambient, and containing concentra-
tions of 1 ppm hydrogen peroxide, can be lethal to plankton
composed of a wide phylogenetic mix of species (Table 1 ). It was

interesting to discover that a concentration of 3-ppm peroxide has
effects comparable to ozone levels (2.2 ppm) when tested on larvae
of the nudibranch mollusc, Hermissenda crassicornis (9). The
short exposures (i.e., mortality times) required at this concentration
of peroxide should encourage the development and implementa-
tion of an onboard electrolytic system capable of generating the
required peroxide levels at rates sufficient to treat ballast water of
ships during uptake at sea or in coastal waters. This device would
provide an efficient, low-energy cost treatment for ballast water,
and would preclude the bulk and danger of storing concentrated
biocide chemicals on board ships.

This research was supported by a Phase I. SBIR/EPA grant
(68-D-01-017) to Eltron Research. Inc.

Literature Cited

1 . Laycock, G. 1966. The Alien Animals. Natural History Press, Garden
City. NY.

2. Mills, E. L., J. H. Leach, J. T. Carlton, and C. L. Secor. 1993. /
Gt. Lakes Res. 19: 1-54.

3. Cohen, A. N., and J. T. Carlton. 1998. Science 279: 555-558.

4. Kideys. A. E. 1994. J. Mar. Sv.vr. S: 171-181.

5. Carlton. J. T. 1992. Pp. 23-26 in Introductions and Transfers of
Marine Species. R. DeVoe, ed. South Carolina Sea Grant Consortium.
Charleston. SC.

6. Laughton, R., T. Moran, and G. Brown, n.d. Polluted! Technical
Papers [Online]. Available: http://www.pollutech.com/papers/p22.htm
[22 August 2001].

7. Oemcke, D. 1999. The Treatment of Ship's Ballast Water. Ecoports
Monography Series 1 8. Ports Corporation of Queensland. Brisbane. P.
102.

8. White, J., M. Schultz. and A. Sammells. 1997. United States Patent.
US-5645700.

9. Kuzirian, A. M.. C. T. Tamse, and M. Heath. 1990. Bio/. Bull. 179:
227.

Reference: Biol. Bull. 201: 300. (October 20011

Published by Title Only

Aldrich, Stephen, R. Gil Pontius, Jr., Takashi Tada, and
Luc Claessens

Influence of land use on nitrate loading in the Ipswich
River Watershed, Massachusetts

Clay, John

Action potentials occur spontaneously in squid giant ax-
ons with moderately alkaline intracellular pH

Denton, Jerod, and J. C. Leiter

Identification of CO^-chemosensitive and non-chemosen-
sitive neurons in the right parietal ganglion of the pul-
monate snail. Helix aspersa

Heck, Diane, Lydia Louis, and Jeffrey Laskin

17-beta-estradiol modulates gastrulation in the sea urchin
Arhacia punctulata

Jaffe, Lionel

Action potential velocities along working heart muscles
are highly conserved and may be calcium based

Martel, David, Rainer Voigt, and Jelle Atema

The Limulus worm (Bdelloitra Candida) prefers individ-
ual horseshoe crab (Limulus polyphemus) odor

Unis, Jennifer, Christopher Neill, and Richard McHorney

Predicting groundwater flow rate at Edgartown Great
Pond on Martha's Vineyard, Massachusetts: salinity
and groundwater flow at the seepage face of a coastal
pond

300

Biomimetic Engineering Conference

March 3-8. 2002 Sandestin, Florida

Biological organisms exhibit sophisticated crystal engineering
capabilities that underlie the remarkable material properties of
mineralized tissues such as bone and teeth, and the beautiful and
functional nacre of molluscs and abalone. Increasing interest is
being paid to nature's processing strategies, particularly by
materials scientists looking for bio-inspired methods to engineer
uniojje ceramics coatings or composites for use in magnetic,
optical, biomedical. and protective coatings applications. In
particular, the engineering of hard tissues may benefit from
biomimetic approaches since the benign conditions allow for the
incorporation of biomolecular compounds into the organic/
inorganic composite during fabrication.

On the other hand, there is great interest from the biomedical
community because the disruption of normal biomineralization
processes may lead to pathological conditions, such as in
arteriosclerotic plaque formation, encrustation of biomaterials
(such as urinary catheters and artificial heart valve calcification),
kidney stone buildup, dental calculus formation, or bone and tooth
demineralization.

The main objective of this conference is to bring together
scientists, physicians, and engineers in a relaxed environment, with
talks designed to bridge the gap between researchers in this
interdisciplinary field.

This groundbreaking conference will have sessions that deal with
(a) Biomineralization in Nature: Vertebrates and Invertebrates
(Inspiration for Design Principles), (b) Organic Modulators of
Cystallization: Templated Nucleation and Crystal Growth
Modification, (c) Engineering Strategies: Bioinspired Materials
and Novel Physicochemical Properties, and (d) Applications of
Biomimetic Materials: Devices and Processes.

Each day will open with a keynote address to highlight the day's
topics; each of the sessions will include eight to ten presentations.
with afternoons left free for ad hoc meetings and informal
discussions. An evening poster session will also promote dialogue
among the attendees.

The Chair of the Conference is Dr. Allison A. Campbell of the
Pacific Northwest National Laboratory in Richland. Washington.
and the Co-Chair is Prof. Laurie Cower of the University of
Florida in Gainesville.

Additional information about this Conference — and a registra-
tion form — can be found at

The United Engineering Foundation is located at

Three Park Avenue, 27th Floor, New York. NY 10016-5902;

Tel: 212-591-7836. Fax: 212-591-7441. E-mail: engfnd@aol.com

ARINE IXESOURCES CENTER

MARINE BIOLOGICAL LABORATORY • WOODS HOLE, MA 02543 • (508)289-7700
WWW.MBL.EDU/SERVICES/MRC/INDEX.HTML

Animal and Tissue Supply for
Education & Research

• 150 aquatic species available for shipment via
online catalog: <http://www.mbl.edu/animals/
index.htmb: phone: (508)289-7375; or
e-mail: specimens@mbl.edu

• zebrafish colony containing limited mutant strains

• custom dissection and furnishing of specific organ
and tissue samples

zebrafish facilities

MRC Services Available

• basic water quality analysis

• veterinary services (clinical, histopathologic,
microbial services, health certificates, etc.)

• aquatic systems design (mechanical, biological,
engineering, etc.)

• educational tours and collecting trips aboard
the R/V Gemma

Using the MRC for Your Research

• capability for advanced animal husbandry (temperature, light control, etc.)

• availability of year-round, developmental life stages

• adaptability of tank system design for live marine animal experimentation

What makes our bio-imaging software great?
The people behind the product.

Senior Application Scientist Dr. Neal
Cliksman assists customers during the
Mai/ 2001 Customer Training Course at
Universal Imaging Institute.

Come see us at

the Society for Neuroscience
30th Annual Meeting
San Diego, CA
Booth 1042

With the growing complexity of
applications and devices, bio-imaging
has never been more challenging.
That's why educators call on us to
support imaging courses at Marine
Biological Laboratory, Cold Spring
Harbor Laboratory, Mt. Desert Island
Biological Laboratory and other
locations. They want the best for their
students and we give them our best.
Only Universal Imaging provides this
level of complete coverage to so many
courses. Our customers see this
dedication every day. In the world
of bio-imaging support, no one is as
Universal.

To register for our next course, go to
www.universal-imaging.com/
resources/training.cfm

UNIVERSAL IMAGING CORPORATION
Working to Improve Your Image
www.universal-imaging.com/people

610-873-5610

Software and
Systems for:

~~l 6-D Imaging
Brightness
Measurements
Colocalization

Cell-based
Screening

• Intracellular
Calcium, pH, etc.
Ratio Imaging
FRET

Morphometry
Motion
Analysis

• Time-Lapse
and more...

Makers of

MetaMorph

Marine Biological Laboratory

2002 Course Offerings

Advances In Genome Technology &
Blolnformatlcs

October 6 - November 1

Analytical & Quantitative Light Microscopy
May 9- May 17

Biology of Parasitism: Modern Approaches

June 13- August 10

Embryology: Concepts & Techniques In
Modern Developmental Biology

June 16- July 27

Frontiers In Reproduction: Molecular &
Cellular Concepts & Applications

May 19 - June 29

Fundamental Issues In Vision Research

August 11 - August 24

Medical Informatics

1st Session May 26 - June 2
2nd Session September 29 - October 6

Methods In Computational Neuroscience

August 4 - September 1

Mlcroblal Diversity

June 16 - August 2

Mlcrolnjectlon Techniques In Cell Biology

May 21 - May 28

Molecular Biology of Aging

July 30 -August 17

Molecular Mycology: Current Approaches
to Fungal Pathogenesls

August 12 - August 30

Substantial financial assistance is
available for many of our courses!

For more information contact:

Carol Hamel,
Admissions Coordinator

(508) 289-7401

admissions@mbl.edu

http://courses.mbl.edu

The MBL is an EEC/Affirmative Action Institution

Neural Development & Genetics
of Zebraflsh

August 18 - August 31

Neural Systems & Behavior

June 16 - August 10

Neuroblology

June 16- August 17

Neurolnformatlcs

August 17 - September 1

Optical Microscopy & Imaging In the
Biomedlcal Sciences
October 9 - October 18

Physiology: The Biochemical & Molecular
Basis of Cell Signaling

June 16- July 27

Rapid Electrochemical Measurements In
Biological Systems

May 9 - May 13

Summer Program In Neuroscience, Ethics,
& Survival (SPINES)

June 15- July 13

Workshop on Molecular Evolution

July 28 -August 9

Marine Biological Laboratory, 7 MBL Street, Woods Hole, MA 02543

doors to new world

The Next Generation:

Laser Scanning Microscope LSM 51
META for multi-channel fluorescen
in single and multiphoton microsa

META - a unique confocal
microscope detector will separate
overlapping fluorescence dyes
for sharp, crosstalk free images.

Carl Zeiss Microlmaging, Inc.

One Zeiss Drive
Thornwood, NY 10594

1.800.233.2343

micro@zeiss.com

www.zeiss.com/micro

ZEISS

Volume L'Ol

\ umber !>

BIOLOGICAL
BULLETIN

DECEMBER 2001

Published by the Marino Biological Laboratory

http://www.biolbull.ofg

THE BIOLOGICAL BULLETIN

ONLINE

The Marine Biological Laboratory is

pleased to announce that the full text of

The Biological Bulletin is available online at

http://www.biolbull.org

Published since 1897 by the Marine
Biological Laboratory (MBL) in Woods Hole,
Massachusetts, The Biological Bulletin is one
of America's oldest peer-reviewed scientific
journals.

The journal is aimed at a general readership,
and especially invites articles about those
novel phenomena and contexts characteristic
of intersecting fields.

http://www.biolbull.org

Made to my exact
specifications.

Here's how the BX2's modular design came
through for me. First, the 8 position universal
condenser offers the flexibility to choose
from hrightfield, darkfield and phase. Next,
it's assortment of DIC prisms makes it possi-
ble to match the optical image shear to the
specimen, achieving the optimal balance of
contrast and resolution. Finally, the motor-
ized nosepiece, Z-drive, condenser, illuminator
and filter wheels are fully integrated
through the user-friendly software package.
So digital images can now be acquired,
processed and analyzed faster than before.
Now let's move on.

Mai

DtL 2 6

And yours.

Rcture yourself sitting here, looking into your
Olympus BX2 research microscope, your
fluorescence requirements having been met.
Specifically: The aspherical collector lens
produces a fluorescence intensity that's
twice as bright as others. The unique excita-
tion balancers improve visualization of
multiple labels. The six-position filter turret
makes single and multiband imaging faster
and simpler. And the rectangular field stop,
another Olympus exclusive, protects the
specimen by exposing only the precise area
being imaged. With all this modularity and
flexibility, my BX2 microscope is also your
BX2 microscope.

Andy

Now modularity re£
Olympus FLUOVlEi
laser scanning micr
5 imaging channels
intuitive operation I
productivity is grea
the BX2 is the only
a Metal Matrix Corn
static and thermal r
use of 3D microsco
time-lapse observat
high-end digital ima
resulting in a compl
confocal system. It
optimal solution.

Research Microscope Series

OLYMPUS

FOCUS ON LIFE

Cover

Two striking features of the open ocean are the lack
of obvious physical features and the extraordinary
clarity of the water, in which visibility extends to
about 100 meters. Animals that live in this pelagic
realm are, of course, extremely vulnerable to being
seen, identified as food, and eaten. To meet the
challenge of predation, oceanic animals have
evolved various tactics for hiding in an environment
that precludes hiding. Effective adaptations include
counterillumination, countershading, and mirrored
sides; but the most apt mechanism is surely whole-
body transparency — an emulation of the habitat.
Therefore, although transparency is rare on land or
in coastal waters, it is extremely common in the
open ocean and is closely tied to the pelagic life-
style.

On the cover of this issue is a photograph of an
octopus, Vitreledonella richardi, which spends its
entire life swimming in the subtropical and tropical
regions of the world's oceans at depths of 200-
1000 meters. Not only are most of its tissues highly
transparent, but it is further modified for camou-
flage: its non-transparent elongated gut and eyes are
continually oriented vertically and thus cast a min-

imal shadow toward potential predators that may be
cruising below. (Credits: photo by David Wrobel,
Monterey Bay Aquarium).

For many years, the biology of animals like Vitrele-
donella was completely overlooked. Most transpar-
ent species are fragile, and they are therefore de-
stroyed by the sampling nets deployed from
oceanographic vessels. In a sense, therefore, these
cryptic organisms were hidden from biologists, as
well as from their predators. The development of
manned and robotic submersibles, blue water div-
ing techniques, and optical equipment that is both
portable and reliable has greatly increased our
knowledge of these animals and their transparency.

In this issue, Sonke Johnsen reviews our current
understanding of biological transparency. This
Meld — still in its infancy — includes empirical stud-
ies by marine and fresh-water biologists, but espe-
cially work on such camouflage-breaking visual
abilities as ultraviolet and polarization vision. The-
oretical and empirical research into the physical
basis of biological transparency is also being car-
ried out, much of it driven by the need to prevent
and treat lapses in ocular transparency, such as
cataracts.

THE

BIOLOGICAL BULLETIN

DECEMBER 2001

Editor
Associate Editors

Section Editor
Online Editors

Editorial Board

Editorial Office

MICHAEL J. GREENBERG

Louis E. BURNETT
R. ANDREW CAMERON
CHARLES D. DERBY
MICHAEL LABARBERA

SHINY A LNOUE, Imaging and Microscopy

JAMES A. BLAKE, Keys to Marine
Invertebrates of the Woods Hole Region
WILLIAM D. COHEN, Marine Models
Electronic Record and Compendia

PETER B. ARMSTRONG
ERNEST S. CHANG
THOMAS H. DIETZ
RICHARD B. EMLET
DAVID EPEL

KENNETH M. HALANYCH
GREGORY HINKLE
NANCY KNOWLTON
MAKOTO KOBAYASHI
ESTHER M. LEISE
DONAL T. MANAHAN
MARGARET MCFALL-NGAI
MARK W. MILLER
TATSUO MOTOKAWA
YOSHITAKA NAGAHAMA
SHERRY D. PAINTER
J. HERBERT WAITE
RICHARD K. ZIMMER

PAMELA CLAPP HINKLE
VICTORIA R. GIBSON
CAROL SCHACHINGER
WENDY CHILD

The Whitney Laboratory, University of Florida

Grice Marine Biological Laboratory. College of Charleston
California Institute of Technology
Georgia State University
University of Chicago

Marine Biological Laboratory

ENSR Marine & Coastal Center, Woods Hole

Hunter College, City University of New York

University of California, Davis

Bodega Marine Lab., University of California, Davis

Louisiana State University

Oregon Institute of Marine Biology, Univ. of Oregon

Hopkins Marine Station, Stanford University

Woods Hole Oceanographic Institution

Cereon Genomics. Cambridge, Massachusetts

Hiroshima University of Economics, Japan

Scripps Inst. Oceanography & Smithsonian Tropical Res. Inst.

University of North Carolina Greensboro

University of Southern California

Kewalo Marine Laboratory, University of Hawaii

Institute of Neurobiology, University of Puerto Rico

Tokyo Institute of Technology, Japan

National Institute for Basic Biology, Japan

Marine Biomed. Inst., Univ. of Texas Medical Branch

University of California, Santa Barbara

University of California, Los Angeles

Managing Editor

Staff Editor

Editorial Associate

Subscription & Advertising Secretary

Published by

MARINE BIOLOGICAL LABORATORY
WOODS HOLE, MASSACHUSETTS

http://www.biolbull.org

Genomic Research Leaders Choose
Microway® Scalable Clusters

Eos Biotechnology, Marine Biological Laboratory, Millennium Pharmaceu-
ticals, Mount Sinai Medical School, NIH, Pfizer, and Rockefeller University
All Choose Microway Custom Clusters and Workstations for Reliability,
Superior Technical Support and Great Pricing.

1.4 GHz Dual Athlon, 1.7 GHz Pentium 4,
or 1 GHz Dual Pentium in in 1U or 2U
Clusters

Dual Alpha 833 MHz Clusters and Towers

For maximum price/performance choose our
Alpha 1U 833 MHz, 4 MB DDR Cache CS20,
4U UP2000+ or 4U 264DP RuggeclRack ™

Myrinet, Gigabit Ethernet or Dolphin Wulfkit
High Speed Low Latency Interconnects

RAID and Fibre Channel Storage Solutions

Microway1"1 Screamer™
Dual Alpha UP20001
833 MHz, 4MB Cache in
RuggedRack™ Chassis
with RRR™ Redundant
Power Supply

Microway has delivered high-performance computing prod-
ucts since 1982, when our pioneering software made it pos-
sible to use an 8087 in the IBM-PC In 1987 we created the
world's first PC parallel processing systems. Since then, our
QuadPuter™ architecture has migrated from Transputers to
i86()s and finally to Alphas in 1995. Over the past three years,
we engineered and delivered over 300 clusters that utilized
MPI running on Linux. As a software developer and hard-
ware manufacturer, we know the value of extensive testing
and validation. We are experts at configuring and validating
the low latency interconnects we employ in our clusters.
Our technical support is legendary — the systems we sell
arrive at your site and WORK! Los Alamos chose Microway to
maintain and upgrade its 144 node Alpha Avalon Cluster
because of our reputation. Large clusters we have sold
include 400+ nodes at the University ofWisconsin and 250+
nodes at Rockefeller University.

Microway offers three Athlon/
Pentium enclosures — 1U, 2U and
tower, plus five Alpha configura-
tions—1U, 3U, 4U RuggedRack™,
QuadPuter™ and full tower. Our
264DP includes two 21264's with
up to 4 GB of memory in our cus-
tom 4U RuggedRack. which features
front accessible redundant power
supplies and hard disks. This rugged
configuration was chosen by the
U.S. Navy for onboard use. We also
offer a dual Alpha UP2000+ running
at 833MHz with 2GB of memory.
Our QuadPuter chassis holds 4
Alpha processors and up to 4GB
memory. The 1U CS20 dual Alpha
(at right) is the highest density
computational platform available.

Microway is API-Networks' Top
North American

Channel
Partner.

* ^iNetWorks

"Most Powerful, Highest Density
Computational Platform On the Planet'

Microway Scalable 25 Mode

50 Processor Cluster Using CS20 Dual

833 MHz Alphas and Myrinet Interconne

Yielding Peak Throughput of 82,5 Gigaflo

I have ordered numerous Alpha and Intel-based servers and work
stations from Microway running both Tru64 UNIX and Linux. W<
have been very happy with both the performance and great valuf
of Microway's products. The major UNIX vendors don't come clos<
to Microway in this regard, and we have also found that Microwa\
provides better value than other Linux hardware vendors. I hav<
also used Microway's tech support and was pleased with thei
response. We've been using their systems for over a year anc
have had only a couple of minor incidents which were dealt witl
promptly."

— David Kristofferson, Ph.D., MBA,
Director of Information Systems, Eos Biotechnology, Inc.l

Find out why over 75% of Microway's

sales come from repeat customers.

Please call 508-746-7341 for a

technical salesperson -who

speaks your language!

Visit us at www.microway.com

Technology* yOU Can COUnt OH" M Research Park Box 79, Kingston. MA 02364 • 508-746-7341 • info@microway.co

CONTENTS

VOLUME 201. No. 3: DECEMBER 2001

REVIEW

ECOLOGY AND EVOLUTION

Johnsen, Sonke

Hidden in plain sight: the ecology and physiology of
organismal transparency 301

RESEARCH NOTE

Hibbett, David S., and Manfred Binder

Evolution of marine mushrooms

Helmuth, Brian S. T., and Gretchen E. Hofmann

Microhabitats, thermal heterogeneity, and patterns
of physiological stress in the rocky intertidal zone. . . . 374
Rossi, Sergi, and MarkJ. Snyder

Competition for space among sessile marine inverte-
brates: changes in HSP70 expression in two Pacific
cnidarians . 385

319

CELL BIOLOGY

Leys, Sally P., and Bernard M. Degnan

Cytological basis of photoresponsive behavior in a
sponge larva 323

PHYSIOLOGY AND BIOMECHANICS

Shimomura, Osamu, Per R. Flood, Satoshi Inouye,
Bruce Bryan, and Akemi Shimomura

Isolation and properties of the luciferase stored in
the ovarv of the scyphozoan medusa Periphylla pe-
riphylla 339

DEVELOPMENT AND REPRODUCTION

Bishop, Cory D., and Bruce P. Brandhorst

NO/cGMP signaling and HSP90 activity repress
metamorphosis in the sea urchin Lftfchinus pictus. . . . 394

Furuya, Hidetaka, F. G. Hochberg, and Kazuhiko Tsuneki
Developmental patterns and cell lineages of vermi-
form embrvos in dicyemid mesozoans 405

Kossevitch, Igor A., Klaus Herrmann, and Stefan Berking
Shaping of colony elements in Laomedea flexuosa
Hinks (Hydrozoa, Thecaphora) includes a temporal
and spatial control of skeleton hardening 417

SYMBIOSIS AND PARASITOLOGY

Toller, W. W., R. Rowan, and N. Knowlton

Zooxanthellae of the Montastraea annularis species
complex: patterns of distribution of four taxa of Sym-

biodinium on different reefs and across depths 348

Toller, W. W., R. Rowan, and N. Knowlton

Repopulation of zooxanthellae in the Caribbean cor-
als Montastraea annularis and M. faveolata following
experimental and disease-associated bleaching 360

NEUROBIOLOGY AND BEHAVIOR

Dufort, Christopher G., Steven H. Jury, James M. Newcomb,
Daniel F. O'Grady HI, and Winsor H. Watson m

Detection of salinity by the lobster, Homarus america-
nus. . 424

Index for Volume 201 435

THE BIOLOGICAL BULLETIN

THE BIOLOGICAL BULLETIN is published six times a year by the Marine Biological Laboratory, 7 MBL Street,
Woods Hole, Massachusetts 02543.

Subscriptions and similar matter should be addressed to Subscription Manager, THE BIOLOGICAL BULLETIN,
Marine Biological Laboratory, 7 MBL Street, Woods Hole, Massachusetts 02543. Subscription per year (six
issues, two volumes): $260 for libraries; $105 for individuals. Subscription per volume (three issues): $130 for
libraries; $52.50 for individuals. Back and single issues (subject to availability); $45 for libraries; $20 for
individuals.

Communications relative to manuscripts should be sent to Michael J. Greenberg, Editor-in-Chief, or Pamela
Clapp Hinkle, Managing Editor, at the Marine Biological Laboratory, 7 MBL Street, Woods Hole, Massachusetts
02543. Telephone: (508) 289-7428. FAX: 508-289-7922. E-mail: pclapp@mbl.edu.

http://www.biobull.org

Printed on acid free paper,
effective with Volume 180, Issue 1, 1991.

POSTMASTER: Send address changes to THE BIOLOGICAL BULLETIN, Marine Biological Laboratory,

7 MBL Street, Woods Hole, MA 02543.

Periodicals postage paid at Woods Hole, MA, and additional mailing offices.

ISSN 0006-3185

INSTRUCTIONS TO AUTHORS

Very short, especially topical papers (less than 9 manuscript
pages including tables, figures, and bibliography) will be published
in a separate section entitled "Research Notes." A Research Note
in The Biological Bulletin follows the format of similar notes in
Nature. It should open with a summary paragraph of 150 to 200
words comprising the introduction and the conclusions. The rest of
the text should continue on without subheadings, and there should
be no more than 30 references. References should be referred to in
the text by number, and listed in the Literature Cited section in the
order that they appear in the text. Unlike references in Nature,
references in the Research Notes section should conform in
punctuation and arrangement to the style of recent issues of The
Biological Bulletin. Materials and Methods should be incorpo-
rated into appropriate figure legends. See the article by Loh-
mann et al. (October 1990, Vol. 179: 214-218) for sample
style. A Research Note will usually appear within two months
after its acceptance.

The Editorial Board requests that regular manuscripts con-
form to the requirements set below; those manuscripts that do
not conform will be returned to authors for correction before
review.

1. Manuscripts. Manuscripts, including figures, should be
submitted in quadruplicate, with the originals clearly marked.
(Xerox copies of photographs are not acceptable for review pur-
poses.) The submission letter accompanying the manuscript should
include a telephone number, a FAX number, and (if possible) an
E-mail address for the corresponding author. The original manu-
script must be typed in no smaller than 12 pitch or 10 point, using
double spacing (including figure legends, footnotes, bibliography,
etc.) on one side of 16- or 20-lb. bond paper, 8 by 1 1 inches.
Please, no right justification. Manuscripts should be proofread
carefully and errors corrected legibly in black ink. Pages should be
numbered consecutively. Margins on all sides should be at least 1
inch (2.5 cm). Manuscripts should conform to the Council of
Biology Editors Style Manual, 5th Edition (Council of Biology
Editors. 1983) and to American spelling. Unusual abbreviations
should be kept to a minimum and should be spelled out on first
reference as well as defined in a footnote on the title page.
Manuscripts should be divided into the following components:
Title page. Abstract (of no more than 200 words). Introduction.
Materials and Methods. Results, Discussion, Acknowledgments.
Literature Cited. Tables, and Figure Legends. In addition, authors
should supply a list of words and phrases under which the article
should be indexed.

2. Title page. The title page consists of a condensed title or
running head of no more than 35 letters and spaces, the manuscript
title, authors' names and appropriate addresses, and footnotes

listing present addresses, acknowledgments or contribution num-
bers, and explanation of unusual abbreviations.

3. Figures. The dimensions of the printed page, 7 by 9
inches, should be kept in mind in preparing figures for publication.
We recommend that figures be about 1 times the linear dimensions
of the final printing desired, and that the ratio of the largest to the
smallest letter or number and of the thickest to the thinnest line not
exceed 1:1.5. Explanatory matter generally should be included in
legends, although axes should always be identified on the illustra-
tion itself. Figures should be prepared for reproduction as either
line cuts or halftones. Figures to be reproduced as line cuts should
be unmounted glossy photographic reproductions or drawn in
black ink on white paper, good-quality tracing cloth or plastic, or
blue-lined coordinate paper. Those to be reproduced as halftones
should be mounted on board, with both designating numbers or
letters and scale bars affixed directly to the figures. All figures
should be numbered in consecutive order, with no distinction
between text and plate figures and cited, in order, in the text. The
author's name and an arrow indicating orientation should appear
on the reverse side of all figures.

Format. Acceptable graphic formats are TIFF and EPS. Color
submissions must be in EPS format, saved in CMKY mode.

Sofhvare. Preferred software is Adobe Illustrator or Adobe
Photoshop for the Mac and Adobe Photoshop for Windows. Spe-
cific instructions for artwork created with various software pro-
grams are available on the Web at the Digital Art Information Site
maintained by Cadmus Professional Communications at http://
cpc.cadmus.com/dal/

Resolution. The minimum requirements for resolution are
1200 DPI for line art and 300 for halftones.

Size. All digital artwork must be submitted at its actual
printed size so that no scaling is necessary.

Multipanel figures. Figures consisting of individual parts
(e.g., panels A, B, C) must be assembled into final format and
submitted as one file.

Hard copv. Files must be accompanied by hard copy for use
in case the electronic version is unusable.

Disk identification. Disks must be clearly labeled with the
following information: author name and manuscript number; for-
mat (PC or Macintosh); name and version of software used.

Color: The Biological Bulletin will publish color figures and
plates, but must bill authors for the actual additional cost of
printing in color. The process is expensive, so authors with more
than one color image should — consistent with editorial concerns,
especially citation of figures in order — combine them into a single
plate to reduce the expense. On request, when supplied with a copy
of a color illustration, the editorial staff will provide a pre-publi-
cation estimate of the printing cost.

4. Tables, footnotes, figure legends, etc. Authors should
follow the style in a recent issue of The Biological Bulletin in

preparing table headings, figure legends, and the like. Because of
the high cost of setting tabular material in type, authors are asked
to limit such material as much as possible. Tables, with their
headings and footnotes, should be typed on separate sheets, num-
bered with consecutive Roman numerals, and placed after the
Literature Cited. Figure legends should contain enough informa-
tion to make the figure intelligible separate from the text. Legends
should be typed double spaced, with consecutive Arabic numbers.
on a separate sheet at the end of the paper. Footnotes should be
limited to authors' current addresses, acknowledgments or contri-
bution numbers, and explanation of unusual abbreviations. All
such footnotes should appear on the title page. Footnotes are not
normally permitted in the body of the text.

5. Literature cited. In the text, literature should be cited by
the Harvard system, with papers by more than two authors cited as
Jones et al., 1980. Personal communications and material in prep-
aration or in press should be cited in the text only, with author's
initials and institutions, unless the material has been formally
accepted and a volume number can be supplied. The list of
references following the text should be headed Literature Cited,
and must be typed double spaced on separate pages, conforming in
punctuation and arrangement to the style of recent issues of The
Biological Bulletin. Citations should include complete titles and
inclusive pagination. Journal abbreviations should normally follow
those of the U. S. A. Standards Institute (USASI), as adopted by
BIOLOGICAL ABSTRACTS and CHEMICAL ABSTRACTS, with the minor
differences set out below. The most generally useful list of bio-
logical journal titles is that published each year by BIOLOGICAL
ABSTRACTS (BIOSIS List of Serials; the most recent issue). Foreign
authors, and others who are accustomed to using THE WORLD LIST
OF SCIENTIFIC PERIODICALS, may find a booklet published by the
Biological Council of the U.K. (obtainable from the Institute of
Biology, 41 Queen's Gate, London. S.W.7. England, U.K.) useful,
since it sets out the WORLD LIST abbreviations for most biological
journals with notes of the USASI abbreviations where these differ.
CHEMICAL ABSTRACTS publishes quarterly supplements of addi-
tional abbreviations. The following points of reference style for
THE BIOLOGICAL BULLETIN differ from USASI (or modified WORLD
LIST) usage:

A. Journal abbreviations, and book titles, all underlined (for
italics)

B. All components of abbreviations with initial capitals (not
as European usage in WORLD LIST e.g., J. Cell. Comp. Physiol.
NOT J. cell. comp. Physio! . )

C. All abbreviated components must be followed by a period,
whole word components must not (i.e., J. Cancer Res.)

D. Space between all components (e.g., J. Cell. Comp.
Phvsio/.. not J. Cell. Comp. Pltysiol. )

E. Unusual words in journal titles should be spelled out in
full, rather than employing new abbreviations invented by the
author. For example, use Rit Vi'sindafje/ags Islendinga without
abbreviation.

F. All single word journal titles in full (e.g., Veliger, Ecol-
ogy, Brain).

G. The order of abbreviated components should be the same 6. Reprints, page proofs, and charges. Authors may pur-

as the word order of the complete title (i.e.. Proi: and Trans. chase reprints in lots of 100. Forms for placing reprint orders are

placed where they appear, not transposed as in some BIOLOGICAL sent with page proofs. Reprints normally will be delivered about 2

ABSTRACTS listings). to 3 months after the issue date. Authors (or delegates for foreign

authors) will receive page proofs of articles shortly before publi-

H. A few well-known international journals in their preferred cation. They will be charged the current cost of printers' time for

forms rather than WORLD LIST or USASI usage (e.g.. Nature. corrections to these (other than corrections of printers' or editors'

Science, Evolution NOT Nature. Land., Science, N.Y.: Evolution. errors). Other than these charges for authors' alterations. The

Lancaster, Pa. ) Biological Bulletin does not have page charges.

CONTENTS

for Volume 201

No. 1: AUGUST 2001

RESEARCH NOTE

Seibel, Brad A., and David B. Carlini

Metabolism of pelagic cephalopods as a function of
habitat depth: a reanalvsis using phylogenetically in-
dependent contrasts

NEUROBIOLOGY AND BEHAVIOR

Herberholz, Jens, and Barbara Schmitz

Signaling via water currents in behavioral interac-
tions of snapping shrimp (Alpheus heterochaeits) ....

PHYSIOLOGY AND BIOMECHANICS

Reddy, P. Sreenivasula, and B. Kishori

Methionine-enkephulin induces hvperglycemia through
eyestalk hormones in die estuaiiiie crab Scylla seirata . . .
Mogami, Yoshihiro, Junko Ishii, and Shoji A. Baba
Theoretical and experimental dissection of gravity-
dependent mechanical orientation in gravitactic micro-
organisms 26

SYMBIOSIS AND PARASITOLOGY

Han ten, Jeffrey J., and Sidney K. Pierce

Asexual reproduction in Pygospio elegans Claparede
(Annelida, Polychaeta) in relation to parasitism by
Lepocreadium selij'eroides (Miller and Northup) (Platy-
helminthes, Trematoda)

DEVELOPMENT AND REPRODUCTION

Stewart-Savage, J., Aimee Phillippi, and Philip O. Yund

Delayed insemination results in embryo mortality in

a brooding ascidian 52

CELL BIOLOGY

P.. ill, ii in, Loriano. Antonella Franchini, Enzo Ottaviani,
and Armando Sabbadin

Morula cells as the major immunomodulatory hemo-
cytes in ascidians: evidences from the colonial species
Botryllus schlosseri 59

ECOLOGY AND EVOLUTION

Halanych, Kenneth M., Robert A. Feldman, and
Robert C. Vrijenhoek

Molecular evidence that Sclerolinum brattstromi is
closely related to vestimentiferans, not to frenulate
pogonophorans (Siboglinidae, Annelida) 65

Ponczek, Lawrence M., and Neil W. Blackstone

Effect of cloning rate on fitness-related traits in two
marine hydroids 76

Meidel, Susanne K., and Philip O. Yund

Egg longevity and time-integrated fertilization in a tem-
perate sea urchin (Strongylocentrotus droebafhiensu) .... 84

Wares, J. P.

Biogeography of Asterias: North Atlantic climate
change and speciation 95

SYSTEMATICS

Gershwin, Lisa-arm

Systematics and biogeography of the jellyfish Aurelia
labiata (Cnidaria: Scyphozoa) 104

45 Annual Report of the Marine Biological Laboratory. ... Rl

No. 2: OCTOBER 2001

RESEARCH NOTE

Maier, Ingo, Christian Hertweck, and Wilhelm Boland

Stereochemical specificity of lamoxirene, the sperm-
releasing pheromone in kelp (Laminariales, Phaeo-
phvceae) 121

PHYSIOLOGY AND BIOMECHANICS

Johnson, Amy S.

Drag, drafting, and mechanical interactions in cano-
pies of the red alga Chondnis crispus

CONTENTS: VOLUME 201

Thompson, Joseph T., and William M. Kier

Ontogenetic changes in fibrous connective tissue or-
ganization in the oval squid, Sepioteuthis kssoniana

Lesson, 1830 136

Thompson, Joseph T., and William M. Kier

Ontogenetic changes in mantle kinematics during
escape jet locomotion in the oval squid, Sepioteuthis

lessoniana Lesson, 1830 154

Martinez, Anne-Sophie, Jean-Yves Toullec, Bruce Shillito,
Mireille Charmantier-Daures, and Guy Charmantier
Hydromineral regulation in the hydrothermal vent
crab Bythograea thermydron 167

NEUROBIOLOGY AND BEHAVIOR

Campbell, A. C., S. Coppard, C. D'Abreo, and
R. Tudor-Thomas

Escape and aggregation responses of three echino-

derms to conspecific stimuli 175

Clay, John R., and Alvin Shrier

Action potentials occur spontaneously in squid giant
axons with moderately alkaline intracellular pH ... 186

SYSTEMATICS

Dahlgren, Thomas G., Bertil Akesson, Christoffer Schander,
Kenneth M. Halanych, and Per Sundberg

Molecular phylogeny of the model annelid Ophryotro-
cha.. 193

ECOLOGY AND EVOLUTION

Rondeau, Amelie, and Bernard Sainte-Marie

Variable mate-guarding time and sperm allocation by
male snow crabs (Chionoecetes opilio) in response to
sexual competition, and their impact on the mating
success of females 204

BIOGRAPHY

Inoue, Shinya, and Makoto Goda

Fluorescence polarization ratio of GFP crystals 231

CELL BIOLOGY

Knudson, Robert A., Shinya Inoue, and Makoto Goda

Centrifuge polarizing microscope with dual speci-
men chambers and injection ports 234

Tran, P. T., and Fred Chang

Transmitted light fluorescence microscopy revisited. . . . 235

Hernandez, R. V., J. M. Garza, M. E. Graves,
J. L. Martinez, Jr., and R. G. LeBaron

The process of reducing CA1 long-term potentiation
by the integrin binding peptide, GRGDSP, occurs
within the first few minutes following theta-burst
stimulation 236

Kuhns, William J., Dario Rusciano, Jane Kaltenbach,

Michael Ho, Max Burger, and Xavier Fernandez-Busquets
Up-regulation of integrins a, /3, in sulfate-starved ma-
rine sponge cells: functional correlates 238

Brown, Jeremiah R., Kyle R. Simonetta, Leslie A. Sandberg,

Phillip Stafford, and George M. Langford

Recombinant globular tail fragment of myosin-V blocks
vesicle transport in squid nerve cell extracts 240

Wollert, Torsten, Ana S. DePina, Leslie A. Sandberg,

and George M. Langford

Reconstitution of active pseudo-contractile rings and
myosin-II-mediated vesicle transport in extracts of
clam oocytes 241

Clay, John R., and Alan M. Kuzirian

A novel, kinesin-rich preparation derived from squid
giant axons

Weidner, Earl

Microsporidian spore/sporoplasm dynacdn in Spra-

243

guea

Zottoli, Steven J.

The origins of The Grass Foundation .

218

Conrad, Mara L., R. L. Pardy, and Peter B. Armstrong

Response of the blood cell of the American horse-
shoe crab. Limulus polyphemus, to a lipopolysaccha-
ride-like molecule from the green alga Chhrella. . . .
Silver, Robert

LtB4 evokes the calcium signal that initiates nuclear
envelope breakdown through a multi-enzyme net-
work in sand dollar (Echinaracnius parma) cells ....

245

246

248

SHORT REPORTS FROM THE 2001 GENERAL

SCIENTIFIC MEETINGS OF THE MARINE

BIOLOGICAL LABORATORY

FEATURED REPORT

The Editors

Introduction to the featured report, green fluores-
cent protein: enhanced optical signals from native
crystals 231

DEVELOPMENTAL BIOLOGY

Crawford, Karen

Ooplasm segregation in the squid embryo, Loligo

pealeii

Burbach, J. Peter H., Anita J. C. G. M. Hellemons,
Marco Hoekman, Philip Grant, and Harish C. Pant

The stellate ganglion of the squid Loligo pealeii as a
model for neuronal development: expression of a
POU Class VI homeodomain gene, Rpf-1 252

251

CONTENTS: VOLUME 201

Link, Brian A.

Evidence for directed mitotic cleavage plane reorien-
tations during retinal development within the ze-
brafish 254

Smith, Ryan, Emma Kavanagh, Hilary G. Morrison,

and Robert M. Gould

Messenger RNAs located in spiny dogfish oligoden-
drocyte processes 255

Hill, Susan D., and Barbara C. Boyer

Phalloidin labeling of developing muscle in embryos

of the polychaete Capitella sp. 1 257

Rice, Aaron N., David S. Portnoy, Ingrid M. Kaatz,

and Phillip S. Lobel

Differentiation of pharyngeal muscles on the basis of
enzyme activities in the cichlid Tramitichromis interme-
dius . 258

Zottoli, S. J., D. E. W. Arnolds, N. O. Asamoah,
C. Chevez, S. N. Fuller, N. A. Hiza, J. E. Nierman,
and L. A. Taboada

Dye coupling evidence for gap junctions between
supramedullary/dorsal neurons of the cunner, Tau-
togolabms adspersus 277

K.I.II/. Ingrid M., and Phillip S. Lobel
A comparison of sounds recorded from a catfish
(Orinocodoras eigenmanni, Doradidae) in an aquarium
and in the field 278

Fay, R. R., and P. L. Edds-Walton

Bimodal units in the torus semicircularis units of the
toadfish (Opsanus tan) 280

MARICULTURE

Xhl'ROBIOLOGY

Twig, Gilad, Sung-Kwon Jung, Mark A. Messerli,
Peter J. S. Smith, and Orian S. Shirihai

Real-time detection of reactive oxygen intermediates
from single microglial cells 261

Silver, Robert B., Mahlon E. Kriebel, Bruce Keller,

and George D. Pappas

Porocytosis: Quantal synaptic secretion of neuro-
transmitter at the neuromuscular junction through
arrayed vesicles 263

Chappell, Richard L., and Stephen Redenti

Endogenous zinc as a neuromodulator in vertebrate
retina: evidence from the retinal slice 265

Shashar, Nadav, Douglas Borst, Seth A. Ament,

William M. Saidel, Roxanna M. Smolowita,

and Roger T. Hanlon

Polarization reflecting iridophores in the arms of the
squid Loligo pealfii 267

Chiao, Chuan-Chin, and Roger T. Hanlon

Cuttlefish cue visually on area — not shape or aspect
ratio — of light objects in the substrate to produce
disruptive body patterns for camouflage 269

Errigo, M., C. McGuiness, S. Meadors, B. Mittmann,

F. Dodge, and R. Barlow

Visually guided behavior of juvenile horseshoe crabs ... 271

Meadors, S., C. McGuiness, F. A. Dodge,

and R. B. Barlow

Growth, visual field, and resolution in the juvenile
Limulus lateral eye 272

Kozlowski, Corinne, Kara Yopak, Rainer Voigt,

and Jelle Atema

An initial study on the effects of signal intermittency
on the odor plume tracking behavior of the Ameri-
can lobster, Homarus americanus 274

Hall, Benjamin, and Kerry Delaney

Cholinergic modulation of odor-evoked oscillations

in the frog olfactory bulb 276

Mensinger, Allen F., Katherine A. Stephenson,
Sarah L. Pollema, Hazel E. Richmond. Nichole Price,
and Roger T. Hanlon

Mariculture of the toadfish Opsnniis tau 282

Rieder, Leila E., and Allen F. Mensinger

Strategies for increasing growth of juvenile toadfish. . . . 283
Chikarmane, Hemant M., Alan M. Kuzirian, Ian Carroll,
and Robbin Dengler

Development of genetically tagged bay scallops for

evaluation of seeding programs 285

ECOLOGY AND POPULATION BIOLOGY

Williams, Libby, G. Carl Noblitt IV, and
Robert Buchsbaum

The effects of salt marsh having on benthic algal
biomass 287

Hinckley, Eve-Lyn S., Christopher Neill, Richard McHorney,

and Ann Lezberg

Dissolved nitrogen dynamics in groundwater under a
coastal Massachusetts forest 288

Hauxwell, Alyson M., Christopher Neill, Ivan Valiela,

and Kevin D. Kroeger

Small-scale heterogeneity of nitrogen concentrations
in groundwater at the seepage face of Edgartown
Great Pond 290

Novak, Melissa, Mark Lever, and Ivan Valiela

Top down vs. bottom-up controls of microphytobenthic
standing crop: role of mud snails and nitrogen supply
in the littoral of \Vaquoit Bay estuaries 292

Fila, Laurie, Ruth Herrold Carmichael, Andrea Shriver,

and Ivan Valiela

Stable N isotopic signatures in bay scallop tissue,
feces, and pseudofeces in Cape Cod estuaries subject
to different N loads 294

Grady, Sara P., Deborah Rutecki, Ruth Carmichael.

and Ivan Valiela

Age structure of the Pleasant Bay population of Cirfi-
idula fornicata: a possible tool for estimating horse-
shoe crab age 296

CONTENTS: VOLUME 201

Kuzirian, Alan M., Eleanor C. S. Terry,
Deanna L. Bechtel, and Patrick I. James

Hvdrogen peroxide: an effective treatment for ballast
water . 297

ORAL PRESENTATIONS

Published by title only 300

No. 3: DECEMBER 2001

REVIEW

ECOLOGY AND EVOLUTION

Johnsen, Sonke

Hidden in plain sight: the ecology and physiology of
organismal transparency 301

RESEARCH NOTE

Hibbett, David S., and Manfred Binder

Evolution of marine mushrooms. . . .

Helmuth, Brian S. T., and Gretchen E. Hofmann

Microhabitats, thermal heterogeneity, and patterns
of physiological stress in the rocky intertidal zone. . . . 374
Rossi, Sergi, and MarkJ. Snyder

Competition for space among sessile marine inverte-
brates: changes in HSP70 expression in two Pacific
cnidarians . 385

319

CELL BIOLOGY

Leys, Sally P., and Bernard M. Degnan

Cytological basis of photoresponsive behavior in a
sponge larva 323

PHYSIOLOGY AND BIOMECHANICS

Shimomura, Osamu, Per R. Flood, Satoshi Inouye,
Bruce Bryan, and Akemi Shimomura

Isolation and properties of the luciferase stored in
the ovary of the scyphozoan medusa Pmphylla pe-
riphylla. 339

DEVELOPMENT AND REPRODUCTION

Bishop, Cory D., and Bruce P. Brandhorst

XO/cGMP signaling and HSP90 activity repress
metamorphosis in the sea urchin Lytechinus pictus. . . . 394

Furuya, Hidetaka, F. G. Hochberg, and Kazuhiko Tsuneki
Developmental patterns and cell lineages of vermi-
form embryos in dicyemid mesozoans 405

Kossevitch, Igor A., Klaus Herrmann, and Stefan Berking
Shaping of colony elements in Laomedea flexuosa
Hinks (Hydrozoa, Thecaphora) includes a temporal
and spatial control of skeleton hardening 417

SYMBIOSIS AND PARASITOLOGY

Toller, W. W., R. Rowan, and N. Knowlton

Zooxanthellae of the Montaslraea annularis species
complex: patterns of distribution of four taxa of Sym-

biodinium on different reefs and across depths 348

Toller, W. W., R. Rowan, and N. Knowlton

Repopulation of zooxanthellae in the Caribbean cor-
als Montastmea annularis and M. faveolata following
experimental and disease-associated bleaching 360

NEUROBIOLOGY AND BEHAVIOR

Dufort, Christopher G., Steven H. Jury, James M. Newcomb,
Daniel F. O'Grady HI, and Winsor H. Watson m

Detection of salinity by the lobster, Homarus america-

424

Index for Volume 201 435

Notice to Subscribers

2002 SUBSCRIPTION RATES FOR THE BIOLOGICAL BULLETIN
Prices include print and electronic versions

Per year (six issues, two volumes):

Per volume (three issues):

Back and single issues: (subject to availability)

Libraries
$260.00
$130.00
$ 45.00

Individuals
$105.00
$ 52.50
$ 20.00

For additional information, please contact our subscription secretary at the Marine Biological Laboratory. 7 MBL Street, Woods
Hole, MA 02543: tel: (508) 289-7402: e-mail: wchild@mbl.edu. Visit our new website at www.biolbull.org

Reference: Bial. Bull. 201: 301-318. (December 2001)

Hidden in Plain Sight: The Ecology and Physiology
of Organismal Transparency

SONKE JOHNSEN*
Biology Department, Woods Hole Oceanographic Institution, Woods Hole, Massachusetts

Abstract. Despite the prevalence and importance of trans-
parency in organisms, particularly pelagic species, it is a
poorly understood characteristic. This article reviews the
current state of knowledge on the distribution, ecology, and
physical basis of biological transparency. Particular atten-
tion is paid to the distribution of transparent species relative
to their optical environment, the relationship between trans-
parency and visual predation, the physics of transparency,
and what is known about the anatomical and ultrastructural
modifications required to achieve this condition. Transpar-
ency is shown to be primarily a pelagic trait, uncommon in
other aquatic habitats and extremely rare on land. Experi-
mental and theoretical studies in terrestrial, freshwater, and
marine ecosystems have shown that transparency is a suc-
cessful form of camouflage, and that several visual adapta-
tions seem to counter it. The physical basis of transparency
is still poorly understood, but anatomical observations and
mathematical models show that there are various routes to
transparency. Future avenues for research include examina-
tion of the ultrastructure and optical properties of transpar-
ent tissue, exploring the link between transparent species
and special visual modifications in the species they interact
with, and analysis of the evolution of transparency using
comparative methods.

ctenophores to transparent polychaetes, gastropods, and fish
(Fig. 1). Transparency is one of the few forms of camou-
flage possible in a habitat with no surfaces to match or hide
behind. It is also the only form of camouflage, and one of
the few adaptations, that involve the entire organism. Al-
though the importance of transparency has been mentioned
many times by pelagic ecologists, it is a relatively unstudied
characteristic (Hardy, 1956; Fraser, 1962; McFall-Ngai,
1990; Meyer-Rochow, 1997).

This review synthesizes the current knowledge on the
distribution, ecology, and physical basis of biological trans-
parency. It is divided into five sections. The first section
reviews the phylogenetic distribution of transparent species.
The second section reviews and attempts to explain the
relationship between transparent species and their optical
environment. The third section links transparency to visi-
bility; reviews terrestrial, freshwater, and marine studies of
transparency and visual predation, including the use of
special visual adaptations; and lists known active uses of
transparency. The fourth section presents the underlying
optical principles of transparency and then applies these
principles to the various anatomical and ultrastructural mod-
ifications seen in transparent tissues. The final section sug-
gests several avenues for future research.

Introduction

Transparency is a fascinating and surprisingly common
characteristic that has received little attention because the
majority of transparent species are found only in the pelagic
regions of the open ocean. In these regions, however, the
prevalence and diversity of transparent species is remark-
able, ranging from the relatively well-known medusae and

Received 30 May 2001; accepted 30 August 2001.
* Current address: Biology Department. Box 90338, Duke University.
Durham. NC 27708. E-mail: sjohnsen@duke.edu

Phylogenetic Distribution

The phylogenetic distribution of transparent animals is
diverse, uneven, and strongly influenced by environment.
Although significant levels of tissue transparency are found
in a wide array of organisms (Figs. 1, 2), most transparent
species are found in the following 10 groups, all of which
are pelagic: cubozoans. hydromedusae. non-beroid ileno-
phores. hyperiid amphipods, tomopterid polychaeles, ptero-
tracheid and carinariid heteropods, pseudothec"somatous
pteropods, cranchiid squid, thaliaceans, and chaetognaihs.

301

302

S. JOHNSEN

Figure 1. Assemblage of transparent animals. (A) Amphogona apicata (hydromedusa), (B) Amphitretus
pelagicus (octopus), (C) Leprodora kiiullii (freshwater cladoceran). (D) Planclosphaera pelagica (hemichordate
larva), (E) Naiades cantrainii (polychaete), (F) Phyl/iroe bucephala (nudihrunch), (G) Pterasagirta ilium
(chaetognath), (H) Greta oto (neotropical butterfly), (I) Bathochordeus charon (larvacean), (J) Periclimenes
holtlntixi (shrimp), (K) Bathophilus sp. (larva of deep-sea fish), (L) Cardiopoda richardi (heteorpod). Credits as
follows: A, D, E, G. I, K. L— Laurence P. Madin; B, F— Steven Haddock; C— Wim Van Egmond; H— Randy
Emmitt; J — Jeff Jeffords.

ORGAN1SMAL TRANSPARENCY

303

Ponfera

Mertensiidae

Platyctenida

mdae

|O Beroida

!• Haeckeliidae

!• Cestida

• Thalassocalycida
Lobata
Anthozoa

IO Scyphozoa

Cubozoa

Hydrozoa

Platyhelminthes
DO Nemertea

Brachiopoda

Bryozoa

Pogonophora
DO Polychaeta

Hirudinea

Oligochaeta

Polyplacophora

Aplacophora

Monoplacophora

Bivalvia

Scaphopoda
DO Gastropoda

Nautiloidea
O Coleoidea

• Chaetognatha
Nematoda
Onychophora
Crustacea

[U Chelicerata
OO Uniramia
D Echinodermata
•O Larvacea

Styelidae, Pyundae

Molgulidae
O other Ascidea

Thaliacea

Cephalochordata
•O Ichthyes
D Tetrapoda

Ctenophora

Cnidaria

Annelida

Mollusca

Arthropoda

Urochordata

Chordata

Figure 2. Transparency and pelagic existence mapped onto a phylog-
eny of the major phyla in the Animalia. Open square indicates pelagic-
existence is rare within adults of the group: filled square indicates pelagic-
existence is common. Open circle indicates transparency is rare within
adults of the group; filled circle indicates transparency is common. Inter-
relationships of phyla taken from Halanych and Passamaneck (2001).
Relationships within phyla taken from the following: Cnidaria (Bridge el
uL 1995). Ctenophora (Podar el ai. 2001). Annelida (McHugh, 2000).
Mollusca (Wingstrand. 1985: Scheltema. 1993). Urochordata (Swalla ct
a/.. 2000). Chordata (Nelson, 1994). The phylogeny of the Arthropoda is
controversial and so is left as a polytomy. Taxa are resolved to different
levels to maximize information about the distribution of transparency.
Therefore Ctenophora is resolved to family level, while Nematoda, which
has no transparent members, is unresolved. Gastropoda and Polychaeta are
left unresolved because a resolution showing the distribution of transpar-
ency would make the figure too complex.

Most benthic. neustonic, and terrestrial groups have very
few transparent members, although there are exceptions.

The following phyletic review of transparency was com-
piled with the aid of specialists in different taxa and envi-
ronments (see acknowledgments) and is subject to several
constraints. First, because nearly all small, unpigmented
objects are transparent (for reasons described later), this
section considers only species with transparent regions
larger than 5 mm. Therefore certain phyla (e.g.. Rotifera,
Gastrotricha) and most larvae and freshwater taxa are not

covered. Second, because aquatic species from transparent
groups that are found at aphotic depths tend to be strongly
pigmented (usually red, orange, or black) (Hardy, 1956;
Herring and Roe, 1988), only terrestrial taxa and aquatic
taxa at euphotic and dysphotic depths are considered. Eu-
photic and dysphotic regions possess enough solar radiation
for photosynthesis and vision, respectively. In the clearest
waters, the lower bounds of the two regions are 200 and
1000 m. Finally, infaunul or endoparasitic species, in which
transparency could not have any optical function (e.g..
Echiura, Sipuncula, Nematomorpha), are not covered.

Eight phyla — Porifera, Nematoda, Pogonophora, Onyco-
phora, Brachiopoda, Bryozoa, Platyhelminthes, and Echi-
nodermata— appear to have no transparent adults. The first
seven of these are exclusively benthic, neustonic. or terres-
trial (Faubel. 1984; May. 1994). Echinodermata is benthic
with few exceptions (Miller and Pawson, 1990). Possible
examples of transparency in these phyla, such as hexacti-
nellid sponges and certain benthopelagic holothurians (e.g..
Peniagone diaphuna. Irpa ludwigi) are better described as
unpigmented and translucent (i.e., milky) rather than trans-
parent.

With the exception of the beroids, ctenophores at eu-
photic and dysphotic depths are generally transparent
(Mayer, 1912; Harbison el «/.. 1978). Guts, papillae, and
other small features are sometimes strongly pigmented, and
the comb rows iridesce in directional illumination, but the
bulk of the body is often extraordinarily transparent. Beroid
ctenophores tend to be opaque, due to the presence of
thousands of giant muscle fibers within the mesoglea (Her-
nandez-Nicaise. 1991), though smaller specimens of certain
species (e.g.. Beroe gracilis) can be transparent.

Transparency in the Cnidaria is mostly found in cubozo-
ans, hydromedusae, and siphonophores. Cubozoans are all
highly transparent (Matsumoto, 1995). Hydromedusae tend
to be highly transparent, though often with pigmented guts
or gonads (Russell, 1953; Kramp, 1959) (Fig. 1A). Sipho-
nophores follow a similar pattern with the exception of
neustonic species (e.g.. Plivsalia), which are often blue, and
members of the benthic family Rhodaliidae, which are
opaque (Totton, 1965; Herring, 1967; Pugh, 1983). Scypho-
zoans. in contrast and for unknown reasons, are generally
opaque and pigmented (Mayer, 1910; Russell, 1970; Wro-
bel and Mills, 1998). No anthozoans are transparent.

Among the Annelida, transparency is found only among
the pelagic polychaetes (Fig. IE). Five phyllodocidacean
families (Alciopidae, Lopadorrhynchidae, Pontodoridae.
Tomopteridae, and Typhloscolecidae) and two flabelligerid
families (Flotidae and Poeobiidae) are dominated by trans-
parent species (Uschakov, 1972; Glasby el ai, 2000). The
degree of transparency varies between the different families,
with the tomopterids and alciopids highly transparent and
the flabelligerids less so. The remaining pelagic family.

304

S. JOHNSEN

Isopilidae, apparently does not have transparent members
(Uschakov. 1972; Glasby el al, 2000).

Several genera of polystiliferous pelagic nemerteans are
transparent (Pelagonemertes, Pilonemertes) (P. Roe, Cali-
fornia State University Stanislaus, pers. comm.). However,
pigmented food in their highly branched guts often seriously
reduces any cryptic benefit. No species of benthic nemerte-
ans is known to be transparent (Roe, pers. comm.).

Transparency in the Mollusca is complex. Although the
phylum as a whole is overwhelmingly benthic and opaque,
it contains several pelagic groups that are dominated by
transparent species (Van der Spoel. 1976; Lalli and Gilmer.
1989). The Mollusca also contains pelagic groups that are
entirely opaque, and at least one transparent benthic genus.
The Aplacophora. Monoplacophora, Polyplacophora, Bi-
valvia, and Scaphopoda are exclusively benthic and opaque.
Among gastropods, the exclusively pelagic pterotracheid
and carinariid heteropods, pseudothecosomatous pteropods.
and phylliroid nudibranchs are highly transparent (Figs. IF,
L). However, the janthinid snails, atlantid heteropods, eu-
thecosomatous and gymnosomatous pteropods, and glaucid
nudibranchs are all opaque, despite also being pelagic taxa
(Van der Spoel, 1976; Lalli and Gilmer, 1989). Benthic
gastropods are opaque, with the exception of several species
of the nudibranch Melibe, which have transparent oral
hoods that are used to catch crustaceans (Von W. Kjer-
schow-Agersborg, 1921). Among cephalopods. transpar-
ency is found only in octopus and squid. Although no
benthic octopi are transparent, the pelagic families Amphi-
tretidae and Vitreledonellidae are highly transparent (Ijema
and Ikeda, 1902; Joubin, 1418) (Fig. IB). None of the
genera of the four families of the pelagic argonautoid octo-
pods are transparent, and the pelagic Bolitaenidae are better
described as translucent (Nesis, 1982). The benthopelagic
cirrate octopods are all opaque and often strongly pig-
mented. Among the exclusively pelagic squid, only the
Cranchiidae and small specimens of certain chiroteuthids
(e.g., Chiroteuthis) display any significant transparency.
Vampyroteuthis and the Sepioidea are opaque.

Species in the Chaetognatha are pelagic and highly trans-
parent, with the exception of the benthic Spadcllidae and
certain species at the lower end of the dysphotic zone (Fig.
1G). The spadellids are opaque due to the presence of
transverse muscles and pigmentation (Bone and Duvert,
1991).

With the exception of the wings of certain satyrid and
ithomiid butterflies and sphingid moths (e.g., Greta old.
Ceplwnotles hylas) (Papageorgis, 1975; Yoshida et til..
1997) (Fig. 1C) and the large pelagic larvae of certain
freshwater insects (e.g., Chaoborus), transparency in the
Arthropoda appears to be limited to aquatic crustaceans. As
in the Mollusca, the distribution of transparency in crusta-
ceans is complex, with many major groups containing both
transparent and non-transparent forms. The only group that

is truly dominated by transparent forms is the exclusively
pelagic Hyperiidea (Amphipoda) (Bowman and Gruner,
1973: Vinogradov et al., 1996). The hyperiids, which are
commensal on gelatinous /ooplankton (Madin and Harbi-
son, 1977; Laval, 1980), can be extraordinarily transparent
and often have special modifications to increase their trans-
parency (e.g.. Land, 1981: Nilsson. 1982). The generally
benthic or terrestrial groups (e.g., Decapoda, Gammaridea,
Cirripedia. Stomatopoda, Isopoda) are primarily opaque,
but with many exceptions among pelagic and benthopelagic
subgroups (e.g., some Pasiphaeaid shrimp, various species
of cleaner shrimp, the sergestid Lucifer, the isopod As-
tacilla, the phyllosoma larvae of Paliniints, the anemone
shrimp Periclimenes) (Fig. 1J). As is true of cnidarians and
ctenophores, many transparent pelagic crustaceans have
red-pigmented guts and gonads, particularly at dysphotic
depths (Hardy, 1956; Herring and Roe. 1988). Transparency
is fairly common in freshwater crustaceans, but only a few
species, mostly highly modified cladocerans, are larger than
5 mm (e.g., Leptodora, Bythotrephes) (Fig. 1C).

Most transparent urochordates are found in the exclu-
sively pelagic Thaliacea, which comprises the pyrosomids,
salps, and doliolids (Godeaux et al.. 1998). Pyrosomids are
opaque, while salps and doliolids, excepting large individ-
uals of Thefts vagina, are highly transparent. Among the
exclusively benthic Ascidea, transparency is observed in
several genera of the order Enterogona (e.g., dona,
Clcivelina), some of which are predatory (e.g., Megalodico-
I'ia hians) (Sanamyan, 1998). The larvaceans generally
have small opaque bodies and long transparent tails, but
with few exceptions (e.g.. Buthochordeus) are smaller than
5 mm (L. P. Madin. Woods Hole Oceanographic Institution,
pers. comm.) (Fig. II).

Although adults in the Hemichordata are infaunal and
opaque, the larval form of Planctosphaera pelagica has a
diameter of 25 mm and is highly transparent (Hart et al.,
1994) (Fig. ID). This organism, known only in this form,
appears to have a prolonged larval stage and is well adapted
to a pelagic existence.

No tetrapod chordate is transparent, but a number of fish
are. Transparent adults are scattered throughout marine and
freshwater teleosts, but are common only in the freshwater
family Ambassidae (glassfish) (Johnson and Gill, 1995).
Commonly known examples from other families include the
glass catfish Kn'ptopterus hicirrhix (Siluridae) and Parailia
pelliiciilii (Schilbeidae), the cardinalfish genus Rluibdamia
(Apogonidae). the clingfish Alahes pan-nliis (Cheilo-
branchidae). and the glass knifefish Eigenmannia rirescens
(Sternopygidae) (Briggs, 1995; Ferraris, 1995; Johnson and
Gill. 1995). In addition, the pelagic larvae of many fresh-
water and marine fish are often highly transparent (Breder,
1962; Meyer-Rochow, 1974) (Fig. IK). The most striking
of these are the leptocephalous larvae of elopomorphs.
These leaf-shaped larvae incorporate gelatinous material in

ORGANISMAL TRANSPARENCY

305

their bodies and quickly grow to lengths of up to 50 cm
(Pfeiler, 1986). Most larval fish lose their transparency upon
metamorphosis, some within 24 hours. The only possible
tetrapod candidates, the glass frogs (Centrolenidae), have
transparent skin on their ventral side, but opaque organs and
a strongly pigmented dorsal surface (reviewed by McFall-
Ngai. 1990).

Transparency and Environment

As can be seen from Figure 2 and the previous section,
transparency has evolved multiple times and is almost ex-
clusively a pelagic trait. Organismal transparency (rather
than simply ocular) is extremely rare on land, rare in the
aquatic benthos, uncommon in aphotic regions, somewhat
more common in dysphotic and neustonic habitats, and
ubiquitous at euphotic depths in clear water. The rarity of
terrestrial transparency is probably due to the low refractive
index of air, the presence of gravity, and high levels of
ultraviolet radiation. The distribution of transparency in
aquatic habitats appears to be correlated with the distribu-
tion of successful visual predation and crypsis strategies.

Terrestrial transparency

The extreme rarity of terrestrial transparency is probably
due to the problem of reflections. The invisibility of a
transparent object depends in part on the difference between
its refractive index and the refractive index of the surround-
ing medium. A large difference causes surface reflections
that substantially increase visibility. For example, an ice
sculpture, while transparent, is highly visible due to surface
reflections. At normal incidence, the fraction of incident
light that is reflected (R) is

R =

/I, + HI

(1)

where H, and /i-, are the refractive indices of the object and
the surrounding medium. The refractive index of biological
tissue is roughly proportional to density and ranges from
1.35 (cytoplasm) to about 1.55 (densely packed protein)
(Charney and Brackett. 1961; Chapman, 1976). The refrac-
tive index of seawater depends on temperature and salinity.
but is about 1 .34. For these values, the surface reflection of
a transparent organism in air (2%-5%) is roughly 10-fold to
2000-fold greater than its surface reflection in seawater
(0.001 %-0.7%). Although some nongaseous compounds
with refractive indices slightly less than that of seawater
exist (e.g., trifluoroacetic acid, n = 1.28), the refractive
index of water is the lower limit for biological materials.
Therefore successful crypsis using transparency is unlikely
in terrestrial habitats. Other likely contributing factors are
the increased levels of ultraviolet radiation on land, which

require protective pigmentation, and the need for supporting
skeletal structures that are often opaque.

Distribution of nc/iuitic irnnxpurencv

Transparency is common in pelagic species at euphotic
and dysphotic depths. Almost all non-transparent pelagic
taxa are either camouflaged by small size (e.g.. atlantid
heteropods, euthecosomatous and gymnosomatous ptero-
pods, glaucid nudibranchs. copepods, ostracods) or mir-
rored surfaces (e.g., fish, cephalopods), or are protected by
fast swimming speeds (e.g., fish, cephalopods, shrimp) or
chemical or physical defenses (e.g., scyphozoans, janthinid
snails. Nautilus) (Hatnner, 1996). The primary explanation
for the prevalence of transparency in this environment is
that it is the only form of camouflage in the pelagic realm
that is successful from all viewpoints and at all depths.
Cryptic coloration (e.g.. countershading) is generally suc-
cessful only from a given viewpoint and at a given depth
(Munz and McFarland. 1977; Johnsen, 2002). Mirrored
sides are successful at euphotic and upper dysphotic depths
and for most viewpoints, although not from directly above
or below (Herring, 1994; Denton, 1990). Counterillumina-
tion tactics are metabolically expensive and successful only
during moonlit nights or at dysphotic depths.

The relative rarity of transparency in benthic and neus-
tonic habitats is puzzling. Both benthic and neustonic spe-
cies tend to be pigmented to match the surface below
them — benthic animals matching the substrate and neus-
tonic species matching the upwelling radiance (deep blue in
oceanic water, brown in shallow freshwater) (David, 1965:
Herring. 1967; Cheng. 1975; Guthrie. 1989). The rarity of
transparency in benthic habitats is possibly due to two
factors. First, pigmentation may be less costly to the animal
than transparency, since it requires fewer modifications.
However, a varied background requires the ability to detect
and match a range of patterns and colors, a process done
automatically by transparency camouflage. A second possi-
bility is that even perfectly transparent objects tend to cast
highly conspicuous shadows, due to distortion of the light
by the higher refractive index of the tissue. These shadows,
invisible in pelagic habitats, may render transparency inef-
fective for benthic species.

Neither of these factors, however, can account for the
relative rarity of transparency in neustonic species. The two
major hypotheses for the pigmentation of neustonic species
are photo-protection and crypsis (Herring, 1967; Zaitsev.
1970). Although ultraviolet (UV) radiation is quite high at
the surface of any aquatic habitat, there is no evidence that
the pigmentation in neustonic species absorbs strongly at
UV wavelengths. In addition, there are compounds, such as
mycosporine-like amino acids, that strongly absorb at UV
but not visible wavelengths (Karentz et ai. 1991 ). The fact
that neustonic pigmentation often matches the upwelling

306

S. JOHNSEN

radiation strongly suggests that at least part of its function is
crypsis. However, the blue or brown pigmentation is suc-
cessfully cryptic only from above, or possibly from the side
(Munz and McFarland, 1977; Johnsen, 2002), whereas neus-
tonic individuals are most likely to be viewed from below.
From this angle, any individual is silhouetted by the bright
downwelling light, rendering cryptic coloration useless.
Predation from above (e.g., avian) appears to mostly in-
volve larger species (Zaitsev, 1970). As Herring (1967)
concluded, no functional explanation of pigmentation in
neustonic species is entirely satisfactory, and more data on
the UV absorption of the pigments and the structure of the
neustonic food web is needed.

As mentioned above, transparent species are rare at apho-
tic depths, generally being replaced by species with whole-
body red or black pigmentation (Hardy, 1956; Herring and
Roe. 1988: McFall-Ngai, 1990). At these depths, visual
predation by solar light is sometimes replaced by visual
predation based on directed bioluminescence. Because the
spectra of photophores are generally void of red wave-
lengths (Widder et nI., 1983), neither red nor black surfaces
can be seen by bioluminescent "searchlights." If the red or
black coloration absorbs more than 99.5% of the directed
bioluminescence, it may be more cryptic than transparency
because even a perfectly transparent object causes surface
reflections. However, because the reflected light is a small
fraction of a dim source, the background light levels must be
extremely low for the reflection to be visible. For example,
the radiant intensity of the suborbital photophores of the
Panama snaggletooth (Boroslomias piuuiinensis) is on the
order of 101" photons • s~' • sr~' (Mensinger and Case,
1997). If this light strikes a transparent individual with a
refractive index of 1.37 (10% protein), one can determine
from equation (1) that about 0.01% of the photons are
reflected back to the viewer. Therefore the background light

levels must be 10 photons

or lower. For upward

viewing this occurs at about 750 m in oceanic water (using
absorption and attenuation values from the equatorial Pa-
cific (Barnard et ul., 1998) and radiative transfer software
(Hydrolight 4.1, Sequoia Scientific)). At these depths, hor-
izontal and upward radiances are 3% and 0.5% of the
downward radiance (Denton, 1990), so the equivalent
depths for successful viewing using horizontal and down-
ward bioluminescence are 650 and 600 m. For viewers with
brighter bioluminescent "searchlights" or targets with
higher refractive index, the depths are less. For example, the
chitinous cuticle of a transparent hyperiid amphipod (n =
1.55) reflects 0.5% of the light and would be visible at 625,
525. and 475 m for upward, horizontal, and downward-
directed bioluminescence, respectively. Truly opaque ob-
jects, such as guts and digestive organs, reflect a much
higher percentage of light and are visible at even shallower
depths. This may explain why many opaque and high re-

fractive index organs are pigmented at shallower depths
than those at which whole-body pigmentation is observed.

Visibility and Visual Predation

Although some transparent species may only have trophic
interactions with blind taxa, the majority either prey on or
are preyed upon by at least some species with well-devel-
oped visual systems (Harbison et al, 1978; Alldredge and
Madin. 1982; Alldredge, 1984; Madin, 1988; Lalli and
Gilmer, 1989; Pages et al., 1996; Baier and Purcell, 1997;
Madin et al.. 1997; Purcell, 1997; Harbison, unpublished
literature review of gelatinivory in vertebrates). Since trans-
parent animals are often more delicate and less agile than
their visually orienting predators or prey, their success in
predator/prey interactions with these animals depends crit-
ically upon their visibility and in particular their sighting
distance (the maximum distance at which they are detect-
able by an animal relying on visual cues). Prey with short
sighting distances reduce their encounters with visually
orienting predators (Greene, 1983). "Ambush" predators
(e.g., medusae, siphonophores, cydippid ctenophores) with
short sighting distances increase their chances of entangling
visually orienting prey before being detected and avoided.
Raptors (e.g., chaetognaths, heteropods) with short sighting
distances increase their chances of getting within striking
distance without being detected.

Transparency and contrast

The visibility of a transparent individual generally de-
pends more on its contrast than on its size (Mertens, 1970;
Hemmings, 1975: Lythgoe, 1979). The inherent contrast
(contrast at zero distance) at wavelength A is defined as

where L,,(A) is the radiance of the object and L,,(A) is the
radiance of the background, both viewed a short distance
from the object (Hester, 1968; Mertens, 1970; Jerlov, 1976).
The absolute value of contrast decreases exponentially with
distance according to

C,,(\)\

(3)

where |C(A)| is the absolute value of apparent contrast at
distance d from the object, KL( A ) is the attenuation coeffi-
cient of the background radiance, and r(A) is the beam
attenuation coefficient of the water (adapted from Mertens,
1970; Lythgoe, 1979). The maximum distance at which the
object is still detectable is

Cmin(A)

c(A) - KL(\) '

(4)

ORGANISMAL TRANSPARENCY

307

where Cm,n(A) is the minimum contrast threshold of the
viewer. An animal can indirectly affect c(\) — KL(k) by
moving into a different water type or controlling the angle
from which it is viewed, but it can only directly decrease its
sighting distance by decreasing its inherent contrast. The
inherent contrast of a transparent organism from an arbitrary
viewpoint depends on its light-scattering properties and the
characteristics of the underwater light field (Chapman,
1976), so it is difficult to model exactly. In general, how-
ever, pelagic objects have the greatest sighting distances
when viewed from below, and are often viewed from this
angle (Mertens, 1970; Munz, 1990; Johnsen, 2002). The
transparency, 7(A), of an object is the fraction of light of
wavelength A that passes unabsorbed and unscattered
through it. Therefore, for the upward viewing angle

HA) =

L,,(\)~

andds,ghling(A) =

I -7U)

c(\) - K,(\)

(5)

Thus, the relationship between transparency and sighting
distance is not linear and depends also on the contrast
sensitivity of the viewer. Optimal minimum contrast thresh-
olds have been determined for man (0.01), cat (0.01 ), gold-
fish (0.009-0.05), cod (0.02), rudd (0.03-0.07), roach
(0.02). and bluegill (0.003-0.007) (Lythgoe, 1979; Douglas
and Hawryshyn. 1990). It is important to note, however, that
because these values depend on many aspects of the exper-
imental situation (e.g., temperature, target size, position of
stimulus on retina, whether one eye or two was used,
assessment method), they are not directly comparable
(Douglas and Hawryshyn, 1990). For example, the mini-
mum contrast threshold increases as the light level de-
creases. For example, the minimum contrast threshold of
cod (Gadus morluta) increases from 0.02 at the surface to
nearly 0.5 at 650 m in clear water (10~7 W sr~' m"")
(Anthony. 1981). Therefore, animals that are detectable
near the surface may become undetectable at depth.

Empirical studies

The only empirical research on terrestrial transparency is
a study on predation of neotropical butterflies showing that
transparent species were mostly found near the ground,
where they were presumably maximally cryptic (Papageor-
gis. 1975). A subsequent study, however, did not confirm
this (Burd. 1994).

Most of the research on the relationship between trans-
parency and visual predation has been performed in fresh-
water systems. Early studies by Zaret (1972) on fish preda-
tion on two morphs of transparent daphnia (Ceriodaphnia
cornuta) showed that predation was higher on the morph

with larger eyes. When the "small-eye" morph was then fed
India ink, creating a "super eye spot" in the gut, the preda-
tion preferences of the fish switched. Zaret also found that
the small-eye morph had a greatly reduced reproductive
potential and hypothesized that it was maintained in natural
populations due to its reduced visual predation pressure.
Later Zaret and Kerfoot (1975) showed that predation on a
different transparent cladoceran (Bosmina longirostris) did
not depend on body size but on the size of the opaque eye
spot; they concluded that the important variable in visual
predation was not body size, as previously assumed, but
apparent body size. This conclusion has been supported by
several subsequent studies (e.g., Confer et ai, 1978; Wright
and O'Brien, 1982; Hessen, 1985). Kerfoot (1982) mea-
sured the transparency, palatability. and sighting distances
(for pumpkinseed fish) of several species of transparent
freshwater zooplankton and found that transparency was
correlated with palatability and inversely correlated with
sighting distance. He proposed that visual predation by
freshwater fishes has driven zooplankton in two opposing
directions — palatable groups being selected for decreased
visibility through decreased size, increased transparency, or
both; unpalatable groups being selected for increased visi-
bility through increased size, intense pigmentation, or both.
O'Brien and Kettle (1979) examined the countervailing
selective pressures of tactile predation (selecting for large
prey) and visual predation (selecting for small prey) on two
species of Daphnia. They found that these species increased
their actual size, but not their apparent size, by developing
morphs with large transparent armored sheaths. Giguere and
Northcote (1987) repeated the India ink studies of Zaret
( 1972) in a more natural way by examining the effect of a
full gut on the predation of transparent prey. They found
that ingested prey increased the predation of Chaoborus
larvae by 68% and suggested that this increased risk was at
least partially responsible for the sinking of the animals
after nocturnal feeding.

In contrast to the relatively abundant freshwater studies,
fewer feeding studies on transparency exist for marine eco-
systems. Tsuda el at. ( 1998). in a feeding study similar to
Giguere and Northcote's, found that predation on transpar-
ent copepods roughly doubled when their guts were full; he
also suggested that predation risk due to gut visibility may
be an important factor contributing to vertical migration in
transparent zooplankton. Brownell (1985) and Langsdale
(1993) both found that eye pigmentation significantly in-
creased the vulnerability of transparent fish larvae to pre-
dation. Thetmeyer and Kils (1995) examined the effect of
attack angle on the visibility of transparent mysids to her-
ring predators and found that they were most visible when
viewed from above or below and least visible when viewed
horizontally. Finally, Utne-Palm (1999) found that the
sighting distances for transparent copepods (to goby pred-

308

S. JOHNSEN

ators) were significantly lower than the sighting distances
for pigmented copepods.

Most of the research on transparency in marine ecosys-
tems has concentrated on physical measurements of trans-
parency and modeling its relationship to visibility. Greze
(1963, 1964) was the first to describe the importance of
transparency in visual predation. Using relatively crude
equipment, he measured the average transparency of vari-
ous dinofiagellates, siphonophores, copepods, and larva-
ceans and presented a model, which, unfortunately, was
inaccurate, relating the measurements to sighting distance.
Using a spectrophotometer. Chapman (1976) measured the
transparency of several medusae (Polyorcliis. Chrysaora,
Aurelia) as a function of wavelength (from 200 to 800 nm).
He found that transparency was relatively constant over the
visual and infrared range and then dropped dramatically at
ultraviolet wavelengths. Chapman also modeled the rela-
tionship between transparency, reflectivity, and visibility as
a function of viewing angle, showing that the visibility of
any object that is not 100% transparent depends strongly on
the viewing angle and the underwater radiance distribution.
Forward (1976), in a study of shadow responses in crab
larvae, measured the transparency of the larvae's cteno-
phore predator. Mnemiopsis leiilyi, and showed that the
ctenophores were sufficiently opaque to cause a defensive
response in individuals below them. More recently, Johnsen
and Widder (1998. 2001) measured the ultraviolet (280-
400 nm) and visible (400-700 nm) transparency of 50
epipelagic and mesopelagic Atlantic species from seven
phyla (Cnidaria, Ctenophora, Annelida, Mollusca, Crusta-
cea, Chaetognatha, Chordata) and modeled the relationship
between transparency and sighting distance using analyses
similar to those given in the previous section. They found
that transparency is generally constant over the visual range.
with longer wavelengths slightly more transparent. Deep-
water animals tended to have constant and high transpar-
ency at UV wavelengths, whereas near-surface animals
showed rapidly decreasing and low transparency in the UV.
The relationship between transparency and visibility was
complex and depended strongly on the contrast sensitivity
of the viewer. Many mesopelagic animals were found to be
far more transparent than necessary for complete invisibil-
ity.

Visual adaptations to increase contrast of transparent
animals

The importance of transparency in predator/prey interac-
tions is also demonstrated by the special visual adaptations
seen in pelagic animals. The three best studied of these are
UV vision, polarization vision, and viewing at certain an-
gles. In addition to their possible other functions, all three of
these can "break" the camouflage of transparency.

UV \ision (documented down to —320 nm) has been

demonstrated in many aquatic species; it has been conser-
vatively estimated that there is sufficient UV light for vision
down to 100 m in clear ocean water (reviewed by Losey et
ai, 1999, and Johnsen and Widder, 2001). Visual pigments
with UV sensitivity have been found in dozens of species of
marine and freshwater fish (reviewed by Douglas and
Hawryshyn. 1990; Jacobs, 1992; Goldsmith, 1994; and
Johnsen and Widder, 2001). Among arthropods, UV vision
has been demonstrated in stomatopods, cladocerans, cope-
pods, decapods, and horseshoe crabs (Wald and Krainin,
1963; Marshall and Oberwinkler. 1999; Flamarique et al.,
2000). Finally, and surprisingly, UV sensitivity is found in
at least one mesopelagic alciopid polychaete and four me-
sopelagic decapod crustaceans (Wald and Rayport, 1977;
Frank and Case. 1988).

Three primary functions for UV vision have been hypoth-
esized (Losey et ai, 1999): ( 1 ) intraspecific communication,
(2) enhanced detection of opaque prey (silhouetted against
the relatively bright UV background), and (3) enhanced
detection of transparent prey. Due to higher light scattering
or the presence of UV-protective compounds, many visibly
transparent tissues are opaque at UV wavelengths (Douglas
and Thorpe, 1992; Thorpe et ai, 1993; Johnsen and Widder,
2001). Several researchers have hypothesized that UV vi-
sion is primarily used to improve detection of transparent
prey (Loew et ai. 1993; Cronin et ai, 1994; McFarland and
Loew, 1994; Loew et ai, 1996; Sandstroem, 1999), and
Browman et ai ( 1994) have shown that the presence of UV
light improves the search behavior of certain UV-sensitive
zooplanktivorous fish. The presence of UV sensitivity in
planktivorous but not in non-planktivorous life stages of
salmonids (reviewed by Tovee, 1995) and the correlation
between UV vision and planktivory in coral reef fish (Mc-
Farland et ai, unpubl. data) suggest that UV vision is often
used to increase the contrast of transparent planktonic prey.

Therefore, near-surface transparent species may have to
satisfy the conflicting selective pressures of camouflage and
protection from radiation damage. The increased visibility
due to photo-protective carotenoid and melanin pigmenta-
tion in high-UV freshwater environments has been studied
for many years (Hairston, 1976; Luecke and O'Brien, 1981,
1983; Byron, 1982; Hobaek and Wolf, 1991; Hansson,
2000; Miner et ai, 2000). These studies have shown several
novel solutions, such as inducible pigmentation mediated by
the relative levels of UV radiation and visual predation,
restriction of pigmentation to vital organs, and the use of a
photoprotective compound that also decreases visibility.
Only two studies have examined marine systems (Morgan
and Christy, 1996; Johnsen and Widder, 2001 ), and only the
latter has explored the effect of nonvisible UV protective
pigments on UV visibility. In this study, near-surface zoo-
plankton displayed significantly greater UV absorption than
deep-dwelling zooplankton. but the effect of UV absorption
on UV visibility was less than expected because the mea-

ORGANISMAL TRANSPARENCY

309

Figure 3. Copepod (Labidocera) viewed under (A) unpolarized transmitted light, and (B) crossed polarizers.
The copepod is more distinct in (B) due to the presence of birefringent muscle and connective tissue. Because
the background underwater illumination is polarized, a viewer with polarization vision may be able to visualize
the contrast increase from (A) to (B). Courtesv of Nadav Shashar.

sured UV absorption was generally significantly greater in
the UVB than in the UVA (where UV vision occurs), and
because the highest UV absorption was often found in less
transparent individuals.

The conflict between UV protection and UV concealment
may have important ecological implications in light of re-
ports of decreasing ozone levels at polar, temperate, and
tropical latitudes and concomitant increases in UVB radia-
tion (measured at 10%-20% per decade at temperate lati-
tudes) (Solomon. 1990; Smith et ui, 1992: Stolarski et <//..
1992). A responsive increase in UV-protective pigmenta-
tion (at either an individual or population level) increases
visibility at UV and possibly visible wavelengths, poten-
tially resulting in increased predation or decreased feeding
success. A responsive increase in depth may decrease access
to prey, phytoplankton, or warmer water. Given the impor-
tance of transparent zooplankton to the trophic ecology of
the pelagic realm (e.g., Madin et til., 1997; Purcell, 1997),
either response may have significant effects.

A second visual adaptation that can increase the contrast
of transparent predators or prey is polarization vision. Un-
derwater light is polarized, particularly in the horizontal
direction (Waterman, 1981 ). A transparent object can affect
this polarization in two ways: it can depolarize it entirely or,
if the object is birefringent, it can rotate the plane of
polarization (Lythgoe and Hemmings. 1967; Fineran and
Nicol, 1978). Either change is potentially detectable by a
polarization-sensitive visual system (Fig. 3). which may
explain the prevalence of polarization sensitivity in under-
water crustaceans and cephalopods (Waterman, 1981 ). De-
spite the enormous potential of this field, only one study has
tested this possibility (Shashar et ai, 1998). This study
showed that squid (Loligo pealei) preferentially attacked

birefringent plastic beads over non-birefringent beads, al-
though they were otherwise indistinguishable.

The final adaptation is behavioral rather than physiolog-
ical and relies on the special optical properties of the air-
water interface. Due to refraction at the water's surface, the
hemispherical sky is compressed into a region 97 ° across,
known as Snell's window. Any transparent object just out-
side the edge of this window is more conspicuous because
it refracts and reflects some of the light from within the
window, but is seen against the relatively dark background
of water outside the window (Lythgoe, 1979). As with
polarization sensitivity, this contrast enhancer, while poten-
tially quite important, has only been tested once. Janssen
( 1 98 1 ) showed that the attack angles of the blueback herring
(Alosa aestivalis) were closely distributed around the out-
side ed«e of Snell's window.

Active uses of transparenc\

Although transparency seems to be primarily designed
for passive crypsis, a few examples exist of more active uses
of this trait. The physonect siphonophores Athoiybia rosa-
cea and Aglaina okeni are mostly transparent, but they have
pigmented regions mimicking copepods and larval fish that
are apparently used as lures (Purcell, 1980, 1981). There-
fore, animals approaching the small lures cannot detect the
large individual that is also present. Other siphonophores
appear to have exploited temporal changes in transparency
for defense. The calycophoran siphonophores Hippopodius
hippopits and Vogtin are normally transparent, but they
rapidly become opaque when disturbed, presumably as a
defensive startle response (Mackie. 1996).

310

The Physical Basis of Transparency

S. JOHNSEN

Anatomical adaptation*

General principles

Transparency differs from other forms of crypsis and
most adaptations in general in that it involves the entire
organism. Therefore, many or all the tissues must be spe-
cialized for transparency. How this is achieved and how the
modifications are compatible with life are only just begin-
ning to be understood. The following sections explain the
physics of transparency and then discuss the few theoretical
and fewer empirical biological studies that have been per-
formed.

An organism or tissue is transparent if it neither absorbs
nor scatters light (Kerker, 1969). The majority of organic
molecules do not absorb visible light (Tardieu and Delaye.
1988), and measurements of the wavelength dependence of
light attenuation in 52 species of transparent zookplankton
show no evidence of visible absorption bands in the trans-
parent regions (Chapman. 1976; Johnsen and Widder. 1998.
2001). Therefore, except for necessarily opaque tissues
(e.g., gut. retina) and the special case of UV transparency,
the primary barrier to transparency in organic tissue appears
to be light scattering.

Scattering is caused by discontinuities in refractive index.
A nonabsorbing substance with a homogeneous refractive
index is transparent. Biological tissue has many refractive-
index discontinuities, due to the varying proportions and
densities of its components. For example, the refractive
index of lipids is higher than that of cytoplasm (Meyer,
1979). Therefore, plasma membranes, lipid droplets, and
organelles with extensive folded membranes (e.g., mito-
chondria. Golgi apparatus, and endoplasmic reticulum) have
a higher refractive index than the surrounding cytoplasm.
Organelles with dense protein concentrations, such as per-
oxisomes and lysosomes, also have a higher refractive index
than the surrounding cytoplasm, as do nuclei, due to their
high concentrations of nucleic acids. Even gelatinous or-
ganisms containing large amounts of water have sufficient
complexity to scatter light, as evidenced by their opacity
after death. In addition to these internal discontinuities,
there is also the large discontinuity defined by the surface of
the organism. As a photon passes through regions of differ-
ent refractive indices, its direction is altered. Given enough
direction changes, the tissue, though nonabsorbing. will be
opaque. Common examples of nonabsorbing. highly scat-
tering, opaque substances are milk, clouds, snow, and the
sclera (white) of the eye.

Therefore, transparent animals must be adapted to scatter
as little light as possible. Because the refractive indices of
organic molecules are generally closely correlated with den-
sity (Ross, 1967). chemical adaptations are unlikely, and the
problem is essentially a structural one.

Although most of the adaptations for transparency are
observable only at the electron microscopy level, some are
visible to the naked eye. These can be divided into the
cloaking of tissues that cannot be made transparent and
body flattening (Fig. 4).

Eyes and guts cannot be made transparent. Eyes must
absorb light to function and guts are betrayed by their
contents, since even transparent prey become visible during
digestion. The eyes of transparent animals have been cam-
ouflaged in various ingenious ways. Many hyperiid amphi-
pods have enormous eyes, covering most of their head, and
could be betrayed by their large, pigmented retinas. How-
ever, the retinal signature is masked using either of two
strategies. In some hyperiids (e.g., Phronima), the light is
directed from the large eyes to highly compact retinas using
transparent fiber optic cables of complex optical design
(Land, 1981; Nilsson, 1982) (Fig. 4B). Conversely, the
retina of the hyperiid Cystisoma is thinned, expanded, sit-
uated directly behind the cornea, and therefore indistinct
(Fig. 4A). Many transparent molluscs camouflage their eyes
with mirrors, because mirrors in the open ocean reflect only
more ocean and so are invisible (Herring. 1994). Others,
particularly the transparent cranchiid squid, use counteril
lumination to mask the shadows of their eyes seen from
below (Fig. 4D) (Voss, 1980). Land (1992) suggested that
the elongated eyes of transparent octopi function to mini-
mize the shadow of the eye from below. A final adaptation
that has not been explored is the separation of the eyes using
long stalks (e.g.. cranchiid and phyllosoma larvae), thereby
minimizing the characteristic signature of two eyes side by
side (Fig. 4F).

Similarly ingenious adaptations exist for minimizing the
visibility of the opaque guts. Many transparent animals have
elongated, vertically oriented, and sometimes reflective
guts, including pterotracheid heteropods, cranchiid squid,
transparent octopi, and hyperiid amphipods (Seapy and
Young, 1986; Land. 1992; Vinogradov et ol.. 1996; Young
et ul., 1998). The shape and orientation minimizes the
profile of the gut when viewed from above or below. The
reflective coating minimizes the contrast of the gut when
viewed from other angles. Seapy and Young ( 1986) showed
that pterotracheids and cranchiids actively maintained the
vertical orientation of their guts while altering the orienta-
tion of their bodies (Fig. 4C. D). A converse approach, seen
in many salps. ctenophores. and medusae, is the possession
of compact, spherical guts. Although not as cryptic from
below, a sphere has the minimum average projected area
when averaged over all potential viewing angles (Johnsen
and Widder, 1999). Finally, as is found in eyes, the shadows
of the opaque guts of certain species are masked using
counterilluminating bioluminescence. For example, the
mostly transparent midwater shrimp Sergestes similis masks

ORGANISMAL TRANSPARENCY

311

Figure 4. Various anatomical modifications that reduce the visibility of transparent animals. (A) Thin and
extended retina directly behind cornea reduces the opacity of the eyes of the hyperiid amphipod Cysii.wmu. (B)
Although the eyes of the hyperiid Phronima are large, the light is directed to the compact retinae using
transparent fiber optic guides. (C) and (D) The guts of the heteropod Pierotruclieu and the cranchiid squid
Taonius pnvo are elongated, mirrored, and vertical to minimize their visibility. (E) and (F) The bodies of
leptocephalous and phyllosoma larvae are highly flattened to minimize light attenuation. Credits: A, B,
E — Laurence Madin: C. D — Edith Widden F — Tamara Frank.

the shadow of its digestive organs in this fashion (Warner et
<//.. 1979).

Many guts of transparent animals, if not mirrored, are
pigmented. This is hypothesized to mask the presence of
bioluminescent prey but may also serve as cryptic colora-
tion, particularly since the color often approximates the
optimally cryptic shade for a given depth (Johnsen, 2002).

Finally, some animals simply ingest substances that re-
main clear. The highly transparent larva of the phantom
midge (Chaoborus) sucks out clear fluids from its prey
(Kerfoot, 1982). Therefore, the gut remains transparent and
does not need to be camouflaged.

Light attenuation in tissue, whether due to absorption or
scattering, is exponential. For example, if a 1 -cm-thick
section of tissue is 60% transparent, then 2 cm is 36%
transparent, and 3 cm is 22% transparent. Conversely, a
1-mm-thick section of the same tissue is 95% transparent.
Therefore, transparency can be achieved through extreme
body flattening. This adaptation has the additional advan-
tage of also camouflaging the animal when it is observed
edge-on. Flattening is observed in many transparent animals

including cestid ctenophores. phylliroid nudibranchs, many
freshwater cladocerans, hyperiid amphipods. phyllosoma
and stomatopod larvae, and the leptocephalous larvae offish
(Mayer, 1912: Zaret, 1981; Pfeiler. 1986; Lalli and Gilmer,
1989; Vinogradov et ai, 1996) (Fig. 4E. F). In certain cases,
the flattening is extreme. The phyllosoma larvae of Palinu-
rus are about 50 mm across and less than 1 mm thick (Fig.
4F). In many cases, body flattening may serve additional
functions, such as more efficient swimming in fish and
phylliroid nudibranchs, or increased surface area for gas
exchange in cestid ctenophores.

Transparency and ultrastructure

The primary modifications for transparency, however, are
ultrastructural and can only be seen with electron micros-
copy. The modifications depend on the tissue, which can be
divided into three groups: external surface, extracellular
matrix, and cellular tissue.

As mentioned above, the external surface of even a per-
fectly transparent organism reflects light due to the change

312

S. JOHNSEN

.n1A1+"2A2
A+A

n=n2

Figure 5. Photons impinging from above on an irregular surface with
protrusions smaller than half a wavelength of light experience a gradual
change in refractive index rather than a sharp discontinuity, n, is the
refractive index of the external medium, n2 is the index of the surface of
the organism (e.g.. cuticle). The refractive index at a given horizontal plane
within the protrusion layer equals the average refractive index, which is
given by the equation in the figure, where At and A, are the respective
areas of the external and organismal regions in that plane. The gradual shift
in refractive index can reduce or eliminate surface reflections.

in refractive index. These reflections can be reduced or
eliminated by covering the surface with submicroscopic
protrusions (Miller, 1979: Wilson and Hutley, 1982). Be-
cause the protrusions are submicroscopic, they do not scat-
ter light, but instead mimic a material of an intermediate
refractive index. At the tips of the protrusions, the refractive
index is that of the external medium. At the base, the index
is that of the organism. At intermediate heights, the index
varies smoothly and depends on the relative projected areas
of the protrusions and the external medium (Fig. 5). These
structures, known as "moth eye" surfaces, are found on the
eyes of certain, particularly nocturnal, lepidopterans. dipter-
ans, and caddisflies, where they are believed to camouflage
the large eyes and increase sensitivity (by reducing reflected
light) (reviewed by Miller, 1 979; Parker et ul., 1 998). They
are also found on the wings of transparent lepidopterans,
and in certain species (e.g., Cephonodes hylas) have been
shown to reduce their visibility (Yoshida et ai. 1 997).

The transparency of many extracellular tissues may de-
pend on the counterintuitive notion that, although a com-
pletely homogeneous refractive index is sufficient for trans-
parency, it is not always necessary. A transparent tissue can
have components with many different refractive indices, so
long as the average refractive index is constant over dis-
tances equal to half the wavelength of the incident light or
more (Benedek. 1971). More precisely, scattering and light
attenuation are low if the spatial distribution of refractive
index has no Fourier components with wavelengths greater

than one half the wavelength of light. This low scattering is
due to extensive destructive interference of the scattered
light from the various scatterers. What is observed instead is
a slower speed of light through the material. In short,
scattering (in the presence of heavy destructive interf erence )
is the source of refractive index. In glass, for example, each
of the various molecules scatter light, but due to destructive
interference no scattered light is observed and the beam is
not attenuated. This theory has been invoked to explain the
transparency of the mammalian cornea and lens (Benedek,
1971; Tardieu and Delaye, 1988; Vaezy and Clark, 1994).
In both tissues, a substance with a high refractive index
(collasen fibers in the cornea and crystalline proteins in the
lens) is embedded within a substance with a low refractive
index. The substance with the high refractive index is
packed so densely that steric and other repulsive interac-
tions force a local ordering of the scatterers (Tardieu and
Delaye. 1 988). The ordering exists only over distances on
the order of several diameters of the scatterers, but it is
sufficient to drastically reduce scattering. In the case of N
identical scatterers. the total scattering cross-section, Ck)lal,
is given by

Ctota, =

(6)

where Csc;, is the scattering cross-section of an individual
scatterer, ef> is the volume concentration of the scatterers
( VSL.atlt.rers/ V,,,,.,,), and S( </>) is the structure factor. The struc-
ture factor gives the amount of reduction in total scattering
due to destructive interference caused by local ordering. In
general, S(4>) is complex or unknown (see Benedek, 1 97 1),
but in the simpler case of small scatterers (radius < 70 nm)
it. is

( Delaye and Tardieu. 1983). ( 7 )

A concentration of scatterers of 30% reduces the total
scattering to 10% of the value calculated under the assump-
tion of no destructive interference of scattered light. A
concentration of 60% reduces the scattering to less than 1 %
of the value calculated assuming no destructive interfer-
ence. Figure 6 shows the total scattering cross-section ot a
solution of small particles plotted against their volume
concentration. As the volume concentration increases there
are more scatterers. but also more destructive interference.
The maximum light scattering occurs at 13% concentration
and then decreases as the concentration increases (see
Benedek ( 1971 ) and Tardieu and Delaye (1988) for further
details). This theory has been experimentally confirmed
using solutions of lens proteins (Bettleheim and Siew,
1983). The solution becomes cloudier with increasing con-
centration, until a volume concentration of about 13f/r, after
which it becomes clearer.

ORGANISMAL TRANSPARENCY

313

0.05

0.00

0.0

~r
0.2

0,4 0.6

volume fraction of solute

1.0

Figure 6. The amount of light scattering of a solution of small,
identical scatterers plotted against their concentration (by volume). The
scattering peaks when the concentration equals 13%.

Many extracellular and some cellular tissues (e.g., mus-
cle) of transparent organisms may meet these requirements.
Although studies of the extracellular matrices and muscle of
transparent animals are fairly rare, ultrastructural data exist
for hydromedusae, siphonophores, ctenophores, chaeto-
gnaths, transparent ascidians, pyrosomas, doliolids, and
salps (De Leo el ai, 1981; Weber and Schmid, 1985; Franc,
1988; Hernandez-Nicaise, 1991; Shinn, 1997; Hirose el ai,
1999). The fact that all of these appear homogeneous under
light microscopy strongly suggests that they have few Fou-
rier components greater than one half the wavelength of
light. However, rigorous analyses have not been performed.

Although the above theory may explain the transparency
of extracellular structures, it cannot adequately account for
the transparency of cellular tissue. Reduction of scattering
by destructive interference relies on dense packing of sim-
ilar objects. In the two cases where this theory has been
successfully applied (lens and cornea), the tissues are highly
simplified. The mammalian lens, in particular, has been
drastically modified for transparency (Goldman and
Benedek. 1967; Philipson. 1973; Tardieu and Delaye,
1988). Most of the lens cells lack nuclei, mitochondria, and
other organelles and, in fact, are little more than containers
for dense concentrations of a few different proteins. The
cells rely entirely on the surrounding cells for metabolic
support and maintenance. Similarly, the cornea is a tightly
packed array of collagen fibers with very few support cells
and cannot maintain itself. These modifications are obvi-
ously incompatible with life when employed throughout an
entire organism.

The only investigation of the basis of transparency in
more complex cellular tissue is a theoretical treatment by
Johnsen and Widder (1999). This study assumed that a cell
requires given total volumes of various components. It then
determined how to apportion, distribute, and shape the

volumes to minimize light scattering. The study found that
the size of the components was most important, followed by
the refractive index and, distantly, by the shape (Fig. 7;
Table 1 ). A similar analysis was performed assuming that a
cell requires a given total surface area of certain compo-
nents, with similar results. Because a group of smaller
particles within a wavelength of light of each other behave
roughly like one larger particle (Thiele, 1998), clustering
particles can change the total amount of scattering. For
example, if several lysosomes have radii near the critical

10°

5
3,

I
o

10 J
10 3
10-4
10'5
10'6
10 7
10 8

6,
to

0.001 001 0.1 1 10 100

radii of components (microns)

o>
o

4 -

2 -

0.01 0.10 100 10.00 100.00

ratio of cylinder radius to cylinder length

Figure 7. (A) The hiding power (opacity) for a given volume of
material as a function of refractive index and the size of the smaller
volumes into which it is divided. Hiding power is S • ( 1 - (cos (9)). where
S is the total amount of light scattering and (cos 9} is the average cosine of
the angle into which the light is scattered. Therefore, backscattered light
has a higher hiding power than forward scattered light. Material is assumed
to be embedded in cytoplasm (;? = 1.35). The refractive indices are
vacuole — 1.34, mitochondria — 1.42, lipid — 1.49, protein — 1.62. (B) Hid-
ing power plotted against shape for a large cylinder of constant volume
averaged over all possible orientations relative to the incident light. Shape
is given as the ratio between the radius of the cylinder and the length.
Scattering is minimal when the radius equals half the length of the cylinder
(i.e., when the cylinder is most spherical).

314

S. JOHNSEN
Table 1

Ultrastructural predictions for transparent cellular tissue: the left column lists the various parameters in order of their importance to tissue
transparency; the right column lists the predictions for the given parameter under a constant volume constraint; particles are considered clustered if
thi'v are within a wavelength of light of each other

Parameter

Predictions

Size of particles into which substance is subdivided

Clustering or dispersion of particles

Refractive index of particles

Shape for particles with radii less than the wavelength of light

Shape for particles with radii comparable to the wavelength of light

Shape for particles with radii greater than the wavelength of light

Particles will have radii either greater or less than 100 nm
Small particles will be dispersed; large particles will be clustered
All particles will have low relative refractive indices
Particle shape will be arbitrary
Predictions are highly case-specific
Particles will be spherical

radius (see Fig. 7; Table 1 ), they can be clustered to reduce
the total amount of light scattering. Shape is surprisingly
unimportant. For particles smaller than the wavelength of
light, shape is irrelevant (Johnsen and Widder, 1999). For
larger particles, the change in scattering as an object shifts
from needle-shaped to disk-shaped is quite small relative to
the enormous changes due to size (Fig. 7B).

Table 2 lists the predictions for actual cell components to
scatter a minimum amount of blue-green light. For each
component, a range of size and refractive index is given. All
the components are considered to be primarily bound by

constant-volume constraints, with the exception of mito-
chondria. Since mitochondria! functioning depends heavily
on membrane surface, it is considered to be bound by
constant-surface-area constraints (see above). The refractive
index of the cytoplasm is assumed to be 1 .35. The refractive
indices of the components are highly approximate and based
on values of 1.62 for protein, 1.49 for lipid, and 1.34 for
saline. In cases where a given prediction cannot be applied
(e.g., dividing a nucleus into smaller nuclei, changing the
shape of a microtubule), no prediction is made. All predic-
tions assume that the size and refractive index of a given

Table 2

Predictions for a typical cell that scatters a minimum amount of light: the predictions cover the shape, distribution {many and small, few and large),
and refractive index of the cellular components

Component

Constraint

Size

Index

Predictions

Actin filaments, intermediate filaments. Volume
microtubules

4 nm. 5 nm. 12 nm 1.55-1.62 Shape: not applicable

Distribution: dispersed
Refractive index: low

Ribosomes

Volume 15 nm

1.55-1.62

Shape: arbitrary

Distribution: dispersed

Refractive index: low

Transport vesicles

Volume 1 5-50 nm

1.49-1.62

Shape: arbitrary

Distribution: many, small, and dispersed

Refractive index: low

Lysosomes. peroxisomes

Volume 0.1-0.25 (jm

1.49-1.62

Shape: difficult to predict

Distribution: many, small, and dispersed

Refractive index: low

Lipid droplets

Volume 0.1-10 Mm

1.49-1.62

Shape: arbitrary (if droplets are large, then spherical)

Distribution: many, small, and dispersed

Refractive index: low

Mitochondria

Surface area 0. 25-10 MITI

1.42-1.49

Shape: difficult to predict

Distribution: many, small, and dispersed

Refractive index: low

Nucleus

Volume 1 .5-5 Mm

1.42-1.49

Shape: spherical

Distribution: not applicable

Refractive index: low

Large vacuole

Volume 5-15 ju.ni

1.34-1.62

Shape: spherical

Distribution: few. large, and clustered

Refractive index: low

ORGANISMAL TRANSPARENCY

315

component must remain within the range given. None of
these predictions have been tested, although the morpho-
logical techniques are relatively straightforward.

In summary, although the physics of light scattering is
well understood, the field of organismal transparency is still
in its infancy. The few theoretical and empirical studies
suggest that there are several routes to transparency, many
of which probably operate concurrently. For example, the
transparency of leptocephalous larvae may be due to body
flattening, ordered packing within the gelatinous core, a
very thin muscle layer, and possibly modifications within
the cellular tissue itself. Other animals, such as phyllosoma
larvae, may rely entirely on their extreme flattening. How-
ever, the actual modifications and their proximate and ulti-
mate costs are, for the most part, unknown.

Future Directions

Transparency is currently a field with more questions
than answers. Almost every major aspect of its study is a
fruitful avenue for future research, but several topics are
critical for future understanding of this adaptation. First, the
structural predictions must be tested using morphological
and optical measurements of transparent tissue. The un-
likely possibility that organic molecules in transparent or-
ganisms have altered their refractive indices needs to be
tested. More images of transparent animals under UV and
polarized light are needed to evaluate the hypotheses of
special camouflage breakers in planktivores, as are more
feeding studies in both freshwater and marine ecosystems.
Finally, as more phylogenies of pelagic groups become
available, comparative methods should be used to explore
the evolution of this extraordinary trait.

Acknowledgments

I thank the following for information on the transparency
of specific groups: Martin Angel, Daphne Fautin, Tamara
Frank, Steven Haddock, Richard Harbison, Peter Herring.
Dina Leech. Laurence Madin. Marianne Moore, Karen Os-
born. David Pawson, Pamela Roe, Clyde Roper, Michael
Vecchione, Janet Voight. and Edith Widder. I also thank
Ken Halanych and Yale Passamaneck for pointing out rel-
evant phylogenetic literature and software, and Kristina
Fjeld. Tamara Frank, and Laurence Madin for a critical
reading of the manuscript. The images for Figures 1, 3, and
4 were generously provided by Tamara Frank. Steven Had-
dock, Jeff Jeffords, Laurence Madin, Nadav Shashar. and
Edith Widder. This work was funded in part by grants to SJ
from The Rinehart Coastal Research Center, the Reuben F.
and Elizabeth B. Richards Endowed Fund, the Penzance
Endowed Fund, and the Grayce B. Kerr Fund. This is
contribution number 10555 of the Woods Hole Oceano-
graphic Institution.

Literature Cited

Alldredge. A. L. 1984. The quantitative significance of gelatinous

zooplankton as pelagic consumers. Pp. 407 — 134 in Flows of Energy

and Materials in Marine Ecosvsteinx, M. 1. R. Fasham. ed. Plenum

Press, New York.
Alldredge, A. L., and L. P. Madin. 1982. Pelagic tunicates: unique

herbivores in the marine plankton. Bioscience 32: 655-663.
Anthony, P. D. 1981. Visual contrast thresholds in the cod Gadus

morlnui. J. Fish Dial. 19: 87-103.
Baier, C. T., and J. E. Purcell. 1997. Trophic interactions of chaeto-

gnaths. larval fish, and zooplankton in the South Atlantic Bight. Mar.

Ecol. Prog. Ser. 146: 43-53.
Barnard, A. H., W. S. Pegau, and J. R. V. Zaneveld. 1998. Global

relationships of the inherent optical properties of the oceans. J. Geo-

phys. Res. 103: 24955-24968.
Benedek, G. B. 1971. Theory of the transparency of the eye. Appl. Opt.

10: 459-473.

Bettleheim, F. A., and E. L. Slew. 1983. Effect of change in concen-
tration upon lens turbidity as predicted by the random fluctuation

theory. Biophys. J. 41: 29-33.
Bone. Q., and M. Duvert. 1991. Locomotion and buoyancy. Pp. 32-44

in The Biology of Chae tognaths, Q. Bone. H. Kapp, and A. C. Pierrot-
Bulls, eds. Oxford University Press. New York.
Bowman, T., and H. E. Gruner. 1973. The families and genera of

Hyperiidea (Crustacea: Amphipoda). Smithson. Contrib. Zool. 146:

1-60.
Breder, C. M. 1962. On the significance of transparency in osteichthid

fish eggs and larvae. Co/"'"' 1962: 561-567.
Bridge, D., C. W. Cunningham, R. DeSalle, and L. W. Buss. 1995.

Class-level relationships in the phylum Cnidaria: molecular and mor-
phological evidence. Mol. Biol. Evol. 12: 679-689.
Briggs. J. C. 1995. Clmgfishes. Pp. 142-144 in Encyclopedia of Fishes.

1. R. Paxton and W. N. Eschemeyer, eds. Academic Press. New York.
Browman, H. I., I. Novales-Flamarique. and C. W. Hawryshyn. 1994.

Ultraviolet photoreception contributes to prey search behaviour in two

species of zooplanktivorous fishes. J. E.\p. Biol. 186: 187-198.
Brownell. C. L. 1985. Laboratory analysis of cannibalism by larvae of

the Cape anchovy Engraulis capensis. Trans. Am. Fish Soc. 114:

512-518.
Burd, M. 1994. Butterfly wing colour patterns and flying heights in the

seasonally wet forest of Barro Colorado Island, Panama. J. Trap. Biol.

10: 601-610.

Byron, E. R. 1982. The adaptive significance of calanoid copepod pig-
mentation: a comparative and experimental analysis. Ecolog\ 63:

1871-1886.
Chapman, G. 1976. Reflections on transparency. Pp. 491-498 in Coe-

lenterate Ecology and Behavior, G. O. Mackie, ed. Plenum Press, New

York.
Charney, E., and F. S. Bracket!. 1961. The spectral dependence of

scattering from a spherical alga cell and its implication for the state of

organization of the light accepting pigments. Arch. Biochem. Biophys.

92: 1-12.
Cheng, L. 1975. Marine pleuston — animals at the sea-air interface.

Oceanogr. Mar. Biol. Annn. Rev. 13: 181-212.
Confer. J. L.. G. L. Howick, M. H. Corzette, S. L. Kramer, S. Fitzgib-

bon, and R. Landerbert. 1978. Visual predation by planktivores.

Oikos 31: 27-37.
Cronin, T. W., N. J. Marshall, R. L. Caldwell, and N. Shashar. 1994.

Specialization of retinal function in the compound eyes of mantis

shrimps. Vision Res. 34: 2639-2656.
David, P. M. 1965. The surface fauna of the ocean. Endeavour 24:

95-100.
Delaye, M., and A. Tardieu. 1983. Short-range order of crystallin

proteins accounts for eye lens transparency. Nature 302: 415 — H7.

316

S. JOHNSEN

De Leo, G., E. Patricolo, and G. Frittitta. 1981. Fine structure of the
tunic of dona intestinalis L. II. Tunic morphology, cell distribution
and their functional importance. Acta Zool. 62: 259-271.
Denton, E. J. 1990. Light and vision at depths greater than 200 meters.
Pp. 127-148 in Light and Life in the Sea. P. J. Herring. A. K. Campbell,
M. Whitfield. and L. Maddock, eds. Cambridge University Press, New
York.

Douglas, R. H.. and C. W. Hawryshy n. 1990. Behavioral studies of fish
vision: an analysis of visual capabilities. Pp. 373-418 in The Visual
System of Fish. R. H. Douglas and M. B. A. Djamgoz, eds. Chapman
and Hall, New York.

Douglas, R. H., and A. Thorpe. 1992. Short-wave absorbing pigments
in the ocular lenses of deep-sea teleosts. J. Mar. Biol. Assoc. UK 72:
93-112.
Faubel, A. 1984. On the geographical occurrence of pelagic polyclad

turbellanans. Call. Biol. Mai: 25: 153-168.

Ferraris, C. J. 1995. Catfishes and knifefishes. Pp. 106-112 in Ency-
clopedia of Fishes. J. R. Paxton and W. N. Eschemeyer, eds. Academic
Press, New York.

Fineran, B. A., and J. A. C. Nicol. 1978. Studies on the photoreceptors

on Anchoa mirchilli and A. hepsetits (Engraulidae) with particular

reference to the cones. Philos. Trans. R. Soc. Land. B 283: 25-60.

Flamarique, I. N.. H. I. Browman, M. Belanger, and K. Boxaspen.

2000. Ontogenetic changes in visual sensitivity of the parasitic

salmon louse Lepeophtheirus salmonis. J. Exp. Biol. 203: 1649-1659.

Forward, R. B., Jr. 1976. A shadow response in a larval crustacean.

Biol. Bull. 151: 126-140.
Franc, J. M. 1988. The mesoglea of ctenophores. Bull. Soc. Zool. Fr.

113: 347-351.
Frank, T. M., and J. F. Case. 1988. Visual spectral sensitivities of

bioluminescent deep-sea crustaceans. Biol. Bull 175: 261-273.
Fraser, J. 1962. Nature Adrift: The Story of Marine Plankton. G. T.

Foulis, London.

Giguere, L. A., and T. G. Northcote. 1987. Ingested prey increase risks
of visual predation in transparent Chaoborus larvae. Oecologia 73:
48-52.

Glasby, C. J., P. A. Hutchings, K. Fauchald, H. Paxton, G. W. Rouse,
C. W. Russell, and R. S. Wilson. 2000. Polychaeta. Pp. 1-296 in
Polvchaetes and A/lies: The Southern Synthesis. P. L. Beesley, G. J. B.
Ross, and C. J. Glasby, eds. CSIRO Publishing, Melbourne.
Godeaux, J., Q. Bone, and J. C. Braconnol. 1998. Anatomy of Thaha-
cea. Pp. 1-24 in The Biology of Pelagic Tunicates. Q. Bone. ed. Oxford
University Press, New York.

Goldman, J. N., and G. B. Benedek. 1967. The relationship between the
morphology and transparency in the nonswelling corneal stroma of the
shark. Investig. Ophthalmol. 6: 574-600.

Goldsmith, T. H. 1994. Ultraviolet receptors and color vision: evolu-
tionary implications and dissonance of paradigms. Vision Res. 34:
1479-1488.

Greene, C. H. 1983. Selective predation in freshwater zooplankton com-
munities. Int. Rev. Gesamten Hydrobiol. 68: 297-315.
Greze, V. N. 1963. The determination of transparency among planktonic
organisms and its protective significance. Dokl. Akad. Nauk. SSSR 151:
435-438.
Greze, V. N. 1964. The transparency of planktonic organisms in the

equatorial part of the Atlantic Ocean. Okeanologiya 4: 125-127.
Guthrie, M. 1989. Animals of the Surface Film. Richmond Publishing,

Slough, U.K.

Hairston, N. 1976. Photoprotection by carotenoid pigments in the cope-
pod Diaptomus nevadensis. Proc. Null. Acad. Sci. 73: 971-974.
Halanych, K. M., and Y. Passamaneck. 2001. A brief review of meta-
zoan phylogeny and future prospects in Hox-research. Am. Zool. (In
press).

Hamner, W. M. 1996. Predation, cover, and convergent evolution in
epipelagic oceans. Pp. 17-37 in Zooplankton: Sensory Ecology and
Physiology. P. H. Lenz, D K. Hartlme, J. E. Purcell. and D. L.
Macmillan. eds. Overseas Publishers Association. Amsterdam.
Hansson, L. 2000. Induced pigmentation in zooplankton: a trade-off
between threats from predation and ultraviolet radiation. Proc. R. Soc.
Land. B. 267: 2327-2331.

Harbison, G. R., L. P. Madin. and N. R. Swanberg. 1978. On the
natural history and distribution of oceanic ctenophores. Deep-Sea Res.
25: 233-256.
Hardy, A. C. 1956. The Open Sea. Its Natural History: The World of

Plankton. Houghton Mifflin, Cambridge. MA.

Hart, M. W., R. L. Miller, and L. P. Madin. 1994. Form and feeding
mechanism of a living Planctosphaera pelagica (phylum Hemichor-
data). Mar. Biol. 120: 521-533.

Hemmings, C. C. 1975. The visibility of objects underwater. Pp. 359-
374 in Light as an Ecological Factor, G. C. Evans. R. Bainbndge, and
O. Rackhman. eds. Blackwell. Oxford.

Hernandez-Nicaise, M-L. 1991. Ctenophora. Pp. 359-418 in Micro-
scopic Anatomy of the Invertebrates Volume II: Placozoa. Porifera,
Cnidaria. and Ctenophora. F. W. Harrison and J. A. Westfall. eds. John
Wiley, New York.
Herring, P. J. 1967. The pigments of plankton at the sea surface. Symp.

Zool. Soc. Loud. 19: 215-235.
Herring. P. J. 1994. Reflective systems in aquatic animals. Comp. Bio-

chem. Physiol. A 109: 513-546.
Herring, P. J., and H. S. J. Roe. 1988. The photoecology of pelagic

oceanic decapods. Svmp. Zool. Soc. Land. 59: 263-290.
Hessen, D. O. 1985. Selective zooplankton predation by pre-adult roach
{Riiiilnx nitilus): the size-selective hypothesis versus the visibility-
selective hypothesis. Hydrobiologia 124: 73-79.
Hester, F. J. 1968. Visual contrast thresholds of the goldfish (Carassius

auratus). Vision Res. 8: 1315-1335.

Hirose, E., S. Kimura, T. Itoh, and J. Nishikawa. 1999. Tunic mor-
phology and cellulosic components of pyrosomas, doliolids, and salps
(Thaliacea, Urochordata). Biol. Bull. 196: 113-120.
Hobaek, A., and H. G. Wolf. 1991. Ecological genetics of Norwegian
Daplmia. 2. Distribution of Daphnia longispina genotypes in relation
to short-wave radiation and water colour. Hydrobiologia 225: 229-
243.
Ijema, I., and S. Ikeda. 1902. Notes on a specimen of Amphitretus

obtained in the Sagami Sea. Annot. Zool. Jpn. 4: 5-101.
Jacobs, G. H. 1992. Ultraviolet vision in vertebrates. Am. Zool. 32:

544-554.

Janssen, J. 1981. Searching for zooplankton just outside Snell's win-
dow. Limnol. Oceanogr. 26: 1168-1171.
Jerlov, N. G. 1976. Marine Optics. Elsevier, New York.
Johnsen, S. 2002. Cryptic and conspicuous coloration in the pelagic

environment. Proc. R. Soc. Land. B 269(1). (In press).
Johnsen, S., and E. A. Widder. 1998. Transparency and visibility of
gelatinous zooplankton from the northwestern Atlantic and Gulf of
Mexico. Biol. Bull. 195: 337-348.

Johnsen, S., and E. A. Widder. 1999. The physical basis of transpar-
ency in biological tissue: ultrastructure and the minimization of light
scattering. J. Theor. Biol. 199: 181-198.

Johnsen. S., and E. A. Widder. 2001. Ultraviolet absorption in trans-
parent zooplankton and its implications for depth distribution and
visual predation. Mar. Biol. 138: 717-730.

Johnson, G. D., and A. C. Gill. 1995. Perches and their allies Pp
181-196 in Encyclopedia of Fishes. J R. Paxton and W. N. Esche-
meyer. eds. Academic Press, New York.

Joubin, L. 1918. Etudes preliminaries sur les Cephalopodes recueillis au
cours des croisieres de S. A. S. le Prince de Monaco 6e Note: Vitrele-
donella richardi Joubin. Bull. Inst. Oceanogr. 340: 1-40.

ORGANISMAL TRANSPARENCY

317

Karentz, D., F. S. McKuen, M. C. Land, and W. C. Dunlap. 1991.

Survey of mycosporine-like amino acid compounds in Antarctic marine
organisms: potential protection from ultraviolet exposure. Mar. Rial.
108: 1 57-1 (16.

Kerfoot, \V. C. 1982. A question of taste: crypsis and warning coloration
in freshwater zooplankton communities. Ecologv 63: 538-554.

Kerker. M. 1969. The Scattering of Light and Other Electromagnetic
Radiation. Academic Press. New York.

Kramp. P. I.. 1959. The hydromedusae of the Atlantic Ocean and
adjacent waters. Dana-Rep. 46: 1-283.

Lalli, C. M.. and R. VV. Gilmer. 1989. Pelagic Snails. Stanford Uni-
versity Press. Palo Alto. CA.

Land. M. F. 1981. Optics of the eyes of Phronima and other deep-sea
animals. J. Comp. fhyxiol. A 145: 209-226.

Land, M. F. 1992. A note on the elongated eye of the octopus Vitrele-
donella richurdi. J. Mar. Biol. Assoc. UK 72: 89-92.

Langsdale, J. R. M. 1993. Developmental changes in the opacity of
larval herring. Clupea harengits, and their implications for vulnerabil-
ity to predation. J. Mm: Biol. Assoc. UK 73: 225-232.

Larson, R. J. 1976. Cubomedusa: feeding, functional morphology, be-
havior and phylogenetic position. Pp. 237-246 in Coelenterate Ecology
mill Behavior. G. O. Mackie, ed. Plenum Press, New York.

Laval. P. 1980. Hyperiid crustaceans as parasitoids associated with ge-
latinous zooplankton. Oceanogr. Mar. Biol. 18: 11-56.

Loew, E. R.. and W. N. McFarland. 1990. The underwater visual
environment. Pp. 1 — 14 in The Visual System of Fish. R. H. Douglas and
M. B. A. Djamgoz. eds. Chapman and Hall. New York.

Loew, E. R., W. N. McFarland, E. L. Mills, and D. Hunter. 1993. A
chromatic action spectrum for planktonic predation by juvenile yellow
perch, Perca flavescens. Can. J. Zool. 71: 384-386.

Loew, E. R., R. A. McAlary, and W. N. McFarland. 1996. Ultraviolet
visual sensitivity in the larvae of two species of marine atherinid h'shes.
Pp. 195-210 in Zooplankton: Sensory Ecologv and Phvsiologv. P. H.
Lenz, D. K. Hartline, J. E. Purcell, and D. L. Macmillan. eds. Gordon
and Breach, Amsterdam.

Losey, G. S., T. W. Cronin, T. H. Goldsmith, D. Hyde, N. J. Marshall,
and W. N. McFarland. 1999. The UV visual world of fishes: a
review. J. Fixli. Biol. 54: 921-943.

Luecke. C.. and VV. J. O'Brien. 1981. Phototoxicity and fish predation:
selective factors in color morphs in Heterocope. Limnol. Oceanogr. 26:
454-460.

Luecke, C., and VV. J. O'Brien. 1983. Photoprotective pigments in a
pond morph of Daphniu miililcndorffiana. Arctic 36: 365-368.

Lythgoe, J. N. 1979. The Ecology of Vision. Clarendon Press, Oxford.

Lythgoe, J. N., and C. C. Hemmings. 1967. Polarized light and under-
water vision. Niitnn' 213: 893-894.

Mackie, G. O. 1996. Defensive strategies in planktonic coelenterates.
Pp. 435 — 146 in Zooplankton: Sensory Ecology and Physiology. P. H.
Lenz. D. K. Hartline. J. E. Purcell. and D. L. Macmillan. eds. Overseas
Publishers Association. Amsterdam.

Madin. L. P. 1988. Feeding behavior of tentaculate predators: in
situ observations and a conceptual model. Bull. Mai: Sci. 43: 413-
429.

Madin. L. P., and G. R. Harbison. 1977. The associations of Am-
phipoda Hyperiidea with gelatinous zooplankton — I. Associations with
Saipidae. Deep-Sea Rex. 24: 449-463.

Madin, L. P.. J. E. Purcell, and C. B. Miller. 1997. Abundance and
grazing effects of Cyclosalpa bakeri in the subarctic Pacific. Mar. Ecol.
Prog. Ser. 157: 175-183.

Marshall, N. J.. and J. Oberwinkler. 1999. The colourful world of
mantis shrimp. Nature 401: 873-874.

Matsumoto, G. I. 1995. Observations on the anatomy and behavior of
thecubozoan Caryhilea raxtonii Haacke. Mar. Freshw. Behav. Phvxiol.
26: 139-148.

May. R. M. 1994. Biological diversity: differences between land and sea.

Philos. Trans. R. Soc. Loml. B. 343: 105-111.
Mayer, A. G. 1910. Medusae of the World III: The Scyphomedusae.

Carnegie Institution of Washington. Washington. DC.
Mayer, A. G. 1912. Ctenophores of the Atlantic Coast of North America.

Carnegie Institution of Washington. Washington. DC.
McFall-Ngai, M. J. 1990. Crypsis in the pelagic environment. Am. Zool.

30: 175-188.
McFarland, VV. N., and E. R. Loew. 1994. Ultraviolet visual pigments

in marine fishes of the family Pomacentridae. Vision Res. 34: 1393-

1 396.
McHugh, D. 2000. Molecular phylogeny of the Annelida. Can. /. Zool.

78: 1873-1884.
Mensinger, A. F., and J. F. Case. 1997. Luminescent properties of

fishes from nearshore California basins. J. Exp. Mar. Biol. Ecol. 210:

75-90.
Mertens, L. E. 1970. In-Water Photography: Theory and Practice. John

Wiley, New York.
Meyer, R. A. 1979. Light scattering from biological cells: dependence of

backscatter radiation on membrane thickness and refractive index.

Appi Opt. 18: 585-588.
Meyer-Rochow. V. B. 1974. Leptocephali and other transparent fish

larvae from the south-eastern Atlantic ocean. Zool. An-. 192: 240-251.
Meyer-Rochow, V. B. 1997. Wenn Unsichtbares sichtbar wird: durch-

sichtige Tiere — transparente Gewebe. Nat. Mus. 127: 121-127.
Miller, J. E., and D. L. Pawson. 1990. Swimming sea cucumbers

(Echinodermata: Holothuroidea): a survey, with analysis of swimming

behavior in four bathyl species. Smithson. Contrib. Mar. Sci. 35: 1-18.
Miller, VV. H. 1979. Intraocular filters. Pp. 69-144 in Handbook of

Sensory Physiology, Vol. 7/6A, H. Autrum, ed. Springer, New York.
Miner, G. B., S. G. Morgan, and J. R. Hoffman. 2000. Postlarval

chromatophores as an adaptation to ultraviolet radiation. J. Exp. Mar.

Biol. Ecol. 249: 235-24S.
Morgan, S. G., and J. H. Christy. 1996. Survival of marine larvae

under the countervailing selective pressures of photodamage and pre-
dation. Limnol. Oceanogr. 41: 498-504.
Munz, VV. R. A. 1990. Stimulus, environment and vision in fishes. Pp.

491-511 in The Visual System of Fish. R. H. Douglas and M. B. A.

Djamgoz, eds. Chapman and Hall. New York.
Munz, F. VV., and VV. N. McFarland. 1977. Evolutionary adaptations of

fishes to the photic environment. Pp. 194-274 in The Visual System in

Vertebrates. F. Crescitelli, ed. Springer- Verlag, New York.
Nelson, J. S. 1994. Fishes of the World. John Wiley. New York.
Nesis, K. N. 1982. Ccphalopoils of the World. T. F. H. Publications,

Neptune City. NJ.

Nilsson, D. E. 1982. The transparent compound eye of Hvperia (Crus-
tacea): examination with a new method for analysis of refractive index

gradients. J. Comp. Phyxiol. A 147: 339-349.
O'Brien. W. J.. and D. Kettle. 1979. Helmets and invisible armor:

structures reducing predation from tactile and visual planktivores.

Ecology 60: 287-294.
Pages, F., M. G. White, and P. G. Rodhouse. 1996. Abundance of

gelatinous carnivores in the nekton community of the Antarctic polar

frontal zone in summer 1994. Mar. Ecol. Prog. Ser. 141: 139-147.
Papageorgis, C. 1975. Mimicry in neotropical butterflies. Am. Sci. 63:

522-532.
Parker, A. R., Z. Hegedus, and R. A. Watts. 1998. Solar-absorber

antireflector on the eye of an Eocene fly (45 Ma). Proc. R. Soc. Land.

B26S: 811-815.
Pfeiler. E. 1986. Towards an explanation of the developmental strategy

in leptocephalous larvae of marine teleost fishes. Environ. Biol. Fishes

15: 3-13.
Philipson, B. 1973. Changes in the lens related to the reduction of

transparency. Exp. Eye Rex. 16: 29-39.

318

S. JOHNSEN

Podar, M., S. H. D. Haddock. M. L. Sogin, and G. R. Harbison. 20(11.

A molecular phylogenetic framework for the phylum Ctenophora using

I8S rRNA genes. Mol. Biol. Evol. (In press).
Pugh, P. R. 1983. Benthic Siphonophores: a review of the family Rho-

daliidae (Siphonophora, Physonectae). Philos. Trans. R. Soc. Land. B

301: 165-300.
Purcell, J. E. 1980. Influence of siphonophore behavior on their natural

diets; evidence for aggressive mimicry. Science 209: 1045-1047.
Purcell, J. E. 1981. Selective predation and caloric consumption by the

siphonophore Rosacea cymbiformis in nature. Mar. Biol. 63: 283-294.
Purcell, J. E. 1997. Pelagic cnidarians and ctenophores as predators:

selective predation. feeding rates and effects on prey populations. Ann.

lust. Oeeanogr. 73: 125-137.
Ross, K. F. A. 1967. Phase Contrast and Interference Microscopv.

Edward Arnold, London.
Russell, F. R. S. 1953. The Medusae of the British Isles. Cambridge

University Press. Cambridge.
Russell, F. R. S. 1970. The Medusae of the British Isles II: Pelagic

Scypho'oa. Cambridge University Press. Cambridge.
Sanamyan, K. 1998. Ascidians from the north-western Pacific region. 5.

Phlebobranchia. Ophelia 49: 97-1 16.
Sandstroem, A. 1999. Visual ecology of fish — a review with special

reference to percids. Fiskeriverk Rapp. 2: 45-80.
Scheltema, A. H. 1993. Aplacophora as progenetic aculiferans and the

coelomate origin of mollusks as the sister taxon of Sipuncula. Biol.

Bull. 184: 57-7S.
Seapy, R. R., and R. E. Young. 1986. Concealment in epipelagic

pterotracheid heteropods (Gastropoda) and cranchiid squids (Cepha-
lopoda). J. Zool. Land. 210: 137-147.
Shashar, N., R. T. Hanlon, and A. Petz. 1998. Polarization vision helps

detect transparent prey. Nature 393: 222-223.
Shinn. C. L. 1997. Chaetognatha. Pp. 103-220 in Microscopic Anatomy

of Invertebrates. Volume 15: Hemichordata, Chaetognatha, and the

Invertebrate Chordates. F. W. Harrison and E. E. Ruppert. eds. John

Wiley. New York.
Smith. R. C., B. B. Prezelin, K. S. Baker, R. R. Bidigare, N. P. Boucher,

T. Coley. D. Karentz. S. Maclntyre. H. A. Matlick, D. Menzies, M.

Ondrusek, Z. Wan, and K. J. Waters. 1992. Ozone depletion:

ultraviolet radiation and phytoplankton biology in Antarctic waters.

Science 255: 952-959.

Solomon, S. 1990. Progress toward a quantitative understanding of Ant-
arctic ozone depletion. Nature 347: 347-354.
Stolarski, R. S., R. Bojkov, L. Bishop, C. Zerefos, J. Staehelin, and J.

Zawodny. 1992. Measured trends in stratospheric ozone. Science

256: 342-349.
Sualla, B. J., C. B. Cameron. L. S. Corley. and J. R. Garev. 2000.

Urochordates are monophyletic within the deuterostomes. Syst. Biol.

49: 52-64.
Tardieu, A., and M. Delaye. 1988. Eye lens proteins and transparency:

from light transmission theory to solution x-ray structural analysis.

Annn. Rev. Biop/tvs. Biophys. Client. 17: 47-70.
Thetmeyer, H., and U. Kils. 1995. To see and not be seen: the visibility

of predator and prey with respect to feeding behavior. Mar. Ecol. Prog.

Ser. 126: 1-8.
Thiele, K. S. 1998. Light scattering by complex microstructures in the

resonant regime. Ph.D. dissertation. University of Pennsylvania.
Thorpe, A., R. H. Douglas, and R. J. W. Truscott. 1993. Spectral

transmission and short-wave absorbing pigments in the fish lens — I.

Phylogenetic distribution and identity. Vision Res. 33: 289-300.
Totton, A. K. 1965. ,4 Synopsis of the Siphonophora. British Museum,

London.
Tovee, M. J. 1995. Ultra-violet photoreceptors in the animal kingdom:

their d:-,tnbiition and function. Trends Ecol. Evol. 10: 455-460.

Tsuda, A., H. Saito, and T. Hirose. 1998. Effect of gut content on the
vulnerability of copepods to visual predation. Litnnol. Oeeanogr. 43:
1944-1947.

I'schakov, P. V. 1972. Fuunu of the U.S.S.R. Polvchaetes. Vol. 1.
Po/veluietes of the Suborder Phy/lodocifonnia of the Polar Basin and the
Northwestern Pan of the Pacific: Families Phyllodocidae, Alciopidae,
Tomopteridac, Tvphloscoleicidae. and Lucydoniidae. Akademiya NAUK
SSSR. New Series 102. B. E. Bykhorskii, ed. [Translated from Russian by
the Israel Program for Scientific Translations, Jerusalem. 1974.]

Utne-Palm, A. C. 1999. The effect of prey mobility, prey contrast,
turbidity and spectral composition on the reaction distance of Gobius-
cidns flavescens to its planktonic prey. J. Fish Biol. 54: 1244-1258.

Vaezy, S., and J. I. Clark. 1994. Quantitative analysis of the micro-
structure of the human cornea and sclera using 2-D Fourier methods. J.
Microsc. 175: 93-99.

Van der Spoel, S. 1976. Pseudothecosomata, Gymnosomata and Het-
eropoda. Bohn, Scheltema and Holkema. Utrecht.

Vinogradov. M. E., A. F. Volkov. and T. N. Semanova. 1996. Hyperiid
Ainpltipods (Amphipoda, Hyperiidea) of the World Oceans. Smithso-
nian Institution Libraries. Washington. DC.

Von W. KjerschoM-Agersborg, H. P. 1921. Contribution to the knowl-
edge of the nudibranchiate mollusk, Melibe leonina (Gould). Am. Nat.
55: 222-253.

Voss, N. A. 1980. A generic revision of the Cranchiidae (Cephalopoda;
Oegopsida). Bull. Mm; Sci. 30: 365-412.

Wald, G., and J. M. Krainin. 1963. The median eye of Linudiis: an
ultraviolet photoreceptor. Proc. Nat/. Acad. Sci. 50: 1011-1017.

Wald, G., and S. Rayport. 1977. Vision in annelid worms. Science 196:
1434-1439.

Warner, J. A., M. I. Latz, and J. F. Case. 1979. Cryptic biolumines-
cence in a midwater shrimp. Science 203: 1 109-1 1 1(1.

Waterman, T. H. 1981. Polarization sensitivity. Pp. 281-469 in Handbook
of Sensory Physiology. Vol. 7/6B. H. Autrum, ed. Springer, New York.

Weber, C., and V. Schmid. 1985. The fibrous system in the extracellular
matrix of hydromedusae. Tissue Cell 17: 811-822.

Widder, E. A., M. I. Latz, and J. F. Case. 1983. Marine biolumines-
cence spectra measured with an optical multichannel detection system.
Biol. Bull. 165: 791-810.

Wilson, S. J., and M. C. Hutlev. 1982. The optical properties of "moth
eye' antireflection surfaces. Optica Ada 7: 993-1009.

Wingstrand, K. G. 1985. On the anatomy and relationships of Recent
Monoplacophora. Galathea Rep. 16: 7-94.

Wright, D. I., and W. J. O'Brien. 1982. Differential location of Cha-
ohorus larvae and Daphniu hy fish: the importance of motion and
\isihle size. Am. Midi. Nat. 108: 68-73.

Wrobel, D., and C. Mills. 1998. Pacific Coast Pelagic Invertebrates: A
Guide to Common Gelatinous Animals. Sea Challengers, Monterey Bay
Aquarium. Monterey Bay. CA.

Yoshida, A., M. Motoyama, A. Kosaku, and K. Miyamoto. 1997.
Antireflective nanoprotuberance array in the transparent wing of a
hawkmoth. Cepltanodoes hvlas. Zoo/. Sci. 14: 737-741.

Young, R. E.. M. Vecchione, and D. T. Donovan. 1998. The evolution
of coleoid cephalopods and their present biodiversity and ecology. S.
Afr. J. Mar. Sci. 20: 393-420.

Zaitsev, Y. P. 1970. Marine Neustonology. Keter Press, Jerusalem.

Zaret, T. M. 1972. Predators, invisible prey, and the nature of polymor-
phisms in the Cladocera (Class Crustacea). Litnnol. Oeeanogr. 17:
171-184.

Zaret, T. M. 1981. Lateral compression and visibility in cladocerans.
Limnol. Oeeanogr. 26: 965-970.

Zaret, T. M., and W. C. Kerfoot. 1975. Fish predation on Koxmma
longirostrus: body size selection versus visibility selection. Ecology
56: 232-237.

Reference: Bio/. Bull. 201: 314-322. (December 2(101)

Evolution of Marine Mushrooms

DAVID S. HIBBETT* AND MANFRED BINDER
Biology Department, Clark University, 950 Main Street, Worcester, Massachusetts 01610

Fungi make up one of the most diverse, ecologically
important groups of eukaryotes. The vast majority of fungi
are terrestrial, but the chytridiomycetes, a basal group of
fungi, includes flagellated, unicellular, aquatic forms, and it
is likel\ that this was the ancestral condition of the group
( 1 ). The more derived groups of fungi — zygomycetes. asco-
mycetes. and basidiomycetes — are all predominantly fila-
mentous and terrestrial, and lack flagellated cells at any
stage of the life cycle. Within the basidiomycetes, the most
conspicuous group is the homobasidiomycetes, which in-
cludes about 13,000 described species of mushrooms and
related forms. Eleven species of homobasidiomycetes (in
eight genera) occur in marine or freshwater habitats. To
resolve the relationships among terrestrial and aquatic
homobasidiomycetes, we assembled a data set of ribosomal
DNA (rDNA) sequences that includes 5 aquatic species and
40 terrestrial species. Phylogenetic trees obtained using
parsimony and maximum likelihood {ML) methods suggest
that there have been three or four independent transitions
from terrestrial to aquatic habitats within the homobasid-
iomycetes. Three of the marine ta.\a in our data set are
associated with mangroves, suggesting that these ecosys-
tems provide a common evolutionary stepping-stone b\
which homobasidiomycetes have reinvaded aquatic habi-
tats.

One of the major themes in the evolution of eukaryotes is
the repeated transition from aquatic to terrestrial habitats
that has occurred in several major clades. including fungi,
plants, and animals. In a classic paper, Pirozynski and
Mai loch (2) suggested that fungi and plants were the first
eukaryotes to colonize the land, and that this ecological shift
was made possible by the establishment of mycorrhizal
symbioses (associations involving fungal hyphae and plant
roots). This hypothesis has been bolstered by the recent

Received 19 July 2001: accepted 30 August 20(11.
* To whom correspondence should be addressed. E-mail: dhihbettO1
hlack.clarku.edu

discovery of spores of putatively mycorrhizal fungi from the
Ordovician (3). Fungi have radiated extensively in terres-
trial habitats, where they play pivotal ecological roles, as
decayers. pathogens, and symbionts of plants and animals.

One group of fungi that is elegantly adapted to life on the
land is the homobasidiomycetes. Adaptations to terrestrial
habitats displayed by some homobasidiomycetes include
rootlike rhizornorphs that allow the fungi to forage along the
forest floor, drought-resistant sclerotia, and tough, perennial
fruiting bodies. Aerial spore dispersal in homobasidiomy-
cetes is accomplished by a forcible discharge mechanism
termed ballistospory. However, several lineages of terres-
trial homobasidiomycetes have lost ballistospory, including
puffballs and other forms with enclosed spore-bearing struc-
tures.

Aquatic homobasidiomycetes include four species that
have retained ballistospory and seven species that have lost
ballistospory. The ballistosporic forms can be tentatively
assigned to certain terrestrial groups on the basis of the
morphology of spores and fruiting bodies (4-12). However,
the aquatic homobasidiomycetes that have lost ballistospory
have no obvious close relatives among terrestrial groups.
This taxonomically enigmatic assemblage includes several
marine genera that have elongate or appendaged spores,
which superficially resemble the spores of many aquatic
ascomycetes (4, 5; Fig. 1 ).

To resolve the relationships among terrestrial and aquatic
homobasidiomycetes. we assembled a data set that includes
4 marine species, 1 freshwater species, and 40 diverse
terrestrial species (Fig. 2). The heterobasidiomycete "jelly
fungus" Auricularia auricula-jitdae was included for root-
ing puiposes. The data set contains sequences of four rDNA
regions, including nuclear and mitochondria! small and
large subunit rDNA (3.8 kb total). Four species in the data
set lack the mitochondria! large subunit rDNA sequence
(Fig. 2). Sequences from 38 terrestrial species and one
marine species, Nia vibrissa, were drawn from earlier stud-
ies (13, 14).

319

320

D. S. HIBBETT AND M. BINDER

Figure 1. Appendaged spore (A) and enclosed fruiting bodies (B) of
the marine homobasidiomycete Niu vibrissa.

Parsimony analysis (15) resulted in two shortest trees
(5175 steps, consistency index (CD = 0.372. retention in-
dex (RI) = 0.410). and ML analysis resulted in one optimal
tree (-InL = 29962.65066; Fig. 2). In all trees, the aquatic
species occur in four separate lineages (Fig. 2). There are
two equally parsimonious reconstructions of shifts between
terrestrial and aquatic habitats (on all three trees). One
reconstruction suggests that there have been four indepen-
dent transitions from terrestrial to aquatic habitats (Fig. 2 A).
whereas the other reconstruction suggests that there have
been three shifts to aquatic habitats and one reversal from
aquatic to terrestrial habitats (Fig. 2B). Under the latter
scenario, the terrestrial species Cyphellopsis anomala
would be derived from marine ancestors.

All of the aquatic species in our data set are nested in a
strongly supported group (parsimony bootstrap = 90%/ML
bootstrap = 99%) called the euagarics clade. which has
been estimated to contain roughly 7400 species (57%) of
homobasidiomycetes (Fig. 2; 16). Most members of the
euagarics clade are typical mushrooms, with a cap. gills, and
(often) a stalk. Familiar taxa in our data set include the
cultivated button mushroom Agaricus bisporus and the my-
corrhizal "fly agaric" Ainanita muscaria. The ancestor of the
euagarics clade was probably a gilled mushroom ( 14), but
contemporary aquatic fungi bear scant resemblance to such
forms, as described below.

Three marine species in our data set, Halocyphina villosa.
Calathella mangrovei, and Physalacria imiipoensis are bal-
listosporic, have exposed spore-bearing surfaces, and occur
in intertidal mangrove communities. Halocyphina villosa
and Calathella mangrovei produce "cyphelloid" fruiting
bodies, which are minute (a;. 0.3-1.0 mm diameter), cup-
shaped structures, whereas Physalacria imiipoensis pro-
duces a "capitate" fruiting body, which has a globose head
on a short stalk (a/. 0.5-2.5 mm high: 8. 9. 1 1 ). The genera
Calathella and Physalacria each include terrestrial species.
as well as the marine species sampled here (9. 1 1 ). Halo-
cvphina contains only one species, but Ginns and Malloch

(8) suggested that it could be closely related to the terrestrial
cyphelloid genera Henningsomyces or Rectipilns. Consis-
tent with this prediction, our results suggest that the terres-
trial cyphelloid genera Henningsomyces and Cyphellopsis
are closely related to marine homobasidiomycetes (Fig. 2).
The remaining aquatic species in our dataset, Niu vibrissa
(marine) and Limnoperdon incarnatum (freshwater), have
lost ballistospory and produce spores inside minute (ca.
0.3-1.2 mm diameter), enclosed, puffball-like fruiting bod-
ies (5, 7, 17. 1 8; Fig. 1). Niu vibrissa is further distinguished
by having appendaged basidiospores (Fig. 1). Nia ribrissa
and Limnoperdon incarnatum bear a superficial resem-
blance to terrestrial puffballs, but their phylogenetic rela-
tionships have been obscure. Our results indicate that Nia
vibrissa is strongly supported (bootstrap = 99%/100%) as
the sister group of Halocyphina villosa (Fig. 2). The precise
placement of Limnoperdon incarnatwn is not resolved with
confidence, although it is strongly supported as a member of
the euagarics clade (bootstrap = 90%/99%: Fig. 2).

The close relationship of Nia vibrissa and Halocyphina
villosa could not have been predicted based on morphology.
Aside from their small size and marine habit they have no
distinuuishing features in common. Moreover. Halocyphina
villosa occurs in mangroves, whereas Nia vibrissa and the
related species N. epidermoidea and N. globospora have
been collected on fully submerged substrates, including
driftwood and the wreckage of a sunken ship, and have been
isolated by baiting with submerged wooden test panels,
Spurtina culms, horsehair, and feathers (17, 19-22). Nev-
ertheless, the Nia-Halocyphina clade is strongly supported
and is nested in another strongly supported clade (boot-
strap = 100<7f/100%) that includes the mangrove-inhabiting
species Calathella mangrovei and two terrestrial species,
Cvphellopsis anonnila and Favolaschia intermedia (Fig. 2).
With its appendaged spores, enclosed fruiting body, and
habit on submerged substrates, Niu vibrissa is the most
derived member of this clade. Transformations leading to
the evolution of this unusual basidiomycete probably in-
volved a shift from terrestrial to periodically immersed to
fully submerged substrates, loss of ballistospory. and evo-
lution of appendaged spores and an enclosed fruiting body.
Significantly, the cyphelloid fruiting body of Halocyphina
villosa is enclosed during parts of its ontogeny, and at
maturity the opening of the fruiting body is partially cov-
ered by interwoven hyphae (8, 18). Thus, the mangrove-
inhabiting Halocyphina villosa appears to be morphologi-
cally and ecologically intermediate between Nia vibrissa
and terrestrial cyphelloid forms, such as Cyphellopsis
anomala.

In the mangroves where they occur. Calathella man-
grovei, Halocvplumi villosa. and Physalacria imiipoensis
are all periodically submerged in seawater (4. 5, 8, II).
Phvsalacria imiipoensis. however, has also been found in
adjacent upland forests that are not inundated (9).

EVOLUTION OF MARINE MUSHROOMS

321

Calathella mangrove! -mi
Halocyphina villosa
Nia vibrissa M
Cyphellopsis anomala
Favolaschia intermedia
Henningsomyces candidus
Physalacria maipoensis
Fistulma hepatica -ml
Schizophyllum commune
Amanita muscaria
Cortinanus iodes
Strophana rugosoannulata -ml
Crucibulum laeve
Laccaria amethystina
• Typhula phacorhiza
Cyathus striatus -ml

IAgaricus bisporus
v-ycoperdon sp
Entoloma strictius
Pluteuspetasatus
Limnoperdon jncarnatum F
Pleurotus ostreatus
Pleurotus tuberregium
. Humidicutis marginata

Hygrophorus sordidus

Boletus satanas

I Boletus satanas

— pS" Suillus sinuspaulianus

98 1 — Tapinella panuoides

P^'oeocystidiellum leucoxantha
terobasidion annosum
Russula compacta
fyphodontia alutarla
Phellmus ignianus
Albatrellus syringae
Panus rudis

Fomitopsis pinicola
Polyporus squamosus
Bjerkandera adusta
Phlebia radiata
Bankera fuligineoalba
Thelephora sp.

Gomphus floccosus
Sphaerobolus stellatus
Hydnum repandum
Sistotrema eximum
Auricularia auricula-judae

Calathella mangrove! -mi M
Halocyphina villosa M
Nia vibrissa M

Cyphellopsis anomala
Favolaschia intermedia
Fistulina hepatica -ml

Schizophyllum commune
Henningsomyces candidus
Physalacria maipoensis M

Amanita muscaria
Agaricus bisporus
Lycoperdon sp
Cortinarius iodes
Laccaria amethystina

Stropharia rugosoannulata -ml
Crucibulum laeve

Cyathus striatus -ml
Limnoperdon incarnatum F

: Entoloma strictius
Pluteus petasatus

Typhula phacorhiza
Pleurotus ostreatus
Pleurotus tuberregium
Humidicutis marginata

Hygrophorus sordidus

Boletus satanas

Suillus sinuspaulianus
Tapinella panuoides
Gloeocystidiellum leucoxantha
Heterobasidion annosum
Russula compacta
Hyphodontia alutaria
Phellinus igniarius
Albatrellus syringae
Panus rudis
Bjerkandera adusta
Phlebia radiata
Fomitopsis pinicola
Polyporus squamosus

Bankera fuligineoalba
Thelephora sp.

Gomphus lloccosus

Sphaerobolus stellatus

Hydnum repandum

100 > Sistotrema eximum

• Auricularia auricula-judae

— 50 changes

• 0.05 substitutions/site

Figure 2. Phylogenetic relationships of terrestrial, marine, and freshwater homobasidiomycetes interred from
nuclear and mitochondria! ribosomal DNA (rDNA) sequences, and alternative reconstructions of the history1 of shifts
between terrestrial and aquatic habitats. ( Al One of two phylogenetic trees inferred using parsimony (asterisk indicates
the one node that collapses in the strict consensus tree). (B) Phylogenetic tree interred using maximum likelihood
(ML). Names of aquatic taxa are in bold type; M = marine. F = freshwater. Taxa marked —ml lack mitochondria]
large subunit rDNA sequences. Bootstrap values are indicated next to branches (only values above 70% are shown).
Branch shading indicates reconstruction of ancestral habitats; thin lines = terrestrial, thick lines = aquatic. The
parsimony tree (A) shows a reconstruction of habitat shifts that involves four independent transitions from terrestrial
to aquatic habitats. The ML tree (B) shows an equally parsimonious reconstruction of ancestral states that involves
three transitions from terrestrial to aquatic habitats, and one reversal. Methods: DNA was isolated from cultured
mycelium, and nuclear and mitochondnal rDNA regions were amplified and sequenced using protocols and primers
that have been reported elsewhere (13. 14). Sequences were aligned by eye in the PAUP* ( 15) data editor. After
excluding 185 positions that were deemed to be ambiguously aligned, the data set included 3574 aligned positions,
of which 1267 were variable and 827 were parsimony-informative. Parsimony analysis used 1000 heuristic searches
with random taxon addition sequences, tree hisection-reconnection (TBR) branch-swapping, and MAXTREES set to
autoincrea.se. with all characters and transformations equally weighted. Bootstrapped parsimony analysis used 1000
replicates with one heuristic search per replicate, with other settings as in the baseline analysis. ML analysis used the
HKY85 model of sequence evolution, with empirical base frequencies, transition-trans\ ersion bias set to 2. and
among-site rate heterogeneity modeled on a discrete gamma distribution, with four rate classes and shape parameter
a set to 0.5. The ML analysis used a heuristic search, with the trees obtained in the parsimony analysis used as starting
trees for branch swapping with TBR. Bootstrapped ML analyses used 100 replicates, with one heuristic search per
replicate, using a starting tree generated with neighbor-joining (Kimura two-parameter distances), and TBR branch
swapping. A time limit of 1 hour per bootstrap replicate was enforced. Sequences have been deposited in GenBank
(accession numbers AF426948-AF426970. which should be consulted for strain data) and the data set has been
deposited in TreeBASE (accession number S666).

Inderbitzin and Desjardin (9) regarded Physalacria nuii-
poensis as "halotolerant." and suggested that it is closely
related to certain terrestrial species of Physalacria. It is

tempting to speculate that Physalacria maipoensis repre-
sents an early stage in the transition from terrestrial to
marine environments in homobasidiomycetes.

D. S. HIBBETT AND M. BINDER

Acknowledgments

We are indebted to E. B. Gareth Jones, who provided a
collection of Ciilntlu'llii imingrovei; Patrick Inderbitzin,
who provided a culture of Physalacria maipoensis; and
Karen Nakasone, who provided a culture and confirmed the
identification of Favolaschia intermedia. This work was
supported by the National Science Foundation.

Literature Cited

1. Barr, D. J. S. 20IH. Chytridiomycota. Pp. 93-112 in The Mycota
VII. Part A. Systematic! and Evolution. Springer- Verlag. Berlin.

2. Pirozynski, K. A., and D. W. Malloch. 1975. The origin of land
plants: a matter of mycotrophism. Biosyslems 6: 153-164.

3. Redecker, D., R. Kodner, and L. Graham. 20(10. Glornalean fungi
from the Ordovician. Science 289: 1920-1921.

4. Hyde, K. D., V. V. Sarma, and E. B. G. Jones. 2000. Morphology
and taxonomy of higher marine fungi. Pp. 172-204 in Marine Mycol-
ogy— A Practical Approach. Fungal Diversity Press. Hong Kong.

5. Kohlmeyer, J., and E. Kohlmeyer. 1979. Marine Mycology — The
Higher Fungi. Academic Press. New York.

6. Desjardin, D. E., L. Martinez-Peck, and M. Rajchenberg. 1995.
An unusual psychrophilic aquatic agaric from Argentina. Mycologia
87: 547-550.

7. Escobar, G. A., D. E. McCabe, and C. W. Harpel. 1976. Lim-
noperdon. a floating gasteromycete isolated from marshes. Mycologia
68: 874-880.

8. Ginns, ,).. and I). W. Malloch. 1977. Halocyphina, a marine basid-
iomycete (Aphyllophorales). Mycologia 69: 53-58.

9. Inderbitzin, P., and D. E. Desjardin. 1999. A new halotolerant
species of Physalacria from Hong Kong. Mycologia 91: 666-668.

10. Jones, E. B. G. 1986. Digitalispora lignicola sp. nov. A new marine
lignicolous basidiomycotina. Mycotaxon 27: 155-150.

1 1 . Jones, E. B. G., and R. Agerer. 1992. Calaihella mani>rovii sp. nov.
and observations on the Mangrove fungus Halocyphina villosa. Bot.
Men: 35: 259-265.

12 Porter. D., and VV. F. Farnham. 1986. Mycunreola dilseae, a
marine basidiomycete parasite of the red alga. Dilsea carnosa. Trans.
Br. Mycol. Soc. 87: 575-582.

13. Binder, M., D. S. Hibbett, and H. P. Molitoris. 2001. Phylogenetic
relationships of the marine gasteromycete Nia vibrissa. Mycologia 93:
679-688.

14. Binder, M., and D. S. Hibbett. 2001. Higher level phylogenetic
relationships of homobasidiomycetes (mushroom-forming fungi) in-
terred from four rDNA regions. Mol. Phylogen. Evol. (in press).

15. Swoffbrd, D. L. 2001. PAUP* Phylogenetic Analysis Using Parsi-
miHiy and Other Methods. Version 4.0b8. Smithsonian Institution and
Sinauer Associates. Sunderland. MA.

16 Hibbett, D. S., and R. G. Thorn. 2001. Basidiomycota: Homobasi-
diomycetes. Pp. 121-168 in The Mycota VII. Part B. Systemulics and
Evolution. Springer- Verlag. Berlin.

1 7. Jones, A. M., and E. B. G. Jones. 1993. Observations on the marine
gasteromycete Ma vihrissa. Mycol. Res. 97: 1-6.

18. Nakagiri. A., and T. Ito. 1991. Basidiocarp development of the
cyphelloid gasteroid aquatic basidiomycetes Halocyphina rillosa and
Lnnnoperdon incarnatiim. Can. J. Bot. 69: 2320-2327.

19. Barata, M., M. C. Basilo. and J. L. Baptista-Ferreira. 1997. Ma
globospora. a new marine gasteromycete on baits of Spartina mari-
tima in Portugal. Mycol. Res. 101: 687-690.

20. I.eightley, L. E., and R. A. Eaton. 1979. Ma vibrisxa — a marine
white rot fungus. Trans. Br. Mycol. Soc. 73: 35-40.

2 1 . Rees, G., and E. B. G. Jones. 1985. The fungi of a coastal sand dune
system. Bot. Mar. 28: 213-220.

22. Rossello, M. A., and E. Descals. 1993. Ma epidennoidea. a new
marine gasteromycete. Mycol. Res. 97: 68-70.

Reference: Binl. Bull. 201: 323-338. (December 20(11)

Cytological Basis of Photoresponsive Behavior

in a Sponge Larva

SALLY P. LEYS1'* AND BERNARD M. DEGNAN-

1 Department of Biologv, University of Victoria, British Columbia, Canada, V8W 3N5; and Department
of Zoology and Entomology, University of Queensland, Brisbane. QLD 4072 Australia

Abstract. Ontogenetic changes in the photoresponse of
larvae from the deinosponge Reneira sp. were studied by
analyzing the swimming paths of individual larvae exposed
to diffuse white light. Larvae swam upward upon release
from the adult, but were negatively phototactic until at least
12 hours after release. The larval photoreceptors are pre-
sumed to be a posterior ring of columnar monociliated
epithelial cells that possess 1 20-jum-long cilia and pigment-
filled protrusions. A sudden increase in light intensity
caused these cilia to become rigidly straight. If the light
intensity remained high, the cilia gradually bent over the
pigmented vesicles in the adjacent cytoplasm, and thus
covered one entire pole of the larva. The response was
reversed upon a sudden decrease in light intensity. The
ciliated cells were sensitive to changes in light intensity in
larvae of all ages. This response is similar to the shadow
response in tunicate larvae or the shading of the photore-
ceptor in Euglena and is postulated to allow the larvae to
steer away from brighter light to darker areas, such as under
coral rubble — the preferred site of the adult sponge on the
reef flat. In the absence of a coordinating system in cellular
sponges, the spatial organization and autonomous behavior
of the pigmented posterior cells control the rapid responses
to light shown by these larvae.

Introduction

Light, gravity, current, and chemical cues enable the
larvae of many marine invertebrates to locate the habitat
that will best ensure their success as adults (Grave, 1926;
Ryland. I960; Thorson, 1964; Forward and Costlow. 1974;
Brewer, 1976; Young and Chia, 1982; Miller and Hadfield.
1986; Svane and Young. 1989; Pawlik, 1992). Thus, eye-

Received 21 December 2000; accepted 22 October 2001.

* To whom correspondence should be addressed. E-mail: spleys@uvic.ca

spots are well developed in many bilaterian larvae (see
Eakin, 1968, 1972; Burr. 1984), and signals received by
these and other sensory organs are apparently translated into
behavior via the larval nervous system (Thomas et al., 1987;
KempfetaL. 1997; Murphy and Hadfield, 1997; Hadfield et
al.. 2000). The role of photosensory systems in the larval
behavior of basal metazoans is less well documented. Al-
though ocelli are well developed in cnidarian medusae and
polyps (Thomas and Edwards, 1991), the putative photore-
ceptors that have been identified in planulae are simple
monociliated sensory cells with electron-dense granules
(Weis et al.. 1985; Thomas et al.. 1987). Presumably the
neurons underlying the ciliated epithelium of cnidarian
planulae are involved in assessing the environment (Chia
and Koss, 1979; Martin and Chia, 1982; Thomas et al..
1987). but there is currently no evidence for synaptic sig-
naling between presumptive photoreceptors and other cells.
Poriferan larvae are considered to be even more simply
constructed than planulae in that they lack neurons.

Porifera is the only metazoan phylum that lacks neurons
(Pavans de Ceccatty, 1974a, b; Mackie, 1979). Furthermore,
despite one report suggesting electrical coupling between
two reaggregated cells from dissociated adult tissue (Loe-
wenstein. 1967), there is no evidence that sponges have gap
junctions, which would allow the rapid conduction of be-
havioral signals between cells (Green and Bergquist, 1982;
Lethias et al., 1983). Members of the subphylum Sym-
plasma, the Hexactinellida, are the only sponges known to
be capable of rapid behavior (Lawn et al.. 1981; Mackie et
al., 1983). Because hexactinellid tissue is mostly syncytial
(Leys, 1995), the electrical signals that cause concurrent
shutdown of flagellar activity propagate along the mem-
brane of the continuum (Leys and Mackie, 1997; Leys et al..
1999).

Behavior in cellular sponges, the Demospongiae and Cal-

323

324

S. P. LEYS AND B. M. DEGNAN

carea, is limited to gradual contraction of the tissues (Mc-
Nair, 1923: Vacelet, 1966; Pavans de Ceccatty, 1969, 1976;
Mackie. 1979; Lawn, 1982) and variations in pumping
patterns (Reiswig, 1971 ). for which chemical or mechanical
coordination are invoked. Although the mechanisms for
coordinated behaviors are apparently absent, cellular
sponge larvae do exhibit rapid responses to external stimuli.
The responses of sponge larvae to light, gravity, and current
have been reported since the early 1900s (reviewed in
Wapstra and van Soest, 1987).

Photokinesis. one of the most tangible aspects of sponge
larval behavior, is best known from studies on parenchy-
mellae larvae of demosponges (Warburton, 1966; Bergquist
and Sinclair. 1968: Bergquist et ai, 1970; Wapstra and van
Soest, 1987; Woollacott, 1990, 1993: Maldonado and
Young, 1996, 1999). Typically, these larvae are oblong and
heavily ciliated. The parenchymellae of different species are
distinguished primarily by the presence or absence of cilia
at the poles of the larva, of a ring of longer cilia at one end,
or of pigmented cells at one end. Unfortunately, the pattern
of ciliation or pigmentation on larvae smaller than 500 /j,m
is difficult to determine accurately by light microscopy, and
relatively few larvae have been characterized by electron
microscopy (Evans, 1977; Simpson, 1984; Woollacott and
Hadfield, 1989: Harrison and De Vos, 1991; Kaye and
Reiswig, 1991; Amano and Hori, 1992; Woollacott, 1990,
1993; Fell, 1997). Furthermore, only a few investigators
have taken an experimental approach to sponge larval be-
havior (Jaeckle, 1995; Woollacott and Hadfield. 1996; Mal-
donado and Young, 1996. 1999; Maldonado et al., 1997);
most studies report only anecdotal observations.

The cellular mechanisms underlying sponge larval behav-
ior have yet to be addressed: how does an animal lacking
nerves and communicating junctions between its cells re-
spond so agilely to light and other stimuli? This paper
addresses the ontogenetic change in the light response and
its cytological basis in the parenchymella larva of the demo-
sponge Reneiru sp.

Materials, Methods, and General Observations

Collection and maintenance of specimens

Adult specimens of the sponge Reneiru sp. (Porifera.
Demospongiae. Haplosclerida. Chalinidae) were collected
in February. April. August, and December, 1999. from the
reef flat in Shark Bay on Heron Is. Reef, Great Barrier Reef
(23°26'N, 151°03'E). The sponges were maintained in
shaded aquaria in seawater pumped from the reef slope.

Systematics

The identification of this sponge as Reneira sp. was
confirmed by taxonomists at the Queensland Museum.
However, as this species has not yet been formally de-

scribed, a brief description is given here. The sponge is grey
or olive brown, and its texture is firm due to a well-
developed anisotropic reticulate network of primary spon-
gin that is cored by paucispicule to multispicule tracts of
oxeas 80-100 jitm long by 1 ;u,m wide. Oscula are slightly
raised above the surface of the sponge, which is formed by
a typical chalinid isodictyal reticular network that is tangen-
tial to the surface. We have deposited a voucher specimen
and photograph in the Poriferan Collection in the Queens-
land Museum (QM G3 15611). The North Atlantic genus
Reneira has been variously called Haliclona or Adocia in
the past, and most recently taxonomists have formally trans-
ferred the genus Reniera to Haliclona (de Weerdt, 1986).
Although the Pacific species of these genera have not been
revised recently (J. Hooper, Queensland Museum, Austra-
lia; pers. comm.), the behavior and structure of the Reneira
sp. larvae studied here appear to be very similar to those
reported for other chalinids, and even most haplosclerids
(Wapstra and van Soest. 1987).

Huhiuit tinil description of adult sponges

The sponge forms encrustations 1-3 cm thick on the
underside of coral rubble, which is home to numerous other
encrusting and grazing animals. The coral is just submerged
at low tide and is approximately 3 m deep at high tide.

The brood chambers of Reneira sp. are typically located
in the lowest portion of the sponge closest to the coral
substrate (Fig. la). Reneira sp. is reproductive year round
(Leys and Degnan, unpubl. data), but although sponges
collected in all seasons contained brood chambers, sponges
collected in August had the least number and released the
fewest larvae. The chambers are up to 1 cm2 in diameter and
contain 20 to 150 embryos, 600-900 jam long, in a wide
range of developmental stages (Fig. Ib). Spermatocysts
were found in only 2 of more than 100 sponges that were
collected and sectioned during all collection periods.

Description of the larvae

The larvae of Reneira sp. are cream colored with a dark
ring of pigment-containing cells around the posterior end; in
fact, the dark pigmented ring defines the posterior end (Fig.
Ib. c). The outer layer of the larva consists primarily of
monociliated cells possessing 20-ju,m-long cilia (hereafter
called short lateral cilia), but there are two protruding bare
patches, one each at the poles of the larva. The bare patch at
the anterior end is 55-60 /u,m in diameter, and that at the
posterior end is 140-160 /urn in diameter and lies inside the
pigmented ring (Fig. Ic, d). The anterior border of the
pigmented ring is marked by a ring of cells that contain
pigment vesicles but also give rise to 120-150-/xm long
cilia (hereafter called long posterior cilia) (Fig. Ic, d). These
latter structures are more appropriately described as cilia

PHOTOKINES1S IN SPONGE LARVAE

325

Figure 1. Brood chambers and embryos of Reneira sp. in various stages of development (a-c: light
microscopy; d: scanning electron microscopy), (a) A section of the adult sponge that was attached at its lower
edge to the coral substrate (arrowhead) shows a brood chamber (B ch) with embryos and larvae. Bar: 1 cm.
Dermal surface, ds. (b) Embryos and larvae in a brood chamber clearly showing the pigment ring (PRg) at one
pole. Bar: I mm. (c) A swimming larva showing the dark pigmented ring (PRg) and long posterior cilia (LPC)
at the posterior swimming pole (PP). and a protrusion at the anterior swimming pole (AP). Bar: 100 ;xm. (d) A
larva showing the long posterior cilia (LPC). the unciliated posterior pole (PP). and the lines of short lateral cilia
(SLC) arrested by fixation during their beat in metachronal waves. Bar: 250 fjm.

rather than flagella because their motion is whiplike; they do
not propagate quasi-sinusoidal waves (Alberts ct <//., 1989).

Laboratory experiments on lan'iil phototaxis

The larvae were maintained individually in 2 ml of 0.2-
/Lun-filtered seawater in 12-well multiwell dishes at room
temperature (about 22 °C). At various times after release —
i.e.. 0. 2. 4, 6. 12. 24. and 48 hours — individual larvae were
pipetted into a rectangular aquarium (15 X 20 cm) contain-
ing 0.2-jam-filtered seawater (Fig. 2a). Pipetting was not
observed to affect the swimming behavior of the larvae. The
rectangular aquarium (the test chamber) was immersed in
seawater in a second aquarium, which was blackened on all
but one side to reduce reflected light (after Wendt and
Woollacott. 1999). Light from a cold light source (Volpi
Intralux 5000) was passed through a diffuser made of
acrylic plastic into the inner test chamber, such that a

gradient of light was created in the horizontal direction from
the front to the back of the test chamber (950 ju,A/ • photons •

s to < 1 p.M • photons • in

). The radiance at

the side closest to the light was at the same level recorded
at the edge of the underside of a coral boulder at low tide on
the reef flat in bright sunlight during the day (Fig. 2b). Light
measurements were made in the field and in experimental
aquaria with a LI-COR underwater quantum sensor (LI-
192SA. LI-COR Inc.. Nebraska). Ambient light in the room
where measurements were made was 1.8 /iM • photons •
m~2 • s~'. A glass plate was placed above the test chamber,
and the changing position of the larva in the test chamber
was recorded for one minute with a nonpermanent felt
marker; these records were later transcribed onto paper.
Between tests, the larvae were maintained away from direct
light in their multiwell dish, at 22 °C.

The initial direction swum by each larva was recorded

326

S. P. LEYS AND B. M. DEGNAN

Cold
Light
Source

Diffuser

/ Outer Chamber

30cm

Figure 2. (a) The experimental apparatus for measuring the phototaxis
of individual larvae in response to horizontal light from a cold light source
shining through a diffuser of acrylic plastic. A test chamber containing
tillered seawater is immersed in seawater contained in an outer chamber,
which is blackened on all sides except that facing the light source. Larvae
were dropped by pipette into the inner test chamber in which there was a
gradient of light in the horizontal direction (left to right in the diagram) of
950 fj.M- photons- m~- -s~' to <1 jtiM • photons- m~2-s~'. See methods
for further details, (b) Light intensities on the reef flat during a sunny day
were recorded at 5 positions (A-E) around coral rubble that was in 30 cm
of water at low tide. The average of 10 measurements at each position is
given in jiM • photons • m~2 • s~l. A: 1906.5: B: 1354; C: 785.8: D: 57.9;
E: 9.4. The substrate below the coral was sand.

and plotted as a circular distribution. The mean angle swum
by the larvae in each age group (i.e.. 0 — 1-8 h) was calcu-
lated, and the measure of randomness was tested using the
nonparametric Rayleigh test [a high - value, or an r value
approaching 1 . indicates the data are highly grouped (Zar,
1984)].

Video and light microscopy

Live larvae were observed with an Olympus SZH dis-
secting microscope with a IX plan lens, and with an Olym-
pus BX60 compound microscope equipped with an Olym-
pus C35 AD4 photoautomat. New glass coverslips (22 X 22
mm) were placed on the bottom of a 5-cm-diameter plastic
petri dish, and individual larvae were pipetted forcefully
onto this surface, causing them to adhere by their anterior
end for up to 5 min. During this period, light levels could be
manipulated, and the cilia could be observed. Cold light was
shone on the posterior end of the larva. Other larvae that had
adhered to the dish or coverslip were transected medially,
creating an anterior portion and a posterior portion with its
pigment ring and long posterior cilia intact. Although mu-
cus and cellular material from the wound was initially
caught in the cilia, these debris disappeared after several
minute^: then cilia on both the anterior and posterior por-

tions continued their normal beating, and both halves ro-
tated as they did prior to being cut. If the posterior half of
a bisected larva was placed with the pigment ring facing
upward, it would continue to rotate on the spot indefinitely.
Light from a cold light source was shone at the pigmented
ring and long posterior cilia on the posterior end of the
bisected larva from either the left or right side of the
microscope stage. The ciliary beat was recorded using a
Panasonic digital CCD video camera and a National time-
lapse VCR (AG6010) in real-time recording mode. The
intensity of light from the cold light source was measured in
seawater on the dissecting microscope base with a LI-COR
underwater quantum sensor.

The effect of elevated KC1 (10-50 mAf) on beating of
cilia was tested; ciliary beating was recorded by video CCD.

Electron microscopy

Larvae were fixed for ultrastructural observations in a
fixative cocktail consisting of 1% OsO4 and 27c glutaralde-
hyde in 0.45 M sodium acetate buffer (pH 6.4) with 10<7r
sucrose (Leys and Reiswig. 1998). For scanning electron
microscopy, fixed larvae were dehydrated in a graded eth-
anol series, critical-point-dried with CO-,, and coated with
gold in an Edwards S150B sputter coaler. Up to five larvae
were mounted on each stub with clear nail polish and
viewed in a Hitachi S-3500N scanning electron microscope
at the University of Victoria.

For transmission electron microscopy, the fixed larvae
were dehydrated in a graded ethanol series to 10c/c. stained
with 0.57r uranyl acetate in 70% ethanol en bloc overnight,
desiliciried in 47c hydrofluoric acid in 70% ethanol. and
then embedded in Epon (Taab 812). Semithin and thin
sections were cut on either a Reichert UM2 or a Leica
Ultracut T ultramicrotome. Semithin sections were stained
with Richardson's (Richardson et ai. 1960). mounted in
Histoclad, viewed with a Zeiss Axioskop compound micro-
scope, and photographed with a digital DVC camera using
Northern Eclipse software. Thin sections were stained with
lead citrate and viewed with a JOEL 1010 transmission
electron microscope at the University of Queensland, or
with a Hitachi 7000 transmission electron microscope at the
University of Victoria.

Results

Larval release and swimming behavior

If sponges were placed in an aquarium without flowing
seawater. larvae were released at all times of the day. either
within 30 min of collection, or when the brood chambers
were cut open with a scalpel. Upon release, the larvae swam
out of the oscula and directly upward until they reached the
surface of the aquarium. In the presence of light, the larvae
generally swam forward continuously, corkscrewing or ro-

PHOTOKINESIS IN SPONGE LARVAE

Table 1

ufU'iietic change in swimming speed of Reneira sp. larvae

327

Age of larva (h post release)

Mean swimming speed (cm/s)

0.14

0.18

0.16

0.12

0.07

Number

0.081

0.087

0.099

0.110

0.095

0.072

Variance

0.0066

0.0076

0.0098

0.0122

0.0091

0.0052

; test

Zero-h larvae vs.

24-h larvae

2-h larvae vs.

24-h larvae

0.0043

0.0001

tating clockwise (as observed from the posterior end of the
larva), with occasional bursts of acceleration for periods of
several seconds. Larvae responded to light in an identical
manner in all seasons.

If undisturbed, larvae in the laboratory would swim at the
surface of the seawater for the first 2-3 h after release.
Thereafter, they tended to remain at the bottom of their dish
moving forward slowly or rotating in one spot with the
anterior end upward. However, as soon as the dish was
disturbed by light or movement, larvae younger than 2 days
old would begin swimming vigorously forward, often at the
surface of the water. They were energetic swimmers for
1 2 h. until they began a creeping phase along the substrate
prior to settlement and metamorphosis. Whereas undis-
turbed larvae metamorphosed 12 to 24 h after release, larvae
that were disturbed periodically generally did not metamor-
phose until 48 hours after release or longer. Some disturbed
larvae never metamorphosed and eventually died after one
week. Of more than 100 larvae observed, three swam in the
reverse direction with the long posterior cilia leading.

Ontogenetic response of Reneira sp. lan-ae to
unidirectionul lix/it

Young larvae (<12 hours old) stimulated by light swam
energetically in the mid-water column or on the surface of
the test aquarium and stopped when they reached a point at
which the light intensity fell, from approximately 10%, to
0.1% of the original intensity (from 73 juA/ • photons • m~2 •

s~' to 1 n,M • photons • m~2 • s~'). Older larvae (>12 h old)
swam slowly along the substrate away from the light source
and continued swimming until they reached the end of the
test aquarium, regardless of light intensity. The mean ve-
locities of newly released larvae (0 h) and of 2-h-old larvae
stimulated by light were significantly faster than those of
day-old larvae (Table 1 ).

The great majority of newly released larvae (0 h old)
were negatively phototactic in response to unidirectional
light | mean angle swum (a) = 193 °], but a few larvae in
this age cohort swam erratically, showing no preference for
swimming direction (/• = 0.6) (Fig. 3). Larvae aged 2. 4.
and 6 h were all strongly negatively phototactic (Fig. 3).

The mean angle swum by larvae in response to light shone
from zero degrees was 163 ° (r = 0.9) for 2-h-old larvae,
160 ° (/• = 0.82) for4-h-old larvae, and 174 ° (/- = 0.85)
for 6-h-old larvae. At 12 h after release, active larvae were
still swimming directly away from light [mean angle swum
(o) = 187 °], while less active larvae swam in spirals in
one place and were only weakly phototactic, if at all. By
24 h after release from the brood chambers, the swimming
directions of larvae were highly varied (- =- 1.637: ;• =
0.233). All 48-h-old larvae showed little swimming activity
and sank to the bottom of the test aquarium rotating gently
in one spot (Fig. 3).

Response of lan'al cilia

Most larval cilia are 20 /u,m long and beat in a pattern of
metachronal waves that proceeds obliquely around the larva
from anterior to posterior pole (Fig. Id). This beat is un-
ceasing, and the pattern of beat did not change when the
larva was prodded or even cut in half. Moreover, these cilia
did not respond to changes in light intensity or increased
levels of KC1.

The circular band or ring of long posterior cilia circum-
scribing the unciliated posterior pole beat either intermit-
tently or in a single wave in a counterclockwise direction (as
viewed from the posterior of the larva). The beat of these
long cilia was unaffected by mechanical stimuli, but when
the larva was transected medially, so as to isolate the
posterior portion, these cilia stopped beating, apparently
because they were tangled in mucus and cellular debris
released from the wound. The debris disappeared within a
few minutes, and the long posterior cilia resumed their beat.
Treatment with seawater containing 10 and 30 mM KC1 had
no effect on the long cilia, but treatment with seawater
containing 50 mA/ KC1 caused the long posterior cilia to
stop beating and the larva to stop swimming for several
seconds.

The beat of the long posterior cilia halted instantly when
the light intensity abruptly increased or decreased. With a
sudden increase in light intensity (2.3 to 19.5 ^M • photons •

19.5 to 57.7 fjiM • photons • m"

57.7 to

100.9 |U.A/ • photons • m

IOQ.9 to 144.2

328

O h larvae

120

S. P. LEYS AND B. M. DEGNAN

,0 2 h larvae

330

180

12h

48 h

300

270

240

a = 193
r = 0.601
z = 10.41***
n = 28

a = 160
r = 0.825
z = 24.50***
n = 36

a = 174
r = 0.854
z = 26.25***
n = 36

24 h

a = 187
r = 0.128
z = 0.52
n = 32

a = 295
r = 0.887
z = 15.73***
n = 20

Figure 3. Circular histograms showing the directions swum by individual larvae in response to diffuse light
shining from zero degrees (see methods for a complete description). The mean angle swum by larvae of an age
cohort is given (a) and is shown with an arrow . A Raleigh's test ( .- 1 determined the degree of dispersion of the
data; highly grouped data [a high value of ;, or a regression (;•) approaching I] are significant (***) at P <
0.001. The number of larvae (») used at each time point is given. The distance swum by larvae is given in
centimeters and displayed as distance from the center of the circle. The great majority of larvae younger than
12 h old swam directly away from the light source, while 12-h-old larvae either swam directly away from the
light or were indifferent. The majority of day-old larvae showed no clear phototaxis. while 2-day-old larvae sank
to the bottom of the test aquarium and rotated in one spot.

photons • m 2 • s '), these cilia immediately straightened
and remained straight for several seconds (Fig. 4). If the
light intensity remained high for longer than 5 s. the ring of
long posterior cilia gradually bent down over the bare
posterior pole; the cilia constituting the ring responded
sequentially, producing a wavelike motion. The ciliary ring
remained bent until the light intensity was gradually re-

duced, whereupon the cilia began to beat freely again, as
though swimming. If the light intensity was suddenly re-
duced by reversing the gradients described above, the ciliary
ring rapidly bent over the bare posterior pole. If the light
remained low for more than 5 s, the cilia slowly straightened
again in a wavelike motion and remained rigidly extended
until the light intensity was gradually increased. The re-

IMIOTOKINESIS IN SPONGE LARVAE

329

Figure 4. The response of the ring of long posterior cilia to a rapid increase or decrease in light intensity
(video microscopy). The time that each video frame was captured is shown in the bottom right-hand corner of
each image in hours, minutes, and seconds. The rate of straightening and bending of the ring of long posterior
cilia shown in all parts of this figure was controlled by the rate that light intensity was increased and decreased.
See methods for details, (a) Frame 1: The cilia are bent over the pigment ring (the dark line indicated by the
arrow) in response to a previous sudden reduction of light intensity. Frames 2-3: Upon an abrupt increase in light
intensity, the long posterior cilia that constitute the ciliary ring (arrowheads) rapidly straighten and remain rigidly
extended (frame 3). Frames 4-6: When the light intensity is suddenly reduced, the ciliary ring (arrowheads)
rapidly bends down over the pigment ring, (b, c) The ciliary response was viewed with a compound microscope.
The long posterior cilia (arrowheads) are bent over the pigment ring when light is abruptly reduced (b). and
straighten when the light intensity is rapidly increased (c). (d. e) The long posterior cilia on the posterior portion
of a bisected larva still respond to an abrupt increase and decrease in light intensity, (d) The cilia are bent over
the pigment ring (arrow) after a previous sudden decrease in light intensity, (e) With an abrupt increase in light
intensity the cilia (arrowhead) straighten. Bar: a, d. e: 100 ^im; b. c: 50 jum.

sponse of the long posterior cilia to changes in light inten-
sity was instantaneous, and the ciliary ring could be made to
straighten and bend in unison as fast as a shutter in front of
the cold light source could be opened and closed. If the
shutter was opened and closed at a slower rate, the cilia
straightened and bent more slowly, but still in unison.
The long posterior cilia on isolated posterior portions of

the larva, or on posterior portions in which the ciliary ring
had been completely bisected, responded in an identical
manner. The response of these cilia became increasingly
slow in larvae older than 24 h, but even a larva that had
settled on its anterior end and was undergoing metamorpho-
sis would continue to move its long posterior cilia in re-
sponse to changes in light intensity.

330

S. P. LEYS AND B. M. DEGNAN

Direction
of Larval
Rotation

Figure 5. Ciliary movement in "hall" (bisected) larvae that were rotating while illuminated from the left (I)
or right (III side. (a. b) Frames from a video recording of the posterior portion of a bisected larva rotating in the
same spot while illuminated from the left (a) and right (b) as diagrammed in (c). As the larva rotates in a
clockwise direction the cilia straighten (arrowheads) when they are closest to the cold light source (CLS). and
bend (arrows) over the pigment ring and bare posterior pole when they are farthest from the light source.
Magnification of (a) and (b) is the same, (d) A scanning electron micrograph of a larva that was fixed while
rotating in illumination from the right shows that the cilia are straight (arrow) on the right and bent (arrowhead)
over the pigmented ring on the left. Bar: a, b, d: 100 fj.ni.

Light shining parallel to the bench top, from either the left
or right side of the microscope stage, onto the posterior end
of a bisected larva that was rotating in one spot, caused the
long posterior cilia closest to the light source to straighten,
and those farthest from the light source to bend (Fig. 5 ). The
bisected larvae completed a full rotation once every 1 .5-2 s;
each long posterior cilium straightened at the instant it
reached the side closest to the light source, and bent at the
instant it reached the side furthest from the light source.
This experiment was readily repeatable with any number of
bisected larvae.

Lan'dl ultrastriicture

Semithin longitudinal sections of the larva revealed three
layers (Fig. 6a, b). Uniciliated columnar epithelial cells
form the outer layer that constitutes all but the anterior and
posterior poles. These cells have two zones: a basal region
with a nucleus (2 /am long) and electron-lucent inclusions
(0.66 ju,m in diameter), and an apical region that is rich in
mitochondria and gives rise to a 20-jum-long cilium (Fig.
6c). Large mucous cells occur throughout the epithelial

layer (Fig. 6c). In the anterior third of the larva, flask-shaped
ciliated cells are regularly interspersed among the columnar
epithelial cells. These cells have a large, centrally located
nucleus, numerous small clear vesicles in the cytoplasm,
and possess a cilium that arises from a deep indentation in
the apical surface of the cell (Fig. 6d).

Underlying the layer of columnar epithelial cells is a
region of cells and collagen that is arranged circumferen-
tially around the larva, perpendicular to the longitudinal
axis, giving the impression of a belt or girdle of cells (Fig.
6b). This sheet of cells is interrupted only at the posterior
end of the larva. These long, narrow cells contain spheru-
lous inclusions (Fig. 6f). The interior of the larva is com-
posed of at least four cell types, which are aligned along the
anterior-posterior axis of the larva and are surrounded by a
thick layer of collagen fibers and a single type of rod-shaped
bacteria that was present in all specimens sectioned (Fig. 6e,
inset). The anterior end of the larva is bare (Fig. 7) and is
formed of large, almost cuboidal cells filled with very small
(0.08-0.25 jLim). clear vesicles (Fig. 7b, c). Although most
of these cells appear to lack cilia, occasional cilia were seen

PHOTOKINES1S IN SPONGE LARVAE

331

Figure 6. The structure and ultrastructure of Reneira sp. larvae (a, b: light microscopy; c-f: electron
microscopy), (a) A longitudinal section through a 2-h-old larva shows that short (20-/xm-long) lateral cilia (SLC)
arise from columnar epithelial cells (CEC) except at the anterior pole (AP) and posterior pole (PP), which are
bare. Long posterior cilia (LPC) arise from pigment-tilled columnar epithelial cells primarily in the anterior
portion of the pigment ring (PRg). Inside the CECs is a layer of subepithelial cells (SECl that run circumfer-
entially around the larva. Cells in the central region (inner cell mass, ICM) are aligned along the anterior-
posterior axis of the larva. Spicules (sp) are evident at the posterior pole. The region in the box is shown in (c).
Bar a. b: 100 /xm. (b) A tangential longitudinal section through the edge of a 2-h-old larva shows that the
subepithelial cells (SEC) are aligned perpendicular to the A-P axis of the larva. The region in the box is shown
in (f). Pigment ring, PRg. (c) Columnar epithelial cells from the region of the larva shown in the box in (a):
mitochondria, m: mucous cell, me: basal body of the cilia, bb; nucleus, m light inclusions, li. Bar: 4 /Mm. (d)
Flask-shaped epithelial cells that occur towards the anterior end of the larva possess a large centrally located
nucleus (n) and a cilium that arises from a deep invagination in the cell (arrowheads). Bar: 2 /j.m. (e) Cells of
the inner cell mass. Spicules. sp; nucleus, n; extracellular rod-shaped bacteria, b (inset). Bar: 5 /^m; inset: 2 /j.m.
(f) Subepithelial cells (SEC) from the region shown in the box in (b) lie in a dense bed of collagen (co). Light
inclusions (li) can be seen in the bases of the columnar epithelial cells (CEC). Bar: 10 /j,m.

332

S. P. LEYS AND B. M. DEGNAN

«»•»• (a\ ~.-Jt~*>* ' ; '

-*?£T- -

Figure 7. Ultrastructure of the anterior pole of Rcncira sp. larvae, (a) A scanning electron micrograph of a
2-h-old larva shows that the anterior pole (AP) is bare of cilia, and that the short lateral cilia (SLC) are preserved
in bands illustrating the metachronal waves entrained by their beating when alive. Bar: 100 /xm. (b) A
transmission electron micrograph of a region near the edge of the anterior pole of a 48-h-old larva. The short
lateral cilia (SLC) mark the end of the columnar epithelial cells at the anterior pole (AP). The anterior-most cells
are generally nonciliated, but the occasional cilium (ci. arrow; inset) can be found deep within the cells. Mucous
cells, me: mitochondria, m; light inclusions, li. Bar: 10 jum. (c) Magnification of the cuboidal cells at the anterior
end of a newly released (zero hour) larva shows numerous clear vesicles (arrowheads), n, nucleus. Bar: 2 ;um.

arising from deep imaginations in the apical surface of the
cuboidal cells (Fig. 7b. inset).

At the posterior end of the larva, large cells containing
electron-dense, mucus-like inclusions protrude slightly
from the bare posterior pole (Fig. 8). At the boundary
between these large posterior cells and the columnar epi-
thelial cells with short cilia lie the pigmented cells bearing
the long cilia (Fig. 8a). Electron-dense pigment vesicles
occur throughout the length of these cells and in the pro-

trusions of their apical surfaces that extend over the base of
the neighboring cells, covering the basal portions of the long
posterior cilia (Fig. 8, 9a). The posterior-most pigment-
tilled cells appear to lack cilia, but otherwise most pig-
mented cells also give rise to a long posterior cilium (Fig. 8,
l>u). No obvious changes in the number or size of pigment
vesicles, or the area they occupy in the cell protrusions,
could be found in thin sections of the posterior of newly
released larvae and 2- to 3-day-old larvae. Further, neither

PHOTOKINESIS IN SPONGE LARVAE

333

r>. •' ;*•'- • . ;.;-jil •-'.

*"•' ' ' ^ l^&

6fr^ V: yi&iferi

-i «i -:;1 . , ' JM1* ^ _-- v. >' »• .- , -. i JjfcfT

3^#*-if ^M^K^B

|p^.; •«.-. i^yf ; ' vf ;
C^^, ^ • ^f]

jT ' ,C

» A .«.•>•» ,->. -j ;i .

Figure 8. Ultrastructure of the posterior pole of Reneira sp. larvae, (a) The posterior pole (PP) is formed in
pan of large mucus-like cells (me) that protrude slightly from the posterior end (arrowheads). The pigment ring
(PRg. dashed line) is formed of columnar epithelial cells with protrusions (arrow) at their apical surface. These
cells contain numerous pigment vesicles (pv) throughout their length and in the apical protrusions. The long
posterior cilia (LPC) arise primarily from the anterior-most of these cells. Magnification of the region in the box
is shown in Figure 9a. Mitochondria, in; basal bodies of the cilia, bb: nucleus, n. Bar: 5 jxm. (b) A scanning
electron micrograph of the posterior pole of a larva that has been fixed while exposed to an abrupt increase in
light intensity to cause the long posterior cilia to straighten. Note also that the mucus-like cells (me) protrude
slightly from the posterior end (arrowhead), and that most long posterior cilia (LPC) are anterior to the
pigment-filled protrusions (arrow ). Magnification of the region in the box is shown in (c). Bar: 100 /urn. (c) A
scanning electron micrograph of the region of the pigment ring shown in the box in (b). Mucus-like cells (me)
protrude from the posterior pole, and pigment-tilled protrusions (arrow ) lie at the base of. and slightly posterior
to, the long posterior cilia (LPC). Bar: 20 /urn.

the number of pigmented cells nor the general histology of
the posterior end in older larvae changed. The structure of
the basal bodies and of the basal portions of the long
posterior cilia did not appear to be different from those of
the short lateral cilia (Fig. 9b, c).

Discussion

This report presents the first demonstration that sudden
changes in light intensity cause an instantaneous response in
the cilia of a sponge larva. This, together with the demon-

334

S. P. LEYS AND B. M. DEGNAN

Figure 9. Details of the pigmented cells and ciliated cells in and near the pigment ring ( transmission electron
microscopy), (al Magnification of the boxed region in Figure 8a showing that at least some pigment-filled
vesicles (pv; arrow) are in protrusions of the same cells that give rise to the long posterior cilia (LPC,
arrowheads). Protrusions of the apical surface of other cells in the ciliated ring are also in view in this section.
Bar: 1 ju.m. (b) Basal bodies (arrows) of the long posterior cilia, (c) Basal bodies (arrow) of the short lateral cilia.
Bar b. c: 0.5 /urn.

stration that light shining at an oblique angle on the long
posterior cilia of a rotating larva causes the cilia nearest the
light to straighten and those furthest from the light to bend
as the larva rotates, implicates the posterior pigment ring
and the band of long cilia in steering the sponge larva away
from bright light.

Sponge In mil "hcluivior"

Given that cellular sponges lack neurons and gap junc-
tions (Pavans de Ceccatty, I974a; Mackie, 1979; Lethias et
til.. 1983: Green and Bergquist, 1982; Woollacott, 1993),
sponge larval behavior is usually explained as being due to
the physical attributes of the larva. For example, many
sponge larvae are reported to swim directly upward after
release from the adult (Bergquist and Sinclair, 1968; Wap-
stra and van Soest, 1987), although there is no evidence that
sponge larvae possess gravity or pressure sensors, such as
statocysts, or a conduction system that would allow them to
translate such messages rapidly into behavior. However.
Warburten ( 1966) suggested that the ability of young larvae
to swim to the top of a tube of seawater each time it was
inverted, whether illuminated from above or below, could
be caused by a differential weighting of the larva at the
posterior end. Indeed, as in many species, spicules develop
at the posterior end of Reneiru sp. larvae after their release
from the adult, and Maldonado et al. (1997) provided ex-

perimental evidence that differential weighting, caused by
the presence of spicules at the posterior end in some larvae,
is correlated with positive geotaxis and rheotaxis.

The beating of cilia in metachronal waves that run ob-
liquely around the long axis of the larva is often thought to
be a result of coordinated behavior (Borojevic. 1969). How-
ever, the entrainment of cilia into metachronal waves in
many animal systems has been demonstrated to be caused
by viscous coupling among cilia (Sleigh, 1974). A very
small number of Reneiru sp. larvae do swim backwards,
suggesting that reversal of the direction of metachronal
waves is possible in Reneira.

Photntaxis and the shadow response

Neither of the above examples of sponge larval behavior
suggests that sensory receptors are involved. For this rea-
son, the role of the long posterior cilia in Reneiru sp. larvae
in responding to changes in light intensity, and thus in
steering the larva away from the light, is intriguing. Al-
though photoreceptors have often been implicated in the
phototaxis of sponge larvae (Kaye and Reiswig, 1991;
Woollacott, 1990. 1993; Maldonado and Young, 1996,
1999), the mechanism by which this might occur has not
been explored by any of these authors.

The different responses of old and young larvae when
swimmine into a shaded region of a test chamber has been

PHOTOKINESIS IN SPONGE LARVAE

335

noted previously. Maldonado et ul. (1997) suggested that
older larvae are more sensitive to light than newly released
larvae, because they continue to swim long after they have
moved into a shaded region. Reneira sp. larvae exhibited u
similar behavior. However, qualitative analysis of the ultra-
structure of the posterior end of larvae of all ages revealed
no changes in the number of pigment vesicles, the area of
the cells occupied by pigment vesicles, or the number of
long posterior cilia. Furthermore, the long posterior cilia
responded to changes in light intensity in larvae of all ages,
including those undergoing metamorphosis, although the
response became more sluggish in older larvae. Another
interpretation is that, upon entering a shaded area, the
younger larvae exhibit a "shadow response" — a photoki-
netic response that changes the level of activity rather than
the direction of movement. The function of the shadow
response has been examined in some detail in ascidian
tadpole larvae (Woodbridge, 1924; Grave. 1944; Young and
Chia, 1985; Svane and Young, 1989) where it appears to
influence the settlement patterns of larvae, and in the hy-
drozoan medusa Polyorchis penicillatus where it is involved
in vertical diurnal migration (Spencer and Arkett, 1984;
Arkett, 1985). The immediate response of the long posterior
cilia of Reneira sp. larvae to a sudden decrease in light —
bending to cover the pigmented ring and posterior pole — is
also suggestive of a shadow response. If larvae exhibited
this response when entering a region of greatly diminished
light (such as under a rock on the reef flat), the larva would
stop swimming forward. This suggests that, contrary to the
conclusion drawn by Maldonado et al. (1997). older larvae
are. in fact, less sensitive than younger larvae to changes in
light intensity.

The light receptor

Ciliary or rhabdomeric photoreceptors have been de-
scribed in all invertebrate phyla except Porifera (Eakin,
1968. 1972; Burr. 1984). Both Tuzet (1973) and Amano and
Hori (1992) have suggested that the cruciform cells in
developing amphiblastulae, the larvae of calcareous
sponges, are photoreceptive, but no studies have confirmed
this function in larval behavior. The morphology of photo-
receptors in basal metazoan groups is unstudied recently,
but the work of Eakin (1968, 1972) suggests that the sim-
plest photoreceptors, known from the Cnidaria, are mono-
ciliated cells surrounded by cells containing pigment vesi-
cles. The pigment cells in Reneira sp. that give rise to the
long posterior cilia are similar in structure to the simple
photoreceptors described in the hydromedusan Leuckartiara
octona (Singla, 19741 and to sensory cells that may be
photoreceptors in the planulae of Hydractinia echinata
(Thomas et al., 1987: Weis et al., 1985) and Phuiluliiim
(C/ytia) gregarium (Thomas et al., 1987).

The surface of the pigment cells protrudes out over the

surrounding epithelium forming a dark ring on the posterior
side of the long cilia. This band of pigment would effec-
tively block light coming from across the bare posterior pole
from reaching the basal portion of the cilium (Figs. 8, 9). It
appears that although the posterior-most pigment-contain-
ing cells may lack cilia, most cells possess both pigment and
a long cilium. Although the location of the photoreceptor is
currently unknown (future work using microspectrophotom-
etry to determine its location being planned), the cilium is
probably both the receptor and effector, as in the well-
studied green unicell Eitglena (Eakin, 1972; Naitoh and
Eckert, 1974; Neuman and Hertel, 1994). The effect of
increased external potassium ion concentration in causing
temporary arrest of the long cilia in Reneira sp. larvae
suggests that depolarization of the membrane potential, and
possible influx of calcium into the cilium. is the mechanism
behind the shadow response of sponge larvae. The phenom-
enon of reversal or inhibition of ciliary beating due to
calcium ion influx resulting from membrane depolarization,
is well known in protists (reviewed by Naitoh and Eckeit,
1974; Eckert et al.. 1976), ctenophores, anthozoans, bivalve
gills, echinoderm pleutei, and pelagic tunicates (reviewed in
Aiello, 1974).

As indicated earlier, unlike planulae, parenchymellae
lack neurons or gap junctions that would allow coordi-
nation of signals between the cells with long cilia. In-
stead, each posterior ciliated cell probably responds in-
dependently to changes in light intensity. On the basis of
the overt responses of the long posterior cilia to abrupt
changes in light intensity, and the asymmetric response
of the long posterior cilia to light shining on the cilia
from the side, as shown in Figure 5, we hypothesize that
the larva steers by the subtle photokinetic responses of
each ciliated cell to the light, as diagrammed in Figure
10. As the larva rotates through the water, the base of
those cilia on the side opposite the direction of the light
would be shadowed by pigment, thus triggering a shadow
response, which would cause those cilia to bend and
cover the pigmented ring (Fig. 10 B arrowhead). Again,
as the larva rotates, cilia whose bases are exposed to light
would straighten and beat (Fig. 10 B arrow), thus steering
the larva away from the light. In this manner, no coor-
dination between cells is required to steer the larva.
Rather, a cumulative effect is achieved by the slightly
different angle at which each cilium is exposed to
or shaded from light. Phototaxis in Euglena is thought to
be based similarly on the shading of its photoreceptor
(Doughty, 1993). However, as steering in Euglena has
also been shown to depend largely upon polarized light
(Creutz and Diehn. 1976; Hiider. 1993), such mecha-
nisms of receiving light cues should also be considered in
further investigations of the photoreceptor in sponge
larvae.

336

S. P. LEYS AND B. M. DEGNAN

Figure 10. Diagram describing the suggested mechanism by which the
pigment ring and long posterior cilia allow Reneira sp. larvae to steer away
from a light source. (A) As the larva rotates, light from one side of the larva
impinges on the base of the cilia closest to the light, but is blocked by the
pigment ring from the cilia furthest away from the light. (B) Cilia exposed
to the light (arrow) straighten or beat rapidly, depending on the extent of
their exposure; those hidden from the light by pigment (arrowhead) un-
dergo a shadow response and bend over the pigment ring. (C) The indi-
vidual response of each cilium to light as the larva rotates causes a graded
response taking the larva away from the source of light.

Coordination of behavior and cellular differentiation in
sponge larvae

Cellular differentiation is integral to the behavior of Rc-
neira sp. larvae. Five regions of the larva are distinctly
differentiated (Fig. 1 1 ). The outer ciliated columnar epithe-
lial layer of the larva is separated from the cells in the
central region by a sheath or band of circumferential cells.
A radial or circumferential sheath has been described in
many parenchymella larvae as a subepithelial cell layer
(Meewis, 1941; Brien, 1973; Woollacott, 1993). Although it
has been suggested that the cells in this layer have a secre-
tory function (Meewis, 1941), it is equally possible that, in
light of the paucity of cell-cell junctions in these larvae, the
circumferential subepithelial cells give structural support to
the larva during release from the parent and during swim-
ming. The cells of the anterior pole are differentiated in
Reneira sp. larvae as well. Both the monociliated ciliated
flask-shaped cells that occur towards the anterior end of the
larva and the large cuboidal cells at the anterior pole have
numerous small clear vesicles and may therefore have a
secretory function. However, the presence of a cilium aris-
ing from a deep invagination in both cell types is also
reminiscent of some sensory cells in gastropod larvae (e.g.,
Kempf et «/., 1997). This anterior region attaches to the
substratum at settlement in Reneira sp. larvae. Finally,
although the function of the large cells at the posterior pole
remains unclear, the pigmented epithelial cells from which
arise the long posterior cilia are clearly differentiated to
steer the larva away from light.

The resulting picture of the sponge larva is not one
typically conjured up of a parazoan, an "almost metazoan."

The Reneira sp. larva is an ensemble of differentiated and
pluripotential cells arranged in stereotypic patterns along
both central-lateral and anterior-posterior axes (Fig. 11).
The spatial arrangement of differentiated cell types in the
larva, with their specific functions and behaviors, plays a
central role in guiding the larva to a suitable settlement
location. Clearly this grade of multicellular organization is
built by the functioning of multiple transcriptional networks
during embryogenesis and larval development. Although a
variety of regulatory genes are known to exist in sponge
genomes (e.g., Degnan et ai, 1993, 1995; Seimiya et ai,
1994; Coutinho et al, 1994; Kruse et at., 1994; Hoshiyama
et til.. 1998), and even though they may be locally expressed
in the larva, it is unclear whether conserved genes involved
in bilaterian development are operating in a similar manner
in sponges. Analysis of the sponge larva and its embryo-
genesis may enable the identification of developmental
genes and processes that are shared among all metazoans,
helping us to understand the earliest steps in animal evolu-
tion.

Acknowledgments

We thank the director and staff at Heron Island Research
Station for use of facilities for portions of this study. Dr. J.
Hooper and Mr. J. Kennedy for identification of the sponge,
Dr. W. Dennison for use of his LI-COR underwater quan-
tum sensor, and the Great Barrier Reef Marine Park Au-
thority for permission to conduct research on Heron Island
Reef. This research was supported by an Australian Re-
search Council grant to BMD, and a Natural Sciences and
Engineering Research Council (NSERC Canada) Postdoc-
toral Fellowship to SPL.

CEC

SEC Co

LPC

MC SLC

ICM

Figure 11. Schematic diagram of cellular differentiation in Reneira
larvae. PP, posterior pole; AP anterior pole; MC, mucous cell; PRg.
pigment ring; LPC, long posterior cilia; SLC, short lateral cilia; CEC.
columnar epithelial cells; SEC, subepithehal cells; ICM, inner cell mass;
Co, collagen; Sp, spicules.

PHOTOKINKSIS IN SPONGE LARVAE

337

Literature Cited

Aiello. K. 1974. Control of ciliary activity in Metazoa. Pp. 353-376 in

Cilia ami Flagella. M. A. Sleigh, ed. Academic Press. London.
Alberts. B., D. Bray, J. Lewis, M. Raff, K. Roberts, and J. D. Watson.
1989. ,!/,./<•( ular Biology of I he Cell. Garland Publishing, New York.
Aninno. S., and I. Hori. 1992. Metamorphosis of calcareous sponges I.
Infrastructure of free-swimming larvae. Invertebr. Reprod. Dev. 21:
81-90.

Arkett, S. A. 1985. The shadow response of a hydromedusan (Poly-
orchis penicillatus): behavioral mechanisms controlling diel and onto-
genetic vertical migration. Biol. Bull. 169: 297-312.
Bergquist. P. R., and M. E. Sinclair. 1968. The morphology and
behavior of larvae of some intertidal sponges. N. Z. J. Mm: Fre\h\\:
Res. 2: 426-437.

Bergquist, P. R., M. E. Sinclair, and J. J. Hogg. 1970. Adaptation to
intertidal existence: reproductive cycles and larval behavior in Demo-
spongiae. Zoo/. Soc. Loud. 25: 247-27 1 .

Borojevic, R. 1969. Etude due developpement et de la differenciation
cellulaire d'eponges calcaires calcineennes (generes Clathrina et As-
candra\. Ann. Embryo!. Morphog. 2: 15-36.
Brewer, R. H. 1976. Larval settling behavior in Cyuncu capillata (Cni-

daria: Scyphozoa). Biol. Bull. 150: 183-199.
Brien, P. 1973. Les Demosponges. Pp. 133-461 in Traite de Zoologie.

P.-P. Grasse. ed. Mason Cie. Paris.

Burr, A. H. 1984. Evolution of eyes and photoreceptor organelles in the
lower phyla. Pp. 131-178 in Photoreceptioi: and Vision in Invertc-
hmtes, M. A. Ali, ed. Plenum Press, New York.

Chia, F.-S., and R. Koss. 1979. Fine structural studies of the nervous
system and the apical organ in the planula larva of the sea anemone
Anthop/eiira elegantissima. J. Morphol. 160: 275-298.
Coutinho, C. C., J. Seack, G. Van de Vyver, R. Borojevic, and W. E. G.
Mueller. 1994. Origin of the metazoan bodyplan: characterization
and functional testing of the promoter of the homeobox gene EmH-3
from the freshwater sponge Ephydatia muelleri in Mouse 3T3 cells.
Biol Chem. 379: 1243-1251.
Creutz, C., and B. Diehn. 1976. Motor responses to polarized light and

gravity sensing in Euglena gracilis. J. Protozoa!. 23: 552.
de \Veerdt, W. H. 1986. A systematic revision of the North-Eastern
Atlantic shallow-water Haplosclerida (Porifera. Demospongiae). Part
II. Chalinidae. Beaufortia 36: 81-165.

Degnan. B. M., S. M. Degnan, T. Naganuma, and D. E. Morse. 1993.
The els multigene family is conserved throughout the Metazoa. Nucleic
Acids Res. 21: 3479-3484
Degnan, B. M., S. M. Degnan, A. Giusti, and D. E. Morse. 1995. A

hox/hom homeobox gene in sponges. Gene 155: 175-177.
Doughty, M. J. 1993. Step-up photophobic response of Euglena gracilis

at different irradiances. Ada Protozoa!. 32: 73-77.
Eakin, R. M. 1968. Evolution of photoreceptors. Pp. 194-242 in Evo-
lutionary Biology. T. Dobzhansky. M. K. Hecht, and W. C. Steere. eds.
Appleton-Century-Crofts. New York.

Eakin. R. M. 1972. Structure of invertebrate photoreceptors. Pp. 625-
684 in The Photochemistry of Vision, }. A. Dartnall, ed. Springer.
Berlin.

Eckert. R., Y. Naitoh, and H. Machemer. 1976. Calcium in the bioel-
ectric and motor functions of Paramecium. Pp. 233-256 in Calcium in
Biological Systems. C. 1. Duncan, ed. Cambridge University Press.
London.

Evans, C. W. 1977. The ultrastructure of larvae from the marine sponge
Halicondria moorei Bergquist (Porifera, Demospongiae). Call. Biol.
Mar. 18: 427-433.

Fell, F. E. 1997. Poriferans. the sponges. Pp. 39-54 in Embryology.
Constructing the Organism. S. F. Gilbert and A. M. Raumo. eds.
Sinauer Associates. Sunderland. MA.

Forward, R. B. J., and J. D. J. Costlow. 1974. lh« ontogeny of

phototaxis by the crab Rhitliiopcinopeus Inirrisii. Mar. Biol. 26: 27-33.

Grave, C. 1926. Mogida ciirina (Alder and Hancock!. Activities and

structure of the free-swimming larva. 7. Morphol. Ph\siol 42: 451-

471.

Grave, C. 1944. The larva of Styela (Cynthia) partita: structure, activities

and duration of life. 7. Morphol. 75: 173-191.

Green, C. R.. and P. R. Bergquist. 1982. Phylogenetic relationships

within the invertebrata in relation to the structure of septate junctions

and the development of occluding junctional types. 7. Cell Sci. 53:

27(1-305.

Hader, D.-P. 1993. Simulation of phototaxis in the flagellate Eiiglenu

gracilis. 7. Biol. Phys. 19: 95-108.

Hadfield, M. G., E. A. Meleshkevitch, and D. Y. Boudko. 200(1. The
apical sensory organ of a gastropod veliger is a receptor for settlement
cues. Biol. Bull. 198: 67-76.

Harrison, F. W., and L. De Vos. 1991. Porifera. Pp. 29-89 in Micro-
scopic Anatomy of Invertebrates. Volume 2. Placo-oa. Porifera. Cni-
daria, and Ctenophora. F. W. Harrison and J. A. Westfall, eds. Wiley-
Liss. New York.

Hoshiyama, D., H. Suga, N. Iwabe, M. Koyanagi, N. Nikoh, K. Kuma,
F. Matsuda, T. Honjo, and T. Miyata. 1998. Sponge Pax cDNA
related to Pax 2-5-8 and ancient gene duplications in the Pa.\ family. 7.
Mol. Evol 47: 640-648.

Jaeckle, W. B. 1995. Transport and metabolism of alanine and palmitic
acid by field-collected larvae of Tedania ignis (Porifera, Demo-
spongiae): estimated consequences of limited label translocation. Biol.
Bull. 189: 159-167.

Kaye, H. R., and H. M. Reiswig. 1991. Sexual reproduction in four
Caribbean commercial sponges. III. Larval behavior, settlement and
metamorphosis. Invertebr. Reprod. Dev. 19: 25-35.
Kempf, S. C., L. R. Page, and A. Pires. 1997. Development of seroto-
nin-like immunoreactivity in the embryos and larvae of nudibranch
mollusks with emphasis on the structure and possible function of the
apical sensory organ. 7. Comp. Neiirol. 386: 507-528.
Kruse, M., A. Mikoc, H. Cetkovic, V. Gamulin, B. Rinkevich, I. M.
Mueller, and W. E. G. Mueller. 1994. Molecular evidence for the
presence of a developmental gene in the lowest animals: identification
of a homeobox-like gene in the marine sponge Geodia cvdonium.
Mech. Ageing Dev. 77: 43-54.

Lawn, I. D. 1982. Porifera. Pp. 49-72 in Electrical Conduction ami
Behavior in 'Simple' Invertebrates. G. A. B. Shelton, ed. Clarendon
Press. Oxford.
Lawn, I. D., G. O. Mackie, and G. Silver. 1981. Conduction system in

a sponge. Science 211: 1169-1171.

Lethias, C., R. Garrone, and M. Mazzorana. 1983. Fine structure of
sponge cell membranes: comparative study with freeze-fracture and
conventional thin section methods. Tissue Cell 15: 523-535.
Leys, S. P. 1995. Cytoskeletal architecture and organelle transport in
giant syncytia formed by fusion of hexactinellid sponge tissues. Biol.
Bull. 188: 241-254.
Leys, S. P., and G. O. Mackie. 1997. Electrical recording from a glass

sponge. Nutiire 387: 29-31.
Leys, S. P., and H. M. Reiswig. 1998. Nutrient transport pathways in the

neotropical sponge Ap/ysina. Biol. Bull. 195: 30-42.
Leys, S. P., G. O. Mackie. and R. W. Meech. 1999. Impulse conduction

in a sponge. 7. E.\p. Biol. 202: 1 139-1 150.
Loewenstein, W. R. 1967. On the genesis of cellular communication.

Dev. Biol. 15: 503-520.
Mackie, G. O. 1979. Is there a conduction system in sponges'1 (, >//,;,/

Int. Cent. Nail. Rech. Sci. 291: 145-151.

Mackie, G. O., I. D. Lawn, and M. Pavans de Ceccatty. 1983. Studies
on hexactinellid sponges. II. Excitability, conduction and coordin:

338

S. P. LEYS AND B. M. DEGNAN

of responses in Rliahtlocal\/>ti<!, ilawstmi (Lambe 1873). Pliilos. Trans.

R. Soc. Loud. B 301: 40I-4IS.
Maldonado, M., and C. M. Young. 19%. Effects of physical factors on

laruil behavior, settlement and recruitment of four tropical demo-
sponges. Mar. Ecu!. Prog. Ser. 138: 169-180.
Maldonado, M., and C. M. Young. 1999. Effects of the duration of

larval life on postlarval stages of the demosponge Sigmadocia caer-

n/ea. J. Exp. Mm: Biol. Ecol. 232: 9-21.
Maldonado, M., S. B. George, C. M. Young, and I. Vaquerizo. 1997.

Depth regulation in parenchymella larvae of a demosponge: relative

roles of skeletogenesis. biochemical changes and behavior. Mar. Ecol.

Prog. Ser. 148: 115-1 24.
Martin, V. J., and F.-S. Chia. 1982. Fine structure of a scyphozoan

planula, Cassiopeia xamachana. Biol. Bull. 163: 320-328.
McNair, G. T. 1923. Motor reactions of the fresh-water sponge Ephy-

datia fiuviatilis. Bml. Hull. 44: 153-166.
Meewis, H. 1941. L'embryogenese des eponges siliceuses. Ann. Soc. R.

Zool. Belg. 72: 126-149.
Miller. S. E., and M. G. Hadtield. 1986. Ontogeny of phototaxis and

metamorphic competence in larvae of the nudibranch Pheslilla sibogae

Bergh (Gastropoda: Opisthobranchial. J. Exp. Mar. Biol. Ecol. 97:

95-112.
Murphy. B. F.. and M. G. Hadfield. 1997. Chemoreception in the

nudibranch gastropod Phestilla sibogae. Comp. Biochem. Physiol. A

Comp. Physiol. 118: 727-735.
Naitoh. Y., and R. Eckert. 1974. The control of ciliary activity in

Protozoa. Pp. 305-352 in Cilia ami Flagel/a. M. A. Sleigh, ed. Aca-
demic Press. New York.
Neumann, R., and R. Hertel. 1994. Purification and characterization of

a riboflavin-binding protein from flagella of Eiiglena gracilis. Photo-

chfin. Phorohiol. 60: 76-83.
Pavans de Ceccatty, M. 1969. Les systemes des activites motrices,

spontanees et provoquees des Eponges. C. R. Acad. Sci. Ser. III Sci.

Vie. 269: 596-599.

Pavans de Ceccatty, M. 1974a. Coordination in sponges. The founda-
tions of integration. Am. Zool. 14: 895—903.
Pavans de Ceccatty, M. 1974b. The origin of the integrative systems: a

change in view derived from research on coelentrates and sponges.

Perspect. Biol. Med. 17: 379-390.
Pavans de Ceccatty, M. 1976. Cellular movements and pathways of

coordination in the sponges. Bull. Soc. Zool. Ft: 100: 154.
Pawlik, J. R. 1992. Chemical ecology of the settlement of benthic

marine invertebrates. Oceanogi: Mar. Biol. Aniui. Rev. 30: 273-335.
Reiswig, H. M. 1971. In situ pumping activities of tropical demo-

spongiae. Mar. Biol. 9: 38-50.
Richardson, K. C., L. Jarett, and E. H. Finke. 1960. Embedding in

epoxy resins for ultrathin sectioning in electron microscopy. Stain

Technol. 35: 313-323.
Ryland, J. S. 1960. Experiments on the influence of light on the behavior

of polyzoan larvae. J. Exp. Biol. 37: 783-800.
Seimiya. M., H. Ishiguro. K. Miura, Y. Watanahe, and Y. Kurosawa.

1994. Homeobox-containing genes in the most primitive metazoa. the

sponges. Em: J. Biochem. 221: 219-225.
Simpson, T. L. 1984. The Cell Biology of Sponges. Springer-Verlag.

New York.

Singla. C. I.. 1974. Ocelli of hydromedusae. Cell Tissue Res. 149:

413-429.
Sleigh, M. A. 1974. Metachronism of cilia of Metazoa. Pp. 287-304 in

Cilia ami Flagcllti, M. A. Sleigh, ed. Academic Press. New York.
Spencer, A. N.. and S. A. Arkett. 1984. Radial symmetry and the
organization of central neurones in a hydrozoan jellyfish. J. Exp. Biol.

110: 69-90.
Svane, I., and C. M. Young. 1989. The ecology and behavior of ascidian

larvae. Oceanogi: Mar. Biol. Annu. Rev. 27: 45-90.
Thomas, M. B., and N. C. Edwards. 1991. Cnidaria: Hydrozoa. Pp.

91-183 in Microscopic Anatomy of Invertebrates. F. W. Harrison and

J. A. Wesifall, eds. Wiley-Liss. New York.
Thomas, M. B., G. Freeman, and V. J. Martin. 1987. The embryonic

origin of neurosensory cells and the role of nerve cells in metamor-
phosis in P/iialiilinin grcgariiini (Cnidaria, Hydrozoa). Invenebr. Re-
prod. I)c\: 11: 265-287.
Thorson, G. 1964. Light as an ecological factor in the dispersal and

settlement of larvae of marine bottom invertebrates. Ophelia 1: 167-

208.
Tuzet, O. 1973. Eponges calcaires. Pp. 27-132 in Traite de Zoologie.

P.-P. Grasse. ed. Mason Cie. Paris.
Vacelet. J. 1966. Les cellules contractiles de 1'eponge cornee Verongia

carernicola Vacelet. C R. Acini. Sc. Ser. III. Sci. Vie 263: 1330-1332.
Wapstra, M., and R. W. M. van Soest. 1987. Sexual reproduction,

larval morphology and behavior in demosponges from the southwest of

the Netherlands. Pp. 281-307 in Taxo/ioim of Porifera. i. Vacelet and

N. Boury-Esnault, eds. Springer-Verlag. Berlin.
Warhurton, F. E. 1966. The behavior of sponge larvae. Ecology 47:

672-674.
Weis, V. M., D. R. Keene, and L. W. Buss. 1985. Biology of hydrac-

tiniid hydroids. 4. Ultrastructure of the planula of Hyilraclinia cchinata.

Biol. Hull. 168: 403-4 IN.
Wendt, D. E., and R. M. Woollacott. 1999. Ontogenies of phototactic

behavior and metamorphic competence in larvae of three species of

Bugit/a (Bryozoa). Invertebr. Biol. 118: 75-84.
Woodbridge, H. 1924. Botrvlliis sc/ilosseri (Pallas): the behavior of the

larva with special reference to the habitat. Biol. Bull. 47: 223-230.
Woollacott, R. M. 1990. Structure and swimming behavior of the larva

of Halichom/ria melanadocia (Porifera: Demospongiae). J. Morphol.

205: 135-145.
Woollacoll. R. M. 1993. Structure and swimming behavior of the larva

of Haliclona inhifcra (Porifera: Demospongiae). J. Morphol. 218:

301-321.
Woollacott, R. M., and M. G. Hadfield. 1989. Larva of the sponge

Di'iulrilla cactus (Demospongiae: Dendroceratida). Trans. Am. Mi-

crosc. Soc. 108: 410-413.

Woollacott, R. M., and M. G. Hadfield. 1996. Induction of metamor-
phosis in larvae of a sponge. Invertebr. Biol. 115: 257-262.
Young, C. M., and F.-S. Chia. 1982. Ontogeny of phototaxis during

laival development of the sedentary polychaete, Serpnla venniciilaris

(L.). Biol. Bull. 162: 457-468.
Young, C. M., and F.-S. Chia. 1985. An experimental test of shadow

response function in ascidian tadpoles. J. Exp. Mar. Biol. Ecol. 85:

165-175.
Zar, J. H. 1984. tSiostutistical Analysis. Prentice Hall, Englewood Cliffs.

NJ.

Reference: Bid. Hull. 201: 339-347. (December 2001 >

Isolation and Properties of the Luciferase Stored
in the Ovary of the Scyphozoan Medusa

Periphylla periphylla

OS AMU SHIMOMURA1 *. PER R. FLOOD2. SATOSHI INOUYE3,
BRUCE BRYAN4. AND AKEMI SHIMOMURA1

lMarine Biological Laboratory, Woods Hole, Massachusetts 02543: Bathybiologica A.S.. N-5081

Bergen, Nonvay; ^Yokohama Research Center. Chisso Corporation, 5-1 Okawa, Kanazawa-kit.

Yokohama 236. Japan; and 4Prolume Ltd., 1085 William Pitt Way, Pittsburgh. Pennsylvania 15238

Abstract. Bioluminescence of the medusa Periphylla is
based on the oxidation of coelenterazine catalyzed by lucif-
erase. Periphylla has two types of luciferase: the soluble
form luciferase L, which causes the exumbrellar biolumi-
nescence display of the medusa, and the insoluble aggre-
gated form, which is stored as paniculate material in the
ovary, in an amount over 100 times that of luciferase L. The
eggs are especially rich in the insoluble luciferase, which
drastically decreases upon fertilization. The insoluble form
could be solubilized by 2-mercaptoethanol. yielding a mix-
ture of luciferase oligomers with molecular masses in mul-
tiples of approximately 20 kDa. Those having the molecular
masses of 20 kDa, 40 kDa. and 80 kDa were isolated and
designated, respectively, as luciferase A, luciferase B. and
luciferase C. The luminescence activities of Periphylla lu-
ciferases A, B. and C were 1.2—4.1 X 101(1 photon/mg • s,
significantly higher than any coelenterazine luciferase
known, and the quantum yields of coelenterazine catalyzed
by these luciferases (about 0.30 at 24 °C) are comparable to
that catalyzed by Oplophorus luciferase (0.34 at 22 °C),
which has been considered the most efficient coelenterazine
luciferase until now. Luciferase L (32 kDa) could also be
split by 2-mercaptoethanol into luciferase A and an acces-
sory protein (approx. 12 kDal. as yet uncharacterized. Lu-
ciferases A, B. and C are highly resistant to inactivation:
their luminescence activities are only slightly diminished at
pH 1 and pH 1 1 and are enhanced in the presence of 1 ~2 M
guanidine hydrochloride: but they are less stable to heating
than luciferase L. which is practically unaffected by boiling.

Received 30 April 2001; accepted 1 October 2001.

* To whom correspondence should be addressed: E-mail: shimomurCs'mbledu

Introduction

The bioluminescent deep-sea medusa Periphylla pe-
riphvlla is widely distributed in the oceans. It is especially
abundant in certain Norwegian fjords, where large speci-
mens are commonly found — up to 20 cm in diameter. 25
cm in height, and weighing over 600 g (Fossa, 1992).
Unlike hydrozoan medusas, which contain calcium-sensi-
tive photoproteins. the glow of the scyphozoan Periphylla
periphvlla is due to a luciferin-luciferase reaction involving
coelenterazine (a luciferin) and Periphylla luciferase. The
luciferase of Periphylla occurs as a soluble enzyme and as
an insoluble particulate matter (Shimomura and Flood,
1998). The soluble form is found mainly in the exumbrellar
epithelia of the dome and lappets and in the dome mesoglea.
The particulate matter (about 0.5-1 /urn in size) occurs
abundantly in maturing ovarian eggs, and the total amount
of luciferase activity in this form per medusa is far greater
than the activity of the soluble form. The soluble form,
extracted and purified from the lappets, is an unusually
heat-stable luciferase. called luciferase L (32 kDa). The
highly active particulate matter obtained from the ovary was
partially solubilized and extracted with a buffer containing
2 M guanidine hydrochloride. then purified. The enzyme
obtained was highly resistant to various denaturants. and it
was designated luciferase O (75 kDa).

We recently found that the treatment of the ovarian
particulate with 2-mercaptoethanol solubilizes the luciferase
and markedly increases its activity. Moreover, we also
found that the solubilized luciferase was a mixture of var-
ious molecular species having different molecular weights.
These and other lines of evidence suggested that the partic-

339

340

O. SHIMOMURA ET AL

A;i example of the purification nf lucij

Table 1

. B. and C from 40 g of ovaries, slum-ing the progress of purification

Step

Method

Total activity (10" LU)

Specific activity (10" L\J/A2m lcm)

2-Mercaptoethanol treatment and extraction

Not measured

Ether-650 hydrophobic interaction chromatography

Not measured

Superdex 200 gel filtration

4.3

0.3

4.6

0.7

3.2

0.3

SP-Sepharose anion exchange chromatography

2.6

3.8

2.5

SP-650 anion exchange chromatography

1.9

2.8

2.1

Superdex 200 gel filtration

1.2

1.6

1.2

1 LU (light unit) corresponds to 5.5 x 10s photons/s.

ulate material could be the storage form of aggregated
luciferase. The present work was undertaken to clarify the
chemical nature of luciferase in those aggregates. We have
isolated three new molecular species of Periphylla lucit-
erase from the ovaries of Periphylla, and named them
luciferases A, B, and C, respectively. The methods of ex-
traction and purification, and the properties of these lucif-
erases are described and discussed in this paper; we also
present a new method of preparing luciferase L.

Materials and Methods

Measurement of luminescence activity of luciferase

Luminescence intensity and total light were measured
with an integrating photometer model 8020 (Pelagic Elec-
tronics, Falmouth. Massachusetts) calibrated with the Cyp-
riilimi bioluminescence reaction (Shimomura and Johnson,
1970). In the assay of luciferase activity, 3 ml of 1 M
NaCVO.05% BSA/20 mM Tris-HCl (pH 7.8) containing 10
jal of 0.1 mM methanolic coelenterazine was added to a
luciferase sample (2-50 p.\) at 24 °C (this is the standard
assay condition), and luminescence intensity was measured.
Because coelenterazine was present in large excess and
luciferase was stable, the rate of the reaction (and thus light
emission) was essentially zero-order. One light unit (LU) of
luciferase activity on this instrument corresponded to a
luminescence intensity of 5.5 X 108 photons/s. The specific
activity of a luciferase sample is defined as "luciferase
activity in LU or photons/s, divided by A2SO nm , cm" except
as noted.

Extraction and purification of luciferase from ovaries

The following is only a general plan of the procedure of
purification, because experiments were often modified due

to unavoidable variations in the starting materials. At each
step of column chromatography, active side fractions were
re-chromatographed, and any good fractions recovered were
added to the main fractions. In the purification of luciferases
A, B. and C (Steps 3-5, below), any side fractions of a target
luciferase that contained a different molecular species were
combined with a batch of the corresponding luciferase spe-
cies for further purification. The yields of luciferase at each
purification step are shown in Table 1 .

Step 1: The specimens of Periphylla were collected on
board R/V Hakon Mosby by vertical plankton-net hauls and
midwater trawling in Lurefjolden. western Norway. The
ovaries and other organs were excised from live specimens.
and stored at -75 °C. Frozen ovary (40 g) was thawed and
homogenized with a Bamix mixer Ml 22 (Clark National
Products, San Dimas, California) in 80 ml of 10 mM phos-
phate buffer (pH 6.6). The homogenate was then centri-
fuged at 1 2,000 X g for 1 0 mm at 0 °C. The supernatant was
discarded, and the pellets were homogenized in 80 ml of 20
mM acetate buffer (pH 5.4) containing 1 M KC1 and 25 mM
2-mercaptoethanol. This mixture was left standing at 0 °C
for 3 or 4 h. during which the activity of the sample
increased approximately 4-fold. Centrifugation of the mix-
ture gave a clear supernatant (Extract 1 ). The pellets were
mixed with 80 ml of 20 mM acetate buffer (pH 4.8) con-
taining 1 M KC1 and were left standing at 0 °C for 3 or 4 h,
then centrifuged to give Extract 2. This extraction was
repeated two more times in the same manner, except that the
standing time was increased, each time, to 1 day; Extracts 3
and 4 were thus produced.

Step 2: Extracts 1 and 2 were combined, and ammonium
sulfate was added to make 2.4 M. The solution was ad-
sorbed on a column of Toyopearl Ether-650M (Supelco,
Bellefonte, PA; 2.5 cm X 7 cm). The column was washed

PERIPHYLLA LUCIFERASE STORED IN OVARY

341

50 60 70 80 90 100 110

Elution volume (ml)

Figure 1. An example of the third-step gel filtration on Superdex 200
Prep. Elution curves are shown for luminescence activity (solid line) and
the value of A280 nm i Cm (dashed line). A, B, and C are the peaks of
luciferases A, B. and C. respectively. The fractions constituting each of
these peaks were combined for further purification.

with 2.2 M ammonium sulfate/20 mM acetate buffer (pH
4.8) at room temperature, then luciferase was eluted with
1.8 M ammonium sulfate/20 mM acetate buffer (pH 4.8),
and the active fractions were collected. Luciferase fractions
that were eluted with ammonium sulfate concentrations
lower than 1.8 M were not used in this study. Extracts 3 and
4 were chromatographed on the Ether-650M column in the
same manner. All of the active fractions were combined,
made up to 2.4 M ammonium sulfate, and then adsorbed on
a column of Ether-650M ( 1.5 cm X 3.5 cm). The adsorbed
luciferase was eluted with 0.5 M K.Cl/0.01% lauroylcholine
chloride (LCQ/20 mM acetate buffer (pH 4.8), giving about
6 ml of concentrated luciferase solution.

Step 3: Size-exclusion chromatography was carried out
on a column of Superdex 200 Prep (Pharmacia; 1.5 cm X 72
cm) with 1 M KC1/0.01% LCC/20 mM acetate buffer (pH
4.8) as the eluent. On each run. 3 ml of the sample were
injected and the effluent was collected in 2-ml fractions. The
fractions were separated into 3 groups — luciferases A, B
and C (see Fig. 1) — according to their elution volume, and
all the fractions from the same group were combined.

Step 4: Cation-exchange chromatography was carried out
on a column of SP Sepharose High Performance (Pharma-
cia; 1 cm X 6 cm) at room temperature. The eluate from the
third step was diluted with two volumes of 0.01% LCC/20
mM acetate buffer (pH 5.5), then adsorbed onto the column.
After washing the column with 0.5 M KC1/0.01% LCC/20
mM acetate buffer (pH 5.5), elution was done with a linear
gradient of 1.3 M KC1/0.7 M guanidine hydrochloride/
0.01% LCC/20 mM acetate buffer (pH 5.5), that increased
from 0% to 100% in 23 mm.

Step 5: Cation-exchange chromatography was repeated
with Toyopearl SP-650M (Supelco; 1 cm X 6 cm). The
eluate of the fourth step was diluted with three volumes of
0.01% LCC/20 mM acetate buffer (pH 5.5) and was ad-
sorbed onto the column. The elution was done in the same

manner as in the fourth step. Step 5 effectively eliminated
the tailing UV-absorbing impurities that were seen in the
fourth step.

Step 6: Size-exclusion chromatography was performed
on a column of Superdex 200 Prep ( 1 cm X 48.5 cm) in 1
M KC1/20 mM acetate buffer (pH 4.8); 1 ml of sample was
injected in each run.

An improved alternative to Steps I and 2: Ovarian tissue
(6 g) was briefly homogenized with 20 ml of 10 mM
phosphate buffer (pH 6.8), then centrifuged at 20,000 X g
for 10 min, and the supernatant was discarded. The precip-
itate was mixed with 6 ml of 1 M KC1/1 M guanidine
hydrochloride/50 mM acetate buffer (pH 5.4), heated at
80 °C for 1 min, and then centrifuged again. The superna-
tant, which contained 106 LU of luciferase activity and a
large amount of protein, was not used. The precipitate was
mixed with 4 ml of 1 M KC1/0.025% BSA/0.3% 2-mercap-
toethanol/50 mM acetate buffer (pH 5.4), and left standing
at 0 °C for 20 h. Centrifugation of the mixture produced a
clear supernatant with a luciferase activity of 2.9 X 106 LU.
Extraction of the precipitate with 1 ml of 1 M KC1/50 mM
acetate (pH 5.4) gave an additional luciferase activity of
3.6 X 10s LU. This alternative method has three advan-
tages: ( 1 ) the product obtained has a markedly higher purity
than that obtained in Step 2 above; (2) the ratio of lucif-
erases A:B:C can be changed by altering the concentration
of 2-mercaptoethanol and the reaction time, because lucif-
erase C progressively dissociates into B and A; and (3) a
significant activity loss caused by the use of high concen-
trations of ammonium sulfate can be avoided.

A modified method for preparing luciferase L

Only the dome mesoglea (average weight 300 g each),
with the thin pigmented layer on the surface removed, were
used. The lappets contained greater concentrations of lucif-
erase L, but they were not used because the surface pigment,
which drastically decreases the yield of luciferase, is diffi-
cult to remove. Cleaned dome mesoglea (500 g) were ho-
mogenized in 500 ml of water with 0.3 g of BSA. The
homogenate was mixed with 3 teaspoonfuls of Whatman
CDR (cell debris remover) and filtered on a Biichner funnel.
The filtrate was diluted with two volumes of 10 mM acetate
buffer (pH 4.8) and filtered through a column of SP-650M
(2.5 cm X 8 cm). Luciferase adsorbed at the top of the
column was eluted with 0.5 M NaCl/0.025% BSA/10 mM
acetate buffer (pH 4.8), giving approximately 50 ml of
luciferase solution (40.000 LU; this could be safely stored at
-70 °C, if necessary). The solution was neutralized (pH
7.0) with dibasic sodium phosphate, made up to 2.5 M with
ammonium sulfate, and then adsorbed on an Ether-650M
column, as described in Step 2 above. The column was
washed successively with 15 ml each of 2 M and 1 M
ammonium sulfate made with 10 mM phosphate buffer (pH

342

O. SHIMOMURA ET AL

1. 1 ), and the luciferase L was eluted with 0.5 M ammonium
sulfate/acetate buffer (pH 4.8). The material was further
purified by chromatography on the columns of SP-650M
and Superdex 200, in basically the same manner as reported
previously (Shimomura and Flood, 1998). The final yield of
purified luciferase L from approximately 5 kg of cleaned
domes was 400,000 LU.

Assav of the luciferase in single eggs, embryos, and
juveniles

The specimens were frozen in dry ice on board ship
immediately after collection. A single frozen specimen was
ground thoroughly in a cold aluminum oxide mortar and
pestle with 2 ml of 10 mM phosphate buffer (pH 7.0)
containing 0.05% BSA. Fifty microliters of this ground
suspension was used to measure the total amount of lucif-
erase activity. The rest of the suspension was centrifuged at
20,000 x g for 10 min. and the amount of soluble luciferase
in 50 jul of the supernatant was then determined. After the
supernatant was discarded, the precipitate was mixed with 2
ml of 20 mM acetate buffer (pH 5.5), containing 0.05% each
of BSA and LCC (lauroylcholine chloride) 1 M KC1. and 25
mM 2-mercaptoethanol. This extraction/solubilization with
2-mercaptoethanol continued overnight at 0 °C. The total
amount of luciferase activity in 50 /id of this suspension was
then measured. After this suspension was centrifuged at
20,000 X g for 10 min, the amount of soluble luciferase in
50 jul of the supernatant was assayed.

Results

Purification of the three molecular species of Periphylla
luciferase from ovaiy

The insoluble, aggregated form of luciferase was success-
fully solubilized with 2-mercaptoethanol, presumably by
the splitting of disulfide bonds. However, the purification of
the solubilized luciferase was difficult, mainly because of
certain unusual characteristics of luciferase: its inactivation
by dilution — particularly when high concentrations of am-
monium sulfate were diluted, its irreversible binding to
almost anything, and its expected loss of activity by enzy-
matic degradation in the early stages of purification.

Because the luciferase bound irreversibly to most chro-
matographic adsorbents, these materials were unsuitable for
purification; indeed their use led to large, often complete,
loss of enzyme activity. After extensive tests, a few kinds of
adsorbent were found to be relatively safe under certain
conditions. At first. BSA was used in the purification of
luciferase L to minimize the activity loss, although a protein
additive is clearly undesirable in the purification of a pro-
tein. However, we have recently found that cationic deter-
gents, such as LCC, effectively prevent the inactivation of
luciferase preparations, although the tight binding of the

detergent may cause certain complications by altering the
properties of proteins. Guanidine hydrochloride (1-2 M)
was also highly effective at stabilizing luciferase.

Because of their "sticky" nature, luciferase molecules in
crude and partially purified preparations probably exist in
complexed forms, bound to some impurities in the solution.
Thus, the chromatographic behavior of luciferase in crude
preparations may differ from that of pure luciferase, and the
behavior of luciferase may change as purification
progresses. For example, hydrophobic interaction columns,
such as Ether-650M and Butyl Sepharose (Pharmacia), can
be used with crude luciferase but not for highly purified
luciferase. Similarly, a gel filtration column of Superdex
200 Prep is reliable in most stages of purification, but not
with completely pure luciferase.

The purification of the solubilized luciferases from 40 g
of ovaries involved over 50 column chromatography runs
(summarized in Table 1 ). In Step 5, the elution curve plotted
with luciferase activity and that plotted with A280 nm , cm
value were practically parallel for all three species of lucif-
erase (data not presented), indicating that they were highly
pure. The results of Step 6 show decreases in the specific
activity, possibly due to two combined effects: the loss of
luciferase by adsorption onto the column, and an actual
decrease in specific activity; both decreases were caused by
the omission of LCC from the buffer used. Assuming that a
solution with A280 nm , cm equal to 1.0 contains 1 mg/ml of
luciferase, the yields of purified luciferases at Step 5 (Table
1) are 0.09 mg, 0.12 mg. and 0.07 mg for luciferase A, B,
and C, respectively. When Steps 1 and 2 were replaced by
the alternative method, the specific activities at Step 4
became 2-3 times higher than those at Step 5 in the original
method, although the protein purities remained on compa-
rable levels.

Molecular properties of luciferases A, B, and C

The molecular weights of luciferases A, B, and C were
estimated by gel filtration on the same Superdex 200 Prep
column that was used in Step 6 to purify the luciferases;
0.005% LCC was added to the buffer to minimize adsorp-
tion onto the column. The results (Fig. 2) indicated the
molecular masses of luciferases A, B. and C to be 19 kDa,
40 kDa, and 80 kDa, respectively. After treating the proteins
for 1 day at room temperature with an elution buffer con-
taining 25 mM 2-mercaptoethanol, the main peaks of all the
proteins were found at the same position — that correspond-
ing to 19 kDa. This result suggests that luciferases B and C
are the dimer and tetramer, respectively, of luciferase A, and
that the molecular masses of luciferase A is about 20 kDa.
rather than 19 kDa. During the process of purification, three
other luciferase species, corresponding to 60 kDa. 120 kDa,
and 160 kDa, were observed as relatively minor compo-
nents. These were not purified.

PER1PHYLLA LUCIFERASE STORED IN OVARY

343

100

'<:•

70
60

* 50-

24 26 28 30

Elution volume (ml)

Figure 2. Molecular weight estimation of luciferases A. B, C, and L.
The gel filtration was carried out on a column of Superdex 200 Prep ( 1 X
48.5 cm), in 20 mM acetate buffer (pH 4.8), containing 1 M KCI and 0.01%
lauroylcholine chloride. Calibration standards: aldolase ( 1 ). BSA (2),
ovalbumin (3). carbonic anhydrase (4). myoglobin (5). ribonuclease A (6).

SDS-PAGE (poly aery lamide gel electrophoresis) analy-
sis of luciferases A, B, and C under reducing condition
(with 2-mercaptoethanol) showed only one major band cor-
responding to a molecular mass of 24 kDa (Fig. 3); thus, the
purity of these proteins and the oligomeric nature of lucif-
erases B and C are verified. Although the molecular mass of
the luciferase monomer obtained by gel filtration (19 kDa)
does not match well with that obtained by SDS-PAGE (24
kDa), we chose to use the value of 20 kDa for luciferase A
(and 40 kDa and 80 kDa for luciferases B and C) as a
reasonable approximation, pending the determination of its
precise value in the future. We also note here that luciferase
L (32 kDa) also yielded luciferase A upon treatment with
2-mercaptoethanol (Fig. 3; see Discussion).

The spectral properties of luciferase A, B, and C were
practically identical with those of luciferase L; their absorp-
tion and fluorescence spectra indicate that the luciferases are
simple proteins, without any chromophore that absorbs or
fluoresces in the visible region.

En -\iuutic properties

The enzymatic properties of luciferases A. B, and C are
generally similar to those of luciferase L previously reported
(Shimomura and Flood. 1998), but with some notable differ-
ences. Thus, the luminescence intensity of luciferase A was
highest at 27 °C, and those of luciferases B and C at 30 °C,
whereas the luminescence intensity of luciferase L showed no
maximum, but steadily increased as the temperature was low-
ered to almost 0 °C (Fig. 4). A clear difference was also found
in their heat stabilities (Fig. 5). Luciferase L is extremely stable
to heat, with almost no loss of luminescence activity after

being heated at 95 °C for 2 min. Luciferases A, B, and C are
less stable, showing activity losses of 50% or more under the
same conditions; the loss seems to be greater with the lucif-
erase species of larger size.

The effects of pH on the luminescence of luciferases A, B.
and C were similar to that of luciferase L. showing an optimum
at pH 8.0. All Periplwlla luciferases are highly stable at acidic
and alkaline pHs (Fig. 6). The influence of salt concentration
on the luminescence activities of luciferases A, B, and C
appears essentially the same as that for luciferase L, showing
that the optimum salt concentration is about 1 M (Fig. 7). The
effect of the concentration of coelenterazine on luminescence
intensity is presented in Figure 8, and the Michaelis constants
of luciferases A, B, and C calculated from these data are about
0.2 H.M, which is significantly lower than the value for lucif-
erase L ( 1 . 1 juM).

Like luciferase L. luciferases A, B, and C were strongly
inhibited by Cu:+; but they were not inhibited by thiol
agents. Inhibition of luciferase L by thiol agents reported
previously (Shimomura and Flood, 1998) must be incorrect
because purer preparations showed lesser inhibition; the
inhibition seen earlier could have been caused by the acti-
vation of thiol-activated proteases.

kDa

97.4-
69.0-

55.0-

36.5-
29.0-

20.1-
14.3-

•M

55.0-

36.5-
29.0-

20.1-
14.3-

Figure 3. SDS-PAGE analysis of luciferases A, B. C. and L. The

electrophoresis was carried out under reducing condition for luciferases A.
B, C, and L (left panel, lanes 1, 2. 3. and 4, respectively) and under
nonreducing condition for luciferases B. C, and L (right panel, lanes 1, 2.
and 3, respectively), by the method of Laemmli (1970) using 12% gel; the
protein bands were visualized by silver staining. Approximate amounts of
protein used: luciferases A. B. and C, 0.3-1.2 ;ag; luciferase L. 10 jug.
Marker proteins (not shown): myosin (205 kDa), 0-galactosidase (116
kDa). phosphorylase b (97.4 kDa), BSA (69 kDa), glutamic dehydrogenase
(55 kDa), lactic dehydrogenase (36.5 kDa), carbonic anhydrase (29 kDa).
trypsin inhibitor (20.1 kDa), lysozyme (14.3 kDa). Note that the 15-kDa
band of lane 4 (left panel) corresponds to the 14.5-kDaband of lane 3 (righl
panel), in both color and position; the accessory protein of luciferase 1. f-
shown as the weak 14-kDa band in lane 4, left panel.

344

O. SHIMOMURA ET AL.

350 -

300 -

250 -

c 200 -
§

I 100 -

50 -

0 10 20 30 40 50 60

Temperature (°C)

Figure 4. Effect of temperature on the luminescence intensities of
coelenterazine catalyzed by luciferases A. B. C. and L. The measurements
were done in 20 mM Tris-HCl buffer (pH 7.8), containing 1 M NaCl and
0.05% BSA (the standard buffer)- The luminescence reaction was started
by the addition of 10 /al of 0.1 mM methanolic coelenterazine. The amount
of sample used for measuring each point: luciferase A, 170 LU; luciferase
B, 190 LU: luciferase C, 210 LU; luciferase L, 210 LU.

Luminescence reaction of coelenterazine and its analogs
catalyzed by luciferases A, B, and C

The spectra of the luminescence of coelenterazine cata-
lyzed by luciferase A, B, and C were all identical with that
of luciferase L, showing a peak at 465 nm. The specific
activity (quanta emitted per second, divided by A2SO nm. \ cm)
of the materials obtained in Step 5, Table 1, was 1.21 X
1016 photons/s for luciferase A, 1.32 X 1016 photons/s for
luciferase B, and 1.65 X 1016 photons/s for luciferase C,
under the standard assay conditions. However, significantly
higher specific activities were obtained when the purifica-
tion included the alternative method for Steps 1 and 2: 3.6 X
1016 photons/s and 4.1 X 10lh photons/s for luciferases A
and B, respectively (the yield of luciferase C was low). The
maximum specific activities obtainable with high concen-
trations of coelenterazine (over 2 \iM) should be roughly
twice these values, based on the data of Figure 8 (note that
the coelenterazine concentration in the standard assay is
about 0.3 /u,Af). As a reference to these data, the maximum
specific activity of luciferase L reported previously was 8 X
10'3 photons/s (Shimomura and Flood, 1998). The quantum
yields of coelenterazine in the luminescence reaction cata-
lyzed at 24 °C by luciferases A, B, and C were 0.287, 0.291,
and 0.296, respectively, compared with 0.14 previously
reported for luciferase L.

All known coelenterazine luciferases can catalyze the
luminescent oxidation of various coelenterazine analogs,

causing luminescence in various intensities — from a negli-
gibly low level to a level several times higher than that of
coelenterazine (Inouye and Shimomura, 1997; Nakamura et
al, 1997). Using Periphylla luciferases A, B, and C. none of
more than 20 analogs tested gave a luminescence intensity
higher than that of coelenterazine, and only four analogs
emitted significant levels of luminescence, each giving the
same intensity with the three luciferase oligomers. These
four analogs had a substitution at the 2 or 6 position of the
imidazopyrazinone ring of coelenterazine, and their relative
luminescence intensities, taking coelenterazine as 100%,
were: 2-CH2C6H5, 95%; 2-CH2C6Hn, 21%; 6-C6H4NH2
(p), 20%; 6-C6H4NHCH3 (p), 31%.

Luciferase O

Luciferase O (about 75 kDa) had previously been ob-
tained from the ovary by extraction with a buffer solution
containing 2 M guanidine hydrochloride (Shimomura and
Flood, 1998). On the basis of a chromatographic compari-
son on a Superdex 200 Prep column, this material was found
to be a mixture containing luciferase C (80 kDa) as the main
component; the preparation also contained some luciferase
oligomers of 60 kDa and 120 kDa and impurity proteins.

Discussion

Distribution of luciferase

Periphylla becomes luminescent when coelenterazine is
oxidized in the presence of luciferase in certain tissues of

100 -

80 -

60 -

40 -

20 -

20 40 60 80

Temperature (°C)

100

Figure 5. Stability of luciferases A, B. C, and L at various tempera-
tures in 20 mM Tris-HCl buffer (pH 7.8) containing 1 M NaCl and 0.05%
BSA (solid lines) or 0.01% lauroylcholine chloride (dotted lines). The
buffer (1 ml (containing a luciferase sample was added into a glass test tube
that had been soaked and pre-equilibrated in a water bath of a desired
temperature. After 2 min, the test tube was briefly cooled in cold water, and
the luciferase activity in 10 jul of the sample solution was measured at
24 °C by the standard assay method.

PERIPHYLIA LUCIFERASE STORED IN OVARY

345

200

Figure 6. Effect of pH on the light intensity of the luminescence of
coelenterazine catalyzed by luciferases A, B. and C, and on the stability of the
same luciferases. The effect on light intensity (solid lines) was measured in 50
mM phosphate buffers, pH 4.1-7.25. or 50 mM Tris-HCl buffers. pH 7.1-9.7,
all containing 1 M NaCl, 0.025% BSA. and 10 jil of 0.1 mM methanolic
coelenterazine. For measuring the effect on stability (dotted lines), luciferase
samples were left standing for 30 min at room temperature, in 0. 1 ml of a
solution having various pH values, then luminescence intensity was measured
under the standard condition (pH 7.8). The solutions used were 0.1 M HC1.
0.01 M HC1. 0.01 M acetic acid, 50 mM phosphate, 50 mM Tris-HCl, 0.01 M
NH3, and 0.01 M NaOH. all containing 1 M NaCl and 0.025% BSA. The
amount of luciferases used for measuring each point were luciferase A, 150
LU; luciferases B and C. 170 LU.

this organism. The luciferase occurs in a soluble form
(luciferase L) and also as an insoluble aggregate. The sol-
uble form is responsible for the in vivo bioluminescence of
the animal and is distributed widely, not only in the epithe-
lial photocytes but also in the mesoglea of the large coronal
dome. The insoluble form exists in the paniculate matter
distributed abundantly in the ovary, particularly in the eggs.
The size of the particles, measured by differential filtration,
was larger than 0.2 /urn and smaller than 2 /urn; the actual
size is probably close to the low end of this range according
to previous microscopic observation (Flood et al., 1996).
Like soluble luciferase L, the paniculate matter is highly
active in catalyzing the luminescence of coelenterazine, but
its involvement in the in vivo bioluminescence is uncertain.
The luciferase activity of particulate matter is increased
several times by solubilization using 2-mercaptoethanol,
which yields soluble luciferase oligomers, such as lucif-
erases A, B, and C.

The total luciferase activity existing in one gram of the
dome mesoglea, lappet, and ovary was approximately 100
LU, 1000 LU. and 7 X 105 LU, respectively. Taking ac-
count of the quantity of tissue in each organ in the body,
these figures suggest that the amount of luciferase stored in
the ovary is more than 100 times the total amount of
luciferase L in the whole body of a female medusa. The
facts that the luciferase is complexed in a stepwise fashion

(dimer. tetramer. etc.) and that these oligomers occur in
discrete subcellular particles suggest that the luciferase is
being stored for later use. In the case of the male medusae,
an insoluble aggregated form of luciferase was not found in
the testes, but we are unable to conclude that such a lucif-
erase is absent until all other internal organs have been
tested.

Luciferase in the eggs and during early development

In the eggs, the particulates containing aggregated lucif-
erase are in the cortical layer (Flood et al., 1996). The total
content of luciferase in one egg is extremely large for its
small size ( 1 ju,g or 5 X 10~ ' ' mole/egg: calculated from the
data in Table 2). and the luciferase is mostly the aggregated
form. Unexpectedly, the eggs contained a negligibly small
amount of coelenterazine (1 x 10" l4 mole/egg), but some
coelenterazine may have been spent by the luminescence
reaction that occurs during the preparation of the material.

Unlike most medusae, Periphylla periphylla develops
directly from egg to medusa without an intermediary, sessile
polyp stage (Martinussen et al., 1997; Jarms et al., 1999).
The data of Table 2 suggest that the amount of luciferase in
the eggs decreases drastically upon fertilization, reaching a
minimum at a late embryonic or early juvenile stage (about
3% of the initial amount). Therefore, the biosynthesis of
luciferase must start at a later stage of development, because
large adult specimens contain large amounts of luciferase.
We may see the first sign of such biosynthesis in juveniles
with a dome diameter of 8-10 mm. During these juvenile
stages, we also see the first differentiation of exumbrellar
epithelial photocytes with basically the same organization

150 •

100 -

O Luciferase A
D Luciferase B
A Luciferase C

NaCl concentration (M)

Figure 7. Effect of salt concentration on the luminescence intensity of
coelenterazine catalyzed by luciferases A, B. and C. The measurements
were done in 20 mM Tris-HCl buffer (pH 7.8) containing various concen-
trations of NaCl, 0.05% BSA. and 10 ju.1 of 0.1 mM methanolic coelen
terazine. With NaCl concentration lower than 0.2 M, the intensiiy gradually
decreased by the inactivation of luciferase. For the measure-',, nl of ( h
point. 170 LU of each luciferase was used.

346

O. SHIMOMURA ET AL.

300

5 200 -

JS 100 -

O Luciferase A

D Luciferase B

A Luciferase C

O Luciferase L

0123

Concentration of coelenterazine (u.M)

Figure 8. Effect of coelenterazine concentration on the luminescence
catalyzed by luciferases A. B, C, and L. The measurements were done in
the standard buffer. The amounts of sample used were the same as in
Figure 6 for luciferases A, B, and C. The data for luciferase L were taken
from the previous report (Shimomura and Flood. 1998).

as those found in the adult medusae (Flood, unpubl. obs.).
The decrease of luciferase in the eggs, described above, is
puzzling and intriguing. Why does the egg contain a large
amount of luciferase in the first place? What is the function
or purpose of this luciferase?

A similar phenomenon has been observed in the eggs of
bioluminescent hydrozoan medusas that contain a Ca~
sensive photoprotein, a complex of oxygenated coelentera-
zine and an enzyme. In those eggs, the amount of photo-
protein slowly declines during the development of the
planula larva, and then markedly declines when the planula
undergoes metamorphosis to become a polyp (Freeman and
Ridgway, 1987).

Properties of luciferases A, B, and C

The present results indicate that luciferases B (40 kDa)
and C (80 kDa) are the dimer and tetramer. respectively, of

the luciferase A monomer (20 kDa). The specific lumines-
cence activities of luciferases A, B, and C were in a range
of 1.2—4.1 X 1016 photons/s, showing a tendency to in-
crease slightly as the oligomer size increases. This is the
highest specific activity ever reported for a luciferase whose
substrate is coelenterazine; the highest in the past was that
of the luciferase of the deep-sea shrimp Oploplwms (1.75 X
101"1 photons/s) (Shimomura et ai, 1978).

The specific activity of purified luciferases A. B, and C
can vary in a complex manner, depending upon the method
of purification and the history of handling; and an increase
in the purity is sometimes accompanied by a decrease in the
specific activity. Furthermore, the dilution of a solution
containing high concentrations of ammonium sulfate always
causes a marked decrease in the activity, whereas the addi-
tion of 1-2 M guanidine hydrochloride or 0.0 \7c cationic
detergent (such as LCC or hexadecyltrimethylammonium
bromide) to a luciferase solution often results in an increase
in activity. On the other hand, luciferases A, B, and C are
highly resistant to inactivation by heat, extreme pHs, and
denaturants, such as 4 M guanidine hydrochloride. deter-
gents, and organic solvents. Thus, luciferase A, B, and C are
distinctly different from all previously known coelentera-
zine luciferases that are easily inactivated. The facts noted
here may suggest that Periphylla luciferase has a unique and
unusual tertiary structure, and that the luciferases isolated
by us are mixtures of two or more molecular species having
conformationally different structures (having different ac-
tivities) that are not easily separable by chromatography.

The luminescence quantum yields of coelenterazine in
the presence of luciferases A. B. and C were close to 0.30
at 24 °C, one of the highest values among coelenterazine
luciferases. The quantum yields for other coelenterazine
luciferases are Oplophoms luciferase, 0.34 at 22 °C (Shi-
momura et til., 1978), and Renilla (sea pansy) luciferase,
0.11 at 23 °C (Inouye and Shimomura, 1997). The high

Table 2
Average luciferase content in single specimens of eggs, embryos, and juveniles

Before 2-ME treatment

After 2-ME treatment

Sample

Soluble (Luciferase L)

Total

Soluble (Luciferase A, B. C)

Total (mean ± SD)

Egg. dissected from ovary (7)

4760

17.400

19.300 ± 3.000

Egg, liberated^ (5)

1350

4.200

5.010 ± 3.500

Early embryoh (5)

730

2.990

3.380 ± 1.750

Late embryo' (3)

173

620

603 ± 50

Juvenile. 4-5 mmd (6)

205

460

490 ±431

Juvenile. 8-10 mmd (5)

162

370

7X4

824 ± 375

Luciferase activity was assayed before and after treatment with 25 mM 2-mercaptoethanol (2-ME) and shown in light units (LU). The number of samples
tested is shown in parentheses.

a Collected by plankton net. h Yolky throughout, with minute grooves for later development of lappets and tentacles. c Trace of pigmentation around
mouth, still with volk in stomach. d Dome diameter.

PERIPHYLLA LUCIFERASE STORED IN OVARY

347

efficiency of Periphyllu bioluminescence suggests that lu-
ciferase A might be useful as a highly sensitive reporter.

The nature of litciferase L

The specific luminescence activity of luciferase L previ-
ously reported (8 X 1013 photons/s) is more than two orders
of magnitude lower than that of luciferases A, B, orC (more
than 10"1 photons/s), raising doubts about the purity of the
luciferase L sample previously reported (Shimomura and
Flood. 1998). SDS-PAGE analysis of a sample of luciferase
L (Fig. 3) showed that the sample contained only a trace of
luciferase A (shown as a weak 24-kDa band) and a large
amount of other proteins (an intense broad band of 28-34
kDa). indicating that the purity of the sample was indeed
very low, probably about 17r. Such a condition could have
arisen from the extremely small amount of luciferase L, as
well as from the difference in the initial extracts: luciferase
L was purified from an extract containing all soluble pro-
teins, whereas luciferase A. B, and C were purified only
from the proteins that were solubilized by 2-mercaptoetha-
nol.

A pure sample of luciferase L (32 kDa). if available.
could not, on the basis of its molecular weight, be a simple
oligomer of the luciferase A monomer (20 kDa) notwith-
standing that it yielded luciferase A by treatment with
2-mercaptoethanol. Thus, luciferase L must be a complex of
the luciferase A plus another protein (about 12 kDa). The
presence of this accessory protein was confirmed by SDS-
PAGE as a band corresponding to 14 kDa (Fig. 3). One of
the functions of the accessory protein, apparently, is to
solubilize the luciferase, because luciferase L is the only
naturally soluble form of luciferase existing in Periphyllu.
The accessory protein, however, has other important func-
tions as judged from the data presently obtained. One of
these functions pertains to thermal properties, modifying the
temperature-luminescence intensity curve to emit the stron-
gest luminescence at an unusually low temperature, near
0 °C (Fig. 4). and significantly increasing the stability of
luciferase activity at high temperatures (Fig. 5). The adap-
tation of a low-temperature luminescence system as a means
of enhancing light emission is understandable for an organ-
ism that lives in the deep sea at 3 to 7 °C, but it is puzzling
that the accessory protein makes the luciferase L heat-stable
to such an unusual level that it withstands even boiling.

Recently, an accessory protein was also found in the
luciferase of Oplophonts (Inouye et til., 2000). The native
form of this luciferase (about 106 kDa) was found to be a
complex of two proteins, one 19 kDa and the other 35 kDa.
The luciferase function was found in the 19-kDa protein,
whereas the role of the 35-kDa accessory protein remains
unknown.

The quantum yield of coelenterazine in the presence of
luciferase L was previously reported as 0.14 (Shimomura

and Flood. 1998). in contrast to the value of about 0.30 for
luciferases A. B. or C. The quantum yield value of lucif-
erase L should be reexamined because it was obtained by an
unconventional method that is certainly affected by the
impurities that destroy coelenterazine and decreases quan-
tum yield. With a pure preparation of luciferase L. the
specific activity and the quantum yield are probably close to
those of luciferases A. B, and C. Uncertainties concerning
luciferase L and the role of the accessory protein will not be
completely clarified until pure preparations of luciferase L
become available.

Acknowledgments

The Periphyllu material used in this work was collected
aboard the R/V Hdkon Mosby during a cruise organized in
March 2000 by Professor Ulf Bamstedt. University of Ber-
gen. Financial support was received from the National Sci-
ence Foundation (MCB-9722982).

Literature Cited

Flood, P. R.. J.-M. Bassot, and P. J. Herring. 1996. The microscopical
structure of the bioluminescence system in the medusa Periphyllu
periphv/la. Pp. 149-153 in Bioluminescence and Chemiliiminescence:
Molecular Reporting with Photons, i. W. Hastings. L. J. Kncka, and
P. E. Stanley, eds. John Wiley, Chichester. UK.

Fossa, J.-H. 1992. Mass occurrence of Periphylla periphylla (Scypho-
zoa, Coronata) in a Norwegian fjord. Siirsia 77: 237-251.

Freeman, G., and E. B. Ridgway. 1987. Endogenous photoproteins.
calcium channels and calcium transients during metamorphosis in
hydrozoans. Rnux's Arch. Dev. Biol. 196: 30-50.

Inouye, S., and O. Shimomura. 1997. The use of Renilla luciferase.
Oplophoms luciferase and apoaequorin as bioluminescent reporter
protein in the presence of coelenterazine analogues as substrate. Bio-
chem. Biophvs. Res. Commuit. 233: 349-353.

Inouye, S., K. Watanabe. H. Nakamura, and O. Shimomura. 20(10.
Secretional luciferase of the luminous shrimp Oplophonts graciliros-
tris: cDNA cloning of a novel imidazopyrazinone luciferase. FEBS
Lett. 481: 19-25.

Jarms, G., U. Bamstedt, H. Tiemann, M. B. Martinussen, and J. H.
Fossa. 1999. The holopelagic life cycle of the deep-sea medusa
Periplnila periphylla (Scyphozoa, Coronata). Sarsia 84: 55-65.

Laemmli, U. K. 1970. Cleavage of structural proteins during the assem-
bly of the head of bacteriophage T4. Nature 227: 680-685.

Martinussen, M. B., G. Jarms, U. Bamstedt, and P. R. Flood. 1997.
Livssyklusen til Periphylla periphylla. Arsmoie for Norske Havfor-
skeres /oreiiing. Bergen 1997. (Abstract).

Nakamura, H., C. \Vu, A. Murai, S. Inouye, and O. Shimomura. 1997.
Efficient bioluminescence of bisdeoxycoelenterazine with the lucif-
erase of a deep-sea shrimp Oplophoms. Tetrahedron Lett. 38: 6405-
6406.

Shimomura, O., and P. R. Flood. 1998. Luciferase of the scyphozoan
medusa Periphylla periphylla. Biol. Bull. 194: 244-252.

Shimomura, O., and F. H. Johnson. 1970. Mechanisms in the quantum
yield of Cvpridina bioluminescence. Photochem. Photobiol. 12: 291-
295.

Shimomura, O., T. Masugi, F. H. Johnson, and Y. Haneda. 1978.
Properties and reaction mechanism of the bioluminescence ^> stem ol
the deep-sea shrimp Oplophoms gracilomstris. Biochemist '
998.

Reference: Biol. Bull. 201: 34X-359. (December 2001)

Zooxanthellae of the Montastraea annularis Species

Complex: Patterns of Distribution of Four Taxa of

Symbiodinium on Different Reefs and Across Depths

W. W. TOLLER13, R. ROWAN2'*, AND N. KNOWLTON1 3

^Marine Biology Research Division 0202, Scripps Institution of Oceanography, University of California

San Diego, La Jolla, California 92093-0202; 'University of Guam Marine Laboratory, Mangilao, Guam

96923; and Smithsonian Tropical Research Institute, Apartado 2072, Balboa. Republic of Panama

Abstract. Corals of the Montastraea annularis complex
host several different dinoflagellates in the genus Symbio-
dinium. Here we address two questions arising from our
previous studies of these associations on an offshore reef.
First, do the same taxa and patterns of association (Symbio-
dinium A and B found in higher irradiance habitats than
Symbiodinium C) occur on an inshore reef? Second, does M.
franksi at the limits of its depth range host only Symbio-
dinium C, as it does at intermediate depths? In both surveys,
a new Symbiodinium taxon and different patterns of distri-
bution (assayed by analyses of small ribosomal subunit
RNA genes [srDNA]) were observed. Inshore, a taxon we
name Symbiodinium E predominated in higher irradiance
habitats in M. franksi and its two sibling species; the only
other zooxanthella observed was Symbiodinium C. Off-
shore, M. franksi mainly hosted Symbiodinium C, but hosted
Symbiodinium A, B, C, and E in shallow water and Sym-
biodinium E and C in very deep water. Symbiodinium E may
be stress-tolerant. Observed srDNA heterogeneity within
samples of Symbiodinium B, C, and E is interpreted as
variation across copies within this multigene family. Exper-
imental bleaching of Symbiodinium C supported this inter-
pretation. Thus sequences from natural samples should be
interpreted cautiously.

Received 9 February 2000; accepted 5 July 2001.

*To whom correspondence should be addressed. E-mail: rrowan@uog9.
uog.edu

Abbreviations: RFLP. restriction fragment length polymorphism; rDNA,
ribosomal RNA genes; srDNA, small ribosomal subunit RNA genes.

Introduction

Coral reefs are the most biologically diverse marine hab-
itats. Underpinning this diversity are the reef-building cor-
als themselves, which are obligate, mutualistic symbioses
between coral animals and dinoflagellates (commonly
called /.ooxanthellae). This partnership between heterotro-
phic hosts and phototrophic symbionts allows corals to
thrive in shallow, nutrient-poor tropical seas, and deposit
calcium carbonate in amounts large enough to build reefs
(reviewed in Muscatine and Porter. 1977; Falkowski et ai,
1984; Barnes and Chalker, 1990; Muller-Parker and D'Elia,
1997).

Coral taxonomy at the species level, although occasion-
ally frustrating (Knowlton and Jackson, 1994; Veron,
1995). has generally been sufficient to describe overall
diversity and to define experimental subjects. This taxon-
omy seldom, however, has considered zooxanthellae, be-
cause it was widely assumed that one species of coral
associates with only one species of zooxanthella — in other
words, that host taxonomy identified both partners. Zoo-
xanthellae are diverse (e.g., Schoenberg and Trench. 1980;
Rowan, 1998), and it is now recognized that some species of
corals associate with multiple species of zooxanthellae
(Rowan and Knowlton. 1995: Rowan. 1998). Thus corals
identified as members of the same species may not in fact be
equivalent at the whole organism (holobiont) level, and the
taxonomic identities of zooxanthellae may be as ecologi-
cally important as those of their hosts.

As far as is known, zooxanthellae in reef-building corals
are members of the genus Symbiodinium (Rowan, 1998),
which includes four species described as in vitro cultures
(Freudenthal. 1962; Trench and Blank. 1987). Several other

348

ZOOXANTHELLAE IN DIFFERENT HABITATS

349

cultured isolates of Symbiodiniwn have been named infor-
mally, but most members of the genus remain uncultured
and undescribed (Rowan, 1998). Nevertheless, sequences
and restriction fragment length polymorphism (RFLP) of
genes that encode ribosomal RNA (rDNA) can be used to
distinguish some taxa of Symbiodinium and to study eco-
logical relationships among host, symbiont, and habitat
diversity (Rowan and Powers, 199 la, b; Rowan and Knowl-
ton, 1995; Rowan etui. 1997; Baker and Rowan, 1997; Hill
and Wilcox, 1998; Darius el til., 1998; Baker, 1999). The
present study uses genes that encode small ribosomal sub-
unit RNA (srDNA).

Our earlier work concerned zooxanthellae of the sibling
coral species Montastraea annularis, M. faveolata, and M.
franksi, which are the dominant reef-building corals in the
Western Atlantic (Goreau, 1959). On an offshore reef in the
San Bias Islands of Panama, we found that both M. annu-
laris and M. faveolata associate with three distinct taxa of
Symbiodinium (A. B. and C; see Rowan and Knowlton,
1995; Rowan et al., 1997). Symbiodinium A and B. or both,
are predominant in tissue exposed to high irradiance (shal-
lower water or colony tops), Symbiodinium C is predomi-
nant in shaded tissue (deeper water or colony sides), and
mixtures of Symbiodinium A and/or B with C occur between
these extremes. Colonies of M. franksi, in contrast, were
found to host only Symbiodinium C (Rowan and Knowlton,
1995); however, this coral species was not found at shallow
depths on this reef. These observations led to two questions
addressed here. First, do these symbiont taxa and patterns of
association occur on other types of reef? Second, does the
deeper distribution of M. franksi reflect an inability by this
species to host those taxa of Symbiodinium with which M.
annularis and M. faveolata associate in shallow water?

We also discuss some concerns about using srDNA to
identify the Symbiodinium that we collected. Although
srDNA was heterogeneous in samples of Symbiodinium B,
C. and E, we found no evidence to suggest that the zoo-
xanthellae in each of these samples were heterogeneous. We
suspect that srDNA in these Symbiodinium is a heteroge-
neous multigene family, as is rDNA in some other
dinoflagellates (Scholin et al., 1993; Scholin and Anderson,
1994, 1996). We discuss practical implications of this sus-
picion for the use of srDNA as a taxonomic character.

Materials and Methods

Field collections and study sites

Corals were identified in the field by colony-level char-
acters (Weil and Knowlton, 1994). Apparently healthy col-
onies, separated from one another by >2 m. were sampled
with hammer and '/2-in (#12) steel hole punch, yielding a
coral core with about 1.3 cm~ of live colony surface. Cores
were wrapped in aluminum foil and frozen in a cryogenic
dry shipper (chilled with liquid nitrogen). Many colonies of

,....,,,

Cayos o-s ?«• «-w <-,•,, -^,,.'

'vi^Chichlme ~^*ti^:>

Cayos ""
Holandes

Figure 1. Collecting localities in the San Bias Archipelago (upper
panel) and Bocas del Toro (lower panel). Republic of Panama (inset).
Arrows with initials identify places where corals were sampled: AG,
Aguadargana reef; RC, Rio Carti; CL, Cayos Limones; JP, Juan Point; CP,
Cocos Point. Data from Aguadargana reef were reported previously
(Rowan and Knowlton. 1995; Rowan et al., 1997).

Montastraea annularis and M. faveolata were sampled both
on their tops and on their sides to obtain samples from
relatively high- and low-irradiance tissues (respectively)
within a colony (Rowan et al., 1997). Most colonies of M.
franksi were sampled at only one location because their
relatively flat morphologies made a distinction between
colony top and side superfluous.

Coral colonies were sampled at three sites in the Republic
of Panama (Fig. 1) between October 1997 and October
1998:

( 1 ) Rio Carti, San Bias. We sampled from a small coastal
fringing reef adjacent to the mouth of a major river (Rio
Carti Grande). During May to December, such nearshore
sites are periodically subjected to heavy freshwater runoff
and riverine sediments (Clifton et al., 1997; D'Croz et al.,
1999). Montastraea species occur at Rio Carti' from the
barely subtidal to a depth of about 12 m, where hard
substrate is replaced by soft-bottom sediments. We sampled
the tops of all encountered colonies (M. annularis, n = 4;
M. faveolata. n = 20; M. franksi, n = 19); 30 of these were
also sampled on their sides.

(2) Cayos Limones, San Bias. These reefs are located 15

350

W. W. TOLLER ET AL

km north of mainland Panama and are not strongly influ-
enced by terrestrial runoff (D'Croz et al.. 1999). We sam-
pled from a relatively steep, leeward fringing reef that ends
abruptly at depths between 35 and 40 m in soft sediments
(see fig. 9 in Robertson and Glynn. 1977). On this reef, M.
franksi is common below 8 m, and it is the dominant coral
(with Agaricia lamarcki) below 15 m. We sampled M.
franksi throughout its depth range (4 to 38 m, n = 78
colony tops).

(3) Bocas del Toro. Juan Point and Cocos Point reefs are
located in the semienclosed lagoon of Bahfa Almirante in
the Province of Bocas del Toro. Like Rio Carti, these sites
are affected by high rainfall and river outflow throughout
much of the year. On many of the reefs in this area, M.
franksi is the most abundant member of the M. annularis
complex. We made a limited collection at depths of 1-15 m
for comparative purposes, consisting of 1 top sample of M.
annularis, 10 of M. franksi, and 3 of M. faveolata.

Observations of srDNA heterogeneity within samples of
S\mbiodinium C (see Results) prompted us to investigate
the stability of these genotypes under stress. We identified
1 1 colonies of M. annularis (each colony consisting of a
cluster of columns) that hosted heterogeneous RFLP geno-
types of Symbiodinium C. After an initial sample, columns
(one per colony) were transplanted from their natural habitat
(ca. 10-14 m depth) to 1 m depth at either Cayos Limones
(n = 4 transplants) or Aguadargana (n = 1 transplants)
reefs (Fig. 1 ), where they bleached. Columns were sampled
again after 4 days (Cayos Limones) or 40 days (Aguadar-
gana). Transplants and determinations of zooxanthellar
numbers were conducted as described in Toller et al.
(2001). In the present study, however, we did not sample
corals further (i.e.. during zooxanthellar repopulation; see
Toller et al., 2001).

Identification of zooxanthellae

Zooxanthellae were isolated and identified as described
previously (Rowan and Powers. 1991b; Rowan and Knowl-
ton, 1995). srDNAs were obtained by PCR amplification
with a "host-excluding" primer pair (ss5 and ss3Z) or with
universal primers (ss5 and ss3). and then characterized by
restriction enzyme digestion. The host-excluding primer
pair does not amplify known host srDNAs (Rowan and
Powers, 1991b; unpubl. obs.), but does amplify srDNAs
from a phyletic group that is much larger than Symbio-
dinium (McNally et al.. 1994; Toller et al.. 2001). All
samples were assayed using host-excluding primers, and
about one-third of them were also analyzed with universal
primers. Data obtained from the two kinds of amplifications
were always in agreement.

Every sample was analyzed by digesting amplified
srDNA with Dpn II and with Taq I, which differentiate
Symbiodinium A. B, and C by RFLP (Rowan and Powers,

1991a: Rowan and Knowlton, 1995; Rowan et al.. 1997).
RFLPs were diagnosed by comparison to genotype stan-
dards, which were obtained by PCR amplification from
cloned srDNAs of Symbiodinium A, B, and C, all isolated
from M. annularis (Rowan and Knowlton, 1995), and from
Symbiodinium E (from M. faveolata, this study). These
cloned genotype standards are denoted hereafter as A°, B°,
C°, and E° '. We use the superscript zero to indicate srDNA
clones, as opposed to taxa of Symbiodinium'. clones ob-
tained from different samples of the same taxon of Symbio-
dinium are distinguished by numbers (e.g., E0"1 and E°~2; see
below). Because universal PCR primers amplify coral host
srDNA when it is present (Rowan and Powers, 1991b), a
cloned srDNA from M. annularis (clone H°; see below) was
used as an additional standard in RFLP analyses of these
amplifications. Where RFLP analyses indicated mixtures of
Symbiodinium A, B, C, or E in a sample, relative abundance
(greater than or less than 50% of the total) was estimated by
comparison to standard mixtures prepared from cloned
srDNAs (Rowan and Knowlton, 1995; Rowan et al.. 1997;
see Fig. 4).

srDNA was cloned from three samples of Symbiodinium
E: clone E0"' is from M. faveolata (from Rio Carti, 3 m
depth), clone E0"2 is from M. franksi (from Cayos Limones,
38 m), and clone E0"3 is from the coral Siderastrea siderea
(from Portobelo, Panama, 6 m). Amplified srDNAs (DNA
for clone En~' by universal PCR primers; DNAs for clones
E°° and E°~3 by host-excluding PCR primers) were gel-
purified, ligated into pGEM-T Easy Vector (Promega Cor-
poration. Madison, WI), and then transformed into Esche-
richia coli according to manufacturer's recommendations.
From each ligation, 4-12 clones were characterized by am-
plifying srDNAs with host-excluding PCR primers and then
digesting the PCR products with Dpn II, with Taq I, and
with Hae III. Each cloned RFLP genotype was compared to
the RFLP of its corresponding natural sample. srDNA of M
annularis was obtained with universal PCR primers from
sperm DNA (Lopez et al.. 1999) and cloned (clone H°) as
described above.

Clones E° ', E0*2, and E11"3 were sequenced completely, as
were cloned genotype standards A°, B°, and C° (from which
only partial sequences had been obtained previously;
Rowan and Knowlton, 1995) and clone H°. Plasmids were
prepared using QIAprep Spin Miniprep kits (Qiagen, Inc.,
Valencia. CA) according to manufacturer's recommenda-
tions, and sequences were determined for both DNA strands
using Big Dye Terminator sequencing kits (PE Corporation,
Norwalk, CT) with vector sequencing primers T7 and Ml 3-
Re verse, and with srDNA sequencing primers 18F1
(5'-AGCTCGTAGTTGGATTTCTG-3'), 18F2 (5'-TTA-
ATTTGACTCAACACGGG-3'), 18R1 (5'-AGTCAAA-
TTAAGCCGCAGGC-3') or 18-R1X (5'-GTTGAGTCA-
AATTAAGCCGC-3'), and 18R2 (5'-ATATACGCTA-
TTGGAGCTGG-3'). Reactions were analyzed with an ABI

/()()\ AMHI I.I Al IN 1)11 I I.K1-NT HABITATS

351

M A B C E

Figure 2. RFLP genotypes A, B, C, and E of Symbiodinium obtained
from different colonies of Montastraea frank.ii. srDNAs were amplified
with host-excluding PCR primers and digested with Dpn II (left) and with
Taq I (right). Lane M contains DNA fragment size standards of (top to
bottom) 1500 base pairs (bp). 1200 bp, and then 1000 hp to 100 bp in
100-bp increments.

373 sequencer (Applied Biosystems, Foster City, CA) and
complete sequences were assembled using SeqEd software
(Applied Biosystems). RFLP genotypes of cloned srDNAs
were obtained from their sequences using Gene Construc-
tion Kit software (Textco, Inc., West Lebanon. NH). Note
that we used only partial srDNA sequences in some analy-
ses (Fig. 3); the full-length srDNA sequences were depos-
ited in GenBank (http://www.ncbi.nlm.nih.gov/; accession
numbers AF238256-AF238258. AF238261-AF238263, and
AF238267).

For phylogenetic analysis, we aligned partial srDNA se-
quences (Rowan and Powers, 1992) with Clustal X software
(Thompson el al., 1997) and used neighbor-joining recon-
struction (Saitou and Nei, 1987). The following srDNA
sequences were obtained from GenBank: Symbiodinium mi-
croadriaticum (M88521), Symbiodinium #8 (M88509).
Symbiodinium sp. PSP1-05 (AB016578), sll-2xba
(U20961). s20-2xba (U20962). 37-4xba (U20959), 86-5xba
(U20960). a!2-5xba (U20954), a8-5xba (U20955), 175-
5xba (U20952), 178-6xba (U20956), 33-6xba (U20958),
al-5xba (U20953), 178-8xba (U20957), Gymnodinium hi-ii
(U37366), Gvrodininm galatlieamtm (M88511). Gymno-
iliniiini simpk'.\ (M88512), and Polarella glacialis
(AF099183). srDNA sequences from Symbioiliniuin C2
[clone C2(M (AF238259) and clone C2""- (AF238260)] are
from Toller et al. (2001 ). A partial sequence of zooxanthel-
lar srDNA from the coral Montipora patuhi is from a
previous study (Rowan and Powers, 199 la).

To investigate srDNA variation within our samples of
Svmhiodininm in greater detail, we selected representative
samples of each S\mbiodinium taxon from each host coral
species (M. annularis, M. faveolata, M. franksi) and made
additional RFLP analyses. Different samples (from different
colonies) of Symbiodinium A (n = 10), B (n = 12), C (n

-- 12). and E (n = 12) were analyzed with a panel of 12
restriction enzymes, used one at a time. These enzymes
were Dpn II, Taq I. Alw I, BstU I, Hue III, Him I. Hinf\. Mse
I, Msp I, Nci I, Sau96 I, and Sty I. Samples of Symbiodinium

E were investigated further with the enzymes Alii I.
Bsp\286 I, Mae III. Mnl I. 5/i/N I, and Tsp45 I. We chose
the latter enzymes based on RFLP differences among clones
E0"', E00, and E0"3. All enzymes were purchased from New
England Biolabs, Inc. (Beverly, MA) except for Mae III
(Roche Diagnostics Corp., Indianapolis, IN).

Results

Identification of Symbiodinium £

Routine analyses of srDNAs with Dpn II and with Taq I
revealed a zooxanthella in our surveys (see below) that was
different from Symbiodinium A, B, and C (Fig. 2). We call this
new RFLP genotype E (D has been assigned to a sponge
symbiont [Carlos et al., 1999]). Cloned genotype E srDNAs
(E0"1, E0"2, and E0"3 from Montastraea faveolata, M. franksi,
and Siderastrea siderea respectively) were more than 99%
similar in sequence to one another, and more than 96% similar
to srDNAs of Symbiodinium A, B. and C that were cloned
from M. annularis (genotype standards A". B". and C"). A
neighbor-joining analysis of partial srDNA sequences (Fig. 3)
places genotype E srDNAs within Symbiodinium (defined by
cultured S. micnxidriaticum and Symbiodinium #8 [Rowan,

S. microadriaticum

A, U20961

A, U20962 (clone A°)

B. U20955
B U20954
B. U20959 (clone B°)

B. U20960
Symbiodinium #8

R — I |- C U20953 (clone C°)

I ,- C. U20958

[I 1 C, U20952

IT C, U20956

i 1 C; U20957

C2 (M. annularis: clone C20'1)

C2 (S. siderea: clone C2°-2)

D, AB016578 (sponge)

£ (S siderea: clone £°-3)
£ (M. franksi: clone E0'2)

E (M. patula)
E (M faveolata, clone £°-1)

Figure 3. Inferred phylogenetic relationships among srDNAs from
different zooxanthellae. Partial srDNA sequences (Rowan and Powers,
1992) were grouped by the neighbor-joining method (Saitou and Nei,
1987). S\mbiodinium microadriaticum and Symbiodinium #8 are cultured
zooxanthellae (Rowan and Powers. 1992). A. B. and C (followed by
GenBank accession numbers) are from Montastraea annularis (Rowan and
Knowlton, 1995): three of these correspond to standard clones A". B". and
C" (this study). Two srDNAs labeled C2 (hosts and clone numbers in
parentheses) are from Toller el al. (2001 1. D (followed by GenBank
accession number) is from a sponge (Carlos et al.. 1999). srDNAs labeled
E (host and clone numbers in parentheses) are from this study, except for
that from the coral Monlifiow patula, which is from Rowan and Powers
(1991a). The branch labeled R (to the left) indicates the root for lins iree.
obtained by including srDNA sequences from the dinoflagellates G\min>-
dinium heii. Gyrodinium galalheanum, Gynmodiniwn siinpl'.-x. and f'r-
larella glacialis (not shown).

352

W. W. TOLLER ET AL.

Samples Clones (E°~1 : C° Mixtures)

123 I L° 8:1 1^1 LI 11 l:i U 1:8 0:J

Figure 4. RFLP genotypes of mixtures of Symbiodinium E and C.
Samples of zooxanthellae are from Montastraea faveolata (samples 1. 2)
and M. franksi (sample 3): other lanes are clones E" ' and C" singly (1:0
and 0:1. respectively) and mixed together in molar ratios ranging from 8:1
to 1:8. to obtain standards. srDNAs were amplified with host-excluding
PCR primers and then digested with Dpn II (top panel) and with Tmi I
(bottom panel). By visual inspection, samples 1-3 contain both Synihin-
iliiiiiini E and C. in ratios of about 4: 1 . 1:1.5. and 1 :4. respectively. Lane M
contains DNA size standards, as in Figure 2.

1998]): separate from Symbiodinium A. B. and C; and close to
a zooxanthellar srDNA from the coral Montipora panda, an
srDNA that previously could not be assigned to either Syni-
hiodiniwn A. B. or C (Rowan and Powers, 1991a). srDNA
from Symbiodinium D. a dinoflagellate cultured from the
sponge Haliclona koremella (Carlos ct al.. 1999), is not similar
to genotype E (Fig. 3). Thus, genotype E represents a distinct
taxon of zooxanthella — Syinbiodinium E.

Some samples of zooxanthellae (see below) had RFLP
genotypes that implied mixtures of Symbiodinium E and
C, based on comparisons to RFLP genotypes of synthetic
mixtures of cloned genes (srDNA clones E""' and C°; Fig.
4). As with mixtures of Symbiodinium A. B, or C de-
scribed previously (Rowan and Knowlton, 1995; Rowan
ct ill., 1997). the apparent ratio of Symbiodinium E to
Symbiodinium C in different samples varied, and did not
depend on which restriction enzyme was used to differ-
entiate these two genotypes (e.g.. Fig. 4, Dpn II digests
versus Taq I digests).

Distribution of different taxa of Symbiodinium

At Rio Carti, M. franksi was observed with only two taxa
of zooxanthellae — Symbiodinium E and C — and the same
two taxa were obtained from M. faveolata and M. annularis

(Fig. 5) at this reef. Symbiodinium E was the predominant
zooxanthella from all three Montastraea species: it occurred
in 35 of 43 corals and was the only zooxanthella detected in
18 of these. In M. franksi and M. faveolata, Symbiodinium
E was more common in higher irradiance habitats (colonies
at 1-3 in depth, tops of colonies at 3-6 m depth) than in
lower irradiance habitats (colony sides at 3-6 m depth and
generally below 6m); Symbiodinium C exhibited the con-
verse pattern (Fig. 5). Samples from M. annularis (n = 4)
showed the same top and side pattern of zooxanthellar
distribution within colonies (Fig. 5). although our small
sample size precludes an examination across depth. A zo-
nation pattern was often observed in comparisons of tops
and sides from the 16 doubly sampled colonies that had the
two types of zooxanthellae. In 12 of these colonies, the ratio
of Svmbiodinium E:C decreased from top to side, in three
there was no clear difference in the ratios, and in only one
colony did the ratio increase from top to side.

At Cayos Limones, M. franksi associated primarily with
Symbiodinium C (Fig. 6), which was the only taxon of
zooxanthella observed between 6.5 and 33 m depth (;; = 53
colonies); this result is consistent with the previous study
(Rowan and Knowlton, 1995) of M. franksi from depths

.210

I 8
o

Tops

.510

o
o 8

Sides

FA KFA KFA

(1-3m) (3-6m) (6-12m)

Coral Species / Depth Interval

Figure 5. Occurrences of Symbii>iiinium C and E (assayed by RFLP.
see Fig. 4) in colonies of Montastraea franksi (K). M. faveolata (F). and M.
iinnuhirn (A) living in three depth intervals at Rio Carti. Top samples
(upper histogram) were taken from 43 corals; 30 of these were also
sampled on their sides (lower histogram). There are no data (ND) from M.
iiiiniiliiris in the shallowest depth interval because no colonies were en-
countered there.

ZOOXANTHELLAE IN DIFFERENT HABITATS

353

Symb. A or A + C
D Symb. 6 or B + C
D Symb. £ or E + C

Symb. C only

1 8

•5 6

5 8 11 14 17 20 23 26 29 32 35 38
Depth Interval (meters)

Figure 6. Occurrences of Symbiodinium A, B, C. and E in tops of
colonies of Montastraea fnuiksi living at Cayos Limones. Shallow depth
intenaK are 3.5-4.5 m (labeled 4) and 4.5-6 m (labeled 5); other depth
intervals are 3 m wide on the centers indicated. Samples were scored as
containing Svinhiniliiiiiun A. B. C. and/or E. according to the key. More
samples were analyzed at the ends of the depth range, where more than one
taxon of Symbiodinium was observed.

between 6 and 1 1 m at Aguadargana, another nearby off-
shore reef (Fig. 1). However, in the shallowest and deepest
colonies of M. franksi. different taxa of zooxanthellae were
observed. Between 4 and 6 m, colonies contained, in order
of decreasing frequency of occurrence, Symbiodinium B, C.
A, and E. With the exception of Symbiodinium E in one
colony, this distribution of taxa resembles that found in M.
annuliiris at similar depths at Aguadargana reef (Rowan and
Knowlton. 1995; Rowan et al.. 1997). Samples from four of
the six deepest colonies of M. franksi (35-38 m depth)
contained Svmbiodinium E only: the other two colonies
contained Synibituliniiini C only (Fig. 6). At both the shal-
low and deep extremes, colonies of M. franksi were rela-
tively small, encrusting forms (<0.5 m diameter).

To find out if the congeners of M. franksi at Cayos
Limones also host Symbiodinium E at their lower depth
limits, we sampled the deepest colonies of M. annularis
(n = 23) and M. faveolata (n = 5) that we could find.
They were not very deep (12-17 m and 13-15 m. respec-
tively), and like M. franksi at the same depths, contained
Symbiodinium C only (not shown).

In our limited sample of corals from two reefs at Bocas
del Toro (1-15 m depth), M. franksi was found with
Symbiodinium E only (1 colony), with Symbiodinium E
and C (4 colonies), with Symbiodinium C only (2 colo-
nies), or with Symbiodinium A only (3 colonies). M.
faveolata was found with Symbiodinium C only (2 colo-
nies) or with Symbiodinium A only ( 1 colony). The single
encountered colony of M. annularis contained Symbio-
dinium A. We did not observe Symbiodinium B in any of
these samples.

Other diversity in -ooxauthellar srDNAs

The routine RFLP analyses (with Dpn II and Taql) reported
above indicated that all samples of zooxanthellae in this study
contained srDNAs of either Symbiodinium A, B. C. or E. or
mixtures thereof, as defined by our standard, cloned srDNA
genotypes (A°, B°. C", E11"'). However, when zooxanthellar
srDNAs were analyzed in greater detail (with additional re-
striction enzymes; see Methods and Materials), samples of
Svmbiodinium B. C. and E (but none of 10 tested samples of
Svmbiodinium A) were found to contain additional srDNAs
that could not be attributed to genotypes A". B°, C°, E0"1. or to
host srDNA. These other srDNAs appeared as additional DNA
fragments in restriction digests, as described below.

Twelve selected samples of Symbiodinium E and clones
E0"', E0"2. and E"~3 were all indistinguishable in digests with
Dpn II (examples in Fig. 7. Dpn II panel) and with Taq I (not
shown). In digests with Mae III, however, all of these samples
had an additional DNA fragment in relatively low abundance
(arrow in Fig. 7. Mae III panel) that was not part of the RFLP
genotype of clones E° ' and E0"2, but which was in the RFLP
genotype of clone E0"3. Thus, these samples apparently con-
tained at least two srDNAs — one defined in Mae III digests by
clones E0"1 and E0"2, the other by clone E0"3. Similarly, an
additional band in digests of sample srDNAs with Mnl I (arrow
in Fig. 7, Mnl I panel) apparently represents the RFLP geno-
type of clone E° ' (versus clones E0"2 and E0"3). Digestion of
samples with Alii I also yielded an additional DNA fragment
(arrow in Fig. 7. Alu I panel), and digestion of cloned srDNAs
with Alu I showed that the genotype of clone E0"2 is unique. In
all. additional bands like those shown in Figure 7 (arrows)
were observed in 7 of 18 different restriction enzyme diges-
tions (other digests not shown) of the 12 tested samples of
Svmbiodinium E. Therefore. srDNA in these samples of Sym-
biodinium E was clearly heterogeneous. This heterogeneity did
not, however, vary qualitatively nor quantitatively among the
tested samples (e.g.. Samples 1-3 in Fig. 7). Thus, clones E0"1.
E0"2, and E0"3, which are different (Fig. 7; see also Fig. 3), were
obtained from indistinguishable samples of zooxanthellae.

As with Svmbiodinium E, srDNA heterogeneity was ob-
served in all tested samples of Symbiodinium B. Two digests
(out of 12) revealed heterogeneity — Hlui I and Sty I (exam-
ples in Fig. 8). In each of these, the additional fragments
(arrows in Fig. 8) imply an srDNA with one restriction site
gain relative to clone B°. Interestingly, a cloned srDNA
from Symbiodinium B (Symbiodinium #8 isolated from a
Hawaiian anemone [Aiptasia pulchella] Rowan and Powers,
1992) has both additional sites (S8 in Fig. 8; schematic
genotype on the right), suggesting that samples of Symbio-
dinium B from other hosts may also exhibit srDNA heter-
ogeneity. In our samples of Symbiodinium B from Monta-
straea. within-sample srDNA heterogeneity did not van
among the 12 tested samples (e.g.. samples 1-4 in Fig. 8).

In the case of Svmbiodinium C, srDNAs in all "1" 12 tested

354

W. W. TOLLER ET AL

Samples Clones

M 1

Figure 7. Examples of srDNA heterogeneity within samples of Sym-
biodinium E. srDNAs were amplified (with host-excluding PCR primers)
from different samples of Symbiodinium E (lanes 1-3) and from srDNA
clones E0'1, E0'2, and E0"-1 (as indicated) and then digested with Dpn II.
Mae III. Mnl I. and Alu I (indicated on the left). On the right, arrows
identify the positions of additional DNA fragments in lanes 1-3 that
indicate srDNA heterogeneity (see text). In Mae III and Mnl I digestions,
these bands were also observed in one of the three clones; for A/u I
digestions no clone contains the indicated band. Samples are from Mon-
tastraea franksi (lane 1 ), from M.faveolata (lane 2), and fromM. annuluris
(lane 3). Lane M contains DNA size markers as in Figure 2.

samples were also heterogeneous. However, unlike Symbio-
dinium E and B (above), within-sample srDNA heterogene-
ity in Symbiodinium C varied both qualitatively (e.g., com-
pare samples 3-5 in Dpn II panel. Fig. 9) and quantitatively
(e.g., compare Samples 1-5 in Hinfl panel, Fig. 9) among
samples. srDNA heterogeneity was observed in as few as
one or as many as six different digests (examples in Fig. 9)
among the 12 samples tested. That additional variation
suggested that some or all samples might have contained
more than one genotype of Symbiodinium.

We made two analyses that might have supported this
hypothesis. First, because mixtures of Symbiodinium A, B,
C, or E vary in proportion at different locations within a
coral colony (Rowan and Knowlton. 1995; Rowan et al..

1997; Results), we analyzed multiple samples from colonies
of M. iinnuliiris in which Symbiodinium C had been ob-
served previously. In 14 colonies (each consisting of a
cluster of columns), we sampled one column on its top and
on its side; srDNA genotypes were indistinguishable in
every top-versHs-side comparison (not shown). We also
sampled the tops of one or two additional columns in 13 of
these colonies, and again saw no differences in zooxanthel-
lar RFLP genotype within any colony (not shown). Second,
we speculated that if the additional srDNAs did represent
distinct, co-occurring zooxanthellae, their relative abun-
dance might change under stress (e.g., as in Rowan et al.,
1997). Transplantation of columns from deep to shallow

Samples

Figure 8. Examples of srDNA heterogeneity within samples of Sym-
biodinium B. srDNAs were amplified (with host-excluding PCR primers)
from different samples of Symbiodinium B (lanes 1-4) and from srDNA
clone B" and then digested with Dpn II, Taq I. Hha I. and Sry I (indicated
on the left). On the right are schematic RFLP genotypes of clone B" and of
an srDNA clone from Symbiodinium #8 (S8), obtained from its sequence
(Rowan and Powers. 1 992). Arrows next to the schematics identify DNA
fragments that, in digests of srDNA from these samples of zooxanthellae.
are additional to the DNA fragments of clone genotype B". Samples are
from Montuxlmeii imnulnris (lanes I, 2). M. faveolata (lane 3), and M.
franksi (lane 4). Lane M contains DNA size markers as in Figure 2.

ZOOXANTHHLLAE IN DIFFERENT HABITATS

355

Samples

1 2 3 4 5 I C°

Figure 9. Examples of srDNA heterogeneity within samples of S\m-
biodiniitm C. srDNAs were amplified (using host-excluding PCR primers)
from different samples of Symbiodinium C (lanes I -5 1 and from srDNA
clone C" (lane C") and then digested with Dpn II. Taq I. SmMb I. and Hinf
I (indicated on the left). Arrows on the right identify DNA fragments that,
in digests of srDNA from these samples of zooxanthellae. are additional to
the DNA fragments of clone genotype C". Samples are from Montasrraea
annularis (3). M. favvolatu ( I. 2. 4) and M. franksi (5).

habitats resulted in bleaching of all columns, and effectively
reduced zooxanthellar numbers (70% reduction on aver-
age). However, neither acute stress (5 days) nor prolonged
stress (ffl. 40 days) of zooxanthellae altered the RFLP
genotypes that were observed (examples in Fig. 10) — the
relative abundance of distinct srDNAs had not changed
compared to samples taken prior to transplantation.

Discussion

Four ta\(i r;/ Symbiodinium in the Montastraea annularis
complex

Previous surveys of zooxanthellar diversity in Monta-
straea annularis, M. faveolata, and M. franksi (Rowan and
Knowlton, 1 995; Rowan et at., 1997) are now shown to be

incomplete. In surveys of additional habitats and depths. \se
found (i) a fourth taxon of Symbiodinium (E) that was not
previously reported in these corals, (ii) differences in the
distribution of zooxanthellae at offshore and coastal reefs,
and (iii) multiple taxa of zooxanthellae in M. franksi, which
previously had been found to contain only Symbiodinium C.

Groups A, B, C, and E constitute the known diversity of
coral-associated Symhiodin'ntm (Rowan, 1998; this study),
and M. annularis, M. faveolata, and M. franksi all associate
with at least one member of each of these groups. This is a
remarkable amount of taxonomic diversity — at least 12
distinct symbioses — in what was previously (Knowlton et
ai, 1992; Rowan and Knowlton. 1995) regarded as one
species of coral hosting one species of zooxanthella. More-
over, this diversity is not randomly distributed, suggesting
that what was once viewed as a single quintessential gen-
eralist (Connell, 1978) is in fact a complex assemblage of
ecologically more specialized entities.

Our observations from Cayos Limones now enable us to
refute the speculation that M. franksi associates exclusively
with Svmbiodiniuin C — this host coral can and does form
symbioses with Symbiodinium A, B, and E. However, at this
offshore reef, the latter host-zooxanthella combinations are
observed only at the margins of this coral's depth distribu-
tion (Fig. 6): shallow (Symbiodinium B > A > E) and very
deep (Symbiodinium E; discussed further below). Other-

Samples

34 56

Clone

7 8 I C°

Figure 10. srDNA heterogeneity in samples of Symbiodinium C from
four colonies of Montastraea annularis before and after experimental
stress (see text). srDNAs were amplified (using hosi-excluding PCR prim-
ers) from samples of zooxanthellae (lanes 1-8) and from srDNA clone C"
(lane C°) and then digested with Tai/ 1 (top panel) and with S<;»96 I (bottom
panel). Arrows identify DNA fragments that are additional to those of
genotype C". Samples were taken from the same colony before (lam1 i i .nid
after (lane 2) stress. Samples 3 and 4 are from another colony, before and
after stress (respectively), as are samples 5 and 6 and samples 7 and 8. Lane
M contains DNA fragment size standards of (top to b<>iti>,n; 2642 base
pairs (bp), and 1500 bp to 100 bp in 100-bp increments.

356

W. W. TOLLER ET AL

wise, M. franksi hosts Symbiodinium C throughout nearly
all of its depth range (Fig. 6). where colony growth is
robust. Although in shallow water the distributions of zoo-
xanthellae (mostly Symbiodinium A and B) are similar in M.
franksi, M. faveolata, and M. ammlaris, the small size of M.
franksi colonies in shallow water may reflect a relatively
poor physiological fit between this coral host and these
zooxanthellae.

The main question posed by our new results is why all
three species in the Montastraea annularis species complex
at a coastal site (Rio Cartf) host predominantly Symbio-
dinium E at higher it-radiance (Fig. 5), instead of Symbio-
dinium A or B, as found at offshore reefs (Rowan and
Knowlton, 1995; Rowan et til., 1997; Fig. 6). One possible
explanation is that this coastal site is characterized by en-
vironmental stress to which Symbiodinium E is more toler-
ant than are Svmbiodinhim A or B. High irradiance is a
stress that may exacerbate (Brown. 1997) the many other
kinds of stress found in nearshore environments (e.g.. fluc-
tuations in temperature, salinity, nutrients, sediments, and
underwater irradiance; see Bowden. 1983; Kirk. 1994). All
of these factors can affect the stability of coral-algal sym-
bioses (Falkowski et al.. 1993; Brown. 1997; Wesseling et
til., 1999). In the San Bias Archipelago (Fig. 1), nearshore
effects associated with freshwater runoff are limited to a
relatively narrow coastal band and do not reach our offshore
study sites at Aguadargana and Cayos Limones (D'Croz et
til., 1999). Symbiodinium E was also common in Monta-
straea within a large coastal lagoon at Bocas del Toro.
Panama (Fig. 1). an area of exceptionally high rainfall
where water quality is also likely to be dominated by coastal
effects.

A second (and perhaps related) question asks why Sym-
biodinium E was distributed differently at Cayos Limones,
where it was common not at high irradiance but rather in the
very deepest colonies of M. franksi (Fig. 6). Perhaps shal-
low and deep populations of Symbiodinium E are different
species of zooxanthella. although we did not find any evi-
dence to support this (see following section). Instead, we
suggest that Svmbiodinimn E was actually not distributed so
differently at these two sites. In both cases it was associated
with marginal habitat: at great depth where M. franksi
colonies are not large and where the reef itself disappears
into sediment (Cayos Limones). and along the coast near a
large river, where coral reefs are poorly developed or absent
(Rio Cartf). Bleaching-associated stress may be common in
both habitats, due to occasional smothering by sediments in
the former (e.g., Wesseling et ai, 1999) and to near-shore
conditions in the latter (see above). We propose that the
Symbiodinium E we observed represents a taxon of zoo-
xanthella that occurs in certain habitats not because it per-
forms best in those habitats, but because it tolerates them,
whereas Symbiodinium A, B, and C do not. According to
this idea, Svmhiddiniitm E is rare or absent from other

habitats not because it performs poorly in them, but because
S\mbiodinium A. B. and C are better adapted to those
habitats and somehow exclude it.

Anecdotal observations are consistent with our interpre-
tation of Svmbioilinium E as a stress-tolerant zooxanthella.
We observed Symbiodinium E (diagnosed by Dpn II and
Taq I digests of srDNA) inM faveolata in the Bahamas (not
shown), in four of seven colonies that were relatively un-
bleached during a natural bleaching event (D. Zawada,
Scripps Institution of Oceanography, pers. comm.). We also
found that Symbiodinium E — but not Symbiodinium B or
C — was adept at repopulating severely bleached corals in
experiments (Toller et al.. 2001). These experimental re-
sults suggest that, in addition to tolerating stress. Symbio-
dinium E may also be good at colonizing corals whose
zooxanthellar communities have been severely disrupted by
stress.

Observations of zooxanthellae related to Symbiodinium E
in other hosts and seas imply that this taxon, like the taxa
Symbiodinium A, B, and C (Rowan, 1998), may represent a
group (clade) of zooxanthellae. Those observations include
the corals Montipora patula in Hawaii (Rowan and Powers,
199 1 a; Fig. 3), Acropora palifera in Australia (R. R., un-
publ. obs.), Pocillopora damicornis in the eastern Pacific
(Baker, 1999), Goniastrea aspera in Thailand (A. Douglas.
University of York, pers. comm.), and the giant clam Hip-
popits hippopus in Australia (R. R., unpubl. obs.). In the
context of our hypothesis that Symbiodinium E is stress
tolerant in Montastraea, it is notable that G. aspera occurs
on reef flats — an environment that is stressful for corals, and
where coral bleaching events occur regularly (Brown et al.,
2000). Similarly, in P. damicornis, Symbiodinium E was
disproportionately common in unbleached colonies during
an El Nino-related bleaching event (Symbiodinium D of
Baker (1999] has an RFLP pattern that is indistinguishable
from that of Symbiodinium E from Montastraea in three
restriction enzyme digests; A. Baker. Wildlife Conservation
Society, pers. comm.). These observations suggest that
other members of the clade Symbiodinium E may also be
stress tolerant.

The hypothesis that Symbiodinium E is a relatively stress-
tolerant zooxanthella is based on circumstantial evidence,
and should be tested in experiments in which environmental
factors are controlled and physiological responses are mea-
sured. Descriptive studies of unmanipulated corals are,
however, indispensable for framing realistic hypotheses in
the first place.

Taxonomic interpretation of variation in zooxanthellar
srDNA

We recognize the RFLP genotype E as a distinct taxon —
Symbiodinium E — for the following reasons: (i) RFLP ge-
notype E was common, and many samples contained only

ZOOXANTHELLAE IN DIFFERENT HABITATS

357

this genotype (Fig. 5); (ii) the nonrandom distribution of
RFLP genotype E (Figs. 5 and 6) strongly implies that it
represents a distinct organism with distinct ecological at-
tributes; and (iii) phylogenetic analyses of genotype E
srDNAs place them within Svmbiodiniiiin, but distinct from
srDNAs of genotypes A, B, and C (Fig. 3), which, by the
same reasoning, represent distinct taxa of Symbiodinium
(Rowan, 1998). In practice, these four taxa of Symbiodinium
are readily identified by comparison to cloned srDNAs
(RFLP genotypes A", B", C°, and E1'"1) digested with the
enzymes Dpn II and Taq I.

By analyzing zooxanthellar srDNA with additional re-
striction enzymes, we found that samples containing srDNA
of RFLP genotype B°, C°, or E0"' also contained at least one
additional srDNA of a different RFLP genotype (examples
in Figs. 7-9). What do these additional srDNAs represent,
taxonomically? Like an srDNA in genotype C* (Rowan and
Knowlton, 1995), they appear to be from Symbiodinium
(and not some other type of organism) because (i) they were
distinguishable in fewer than one-half of different restric-
tion digests, (ii) many of them seemed to represent simple,
single restriction site changes compared to a cloned srDNA
(not shown), and (iii) different srDNAs from samples of
Symbiodinium E (Fig. 3) or of C* (Rowan and Knowlton,
1995) differed relatively little in sequence.

Do these additional srDNAs represent distinct species or
strains of Symbiodinium''! In the case of Symbiodinium E and
B, no evidence suggests that they do. Specifically, these
srDNAs were not observed by themselves, nor did they vary
in relative abundance from sample to sample (Figs. 7 and 8).
This contrasts with observations on srDNAs of RFLP ge-
notypes A°, B°, C° and E0"1, which occur alone, and also
mix in a range of proportions (e.g., Symbiodinium C and E.
Fig. 4; Rowan and Knowlton, 1995; Rowan et at.. 1997).

Because srDNA is a multigene family in eukaryotes,
srDNA heterogeneity (as seen within samples of Symbio-
dinium B and E) can reside in one organism — including
dinoflagellates (Scholin et al, 1993; Scholin and Anderson,
1994, 1996) — among gene-family members (Hillis and
Dixon, 1991). We favor this as an explanation for our data
because it is parsimonious compared to the alternative of
multiple strains of zooxanthellae that for some reason al-
ways co-occur in the same relative proportion. Testing this
hypothesis requires the analysis either of one dinoflagellate
(e.g., Yeung et al., 1996) or of aclonal culture (e.g., Scholin
et al., 1993; Rowan et al., 1996).

Heterogeneity of srDNA within samples of Symbio-
dinium C was more intriguing because sample-to-sample
variation was observed among colonies (Fig. 9). That ob-
servation suggested that different srDNAs within any one
sample could represent different strains of Symbiodinium. It
so, that sample-to-sample variation might also appear
within one coral colony, either from place to place or time
to time, especially before versus after an environmental

change. We found no such variation (e.g.. Fig. 10) in corals
hosting Symbiodinium C, which again is consistent with the
hypothesis that srDNA heterogeneity is a property of indi-
vidual zooxanthellae. Different patterns of srDNA hetero-
geneity seen among samples of Symbiodinium C from dif-
ferent corals (Fig. 9) are different zooxanthellar genotypes,
but we do not know if these differences are biologically
significant (e.g., Scholin and Anderson, 1994. 1996).

Independent of its source, within-sample srDNA hetero-
geneity limits the information that can be obtained from
srDNA sequences. This limitation is apparent in our analy-
sis of Symbiodinium E. The sequence of clone E" ' implies
that our RFLP analyses, using 18 enzymes (examples in Fig.
7), surveyed about 220 nucleotide positions (not shown).
Heterogeneity was detected with seven enzymes, which
implies a within-sample srDNA sequence diversity of about
3% (7 of 220 nucleotide positions). We do not know how
this diversity is distributed; possibilities range from two
srDNAs that differ at 7 of 220 positions (en. 3% different
srDNAs, similar to the difference between srDNAs of Sym-
biodinium A and B [Rowan and Powers, 1992]) to seven
srDNAs that differ from one another at 1 of 220 positions
(r«. 0.4% different srDNAs). Differences among srDNA
clones E0"1, E0"2, and E0"3 fall within this range, and there is
no reason to expect any cloned srDNA to represent our
samples of Symbiodinium E with any greater precision.
Moreover, the PCR creates chimeric DNA molecules when
mixed templates are amplified, and many clones obtained
from those PCR products will be artifacts (Bradley and
Hillis, 1997; Wintzingerode et al., 1997; Darius et al.,
1998).

Sequences of srDNAs obtained (as clones) from Symbio-
dinium in the M. annularis species complex are summarized
in Figure 3. Because we have evidence for only four
taxa — A. B, C, and E — the multiple branches within groups
B, C, and E represent sequence variation within, not among,
taxa. An exception to this statement is the pair of sequences
labeled C2° ' and C2"°, which came from an experimen-
tally bleached M. unnuluris and from an unmanipulated
colony of the coral Siderastrea siderea. respectively (see
Toller et al., 2001). Ecological data and RFLP analyses
strongly imply that C20'1 and C2°"2 represent a taxon '. Sym-
biodinium C2) that is distinct from the taxon Symbiodinium
C found commonly in unmanipulated Montastraea (Toller
et nl., 2001 ). We stress that this taxonomic difference could
not be inferred from srDNA sequence data alone, given the
levels of srDNA heterogeneity within samples of Synihio-
(liniiim C and C2 (Toller et al.. 2001).

In conclusion, the problem of fully interpreting srDNA
variation in natural samples of Symbiodinium is challeng-
ing. By themselves, srDNA sequence data contributed rel-
atively little to understanding zooxanthellar diversity in
Montastraea. RFLP data were much more informative, not
the least because they revealed the informational inn

358

W. W. TOLLER ET AL

srDNA sequences. Many samples of zooxanthellae from
these species of coral contained more than one taxon of
S\mbiodiniiun (Figs. 4 and 5; Rowan et ai, 1997). a phe-
nomenon that would have been difficult to understand from
srDNA sequences alone. RFLP data are easily obtained, at
reasonable cost, from many samples of zooxanthellae.
which allows ecological data to inform taxonomic deci-
sions.

Acknowledgments

We thank the Kuna Nation and the Republic of Panama
(Autoridad Nacional del Ambiente, Departamento de Cuar-
entena Agropecuaria del Ministerior de Desarrollo Agro-
pecuario. and Recursos Marinos) for permission to collect
and export specimens. Many thanks to Javier Jara for tire-
less field assistance and to Juan Mate for help with the deep
collections of M, franksi. Thanks to Ursula Anlauf and
Suzanne Williams for advice. Thanks to Ralf Kersanach and
David Kline for coral DNA and advice. David Zawada
provided samples from the Bahamas. R. R. thanks Chris
Hein and Uma Narayan for hospitality in California. This
research was supported by the Andrew W. Mellon Founda-
tion, the Smithsonian Tropical Research Institute, the
Scripps Institution of Oceanography, and the National In-
stitutes of Health.

Literature Cited

Baker, A. C. 1999. Symbiosis ecology of reef-building corals. Ph.D.
dissertation. University of Miami. 120 pp.

Baker, A. C.. and R. Row an. 1997. Diversity of symbiotic dinoflagel-
lates (zooxanthellae) in scleractinian corals of the Caribbean and East-
ern Pacific. Proc. Eighth Int. Coral Reef Sym/>. 2: 1301-1306.

Barnes. D. J., and B. E. Chalker. 1990. Calcification and photosynthe-
sis in reef-building corals and algae. Pp. 109-131 in Ecosystems of the
World Vol. 25: Coral Reefs, Z. Dubinsky. ed. Elsevier. New York.

Bowden, K. F. 1983. Ph\sicul Oceanography of Coastal Waters. Ellis
Horwood. Chichester, United Kingdom.

Bradley, R. D., and D. M. Hillis. 1997. Recombinant DNA sequences
generated by PCR amplification. Mol. Biol. E\-»l 14: 592-593.

Brown. B. E. 1997. Coral bleaching: causes and consequences. Coral
Reefs 16: Suppl. S129-S138.

Brown, B. E., R. P. Dunne, M. S. Goodson, and A. E. Douglas. 2000.
Bleaching patterns in reef corals. Nature 404: 142-143.

Carlos, A. A., B. K. Baillie, M. Kawachi, and T. Maruyama. 1999.
Phylogenetic position of Symbiodinium (Dinophyceae) isolates from
tridacnids (Bivulviu). cardiids (Bivalvia). a sponge (Porifera). a soft
coral (Anthozoa). and a free-living strain. J. Phycol. 35: 1054-1062.

Clifton. K. E., K. Kim. and J. L. Wulff. 1997. A field guide to the reefs
of Caribbean Panama with an emphasis on Western San Bias. Proc.
Eighth Int. Coral Reef Symp. 1: 167-184.

Connell, J. H. 1978. Diversity in tropical rain forests and coral reefs.
Science 199: 1302-1310.

Darius. H. T., C. Dauga. P. A. D. Grimont, E. Chungue. and P. M. V.
Martin. 1998. Diversity in symbiotic dinoflagellates (Pyrrhophyta)
from seven scleractinian coral species: restriction enzyme analysis of
small subunit ribosomal RNA genes. J. Eukaryot. Microbiol. 45: 619-
627.

D'Croz, I.., D. R. Robertson, and J. A. Martinez. 1999. Cross-shelf

distribution of nutrients, plankton, and fish larvae in the San Bias

Archipelago, Caribbean Panama. Rev. Biol. Trap. 47: 203-215.
Falkowski, P. G.. Z. Dubinsky, L. Muscatine, and J. W. Porter. 1984.

Light and the bioenergetics of a symbiotic coral. Bioscience 34: 705-

709.
Falkowski. P. G., Z. Dubinsky, L. Muscatine, and L. R. McCloskey.

1993. Population control in symbiotic corals. Bioscience 43: 606-

nl 1.
Freudenthal, H. D. 1962. Svmbiodiniuin gen. nov. and Symbiodinium

microadriaticum sp. nov., a zooxanthella: taxonomy, life cycle, and

morphology. J. Proto-ool. 9: 45-52.
Goreau, T. F. 1959. The ecology of Jamaican coral reefs 1. Species

composition and zonation. Ecology 40: 67-90.
Hill. M., and T. VV'ilcox. 1998. Unusual mode of symbiont repopulation

after bleaching in Anthosigmella various: acquisition of different zoo-
xanthellae strains. Symbiosis 25: 279-289.
Hillis, D. M., and M. T. Dixon. 1991. Ribosomal DNA: molecular

evolution and phylogenetic inference. Q. Rev. Biol. 66: 411-454.
Kirk. .1. T. (). 1994. Light and Photosynthesis in Aquatic Ecosystems.

2nd ed. Cambridge University Press, Cambridge.
Knuwlton. N., and J. B. C. Jackson. 1994. New taxonomy and niche

partitioning on coral reefs: jack of all trades or master of some? Trends

Ecol. Evol. 9: 7-9.
Knowlton, N., E. Weil, L. A. Weigt, and H. M. Guzman. 1992. Sibling

species in Montastraea annu/aris, coral bleaching, and the coral cli-
mate record. Science 255: 330-333.
Lopez, J. V., R. Kersanach, S. A. Rehner, and N. Knowlton. 1999.

Molecular determination of species boundaries in corals: genetic anal-
ysis of the Montiisrnii'u annularis complex using amplified fragment

length polymorphisms and a microsatellite marker. Binl. Bull. 196:

80-93.
McNally, K. L., N. S. Govind, P. E. Thome, and R. K. Trench. 1994.

Small-subunit ribosomal DNA sequence analyses and a reconstruction

of the inferred phylogeny among symbiotic dinoflagellates (Pyrro-

phyta). J. Phycol. 30: 316-329.
Muller-Parker, G., and C. F. D'Elia. 1997. Interactions between corals

and their symbiotic algae. Pp. 96-1 13 in Life and Death of Coral Reefs,

C. Birkeland. ed. Chapman & Hall. New York.
Muscatine, L., and J. W. Porter. 1977. Reef corals: mutualistic sym-

bioses adapted to nutrient-poor environments. Bioscience 27: 454-

460.
Robertson, D. R., and P. W. Glynn. 1977. Field guidebook to the reefs

of San Bias Islands, Panama. Third Int. Symp. Coral Reefs. University

of Miami, Florida. 15 pp.
Rowan. R. 1998. Diversity and ecology of zooxanthellae on coral reefs.

/ Phycol. 34: 407-417.

Rowan, R., and N. Knowlton. 1995. Intraspecific diversity and ecolog-
ical zonation in coral-algal symbiosis. Proc. Nail Acad. Sci. USA 92:

2850-2853.
Rowan. R., and D. A. Powers. 1991a. A molecular genetic classification

of zooxanthellae and the evolution of animal-algal symbioses. Science

251: 1348-1351.
Rowan, R., and D. A. Powers. 1991b. Molecular genetic identification

of symbiotic dinoflagellates (zooxanthellae). Mar. Ecol. Prog. Ser. 71:

65-73.
Rowan, R., and 1). A. Powers. 1992. Ribosomal RNA sequences and the

diversity of symbiotic dinoflagellates (zooxanthellae). Proc. Natl.

Aaul. Sci. USA 89: 3639-3643.
Rowan. R.. S. M. Whitney, A. Fowler, and D. Yellowlees. 1996.

Rubisco in marine symbiotic dinoflagellates: form II enzymes in eu-

karyotic oxygenic phototrophs encoded by a nuclear multigene family.

Plum Cell 8: 539-553.
Rowan, R., N. Knowlton. A. Baker, and J. Jara. 1997. Landscape

ZOOXANTHELLAE IN DIFFERENT HABITATS

359

ecology of algal symbionts creates variation in episodes of coral
bleaching. Nature 388: 265-269.

Saitiiu. N.. and M. Nei. 1987. The neighbor-joining method: a new
method for reconstructing phylogenetic trees. Mel. Biol. Evol. 4: 406-
42?.

Schoenberg, I). A., and R. K. Trench. 1980. Genetic variation in
S\mhn>diiuuin ( = Gviiinoiiiniunrt microadriaticum Freudenthal, and
specificity in its symbiosis with marine invertebrates. III. Specificity
and infectivity of Symbiodinium microadriaticum. Proc. R. Soc. Land.
B 207: 445-460.

Scholin, C. A., and D. M. Anderson. 1994. Identification of group- and
strain-specific genetic markers for globally distributed Alexandriwn
(Dinophyceae). I. RFLP analysis of SSU rRNA genes. ./. Phycol. 30:
744-754.

Scholin, C. A., and D. M. Anderson. 1996. LSU rDNA-based RFLP
assays for discriminating species and strains of Alexandriwn (Dino-
phyceae). ./. Phycol. 32: 1022-1035.

Scholin. C. A., D. M. Anderson, and M. L. Sogin. 1993. Two distinct
small-subunit ribosomal RNA genes in the North American toxic
dinoflagellate Alexandrium fitndyense (Dinophyceae). ./. Phycol. 29:
209-216.

Thompson, J. D.. T. J. Gibson., F. Plewniak, F. Jeanmougin. and D. G.
Higgins. 1997. The ClustalX windows interface: flexible strategies
for multiple sequence alignment aided by quality analysis tools. Nu-
cleic Acids Res. 24: 4876-4882.

Toller. W. W., R. Rowan, and N. Knovvlton. 2001. Repopulation of
zooxanthellae in the Caribbean corals Montaxtruca anniilari* and M.
faveolata following experimental and disease-associated bleaching.
Biol. Bull. 201: 360-373.

Trench. R. K.. and R. J. Blank. 1987. Synihiudiiiiinn iiiicroailriiiticiini
Freudenthal. S. goreaitii sp. nov.. S. kawagutii sp. nov. and S. pilmnm
sp. nov.: gymnodinioid dinoflagellate symbionts of marine inverte-
brates. J. Phycol. 23: 469-481.

Veron, J. E. N. 1995. Corals in Space and Time: The Biogeography and
Evolution of the Scleractinia. UNSW Press, Sydney, Australia.

Weil, E., and N. Knnwllon. 1994. A multi-character analysis of the
Caribbean coral Montaxtraca aniuilaris (Ellis and Solander, 1786) and
its two sibling species, M. faveolata (Ellis and Solander, 1786) and M.
franksi (Gregory. 1895). Bull. Mar. Sci. 55: 151-175.

Wesseling, I., A. J. Uychiaoco, P. M. Alino, T. Aurin, and J. E.
Vermaat. 1999. Damage and recovery of four Philippine corals from
short-term sediment burial. Mar. Ecol. Prog. Ser. 176: 11-15.

Wintzingerode, F. v., U. B. Gobel. and E. Stackebrandt. 1997. Deter
mination of microbial diversity in environmental samples: pitfalls of
PCR-based rRNA analysis. FEMS Microhiot. Rev. 21: 213-229.

Veung. P. K. K.. K. F. Kong, F. T. W. Wong, and J. T. V. Wong.
1996. Sequence data for two large-subunit rRNA genes from an
Asian strain of Alexandrium catenclla. Appl. Em-iron. Microbiol. 62:
4199-4201.

Reference: Binl. Bull. 201: 360-373. (December 2001)

Repopulation of Zooxanthellae in the Caribbean
Corals Montastraea annularis and M. faveolata
following Experimental and Disease-Associated

Bleaching

w. w. TOLLER". R. ROWAN*-* AND N. KNOWLTON

1.3

] Marine Biology Research Division 0202, Scripps Institution of Oceanography, University of California

San Diego. La Jolla, California 92093-0202; 2 University of Guam Marine Laboratory, Mangilao, Guam

96923; and * Smithsonian Tropical Research Institute, Apartado 2072, Balboa, Republic of Panama

Abstract. Caribbean corals of the Montastraea annularis
species complex associate with four taxa of symbiotic
dinoflagellates (zooxanthellae; genus Symbiodinium) in eco-
logically predictable patterns. To investigate the resilience
of these host-zooxanthella associations, we conducted field
experiments in which we experimentally reduced the num-
bers of zooxanthellae (by transplanting to shallow water or
by shading) and then allowed treated corals to recover.
When depletion was not extreme, recovering corals gener-
ally contained the same types of zooxanthellae as they did
prior to treatment. After severe depletion, however, recov-
ering corals were always repopulated by zooxanthellae
atypical for their habitat (and in some cases atypical for the
coral species). These unusual zooxanthellar associations
were often (but not always) established in experimentally
bleached tissues even when adjacent tissues were untreated.
Atypical zooxanthellae were also observed in bleached tis-
sues of unmanipulated Montastraea with yellow-blotch dis-
ease. In colonies where unusual associations were estab-
lished, the original taxa of zooxanthellae were not detected
even 9 months after the end of treatment. These observa-
tions suggest that zooxanthellae in Montastraea range from
fugitive opportunists and stress-tolerant generalists (Sym-
hiodiniiim A and E) to narrowly adapted specialists (Sym-
hiodinium B and C). and may undergo succession.

Received 9 February 2000; accepted 5 July 2001.

*To whom correspondence should be addressed. E-mail: rrowanCa'uog9.
uog.edu

Abbreviations: RFLP, restriction fragment length polymorphism:
srDNA, small subunit ribosomal RNA gene; YBD, yellow-blotch disease.

Introduction

Scleractinian reef-building corals are obligate, mutualis-
tic symbioses involving heterotrophic coral animals (hosts)
and phototrophic dinoflagellate endosymbionts in the genus
Symbiodinium (commonly called zooxanthellae). Sclerac-
tinian corals (Wells, 1956; Veron, 1995; Cairns. 1999) and
zooxanthellae (Trench. 1997; Rowan, 1998) are both taxo-
nomically diverse groups. Their symbioses, however, are
restricted to a small and specific subset of the myriad
combinations that theoretically might exist (Trench. 1988,
1993). Presumably this host-symbiont specificity is shaped
by natural selection, which favors those combinations that
perform well and can perpetuate themselves effectively
(Trench. 1988; Rowan and Powers, 1991a; Buddemeier and
Fautin. 1993). Hypotheses about coral-zooxanthellar speci-
ficity were originally shaped by the belief that corals (as
individuals or as species) associate with only one species of
Symbiodinium (Trench. 1988, 1993; see Buddemeier and
Fautin. 1993). Accordingly, any direct interactions among
different species of Svmbiodinium were thought to result in
one species of zooxanthella consistently "winning" and
therefore specifically and exclusively populating its host
(Fitt. 1985a; Trench. 1988. 1993).

In contrast to this view, we found that individual colonies
of coral in the Montastraea annularis species complex often
contain more than one taxon of Symbiodinium (Rowan and
Knowlton. 1995: Rowan et at.. 1997; Toller et al., 2001 ). At
Aguadargana reef in the San Bias Islands of Panama (see
fig. 1 in Toller et al.. 2001 ). colonies of M. annularis host
Symbiodinium B (or rarely. Symbiodinium A) in tissues
exposed to high irradiance. and they host Symbiodinium C

360

REPOPULATION OF ZOOXANTHELLAE

361

in tissues exposed to low irradiance. Colonies of M. faveo-
lata exhibit a similar pattern except that Symbiodinium A
and B are both common at high irradiance in these corals
(Rowan and Knowlton, 1995; Rowan et al.. 1997). Nearby
at Rio Carti (a near-shore habitat; see fig. 1 in Toller et al..
2001 ), members of the Montastraea annularis complex host
Symbiodinium E in tissues exposed to high irradiance and
host Svmbiodiniiim C otherwise (Toller et al.. 2001). Thus
on these two reefs, corals at shallower depths, which expe-
rience both high (on the colony top, exposed to down-
welling irradiance) and low (on colony sides) irradiance,
typically host both high- (Symbiodinium A and/or B or
Svmhiodinium E) and low- (Symbiodinium C) irradiance-
associated zooxanthellae simultaneously. (On another off-
shore reef. Svmbiodiniiim E also occurs in some of the
deepest colonies of M. franksi. possibly as a result of
sediment-associated stress [Toller et al.. 2001]).

Several observations suggest that interactions among dif-
ferent taxa of Symbiodinium within one colony of Monta-
straea may be dynamic. First, coral growth causes slow
changes in irradiance microenvironments (e.g.. corallites
moving from tops to sides of M. annularis columns), and
the specificity of different zooxanthellae for different irra-
diance environments (above) implies that zooxanthellar
communities will change in response to these irradiance
changes. Second, experimental manipulations of irradiance
gradients within colonies of M. annularis hosting Symbio-
dinium B and C resulted in changes in the distribution of
these zooxanthellae (Rowan et al.. 1997). Finally, the pro-
portions of Svmbiodiniiim A. B, and C in Montastraea
changed during a coral bleaching event (Rowan et al..
1997).

The present study tested the ability of zooxanthellar sym-
bioses in M. annularis and in M. faveolata to reestablish
typical patterns of association after being disturbed. Be-
cause zooxanthellae in unmanipulated corals have such en-
vironmentally predictable patterns of distribution (above),
we hypothesized that disturbed zooxanthellar populations
would re-establish the same patterns of association, directly.
To disturb zooxanthellae, we treated corals with low light
(e.g., Franzisket. 1970) or with high light (e.g., Dustan,
1979). both of which caused corals to lose zooxanthellae (to
bleach). Corals were then allowed to recover. We also
studied the zooxanthellar communities of unmanipulated
corals that exhibited yellow-blotch disease and associated
reductions in zooxanthellar numbers.

Materials and Methods

Experimental manipulations

Experiments were conducted between October 1997 and
October 1998 at Aguadargana reef, San Bias Archipelago.
Republic of Panama (see fig. 1 in Toller et al.. 2001 ). Time
courses of experiments (not always optimal) were dictated

4 or 7 weeks

Donor
(#1-12)

Figure 1. Experimental and sampling designs. (Al Experiment I, in
which three columns (gray is live tissue; black is nonliving base) from each
of 15 donor colonies of Montastraea annularis living at 9-10 m depth
(#1-15. corresponding to Corals 1-15 in Figs. 3 and 4) were transplanted
to 1 m depth (box labeled High Light), to 9 m depth (Control), and to a
cave (Low Light) for treatment for a period of 4 (Colonies 1 1-15) or 7
(Colonies 1-10) weeks; corals were then transplanted to common gardens
at 9 m depth. Samples of zooxanthellae (Initial. Treatment. Recovery.
Final) were taken at the times indicated (open arrows); longer times apply
to Colonies 1-10; shorter times to Colonies 1 1-15. (B) Experiment II. in
which cores cut from 12 colonies of M. faveolata (#1-12. corresponding to
Corals 1-12 in Fig. 5) were transplanted to caves (Low Light) for 6 weeks,
and then to a common garden at 9 m depth. Samples of zooxanthellae
(Initial, Treatment. Recovery. Final) were taken (open arrows) at the times
indicated. (C) Experiment III, in which one column on each of 14 colonies
of M. annularis was half-covered with aluminum foil for 4 weeks, then
uncovered for 12 weeks (nail marks the treatment boundary), and then
sampled on the top of each half (open arrows. Untreated and Treated; data
in Fig. 6). (D) Schematic of yellow-blotch disease (YBD) on Monfastraea
(polygon), showing concentric halos of yellow (light gray color) and
bleached (white color) tissue around dead skeleton (Dead), and surrounded
by normally pigmented tissue (Normal). Black dots and arrows indicate
places where zooxanthellae were sampled (Normal. White. Yellow; data in
Fig. 7).

by the imminent closure of the Smithsonian Institution's
field station. Experiments used parts of apparently healthy,
large colonies of Montastraea annularis (Experiments I and
III) and M. faveolata (Experiment II). as described below
and in Figure 1. These large donor coral colonies were
separated from one another by more than 5 m. ("oral tissues

362

W. W. TOLLER ET AL

were sampled with a #6 hole punch, which yields a sample
(small core) with about 0.24 cm of coral tissue, or with a
#12 hole punch (ca. 1.3 cm2 of tissue). These samples were
wrapped in aluminum foil and frozen in a cryogenic dry
shipper (chilled with liquid nitrogen) in the Held, and then
stored in the laboratory at — 80 °C until analysis.

Experiment 1 (Fig. /A). In December 1997, three col-
umns of similar size (en. 7-10 cm diameter) were collected
from each of 10 colonies of M. annularis living at a depth
of 9-10 m. (Colonies of M. annularis consist of clusters of
columns, each of which is covered distally with living
tissue; see Fig. 1 A.) The columns were broken off at their
nonliving bases, labeled, and a sample (#6 hole punch) was
taken from the top of each one (Initial samples). The three
columns from each colony were then distributed among
three treatments: one was transplanted to an open site on the
reef crest at a depth of about 1 m (high-light treatment; High
Light in Fig. 1A): one was transplanted to a cave at 14m
(low-light treatment; Low Light in Fig. 1A): and one was
transplanted to an open site at 9 m (treatment control;
Control in Fig. 1 A). The cave was a crevice (ca. 2.5 m deep,
a/. 1 .5 m wide, and < 1 m high) that was completely shaded
from downwelling irradiance, largely shaded from other
irradiance. and lacked conspicuous photosynthetic organ-
isms. For low-light treatment, coral columns were mounted
upright on PVC posts set in blocks of concrete, using nylon
cable ties to secure the columns at their nonliving bases.
These blocks then were placed in the back of the cave.
Control and high-light-treated coral columns were affixed in
an upright position to wire grids using cable ties, and these
grids were secured to the reef by wedging them into sub-
strate and covering them with rubble.

After 7 weeks of treatment, all coral columns were col-
lected, assessed visually, and sampled (#6 hole punch;
Treatment samples). They were then mounted on wire grids
in an upright position and placed in a common garden at 9 m
depth, with unobstructed irradiance, for the remainder of the
experiment. There were six grids, each with rive columns
arranged analogously to a Latin square with respect to
treatment. After 17 weeks, all columns were assessed and
sampled again (#6 hole punch; Recovery samples). Nine of
the 30 columns were assessed and sampled once more (#6
hole punch; Final samples) after a total of 37 weeks in the
common garden; the other 21 columns had been lost to
vandals by that time. All samples (Initial. Treatment, Re-
covery, and Final) were taken from the tops of columns,
within an area (ca. 1 cm2) over which zooxanthellar iden-
tities do not vary much or at all in unmanipulated columns
of M annularis (Rowan et <//.. 1997).

A second experiment was done (starting in January 1998)
at a different location (ca. 0.5 km away). The second
experiment differed from the first one only as follows: five
colonies (15 coral columns) of M. annularis were used,
treatment was for 4 rather than 7 weeks. Recovery samples

were obtained after 1 3 rather than 1 7 weeks, and Final
samples were obtained (from all columns) after a total of 33
rather than 37 weeks.

Experiment II (Fig. IB). In October 1997, one core was
removed from each of 12 large (> 1.5 m tall and wide)
colonies of M. faveolata living at depths of 1-9 m. using a
pneumatic drill fitted with a 44-mm hole saw (resulting
cores had ca. 12.6 cm2 of live tissue and were ca. 5 cm in
height). At this time, tissue samples (Initial samples) were
taken immediately adjacent to the coring sites with a steel
hole punch (#12). Coral cores were then transplanted among
three small caves (Low Light, as above) on the reef (7-1 1 m
depth), where they were secured with plastic cable ties to
masonry nails pounded into reef framework. Cores occu-
pied the back (darkest) portion of the caves and were
mounted upside-down on the cave ceilings.

After 6 weeks of low-light treatment in caves, coral cores
were collected and assessed visually; tissue samples were
taken from each coral core at a haphazardly selected loca-
tion away from the core's perimeter (#6 hole punch; Treat-
ment samples). Cores were then attached to cleared reef
substrate at 9 m depth with epoxy putty (Z-Spar Splash
Zone, Kop-Coat, Inc., Pittsburgh, PA), facing upward under
unobstructed natural irradiance. After 8 weeks in this com-
mon garden and then again after another 16 weeks, coral
cores were assessed and sampled again (#6 hole punch;
Recovery samples and Final samples, respectively). In the
latter case (Final samples), only eight coral cores were
sampled — the four others were lost.

Experiment III (Fig. 1C). In January 1998, individual
columns of M. annularis were each half-covered with a
shield of aluminum foil. This treatment bisected each col-
umn vertically into two morphologically equivalent halves,
one of which was covered by foil and therefore low-light
treated (Treated) and the other of which was exposed to
natural irradiance (Untreated). Each shield was molded to
its column, lifted off slightly (<0.5 cm), and secured to the
column's nonliving base with nylon cable ties. A shield was
placed on one column of each of 15 colonies living at depths
of 2-4 m (shallow group), and on one column of each of 15
colonies living at depths of 7-9 m (deep group). Shields
were removed after 4 weeks, at which time treatment
boundaries were marked by gently tapping two small steel
nails into opposite sides of each column. After another 12
weeks, columns were assessed visually and a pair of tissue
samples was taken from the top of each, 2 cm apart and on
either side of the treatment boundary (#6 hole punch;
Treated and LIntreated samples. Fig. 1C).

Yellow-blotch disease

In October 1997 and January 1998 at Cayos Limones.

San Bias (see fig. 1 in Toller ct <;/., 2001), we found a
number of colonies of Montastraea that appeared to have

REPOPULATION OF ZOOXANTHELLAE

363

-yellow-blotch disease" (YBD: Santavy et <;/.. 1999). Some
colonies had only one or two small lesions (fa. 10-30 cm
wide), which usually consisted of a patch of exposed skel-
eton surrounded by a halo (typically fa. 1-3 cm wide) of
yellow living tissue, which in turn was surrounded by a halo
(typically <2 cm wide) of white (bleached) tissue: lesions
were surrounded by apparently healthy tissue (see Fig. ID).
Other colonies were mostly dead, in which case a patch of
normal tissue was surrounded by a band of bleached tissue
inside a band of yellow tissue.

Using a steel hole punch (#12), we took samples from
rive colonies of M. franksi (one lesion per colony), from six
colonies of M. faveolata (one or two lesions per colony),
and from one colony of M. annnlaris (two lesions). Two
samples were taken at every lesion — one of normally pig-
mented tissue and one of yellow tissue nearby (S3 cm
apart: Normal and Yellow, respectively; see Fig. ID). At
rive lesions we also sampled the white tissue that was
between yellow and normal tissue (White: see Fig. ID).

Progression of YBD was monitored in 12 colonies of M.
faveolata. On 25 January 1998, two small nails were driven
into the bare skeleton next to one YBD lesion in each
colony. The two nails defined a line parallel to the lesion
edge, and the distance between that line and the lesion edge
(living, yellow tissue) was measured with a pair of calipers.
We also measured the distance to normally pigmented tis-
sue, along the same vector. These measurements were re-
peated 5 months later (28 May 1998).

Laboratory methods

Zooxanthellae were isolated from frozen samples as de-
scribed previously (Rowan and Powers, 1991b: Rowan and
Knowlton. 1995). except that skeletal cores or fragments,
after being stripped of tissue, were broken apart with a steel
spatula and then washed with isolation buffer. That wash
was combined with the tissue that had been stripped from
the sample previously. At this point, one-tenth of each
sample was fixed in 10% formalin and stored at 4 °C for cell
counts, which were obtained from eight subsamples of each
sample by hemacytometry. The rest of each sample was
used to prepare DNA as described previously (Rowan and
Powers. 1991b: Rowan and Knowlton. 1995).

Zooxanthellae in each sample were identified by restric-
tion fragment length polymorphism (RFLP) genotypes of
small ribosomal subunit RNA genes (srDNA). as described
previously (Rowan and Powers. 1991b; Toller el ai, 2001 ).
Each sample was analyzed at least twice — once by srDNA
amplification with universal PCR primers (ss3 and ss5;
Rowan and Powers. 1991b). and once by srDNA amplifi-
cation with host-excluding PCR prim (ss3Z and ss5; Rowan
and Powers, 1991b: Toller et ai. 2001). All amplified
srDNAs were digested with Dpn II and with Taq I, and then
compared to standard srDNA genotypes of Symbiodiniwn

A, B. C. and E (srDNA clones A". B". C°. and E0"1,
amplified and digested the same way: see Toller et ai,
2001). RFLP genotypes C2 (in two samples) and C (in 12
samples) were compared in greater detail using a total of 12
restriction enzymes: Alii I. B.sfN I. Bst\J I. Dpn II, Hae III.
Hha I. Hinf\, Mho I. Mse I. Msp I. Sau96 I. and Taq I.

Samples that contained more than one zooxanthellar
RFLP genotype were compared to a series of synthetic
mixtures of cloned srDNAs (srDNA clones A". B". C", and
E°~' and srDNA clone C2"~', see below) to estimate the
relative abundance of each genotype in the sample (Rowan
et al.. 1997: Toller et ai, 2001 ). For graphical presentation,
these estimates were multiplied by the sample's total cell
number (see above) to estimate the number of cells of each
genotype in the sample, and these values were then con-
verted to numbers of cells per square centimeter of live
coral surface (number of zooxanthellae/cnr in Figs. 3-7).

Samples with low numbers of Zooxanthellae (<4 X 10s
cells/cm2 of coral) yielded little zooxanthellar srDNA when
srDNAs were PCR-amplified in the usual manner. To obtain
more srDNA from such samples in Experiment I, we used
two rounds of amplification (Roux, 1995) as follows. Sam-
ple srDNAs were amplified with host-excluding PCR prim-
ers over 34 cycles of the PCR. Aliquots (10 ju.1) of those
amplifications were electrophoresed on agarose gels (1.0%
Nuseive GTG: FMC BioProducts, Rockland, ME), and faint
bands of srDNA were excised and added to 100 ju,l of water.
These gel-purified srDNAs were heated to 65 °C for 2 min.
and then 1 /id of each one was PCR-amplified with host-
excluding primers in the same manner. The resulting re-
amplified srDNAs were then analyzed as described above.

srDNA with an RFLP genotype distinct from Synibio-
tliiiinni A". B". C", and E""1, here called C2. was cloned and
sequenced using methods described previously (Toller et
a!., 2001 ). It was amplified with host-excluding PCR prim-
ers from a colony of M. annularis in Experiment III (clone
C2°~') and from an unmanipuluted colony of the coral
Siderastrea siderea (clone C20'2). DNA sequences were
deposited in GenBank [http://www.ncbi.nlm.nih.gov/: ac-
cession numbers AF238259 (C2"-'), AF238260 (C2°~2)].

Results

RFLP genotypes of Zooxanthellae in experimental corals

Using the restriction enzymes Dpn II and Taq I. we
scored six different RFLP genotypes of srDNA in samples
of Zooxanthellae (Fig. 2). As explained below, genotypes A.

B, C, C2, and E represent the taxa Symbioiliniiim A. B. C,
C2, and E. RFLP genotype N (Fig. 2. lane N°~') is not a
taxon of Svmbiodiniuin and is instead related to protozoa of
the phylum Apicomplexa (Toller et al.. in press). Using our
methods (above), genotype N was observed only in six
corals — all of these from Experiment I (low-light treatment)
and only in samples taken immediately after treatment (see

364

W. W. TOLLER ET AL.

Figure 2. RFLP genotypes of Svmhiinliiiiuin. The same srDNAs were
amplified with host-excluding PCR primers (A) and with universal PCR
primers (B). and then digested with Dpn II (upper panels) and with Taq I
(lower panels). Clones are srDNA standards for genotype N (clone N°~l:
see Results). Mtmlastraea annularis (clone H"), Symbiodinium B (clone
B"}. SvmhiiiJinium C (clone C"), Svmbioiliniuni A (clone .4"). Symbio-
dimiiin E (clone £""'), and Symbiodinium C2 (clone C2° '). Samples are
zooxanthellae from Moniastraea scored as Symbiodinium C2 (B; C2).
Symbioilinitim C2 and Symbititlinium E in approximately equal amounts (A
and B; C2~E), more Symbiodinium A than Symbiodinium E (A and B; A >
El. more S\mbiodinium E than Symbiodinium A (A and B; E > A), more
Symbiodinium A than Symbiodinium C2 (A and B; A > C2). more
Syinbimliniiim C2 than Symbiodinium A (A and B: C2 > A), and Symbio-
iliiiium A and C2 in approximately equal amounts with more Symbiodinium
E (A and B; E > A— C2). Lane M contains DNA fragment size standards
of (top to bottom) 1500 base pairs (bp). 1200 bp. and then 1000 hp to 100
bp in 100-bp increments.

below). Further observations on genotype N are presented
elsewhere (Toller et til., in press).

The srDNA genotypes of Symbiodinium A. B, C, and E
are represented by cloned srDNAs A". B°. C". and E°~'
(respectively), and these genotypes differ from one another
in both Dpn II and Tat/ I digests (Toller et til., 2001; Fig.
2 A. B). RFLP genotype C2. represented in Figure 2 A by a
cloned srDNA (C2"~': below) and in Figure 2B by a sample

of zooxanthellae (C2), has not been found in unmanipulated
colonies of Montastraea annularis and M. faveolata
(Rowan and Knowlton. 1995; Rowan et al., 1997; Toller et
til., 2001; this study and unpubl. obs.). Genotype C2 is
distinguished from genotypes C and E only when both Dpn
II and Taq I digests are examined together (Fig. 2A, B).

srDNA of genotype C2 appears to lack a Dpn II restric-
tion site relative to srDNA of genotype C (Fig. 2). Defined
by this character, genotype C2 was found previously in
various other species of host (R. Rowan and W. Toller,
unpubl. obs.). Cloned srDNAs C" (which represents Sym-
biodinium C; Toller et til., 2001). C20'1 (genotype C2 from
Experiment III), and C2°~2 (genotype C2 from Siderastrea
siderea. collected nearby) differed from one another in
nucleotide sequence by about 0.9% (not shown; see also
Toller et til.. 2001). That amount of srDNA sequence dif-
ference could imply that these three clones represent three
species of Symbiodinium (e.g., McNally et al.. 1994), or one
species of Symbiodinium in which srDNA is heterogeneous
(see Toller et n!.. 2001).

We further compared the two samples from which clones
C20"1 and C2°~2 were obtained to 12 samples of Symbio-
dinium C (samples in fig. 9 in Toller et al.. 2001) by
digesting srDNAs with 12 restriction enzymes (listed in
Materials and Methods). The two samples of genotype C2
were indistinguishable and differed from Symbiodinium C
only in Dpn II (above) and Mse I digests (not shown). This
analysis also showed that srDNA was heterogeneous in all
samples, which means that zooxanthellae in the samples
cannot be described precisely by sequences of cloned
srDNA (i.e., clones C". C20"1, and C2""2; see above and
Toller et til.. 2001). Nevertheless, RFLP data indicate that
genotype C2 represents a taxon of zooxanthella, Symbio-
dinium C2, that is distinct from the Symbiodinium C that
occurs commonly in M. annularis and M. faveolata.

Examples of RFLP genotypes that we interpreted as
mixtures of taxa of Svinbiodiniinn are shown on the right
side of Figure 2 (Samples). The figure compares data ob-
tained by amplifying srDNAs with host-excluding (Fig. 2A)
vcrxus universal (Fig. 2B) PCR primers because, using Dpn
II and Taq I, both sets of data are needed to distinguish
mixtures of genotypes A and C2 (e.g.. Fig. 2, A > C2 and
C2 > A) from mixtures of genotypes A, C2, and E (e.g..
Fig. 2, E > A-C2). Universal PCR primers also amplify
coral host srDNA (Fig. 2B, clone H°), but none was de-
tected in the samples of zooxanthellae shown in Figure 2B.

Experiment I

The experimental units were individual coral columns
taken from 15 donor colonies of M. annularis (see Fig. 1 A).
Below, a column is identified by the colony from which it
came and by its treatment group (e.g.. Colony 1, High
Light). Columns 1-10 were in the first experimental group:

REPOPULATION OF ZOOXANTHELLAE

365

Colony Number

Figure 3. Zooxanlhellae observed in Montaslnwa annularis in Exper-
iment I. Panels labeled Control (top). High Light (middle), and Low Light
(bottom) present data from corals in the treatments labeled as such in
Figure 1A. Panel divisions labeled Initial, Treatment, and Recovery (in
ovals at top) present data from samples labeled as such in Figure 1 A (ovals
with open arrows). Colony Number (horizontal axes) identifies data ob-
tained from different coral columns; data with the same Colony Number
within a panel are different samples from the same column: data with the
same Colony Number in different panels are samples from different col-
umns taken from the same donor coral colony (see Fig. I A). Colonies 1-10
are from the first experimental group, and Colonies 1 1-15 are from the
second experimental group (see Methods). Bars indicate the taxa of zoo-
xunthellae (by shade, according to the key in the middle panel) and the
number of /ooxanthellae (by height, normalized to I crrr of coral surface)
observed in each sample. Where bars are too short to be legible, zooxan-
thellar identities are given by the arrows labeled C (Synihiotliiiiiini C), E
(S\mbi<Hlinium E), and 10% E (together with 90% Symbiodinium B).
Samples in which zooxanthellae were not identified are indicated with a
theta (0). Samples that contained RFLP genotype N (see text) are indicated
with asterisks ( ").

Columns 1 1-15 were in the second group (see Materials and
Methods). Before treatment, samples from most coral col-
umns contained Symbiodinium C (Fig. 3, Initial). Excep-
tions were two donor colonies that yielded only Symbio-
dinium B (Fig. 3, Initial; Colonies 14 and 15 in all treatment
groups), and one column that yielded Symbiodinium B with
a small amount of Symbiodinium E (Fig. 3, Initial; Colony
7, Low Light). Initial zooxanthellar numbers in the three
treatment groups (Control, High Light, and Low Light; Fig.
3) were indistinguishable (31.5 ± 10.1 X 105, 31.9 ±

9.80 X 10s. and 31.4 ± 7.10 X 10s zooxanthellae/cm2 of
coral, respectively).

Zooxanthellar numbers decreased after treatment. Sam-
ples from high-light-treated coral columns (Fig. 3, High
Light, Treatment) had. on average. 29% as many zooxan-
thellae as did samples from controls (10.7 ± 9.0 >' l(f
versus 36.6 ± 17.5 X 10s zooxanthellae/cm2 [means ±
standard deviations]; Wilcoxon signed rank test, P •
0.001). Two high-light-treated columns (Colonies 14 and
15; with Symbiodinium B before treatment) appeared nor-
mal; the other 13 (with Symbiodinium C before treatment)
were pale or bleached, but only on their tops and south-
facing (sun-facing) sides. All low-light-treated columns
(Fig. 3. Low Light. Treatment) were white, and samples
from them had, on average, only about 2.5% as many
zooxanthellae as did samples from controls (0.90 ± 1.4 X
[Q? versus 36.6 ± 17.5 X 105 zooxanthellae/cm2; Wilcoxon
signed rank test. P < 0.001). Zooxanthellar identities did
not change in samples from the tops of coral columns in the
control group after 4 or 7 weeks (Fig. 3, Control, Treatment
versus Initial), with one exception. That exception was
Colony 14. which yielded only Symbiodinium B initially but
yielded Symbiodinium B with a small amount of Symbio-
dinium E 4 weeks later. In high-light-treated columns one
change was observed (Fig. 3. High Light, Treatment versus
Initial): Colony 5 initially yielded Symbiodinium C but
yielded roughly equal parts of Symbiodinium C and Sym-
biodinium E immediately after treatment.

Identities of zooxanthellae were difficult to determine in
low-light-treated coral columns at the end of treatment,
presumably because these columns contained so few zoo-
xanthellae (above). Two rounds of PCR amplification (see
Materials and Methods) allowed 13 samples to be analyzed
(Fig. 3, Low Light, Treatment); no srDNA was obtained
from the other two samples. Six samples contained Symbio-
diniuin C. one contained Symbiodinium E (Colony 7. which
contained some Svmbiodinium E initially), and six yielded
only a aon-Symbiodinium RFLP genotype (genotype N; see
above).

After a total time of 17 or 24 weeks, zooxanthellar
numbers and RFLP genotypes in samples from the control
group were similar to initial conditions; Colony 14 once
again yielded only S\mbiodinium B (Fig. 3, Control; Initial.
Treatment. Recovery). High-light-treated columns, which
had then spent 13 or 17 weeks in their original, deeper
environment (Fig. 3, High Light. Recovery) regained col-
oration (1 1 normal, 4 pale on tops only) and zooxanthellar
numbers (Recovery versus Treatment: 45.4 ± 34.2 x 10^
versus 10.7 ± 9.0 X 10s zooxanthellae/cm2: Wikovm
signed rank test. P < 0.001 ). At this time, /.ooxanthellar
numbers in samples from high-light-treated columns were
similar to those in samples from controls (45. •
10? versus 35.2 ± 18.5 X 105 zooxunl!< llai > espec
lively; Wilcoxon signed rank test, P > 0. 1 )

366

W. W. TOLLER ET AL.

Thirteen or 17 weeks after the end of treatment, 10 of 15
high- light-treated coral columns had the same taxa of Sym-
biodinium that they contained before treatment, but 5 of 15
coral columns apparently contained different taxa of Sym-
biodinium than they began with (Fig. 3, High Light, Recov-
ery). One of these taxonomic differences had been observed
at the end of treatment (Colony 5). The other four changes
(Fig. 3. High Light. Initial versus Recovery) were Symbio-
dinium C to Symbiodinium A and C2 (Colony 1); Svmbio-
dinimn C to Symbiodinium A (Colony 2); Symbiodinium C
to Symbiodinium C, A. and E (Colony 3); and Symbiodinium
C to Symbiodinium C and E (Colony 4). These four columns
represented four of the five columns with the lowest num-
bers of zooxanthellae following treatment.

All low-light-treated coral columns experienced major
changes in zooxanthellar populations after 13 or 17 weeks
back in their original environment (Fig. 3, Low Light.
Recovery). Zooxanthellar numbers were about 100-fold
higher than after treatment (Recovery versus Treatment:
111 ± 45.2 X 105 versus 0.90 ± 1.4 X 105 zooxanthellae/
cm2; Wilcoxon signed rank test, P < 0.001), and were
about 3-fold higher than in the control group (35.2 ±
18.5 X 105 zooxanthellae/cm2; Wilcoxon signed rank test,
P < 0.001 ). Only Colony 2 appeared normal; the other 14
columns, despite their large numbers of zooxanthellae, were
still pale at this time (May 1998).

No low-light-treated coral column contained the same
zooxanthellae that it had originally (Fig. 3. Low Light.
Recovery versus Initial). Eight of them contained mixtures
of taxa, and Symbiodinium A was predominant, followed by
Symbiodinium E and C2; Symbiodinium C and B were not
detected (Fig. 3, Low Light, Recovery). The predominance
of Symbiodiniiini A was observed primarily in Colonies
11-15 (sampled 13 weeks after treatment); among Colonies
1-10 (sampled 17 weeks after treatment) only four samples
contained more than 50% Symbiodiniiini A.

Twenty-four of the above coral columns (n = 8 colonies)
were sampled again, for the last time, 33 or 37 weeks after
treatment. At this time ( 18 October 1998), many unmanipu-
lated colonies of M. annularis. M.faveolata, and M. franks i
living at depths of 8-14 m at our study site were pale or
bleached on their upper surfaces. Weekly mean sea-surface
temperatures near our study site ranged between 29.4 °C
and 29.9 °C from 26 August to 7 October (9.5 °N. 78.5 °W;
data from Integrated Global Ocean Services System, http://
ingrid.ldgo.columbia.edu/SOURCES/.IGOSS ). Historically,
temperatures this high are associated with coral bleaching at
our study site (see fig. 3e in Rowan el al, 1997). Moreover,
most of the summer of 1998 was unusually warm: from the
first week of May through the first week of October (23 weeks)
in the years 1981 to 1997 (but excluding 1983 and 1995, when
corals bleached), there were an average of 3.5 weeks of aver-
age sea-surface temperature at or above 29.0 °C near our study
site; for this period in 1998, there were 16 such weeks (data

Colony Number

Figure 4. Zooxanthellae observed in Montastraea annularis at the last
two sampling times of Experiment I. Data are presented as in Figure 3,
using the same system to number colonies. Data under Recovery (oval,
above) are the same data from Figure 3. and data under Final (oval, above)
are from samples from the same corals 20 weeks later (see Fig. 1 A). When
Final samples were collected, corals were scored as normal (N). slightly
pale (SP). or pale iP).

from IGOSS, as above). However, bleaching appeared to be
less severe than in October 1995 (Rowan et ai, 1997; pers.
ohs.) in terms of the number of colonies of Montastraea
affected, the number of species of coral affected, and the extent
to which individual corals were bleached.

Results are presented in Figure 4 (Final), in comparison
to Recovery samples (13 or 17 weeks after treatment) from
the same columns (data from Fig. 3). When final samples
were collected, six of eight control coral columns appeared
pale or slightly pale [labeled (P) and (SP), respectively, in
Fig. 4] on top, as were six of eight high-light-treated col-
umns (Fig. 4, Final); the other two columns in each group
hosted Symbiodinium B and appeared normal [labeled (N)].
Twenty weeks earlier (Recovery) only one of these 16
columns appeared pale (Colony 1. High Light) and all
others appeared normal. As suggested by the increase in
numbers of pale colonies, average zooxanthellar numbers
decreased and were about 5-fold lower than in the previous
samples (Final versus Recovery: control columns, 6.7 ± 4.0
versus 35.2 ± 18.5 X 105 zooxanthellae/cm2; high-light-
treated columns, 9.4 ± 4.3 versus 45.4 ± 34.2 X 105
zooxanthellae/cm2; Wilcoxon signed rank tests, P = 0.01 ).

In contrast, only two low-light-treated coral columns
(Colonies 1 and 12) appeared pale or slightly pale in Octo-
ber, and six appeared normal (Fig. 4, Low Light, Final). All

RFPOPULATION OF ZOOXANTHELl.AE

367

Coral Number

Figure 5. Zooxanthellae observed in Montastraea faveolata before and after low-light treatment (Experi-
ment II). The data are presented as in Figures 3 and 4. for the experiment diagrammed in Figure IB; there was
no control group in this experiment. Corals 8-11 were lost before Final samples were collected (8-11. no data).

eight of these columns were pale 20 weeks earlier. Contrary
to expectation, the overall increase in pigmentation was
accompanied by an average decrease of about 3-fold in
zooxanthellar number (Final versus Recovery: 33.6 ± 16.1
versus 111 ± 45.2 X 105 zooxanthellae/cnr; Wilcoxon
signed rank test. P < 0.02), to numbers comparable to
those in control, normally pigmented columns at previous
sampling times (e.g.. Fig. 4, Low Light, Final versus Con-
trol, Recovery).

Final samples from control columns contained the same
taxa of Symbiodinium that were observed previously in
those columns (Fig. 4; Control, Recovery). This was also
true for most high-light-treated columns, although in one of
these Svmbiodinium B was found with Symbiodinium C
(Fig. 4, High Light. Final; Colony 7), whereas only Sym-
biodinium C was detected in that column previously. In
contrast, we found different zooxanthellae (relative to Re-
covery) in six of eight low-light-treated columns (Fig. 4,
Low Light, Final versus Recovery). In final samples from
low-light-treated corals overall, Symbiodinium A declined,
Symbiodinium E increased to become predominant, and
Svmbiodinium C2 appeared in different columns, compared
to samples taken 20 weeks earlier. Symbiodinium C and B
were not detected.

Experiment II

Cores taken from colonies of M. faveolata (see Fig. IB;
these cores are hereafter referred to as "corals") living at
depths between 1 and 9 m contained a variety of zooxan-
thellar taxa before treatment (Symbiodinium A. B, C; Fig. 5.
Initial), as expected based on earlier surveys (Rowan and
Knowlton, 1995; Rowan et ai. 1997). After low-light treat-
ment, all 12 corals were white and contained, on average,
about 2.6% as many zooxanthellae as they began with (Fig.
5, Treatment versus Initial; 0.8 ± 1.0 X 10? versus 30.8 ±

10.2 X 10^ zooxanthellae/cnr. respectively; Wilcoxon
signed rank test. P = 0.002). Zooxanthellae were identified
in only two samples; both contained Svmbiodiniitm A, and
were from corals that had mixtures of A and B before
treatment (Corals 3 and 4; Fig. 5).

Eight weeks after the end of treatment, zooxanthellar
numbers were about 2-fold higher than before treatment
(Fig. 5, Recovery versus Initial; 71.6 ± 40.5 X 10s versus
30.8 ± 10.2 X 10s zooxanthellae/cnr, respectively; Wil-
coxon signed rank test, P = 0.01 ). Corals 1 and 6 appeared
normal. Coral 12 was bleached (and had very few zooxan-
thellae: Fig. 5, Recovery), and the other eight corals ap-
peared pale. Only Svmbiodinium A was detected at this time
(Fig. 5, Recovery), in contrast to the typical pattern for M.
faveolata in this habitat, which host Symbiodinium C
(Rowan and Knowlton. 1995).

Eight corals were sampled after a further 16 weeks in
their common garden (24 weeks total time after the end of
treatment, at the end of May 1998, prior to the bleaching
event noted above). Six of them appeared normal and two
(Corals 5 and 6) were pale. Zooxanthellar numbers re-
mained high on average (1 1 1.1 ± 55.6 x 105 zooxanthel-
lae/cnr). Samples from six corals contained only Symbio-
dinium A; samples from the other two corals contained
Symbiodinium A and E (Fig. 5, Final).

Experiment III

When the foil treatment shields (see Fig. 1C) were re-
moved, all treated tissues were white. Adjacent tissues that
had not been covered appeared normal, and borders between
the white (treated) tissue and the normal (untreated) tissue
were sharp. Many corals had suffered partial mortality in
covered areas during treatment; further observations were
made only on those in which more than 50% of the treated
tissue appeared healthy (n = 5 in the shallow group: n =

368

W. W. TOLLER ET AL

150

100

!§ 5°

I Symbiodinium A
} Symbiodinium B
] Symbiodinium C
] Symbiodinium C2
| Symbiodinium E

6789
Coral Number

11 12 13 14

Figure 6. Zooxanthellae observed in untreated and in treated parts of the same column of Montastraea
iinniiliiris (Experiment III; see Fig. 1C). Corals 1-5 lived at 2-4 m depth. Corals 6-14 lived at 7-9 in depth. Data
from untreated (Untreated) and from treated (Treated) parts of the same coral column are paired (left bar and
right bar, respectively, as shown for Coral 3). Asterisks (*) identify coral columns in which samples from both
treated and untreated tissue appeared normal (see text). Otherwise, data are presented as in Figures 3-5.

in the deep group). To avoid additional stress to the corals,
no samples were taken immediately after treatment.

The corals had different appearances 12 weeks after
treatment. In four shallow corals and one deep coral (Fig. 6,
asterisks; Corals 1, 2, 3. 5, and 6) it appeared that normal
pigment had spread from untreated into treated tissue by
about 2-3 cm, so that treatment boundaries were no longer
apparent. In these corals, samples taken from either side of
the treatment boundary (see Fig. 1C, Treated and Untreated)
were normally pigmented; they also had similar numbers of
zooxanthellae, of the same taxon of Symbiodinium. The taxa
were those expected in shallower (Corals 1. 2, 3, 5; Sym-
biodininm B or A) and deeper (Coral 6; Symbiodinium C)
colonies of M. unnularis at this location (Rowan and
Knowlton. 1995; Rowan et <//., 1997).

In the other nine corals (Fig. 6: Corals 4, 7-14) treatment
boundaries were still obvious 12 weeks after treatment.
Untreated tissues appeared normal, whereas treated tissues
were unevenly pigmented and pale overall, and samples
taken from either side of the treatment boundary had dif-
ferent taxa of Symbiodinium. Untreated halves contained the
expected taxa (Symbiodinium B, C, or, rarely in deeper
water, some A); treated halves contained, in order of de-
creasing occurrence, Symbiodinium A. C2. E. and B or C'
(Fig. 6; Corals 4, 7-14). In samples from three of these nine
corals (Corals 4, 7, and I 1 ). the taxon of Symbiodinium
found in the untreated half was also found in the treated
half, but — except where that taxon was Symbiodinium A
(Colony 1 1 ) — it was relatively minor in the treated tissue.
Zooxanthellar numbers were variable among samples from
treated halves (Fig. 6); overall, there was no significant
difference in zooxanthellar numbers in samples from treated
versus untreated halves of corals.

Disease-associated disturbance of zooxanthellae

We marked YBD lesions in 12 colonies of M. faveolata
and observed that mortality progressed by II ±6 mm
(mean ± standard deviation) during 5 months. Yellow and
white halos (see Fig. ID) progressed in concert with mor-
tality. Thus, as YBD spreads across a coral, it appears that
tissue first loses zooxanthellae (white), then partially recov-
ers zooxanthellae (yellow), and then dies. Average numbers
of zooxanthellae in samples of normal (31.8 ± 10.1 X 10~
zooxanthellae/cnr), white (3.7 ± 1.8 X 1CP zooxanthellae/
cm2), and yellow (24.6 ± 16.1 X 105 zooxanthellae/cnr)
tissues confirmed that hypothesis.

With one exception (Fig. 7; Colony 9), normal and yel-
low samples from the same lesion contained different taxa
of Svmbiocliniiini. Samples of normal tissues contained the
taxa that are common in unaffected corals at these depths
(Rowan and Knowlton, 1995; Rowan et ai, 1997; Toller et
ul., 2001) — predominantly Symbiodinium C — and yellow
tissues contained predominantly Symbiodinium A (esti-
mated at > 50% of the total in samples from 12 of 15
lesions; Fig. 7). Yellow tissue also often (9 of 15 samples;
Fig. 7) contained the zooxanthellae found in the adjacent
normal tissue. We could identify zooxanthellae in two sam-
ples of white tissue (Colony 3 and Lesion 7-1, Fig. 7); they
contained mixtures of the taxa found in the adjacent normal
(Svmhiotliiiiiim C) and yellow (Symbiodinium A) tissues.

Discussion

Taxonomic identities of zooxanthellae

In laboratory studies of establishment or re-establishment
of symbiosis between Symbiodinium and host animals,
sources of zooxanthellae are under full experimental control

REPOPL'LATION OF ZOOXANTHELLAE

369

Colony (-Lesion) Number

Figure 7. Zooxanthellae associated with yellow-blotch disease (YBD). Lesions of YBD were sampled in 12
affected coral colonies (Colony Number. 1-12). one lesion per colony (most colonies) or two lesions per colony
(Colonies 6. 7. and 12 only; Colony-Lesion Numbers 6-1 and 6-2. 7-1 and 7-2. 12-1 and 12-2). At every lesion,
samples were taken from Normal and Yellow tissues as diagrammed in Figure ID. Samples of White tissue (see
Fig. ID) were obtained from five of the lesions only. Each bar represents data from one sample, presented as in
Figures 3-6. Coral species were Montastraea franksi (Colonies 1-5). M. faveolata (Colonies 6-11 1. and M.
annularis (Colony 12 1.

(e.g., Kinzie and Chee. 1979; Schoenberg and Trench,
1980; Colley and Trench. 1983; Davy et ai. 1997). In
contrast, corals in our field experiments were exposed to
uncharacterized natural populations of Symbiodinium. Here,
the identities of zooxanthellae in re-established symbioses
are certain only to the extent that zooxanthellar taxonomy is
certain.

Our study compared zooxanthellae that were identified by
Dpn II- and Taq I-generated RFLP genotypes of srDNA.
and this method does not discriminate all species of Sym-
biodinium (Rowan. 1998; Toller et al.. 2001 ). For example,
three known species of Symbiodinium A — S. microadriati-
citm (GenBank accession number M88521), S. pilosum
(X62650). and S. corculorum (L13717) — would be indis-
tinguishable in this analysis. Nevertheless, where the same
zooxanthellar RFLP genotype was detected in a coral both
before and after treatment (e.g.. Fig. 3. High Light, Colonies
6-15). parsimony argues that the coral hosted the same
species of Symbiodinium throughout the experiment. On the
other hand, no taxonomic uncertainty affects our observa-
tion that many re-established symbioses involved changes:
when compared samples of zooxanthellae differed with
respect to RFLP genotypes A. B. C. or E (e.g.. Fig. 3. Low
Light; Fig. 5). it is clear that the samples contained different
species of Symbiodinium (Rowan. 1998; Toller et al., 2001 ).

Disturbance and re-establishment of zooxanthellar
symbioses

Where our experiments were conducted. Symbiodinium B
predominates in Montastraea annularis at higher irradiance.

S\mbiodinium A and B predominate in M. faveolata at
higher irradiance, and Symbiodinium C predominates in
both species of coral at lower irradiance (Rowan and
Knowlton, 1995; Rowan et al., 1997). This predictable
pattern suggests that host-symbiont specificity is defined
largely by the interaction of each host-zooxanthella combi-
nation with its environment (sensu Buddemeier and Fautin.
1993). This led us to hypothesize that, under a constant
environment, host-symbiont specificity should be directly
re-established following acute disturbance.

We found some evidence for this in Experiment I, in that
most (10 of 15) of high-light-treated columns of M. annu-
laris were repopulated with the same zooxanthellae that
existed prior to treatment (Symbiodinium C or B; Fig. 3,
High Light. Colonies 6-15). However, the host-zooxan-
thella specificity was not re-established in the other high-
light-treated columns, which contained at least some differ-
ent zooxanthellae (Symbiodinium A. E. and C2; Fig. 3, High
Light. Colonies 1-5) after recovery. In these five corals,
treatment had led to significantly fewer zooxanthellae than
in the other 10 corals (5.0 X 105 cells/cm2 vs. 13.6 X 105
cells/cm2, respectively: P < 0.05. Mann-Whitney test).
This suggests that coral-zooxanthella associations may or
may not be re-established following disturbance, depending
on the magnitude of zooxanthellar depletion.

This conclusion is supported by the results of Experiment
I, in which none of the previously observed coral-zooxan-
thella associations were re-established in low-light-treated
M. annularis. Zooxanthellae were severely depleted in these
corals during treatment (to ca. 1 X 105 cells/cm- on aver-

370

W. W. TOLLER ET AL

age), and all corals were repopulated by completely differ-
ent zooxanthellae (Symbiodinium A, E, and C2; Fig. 3, Low
Light, Recovery), even 9 months after treatment (Fig. 4). A
similar result was obtained by low-light treatment of M.
faveolata in Experiment II. although in that experiment the
re-establishment of symbioses was not tested under a con-
stant environment; after being treated with low light, most
corals were also transplanted to a new environment. In that
new environment (9 m depth), unmunipulated M. faveolata
host S\mhi(>dinium C (Rowan and Knowlton, 1995; unpubl.
obs.), whereas re-established symbioses involved Symbio-
dinium A or A and E (Fig. 5).

We hypothesized that new taxa (Symbiodinium A, E, or
C2) following severe bleaching (Experiments I and II)
would not become established if untreated zooxanthellae
(resident Symbiotiiniiini B or C populations) were abundant
near bleached tissues (zooxanthellae are thought to be trans-
located within colonies, among coral polyps, via their gas-
trovascular systems [e.g., Gladfelter, 1983; Gateno et <;/.,
1998]). In Experiment III. the results from 5 of the 14
half-bleached columns were consistent with this hypothesis:
bleached tissues were repopulated with zooxanthellae that
apparently originated from untreated tissues (Fig. 6, Corals
1. 2. 3, 5. 6). Together with the observations on the spread
of pigmentation (Fig. 6, Results), this indirect evidence
suggests that zooxanthellae are translocated into bleached
tissues in some cases. However, in the majority of tested
cases (9 of 14), new zooxanthellae did become established:
Symbiodinium A, E, and/or C2 repopulated treated tissues
(Fig. 6, Corals 4. 7-14). despite the proximity (<7 polyps
away; see Weil and Knowlton, 1994) of untreated zooxan-
thellae (Symbiodinium C. with one exception). When new
zooxanthellae became established, they were observed more
frequently in the deeper habitat (8 of 9 columns in the deep
group vs. 1 of 5 in the shallow group), and when resident
zooxanthellae were Symbiodinium C (rather than Symbio-
dinium B) — our data do not resolve which factor had the
greater influence. Nevertheless, these observations clearly
show that a reservoir of adjacent zooxanthellae, whether
Symbiodinium C or B, is not sufficient to prevent the estab-
lishment of new host-zooxanthella associations in bleached
tissues.

We do not know where the new Svmbiodinium in re-
established symbioses came from. For most experimental
corals, the fact that these zooxanthellae were not detected
initially or after treatment is not good evidence that they
were truly absent. This is because an srDNA genotype must
be at least 5% of the total to be detected reliably (?.#., for
values of en. 12%, see fig. 4 in Toller et til.. 2001; fig. 2B
in Rowan et <//., 1997). Thus, corals that had ea. 1 >< 105
cells/cm2 of Symbiodinium C after low-light treatment (Fig.
3. Low Light, Treatment; also see Results) also may have
contained up to about 5 X 10? cells/cm" of Symbiodinium
A. E. or C'2 that went undetected. With a hypothetical

doubling time of 5 days (e.g.. Wilkerson et at.. 1988), 5 X
103 zooxanthellae/cm2 become 150 X 10^ zooxanthellae/
cm2 after only 8 weeks. Thus, even where only Symbio-
dinium C was detected right after treatment (Fig. 3. Low
Light. Treatment; Colonies 1, 3, 4, II. 12. and 14), there
might have been enough Symbiodinium A. E, and/or C2
present to found the established symbioses observed 13 or
17 weeks later.

On the other hand, no data show that treated corals did
not acquire Symbiodinium A. E, and C2 for the first time
during recovery. Free-living Symbiodinium may be attracted
specifically to hosts lacking zooxanthellae (Fitt. 1985b).
Juveniles of host species that do not transmit zooxanthellae
vertically (e.g.. Montastraea: Szmant. 1991) must be colo-
nized, and the ability of adult hosts to pick up Symbiodinium
from the environment has been documented for bleached
anemones (Kinzie et <//.. 2001 ) and for juvenile giant clams
originally inoculated with cultured zooxanthellae (Belda-
Baillie et <//., 1999; also see Fitt, 1984).

Regardless of where Symbiodinium A, E. and C2 came
from, they fared well compared to any Svmbiodinium C or
B that remained in corals after low-light treatment. For
example, the about 1 X 10^ cells/cm' of Svmbiodinium C
that six corals in Experiment I contained after treatment
(Fig. 3. Low Light. Treatment; Colonies 1. 3. 4. II, 12, and
14) would have been observed in re-established symbioses
if they had doubled only three or four times during 1 3 or 17
weeks (doubling times of 23-29 days; a slow rate of growth
for zooxanthellae in general [Wilkerson et ul.. 1988]). Thus,
the identities of re-established symbioses in these six corals
resulted not only from the proliferation, acquisition, or both
of Symbiodinium A, E. or C2. but also from the failure of
Symbiodinium C to proliferate.

Competition and succession in zooxanthellar communities

In general, the first phototrophs to colonize disturbed
habitat are transient and eventually replaced by competi-
tively superior species that dominate thereafter. This pro-
cess is called succession (Odum. 1969; Connell and Slatyer.
1977; Huston and Smith, 1987), and it might eventually
have led from Svmbiodinium A. E. and C2 to Svmbiodinium
C or B, and thus restored the host-symbiont specificity
observed in nature. We did not observe this hypothetical
succession of zooxanthellae. However, the only corals we
followed for more than 17 weeks after treatment (Fig. 4,
Final; 33 or 37 weeks after treatment) experienced a natural
bleaching event (Results) that apparently reduced popula-
tions of Symbiodinium C by about 80% in control and
high-light-treated M. annularis; Symbiodinium B may have
been affected also (Fig. 4, Final vs. Recovery). It seems
unlikely that Symbiodinium C or B would have proliferated
in low-light-treated corals during the same period of time.

REPOPULATION OF ZOOXANTHELLAE

371

Thus, unfavorable conditions might explain why succession
was not observed.

Hypotheses on the mechanisms of plant succession in-
voke genetic differences in the abilities of species to com-
pete for resources such as water, light, and nutrients, sup-
plies of which decrease as succession proceeds (e .g., Huston
and Smith, 1987;Tilman, 1988; Wilson, 1999). Zooxanthel-
lae in unmunipuluted corals cannot be water-limited, nor can
they extensively shade one another (Drew, 1972), but they
probably are nutrient-limited (Rees, 1991; Falkowski et ai,
1993). In contrast, severely bleached corals may be nutrient-
rich zooxanthellar habitats because the waste products of
coral heterotrophy go largely unutilized (e.g., Szmant-
Froelich and Pilson, 1977; Muscatine and D'Elia, 1978);
competition among zooxanthellae for nutrients may be min-
imal in this case. Competition should increase, however, as
zooxanthellar biomass increases, and the zooxanthellar ge-
notype that competes for nutrients best should ultimately
prevail, regardless of its rate of growth in the absence of
competition or its initial abundance (e.g.. Huston and Smith,
1987; Tilman, 1988).

In M. annularis and M. faveolata living at 9 m depth at
our study site, those efficient, specialized, but compara-
tively slowly growing zooxanthellae might be Symbio-
diniwn C and B. In contrast, Symbiodinium A, E, and C2
seem to have played the role of early successional, rapidly
proliferating opportunists in our experiments. In Experi-
ment II. S\mbiodinium A reached large numbers in only 8
weeks (Fig. 5. Recovery; see also Fig. 3, Low Light).
Opportunistic behavior by Symbiodinium A has also been
observed in M. annularis and M. faveolata during a natural
coral bleaching event (Rowan et ai., 1997) and in mixed in
vitro cultures of Symbiodinium (Rowan. 1998; Carlos et ai,
1999). Our observation of Symbiodinium E in Montastraea
living in a marginal habitat near our study site (Rio Cartf;
see Toller et ul.. 2001 ) also suggests a weed-like ecology.
Similarly, the unusual association of Symbiodinium C2 with
M. annularis (Results) suggests that our experimental treat-
ment enabled this zooxanthella to exploit a host species with
which it does not commonly associate. We note, however,
in the cases of both Symbiodinium A and E, it is unknown
whether different observations involved one or several spe-
cies of zooxanthella (see discussion on taxonomy, above).

Stabilit\ of zooxanthellar communities and coral
bleaching

During natural episodes of coral bleaching (reviewed in
Brown. 1997), even severely bleached colonies of M. an-
nularis retain at least 10% of their pre-bleaching population
of zooxanthellae (Porter et ai. 1989; Fitt et a!.. 2000),
which represents at least four times as many zooxanthellae
as our low-light-treated corals had. Thus, the dramatic
changes in species of Symbiodinium that we observed fol-

lowing low-light treatment of M. annularis and M. faveolata
are unlikely to be common in nature. Mild coral bleaching
or seasonal fluctuations in numbers of zooxanthellae in
normal years (Stimson, 1997; Fagoonee et ai, 1999; Fitt et
ai, 2000) involve lesser depletions of zooxanthellae, com-
parable to those observed in high-light-treated M. annularis
in which communities of Symbiodinium did not change
(Fig. 3, High Light, Colonies 6-10). Nevertheless, two
high-light-treated corals that retained about 15% and 20%
(Fig. 3, Colonies 4 and 5, respectively) of their zooxanthel-
lae after treatment apparently did acquire detectable
amounts of Symhioilinium E as a result of treatment. This
suggests that severe natural bleaching episodes might mod-
ify communities of Symbiodinium in M. annularis, at least
in part.

If infrequent, natural bleaching events at our study site
(e.g., Lasker et ai, 1984; Rowan et ai, 1997) do allow
Symbiodinium E or C2 to proliferate in M. annularis and M.
faveolata from time to time, the effect must be transient or
slight. These host-zooxanthella combinations are rarely en-
countered in unmanipulated corals from this reef (Rowan
and Knowlton, 1995; Rowan et ai. 1997; this study), de-
spite occasional coral bleaching. On the other hand, at a
nearby coastal site (Rio Cartf), where stresses that can
induce coral bleaching may be severe and frequent (Toller
et ai, 2001), disturbance appears to have a widespread
effect on host-zooxanthella association. In high-irradiance
habitats, M. faveolata and M. annularis at Rio Cartf asso-
ciate predominantly with Symbiodinium E — an opportunis-
tic taxon of zooxanthella (above) that may also be stress-
resistant (Toller et ai. 2001). In this environment, these
host-zooxanthella associations resemble the persistence of
early-to-mid successional phototrophs under conditions of
chronic disturbance (Odum, 1969; Horn, 1974; Huston and
Smith, 1987).

Buddemeier and Fautin (1993) proposed that bleaching
allows corals to replace their zooxanthellae with different,
better ones (see also Baker, 1999, 2001 ; Kinzie et ai, 2001 ).
According to this "adaptive bleaching hypothesis." such
replacements are driven by environmental change, which
simultaneously makes some host-zooxanthella combina-
tions less well adapted and other combinations better
adapted than they had been. Baker (2001) found that the
mortality of corals challenged with prolonged environmen-
tal change (transplantation) was reduced when they ac-
quired new zooxanthellae. but these new host-zooxanthella
associations were only established after coral bleaching.
Similarly, our experiments indicate that severe bleaching
allowed corals to associate with new zooxanthellae: Sym-
biodinium A, E, or C2 replaced Symbiodinium C or B in
Montastraea (see above). However, in our experiments,
environmental change was not a prerequisite, instead, Sym-
biodinium A, E, or C2 proliferated in the very environment
that Symbiodinium C or B apparently thrive in (Fij '^.l

372

\V \V TOLLER ET AL

5: Recovery). .Although our experiments did not include
prolonged environmental change and therefore was not a
direct test of the adaptive bleaching hypothesis, they do
show that testing this hypothesis may not be straightfor-
ward-severe coral bleaching may favor zooxanthellar re-
placements, irrespective of environmental change.

Disease-related disturbance of zooxanthellar symbioses

Symbiodiniiim A especially and also S\mbiodinium E
proliferated in the bleached tissues of corals with yellow-
blotch disease (Fig. 7. Yellow i. These zooxanthellae appar-
ently gave many lesions of YBD their distinctive yellow
color, and the\ were fugitives in the strict sense because
their habitat was ephemeral. Our measurements suggest that
their habitat lasted an average of about 5 months. Nonethe-
less, continuous progression of YBD across a coral would
provide a large amount of habitat for these fugitive zoo-
xanthellae to occupy.

Our findings explain the "yellow" in YBD. but the> do
not address the cause of the pathology. Most colonies we
encountered were in two clusters that were surrounded
widely by unaffected corals, which suggested an infectious
agent with limited dispersal. YBD may have more than one
etiology: our experimental results imply that it would arise
when anything spread through a coral and disturbed stress-
sensitive communities of zooxanthellae ( Symbiodiniiim B or
C) without actually killing the host immediately. One agent
might even be a "rogue" (parasitic) genotype of Symbio-
diniiim A that prospered at the expense of its coral host.

Acknow ledgments

We thank the Kuna Nation and the Republic of Panama
i Autoridad Nacional del Ambiente. Departamento de Cuar-
entena Agropecuaria del Ministerior de Desarrollo Agro-
pecuario. and Recursos Marinosi for permission to collect
and export specimens. Many thanks to Javier Jara for tire-
less field assistance and cell counts. Ursula Anlauf. Ralf
Kersanach. Mike McCartney, and Suzanne Williams pro-
vided valuable advice. Mick Wilson and Dave Wilson as-
sisted with field manipulations. R. R. thanks Chris Hein and
Uma Narayan for hospitality in California. This research
was supported by the Andrew W. Mellon Foundation, the
Smithsonian Tropical Research Institute, the National Insti-
tutes of Health, and the Scripps Institution of Oceanogra-
phy.

Literature Cited

Baker, A. C. 1999. Symbiosis ecology of reef-building corals. Ph.D.

•-ertation. University of Miami. 120 pp.
Baker. A. C. 2001. Reef corals bleach to survive change. \aturt 411:

765-766.
Belda-Baillie. C. A.. M. Sison. V. Silvestre. K. Villamor. V. Monje.

E. D. Gomez, and B. K. Baillie. 1999. Evidence for chanains

symbiotic algae in juvenile tridacnids. J. Exp. Mar. Biol. Ecol. 241:

207-221.
Brown. B. E. 1997. Coral bleaching: causes and consequences. Coral

Reefs 16: S 129- S 138.
Buddemeier. R. \V.. and D. G. Fautin. 1993. Coral bleaching .1- an

adaptive mechanism. Bioscience 43: 320-326.
Cairns. S. D. 1999. Species richness of recent Scleractinia. Atoll Res.

Bull. 459: 1-12.
Carlos. A. A.. B. K. Baillie. M. Kawachi, and T. Marujama. 1999.

Phylogenetic position of Symbiodiniiim I Dinophyceae ( isolates from

tridacnids (BivaJvia). cardiids (Bivalvia). a sponge (Poritera). a soft

coral (Anthozoa). and a free-living strain. J. Ph\col. 35: 1054-1062.
Collej. N. J.. and R. K. Trench. 1983. Selectivity in phagocytosis and

persistence of symbiotic algae by the scyphistoma stage of the jelly fish

Cassiopeia xamachana. Proc. R. Soc. Land. B 219: 61-82.
Connell. J. H.. and R. O. Slatyer. 1977. Mechanisms of succession in

natural communities and their role in community stability and organi-
zation. Am \ .;• Ill: 1119-1144.
Davy. S. K.. I. A. N. Lucas, and J. R. Turner. 1997. L'ptake and

persistence of homologous and heterologous zooxanthellae in the tem-
perate sea anemone Cereus pedunculanis (Pennant). Biol. Bull. 192:

208-216.
Drew. E. A. 1972. The biology and physiology of alga-invertebrate

symbioses. n. The denstt\ of s\mbionc algal cells in a number of

hermatypic corals and alcyonarians from various depths. J. Exp. Mar.

Biol. Ecol. 9: 71-7?.
Dustan. P. 1979. Distribution of zooxanthellae and photosynthetic chlo-

roplast pigment of the reef-building coral Montastrea annularis Ellis

and Solander in relation to depth on a West Indian coral reef. Bull. Mar.

Sci. 29: 79-95.
Fagoonee. I.. H. B. Wilson. M. P. Hassell. and J. R. Turner. 1999. The

dynamics of zooxanthellae populations: a long-term study in the field.

Science 283: 843-845.
Falkowski, P. G.. Z. Dubinsky. L. Muscatine, and L. R. McCIoskey.

1993. Population control in symbiotic corals. Bioscience 43: 606-

611.
titt. \\. K. 1984. The role of chemosensory behavior of Symbiodinium

microadriaticum. intermediate hosts and host behavior in the infection

of coelenterates and molluscs with zooxanthellae. Mar. Biol. 81: 9-17.
Fitt. \V. K. 1985a. Effect of different strains of the zooxanthella S\m-

biodinium microadriaticum on growth and survival of their coelenter-

ate and molluscan hosts. Proc. Fifth Int. Coral ReefCongr. (Tahiti I. 6:

131-136.
Fitt. \V. K. 1985b. Chemosensory responses of the symbiotic dinoflagel-

late Symbiodinium microadriaticum (Dinophyceae). J. Ph\col. 21:

62-67.
Fitt. \\ . K.. F. K. McFarland. M. E. Warner, and G. C. Chilcoat. 2000.

Seasonal patterns of tissue biomass and densities of symbiotic

dinoflagellates in reef corals and relation to coral bleaching. Limnol.

Oceanogr. 45: 677-685.
Franzisket. L. 1970. The atrophy of hermatypic reef corals maintained

in darkness and their subsequent regeneration in light. Int. Re\. Hydro-

biol. 55: 1-12
Gateno. D.. A. Israel. V. Barki. and B. Rinkevich. 1998. Gastrovas-

cular circulation in an octocoral: evidence of significant transport of

coral and symbiont cells. Biol. Bull. 194: 178-186.
Gladfelter, E. H. 1983. Circulation of fluids in the gastrovascular system

of the reef coral Acropora cen-icomis. Biol. Bull. 165: 619-636.
Horn, H. S. 1974. The ecologs of secondary succession. Annu. Re\.

Ecol. Syst. 5: 25-37.
Huston. M.. and T. Smith. 1987. Plant succession: life history and

competition. .Am. \at. 130: 168-198.
Kinzie. R. A.. III. and G. S. Chee. 1979. The effect of different

REPOPULATION OF ZOOXANTHELLAE

373

/ooxanlhellae on the growth of experimentally reinfected hosts. Hint.
Hull. 156: 315-327.

Kinzie, R. A., Ill, M. Takayama. S. R. Santos, and M. A. Coffrolh.
20(11. The adaptive bleaching hypothesis: experimental tests of crit-
ical assumptions. Biol. Bull. 200: 51- 58.

I.asker. H. R., E. C. Peters, and M. A. Coffroth. 1984. Bleaching of
reef coelenterates in the San Bias Islands, Panama. Coral Reefs 3:
183-190.

McNall), K. L., N. S. Govind, P. E. Thome, and R. K. Trench. 1994.
Small-subunit ribosomal DNA sequence analyses and a reconstruction
of the interred phylogeny among symbiotic dinoflagellates (Pyrro-
phyta). ./. Plm-ol. 30: 316-324.

Muscatine. 1... and C. F. D'Elia. 1978. The uptake, retention, and
release of ammonium by reef corals. Limnol. Oceamigr. 23: 725-734.

Odum, E. P. 1969. The strategy of ecosystem development. Science 164:
262-270.

Porter, J. W., VV. K. Fitt, H. J. Spero, C. S. Rogers, and M. \V. While.
1989. Bleaching in reef corals: physiological and stable isotope
responses. Proc. Natl. Acad. Sci. USA 86: 9342-4346.

Rees, T. A. V. 1991. Are symbiotic algae nutrient deficient? Proc. R.
Soc. Land. B 243: 227-234.

Rou\, K. H. 1995. Optimization and troubleshooting in PCR. Pp. 53-62
in PCR Primer: A Laboratory Manual, C. W. Dieffenbach and G. S.
Dveksler, eds. Cold Spring Harbor Laboratory Press, New York.

Rowan, R. 1998. Diversity and ecology of zooxanthellae on coral reefs.
J. Phycol. 34: 407-417.

Rowan, R., and N. Knowlton. 1995. Intraspecific diversity and ecolog-
ical zonation in coral-algal symbiosis. Proc. Natl. Acad. Sci. USA 92:
2850-2S53.

Rowan, R., and D. A. Powers. 1991a. A molecular genetic classification
of zooxanthellae and the evolution of animal-algal symbioses. Science
251: 1348-1351.

Rowan, R.. and D. A. Powers. 1991b. Molecular genetic identification
of symbiotic dinoflagellates (zooxanthellae). Mar. Ecu/. Prog. Ser. 71:
65-73.

Rowan. R.. N. Knowlton. A. Baker, and J. Jara. 1997. Landscape
ecology of algal symbionts creates variation in episodes of coral
bleaching. Nature 388: 265-269.

Santavy. D. L., E. C. Peters. C. Quirolo, J. \V. Porter, and C. N.
Bianchi. 1999. Yellow-blotch disease outbreak on reefs of the San
Bias Islands. Panama. Coral Reefs 18: 97.

Schoenberg. D. A., and R. K. Trench. 1980. Genetic variation in
Svmbitidiniitm (Gymnodinium) microadriaticum Freudenthal. and spec-
ificity in its symbiosis with marine invertebrates. III. Specificity and
infectivity of S. microadriaticum. Proc. R. Soc. Loiul. B 207: 445 — 160.

Stimson. J. 1997. The annual cycle of density of zooxanthellae in the
tissues of field and laboratory-held Pocil/opora damn-omit (Linnaeus).
J. E\p. Mar. Biol. Ecol. 214: 35-48.

Szmant, A. M. 1991. Sexual reproduction by the Caribbean corals \l,,n-
tastrea anniilaris and M. cavernosa. Mar. Ecol. Prog. Ser. 74: 1 3-25.

Szmant-Froelich. A., and M. E. Q. Pilson. 1977. Nitrogen excretion by
colonies of the temperate coral Astrangia danae with and without
zooxanthellae. Proc. Thud Int. Cunil Reef Symp. 1: 418-423.

Tilman, D. 1988. Plant Strategic* ami the Dynamics ami Structure of
Plant Communities. Princeton University Press, Princeton. NJ.

Toller, W. W., R. Rowan, and N. Knowlton. 2001. Zooxanthellae of
the Monta.straea anniilaris species complex: patterns of distribution of
four taxa of Svinbiodininm on different reefs and across depths. Biol.
Bull. 201: 348-359.

Toller, \V. W., R. Rowan, and N. Knowlton. In press. Genetic evi-
dence for a protozoan (phylum Apicomplexa) associated with corals of
the Montastraea anniilaris species complex. Coral Reefs.

Trench, R. K. 1988. Specificity in dinomastigote-marine invertebrate
symbioses: an evaluation of hypotheses of mechanisms involved in
producing specificity. Pp. 325-346 in Cell to Cell Signals in Plant.
Animal ami Microbial Symbiosis. S. Scannerini. D. C. Smith. P. Bon-
fante-Fasolo. and V. Gianninazzi-Pearson. eds. NATO Advanced Re-
search Workshop on Cell to Cell Signals in Plant, Animal, and Micro-
bial Symbiosis (1987: Turin. Italy). Springer-Verlag, Berlin.

Trench, R. K. 1993. Microalgal-invertebrate symbioses: a review. En-
docytobiosis Cell Res. 9: 135-175.

Trench, R. K. 1997. Diversity of symbiotic dinoflagellates and the
evolution of microalgal-invertebrate symbioses. Proc. Eighth Int.
Coral Reef Symp. 2: 1275-1286.

Veron, J. E. N. 1995. Corals in Space anil Time: The Biogeography and
Evolution of the Scleractinia. UNSW Press. Sydney, Australia.

Weil, E., and N. Knowlton. 1994. A multi-character analysis of the
Caribbean coral Montastraea anniilaris (Ellis and Solander. 1786) and
its two sibling species, M. faveo/ata (Ellis and Solander. 1786) and M.
franksi (Gregory. 1895). Bull. Mar. Sci. 55: 151-175.

Wells, J. W. 1956. Scleractinia. Pp. 328-444 in Treatise in Invertebrate
Paleontology. Part F. R. C. Moore, ed. Geological Society of America
and University of Kansas Press. Lawrence. KS.

Wilkerson, F. P., D. Kobayashi, and L. Muscatine. 1988. Mitotic
index and size of symbiotic algae in Caribbean reef corals. Coral Reefs
7: 29-36.

Wilson, S. D. 1999. Plant interactions during secondary succession. Pp.
61 1-632 in Ecosystems of the World. Vol. 16: Ecosystems of Disturbed
Ground. L. R. Walker, ed. Elsevier, New York.

Reference: Biol. Bull. 201: 374-384. (December 2001)

Microhabitats, Thermal Heterogeneity, and Patterns of
Physiological Stress in the Rocky Intertidal Zone

BRIAN S. T. HELMUTH1'* AND GRETCHEN E. HOFMANN2

' Department of Biological Sciences, University of South Carolina, Columbia, South Carolina 29208:
and Department of Biology. Arizona State Universitv. Tempe, Arizona 85287-1501

Abstract. Thermal stress has been considered to he
among the most important determinants of organismal dis-
tribution in the rocky intertidal zone. Yet our understanding
of how body temperatures experienced under field condi-
tions vary in space and time, and of how these temperatures
translate into physiological performance, is still rudimen-
tary. We continuously monitored temperatures at a site in
central California for a period of two years, using loggers
designed to mimic the thermal characteristics of mussels,
Mytilus californianiis. Model mussel temperatures were re-
corded on both a horizontal and a vertical, north-facing
microsite, and in an adjacent tidepool. We periodically
measured levels of heat shock proteins (HspVO), a measure
of thermal stress, from mussels at each microsite. Mussel
temperatures were consistently higher on the horizontal
surface than on the vertical surface, and differences in body
temperature between these sites were reflected in the
amount of Hsp70. Seasonal peaks in extreme high temper-
atures ("acute" high temperatures) did not always coincide
with peaks in average daily maxima ("chronic" high tem-
peratures), suggesting that the time history of body temper-
ature may be an important factor in determining levels of
thermal stress. Temporal patterns in body temperature dur-
ing low tide were decoupled from patterns in water temper-
ature, suggesting that water temperature is an ineffective
metric of thermal stress for intertidal organisms. This study
demonstrates that spatial and temporal variability in thermal
stress can be highly complex, and "snapshot" sampling of
temperature and biochemical indices may not always be a
reliable method for definim: thermal stress at a site.

Received 30 April 2001; accepted 1 1 September 2001.
* To whom correspondence should be addressed. E-mail: helmuthC01
biol.sc.edu

Introduction

Temperature is one of the most important abiotic deter-
minants of organismal distribution and physiological per-
formance in the rocky intertidal zone (Orton, 1929a, b;
Doty. 1946; Hutchins. 1947: Carefoot. 1977; Bertness,
1981; Wethey, 1983, 1984; Menge and Olson, 1990; Wil-
liams and Morritt, 1995). Animals and algae in this envi-
ronment are exposed to rapidly fluctuating and often ex-
treme temperatures, and recent studies have shown that
exposure to high temperatures can have significant physio-
logical consequences to these organisms (Hofmann and
Somero, 1995, 1996a, b; Stillman and Somero. 1996; Rob-
erts et a I., 1997; Chappie et ai. 1998: Tomanek and Som-
ero. 1999; Buckley et al.. 2001; Dahlhoff et ai. 2001;
Snyder et ai. 2001 ). Several studies have further indicated
that thermal stress can have significant ecological conse-
quences, and that exposure to stressful conditions varies
both in space and in time in the rocky intertidal zone. For
example, Wethey ( 1983, 1984) demonstrated that the com-
petitive dominance of one species of barnacle over another
varied with substratum angle, presumably as an indirect
effect of thermal or desiccation stresses on the relative
physiological performance of each species. Menconi et al.
( 1999) found that community structure at an intertidal site in
the Mediterranean varied as much as a function of substra-
tum angle as it did as a function of tidal height. Dahlhoff et
al. (2001 ) showed that temporal variability in physiological
stress had significant effects on the foraging ability of an
intertidal gastropod. However, despite a robust and growing
body of literature on the physiological ecology of intertidal
organisms, we are just beginning to understand on a mech-
anistic basis how body temperature variation influences
physiological performance and. ultimately, how physiolog-
ical performance contributes to the ecological interactions
of intertidal organisms.

374

TEMPERATURE AND HSP70 IN MUSSELS

375

Our understanding of temperature effects on intertidal
organisms is limited by at least three gaps in our know ledge
of the ecological physiology of the rocky intertidal zone.
First, although the ecological community is gaining appre-
ciation and insight into the significance of organismal body
temperatures under natural field conditions (e.g., Elvin and
Conor, 1979: Wethey, 1983. 1984; Bell. 1995; Williams
and Mon-ilt. 1995: Helmuth. 1998. 1999; Dahlhoff et ai.
2001 ). more attention needs to be paid to the complexity of
determining spatial and temporal patterns of body temper-
atures in the intertidal. While it is submerged, an ectother-
mic invertebrate is likely to have a temperature fairly sim-
ilar to that of the surrounding water. In contrast, during
aerial exposure, climatic factors such as air temperature,
wind speed, solar radiation, and relative humidity interact to
drive the flux of heat into and out of an organism's body
(Johnson, 1975: Bell, 1995: Helmuth. 1998, 1999). As a
result, temperature extremes during low tide can far exceed
those experienced during submersion, and an organism's
body temperature can be substantially different from the
temperature of the surrounding air (Helmuth. 1998). Fur-
thermore, heat fluxes are to some extent determined by the
size and morphology of the organism. As a result, organ-
isms exposed to identical climatic conditions can experience
different body temperatures (e.g.. Porter and Gates, 1969;
Porter et ai. 1973: Helmuth. 1998), and an animal's "ther-
mal regime" is determined in part by its own morphology.

Second, even though accurate determinations of body
temperature can be made, the physiologically significant
aspect of the "thermal signal" of environmental temperature
(e.g.. maximum, minimum, average, time history) is un-
known. Investigations of the plasticity of physiological pro-
cesses, such as the environmental induction of heat shock
proteins (e.g.. Buckley et ai. 2001) and the relationship
between oxygen consumption and temperature (Widdows,
1976). have documented that invertebrates are responsive to
a changing thermal environment in a regulatory manner and
therefore must sense environmental temperature. Additional
studies have coupled relatively short-term measurements of
body temperature to physiological indicators of thermal
stress (e.g.. Hofmann and Somero, 1995; Tomanek and
Somero. 1999: Dahlhoff et ai. 2001: Snyder et ai. 2001).
However, we still do not understand what aspect of envi-
ronmentally driven body temperature variation is physio-
logically significant in these ectothermic organisms.

Finally, only recently have advances in technology al-
lowed for measurement of body temperatures as a function
of microhabitat over long time scales. Deploying instrumen-
tation in the rocky intertidal zone is notoriously difficult due
to damage from waves, and only recently have commer-
cially available instruments become sufficiently small and
robust to be deployed for long periods of time. Furthermore,
because of the influence of a temperature logger's (or or-
ganism's) size, mass, and morphology on the temperature

that it records, temperature measurements relevant to inter-
tidal organisms are scarce, and those that exist are not
necessarily accurate proxies for the body temperatures of all
organisms at that site. Thus, there are relatively few data
sets that provide information as to how microscale features
of intertidal substrata influence organismal body tempera-
ture (except see Wethey, 1983. 1984; Williams and Morritt.
1995; Helmuth and Denny. 1999).

As a first step in addressing these complex issues, we
have integrated the fine-scale measurement of organismal
body temperature with the analysis of a bioindicator of
physiological stress. Specifically, in the current study, we
present temperature data recorded by loggers designed to
mimic the body temperatures of a competitively dominant
mussel. Mytilus californianus, and collected over a period
of 2 years at a site in central California (Monterey Bay). We
couple these data with periodic measurements of isoforms
of the 70-kDa heat shock protein (Hsp) gene family, a
molecular chaperone that has been used routinely as a
bioindicator of stress (see Feder and Hofmann, 1999, for a
review), and explore the inherent difficulty in linking pat-
terns in thermal signals in the field to physiological indica-
tors of stress. We further examine the effects of substratum
angle on body temperature and levels of thermal stress to
address the question of how body temperature and thermal
stress vary over small spatial scales in the intertidal.

Our results demonstrate that while biochemical indicators
of stress are potentially a very powerful tool for examining
the role of environmental variation in driving organismal
physiology, we still do not yet have a complete understand-
ing of what aspects of the thermal environment drive the
transcriptional activation of stress protein genes. Similarly,
high spatial and temporal variability in patterns of body
temperatures necessitate caution when extrapolating from
short-term measurements of temperature or Hsp production
in the intertidal. Namely, while the use of biochemical
indicators of stress and concomitant measurements of tem-
perature can potentially serve as an effective link between
the ecology and physiology of intertidal organisms, such
studies require detailed measurements of body temperature,
and an awareness of the potential role of thermal history in
driving physiological stress in the rocky intertidal zone.

Materials and Methods

Temperature measurements and logger design

Mussel temperatures were recorded using temperature
loggers deployed on the shores adjacent to the Hopkins
Marine Station in Pacific Grove. California (37' 18.0" N, 54'
15.5" W), from October 1998 to October 2000. Loggers
were deployed in the centers of small mussel beds at two
microsites in the mid- to high intertidal zone (mean lower
low water + 1.7 in): a horizontal, upward-facing microsite

376

B. S. T. HELMUTH AND G. E. HOFMANN

and a vertical, north-facing site. Sites were located within
20 cm of one another, in an area judged to be moderately
wave-exposed. A third logger was deployed at the bottom of
a nearby tidepool (about 1.5 m x 1 in x 15 cm deep) from
July 1999 to June 2000.

Because the same morphological factors that determine
heat flux to intertidal plants and animals and drive differ-
ences in their body temperatures can also affect heat flux to
temperature loggers, we used loggers imbedded in physical
models of mussels to collect temperature data. Thus, for
example, larger, more massive loggers have a larger thermal
inertia than do smaller loggers, and they may not always
record peaks in body temperature experienced by animals
with a faster thermal response time (Helmuth. 1998).
Matching the thermal characteristics of a temperature logger
to those of the organism in question is therefore critical, and
a single type of logger is unlikely to be an effective proxy
for all organisms at an intertidal site. We therefore deployed
temperature loggers of a size (60-75 mm) and shape com-
parable to those of real mussels, and we matched their
thermal response characteristics (mass X specific heat) to
living animals. From October 1998 to May 2000, empty
shells of Myti/ns califonuanus (-75 mm in length) were
filled with silicone sealant and fitted with a thermistor cable.
The recording tip of the thermistor was placed in the center
of the silicone-filled mussel shell. The thermistor was then
connected to an Onset Corporation Stowaway logger en-
cased in a waterproof housing. Mussels were attached to the
substratum in the middle of small beds, in approximate
growth position, using marine epoxy (Z-spar). Because of
the high rate of damage to thermistor cables, thermistor
loggers were replaced in May 2000 with a similar logger
designed entirely of epoxy plastic, where the logger (an
Onset Corporation Tidbit logger) was encased inside of the
fake mussel. Again, the product of mass X specific heat of
the plastic logger in the fake mussel was similar to that of a
living mussel. Both loggers recorded temperatures to an
accuracy and resolution of 0.3 °C. and recorded average
temperatures at intervals of 5 to 10 min (preliminary studies
indicated that changes in body temperature were slow
enough that this sampling interval would capture all peaks).
Because logger design was thought to have a significant
effect on the temperature recorded only while the logger
was exposed to air and not while completely submerged,
unmodified Onset Corp. Tidbit loggers were used to record
tidepool temperatures.

On 25 days from October 1998 to May 1999. the external
logger temperature was compared to the temperatures of
living mussels. An infrared thermocouple (Omega Corp.)
was used to record the external temperature of the logger,
and of 5-10 mussels in the surrounding bed. Results of the
82 comparisons indicated that loggers recorded tempera-
tures that were, on average, within 0.75 °C of those of living
mussels, and were usually within 1 standard deviation of the

average of the living mussels (correlation analysis indicated
a 1:1 curve fit with an R~ value of 0.94). Temperature data
were collected on days in which logger and mussel temper-
atures ranged from —11 °C to 27 °C. The loggers were thus
thought to serve as a reliable proxy for body temperature,
although comparisons were not made for the uppermost
range of temperatures observed throughout the year (30-34
°C). Furthermore, because loggers were sealed, they poten-
tially ignored any effects of evaporative cooling due to
mussel gaping. However, Bayne et al. (1976) showed that
aerial respiration by M. californianits is generally only
effective when relative humidity approaches 100%, when
evaporative cooling cannot occur (Helmuth. 1998, 1999).
Preliminary experiments (T. Fitzhenry and Helmuth, un-
publ. data) also suggest that this species does not gape as a
means of evaporatively cooling; nonetheless, this potential
complication requires further investigation.

To compare the effects of logger design on temperature
recorded, an unmodified Tidbit logger was deployed in the
horizontal mussel bed from July 1999 to October 2000.
Average and maximum daily temperatures recorded by the
unmodified logger were then compared to those recorded by
the adjacent physical model.

Tempi-future analyses

Because of the large number of data points collected by
the loggers, temperatures were summarized for each micro-
site on a monthly basis. Monthly maxima were divided into
two categories, each broadly representing a different poten-
tial source of thermal stress. "Acute" exposure to high
temperature was defined as the absolute maximum temper-
ature experienced by a logger at each site, on a monthly
basis (Fig. 1). In contrast, as a measure of "chronic" or
repeated exposure to high temperature, the average daily
maximum was calculated (Fig. 1 ). Similarly, the monthly
extreme minimum was recorded and average daily mini-
mum was calculated. Other metrics included the daily av-
erage temperature (including both aerial and submerged
temperatures) and the temperature at high tide (a measure of
water temperature). Except for monthly maxima and min-
ima, in which a single point was used for each month,
standard deviations of daily average, average daily maxi-
mum, average daily minimum, and temperature at high tide
were recorded as a metric of variability between days within
a month.

Western blot analysis of Hsp70 isoforms

Five specimens of Mytitus californiantis (length —50
mm) were collected at each microsite on four dates: 6 July

1999, 24 September 1999. 21 January 2000, and 8 May

2000. Mussels were immediately dissected, and samples of
"ill tissue were stored at -80 °C until they could be ana-

TEMPERATURE AND HSP70 IN MUSSELS

377

S 25

lO-

Monthly extreme

Average Daily
Maximum

1-Aug

8-Aug

1 5-Aug

22-Aug 29-Aug

Figure 1. Example of fluctuations in temperature experienced over one
month (August 1999) at the horizontal microsite. Daily maxima were
calculated from temperature data collected every 5 to 10 min. The highest
daily maximum was recorded as the monthly extreme ("acute") high
temperature at each site. The average of the daily maxima was calculated
as a measure of "chronic" high temperature exposure. Similarly, average
daily minima and monthly minima were calculated.

lyzed. Western blotting was employed to determine the
levels of both the constitutive and inducible isoforms of
Hsp70 in the samples. Hsp70 western blots were performed
as described by Hofmann and Somero (1995) except that
wet electrophoretic transfer at 30 V for 15 h was used
during the western protocol (transfer buffer = 20 mM Tris.
192 mA/ glycine, 207r methanol). Equal amounts of protein
(10 fj.g total protein) were separated on 7.57r polyacryl-
amide gels. A sample of purified Hsc70 (10 ng of bovine
Hsc70; Stressgen) was included on each gel as a positive
control, and as an internal standard to allow comparison of
multiple western blots. Immunodetection was performed
using an anti-Hsp70 rat monoclonal antibody that cross-
reacts with the cognate and inducible forms of Hsp70 (Af-
finity Bioreagents; MA3-001). Western blots were devel-
oped using an enhanced chemiluminescence protocol
according to the manufacturer's instructions (ECL Western
Blot Reagent; Amersham) and visualized on a Fluor-S Mul-
tilmager (BioRad). Band intensity from each western blot
was quantified using Quantity One software. Protein deter-
minations of the gill extracts were performed using a mod-
ified Bradford protein assay (Pierce Coomassie Plus).

Levels of Hsp as a function of microsite and of collection
date were compared using a two-way analysis of variance.
Post-hoc comparisons of the effect of season within site,
and the effect of site within season, were conducted using a
series of one-way ANOVAs with Fisher's PLSD test. Lev-
els of the two isoforms of Hsp70 (Hsp72 and Hsc75. see
below) were analyzed separately.

Results

Temperature analysis

Although the study area was superficially judged to be
only moderately wave-exposed, wave forces at the sites
were often severe (Helmuth and Denny, 1999) and fre-
quently resulted in the loss of or damage to loggers. Gaps in
the data sets are therefore present, particularly during winter
months when wave forces were greatest. Summary statistics
for months in which fewer than 3 weeks of data were
collected are thus not reported.

From 1 1 December 1999 to 6 May 2000. the only loggers
recovered at the horizontal site were the unmodified Tidbit
loggers. A correlation analysis from days on which both the
unmodified and modified loggers were present at the hori-
zontal site (n •-- 329 days) indicated that temperatures
recorded by the unmodified logger could be used to predict
those recorded by the physical models (R2 = 0.96). An
offset value (+0.46 °C) calculated from the correlation
analysis was used to predict maximum daily temperature for
the missing 149 days, and an offset of +0.15° was used to
predict average daily temperature. No correction was re-
quired for predicting minimum temperatures. On any given
day, however, maximum temperatures recorded by the two
loggers differed by as much as 4.7 °C, with an average
difference of 1.3 °C. The correlation between the unmodi-
fied logger and the logger on the north-facing substratum
was too poor to be useful for days in which the logger at that
site was missing.

The highest annual temperatures at the horizontal micro-
site (Fig. 2a) were recorded in May 1999 (33.8 °C on 23
May) and August 2000 (33.8 °C on 10 August). The highest
levels of "chronic" high temperature exposure (average
daily maxima) at this microsite were recorded in August
1999 (24.4 °C), and in June 2000 (24.2 °C; Fig. 2a). Thus,
the levels of these two metrics of temperature exposure
were out of phase with one another, most obviously in 1999
(Fig. 2a). In contrast, on the north-facing site, both the
highest average daily maximum and the yearly extreme high
temperature (29.1 °C on 10 August) occurred in August
1999 (fig. 2b): insufficient data were collected to assess the
timing of the extremes at the north-facing microsite in 2000.
Minimum temperatures were comparable between the
north-facing and horizontal microsites, and tended to occur
during aerial exposure after sunset. Notably, two freeze (or
near freeze) events were recorded on the early evenings of
22 and 23 December 1998, with loggers on the north-facing
sites recording temperatures of about —0.6 to —0.9 °C. A
large disturbance in the mussel bed was recorded a few
weeks later; whether it was precipitated by the freeze is
unknown (Helmuth and M. W. Denny. Stanford University.
unpubl. data).

Temperatures were consistently higher at the horizonta
site than at the north-facing site (Figs. 3 and 4). On average.

378

B. S. T, HELMUTH AND G. E. HOFMANN

(a) Horizontal microsite

35 n

•25-

— ->— Extreme Maximum
--•--Average Daily Maximum
— s — Average

» Average Daily Minimum
- •• Extreme Minimum

3
2
8.20

15-

10-

u >> >,

I 1

a
«

(b) Vertical, North-facing microsite

Iarch99

"5. ^
1 f

tember99

ember99

o
o

temberOO

i/>

Figure 2. Temperature statistics recorded at the (a) horizontal micro-
site and (b) vertical, north-facing microsite from November 1998 to Oc-
tober 2000. Temperature data from January to May 2000 at the horizontal
microsite were extrapolated from an unmodified logger placed in the bed.
Yearly maxima at the horizontal site occurred in May 1999 and August
2000. In contrast, peaks in the average daily maximum ("chronic" temper-
ature exposure) at (his site occurred in August 1999 and June 2000.
Standard deviations indicate the amount of variability within each month,
except for monthly extremes and mimmums, for which a single point was
recorded during each month-long interval. At the north-facing site (h) the
yearly maximum and the highest average daily maximum occurred in
August 1999. Note the incidence of an unusual freeze event in December
1998.

Horizontal microsite
North-facing microsite

^30

|25

I-

^ 20

I15:

>> 10

5 5

CD X,

? I

-5 1

•£
f I

3 O) <U O

Figure 3. Comparison of monthly high extreme temperatures recorded
on horizontal and north-facing microsites. Monthly extrema were always
highest on the horizontal substrate, in some months by 10 °C or more. The
seasonal timing of temperature maxima varied between sites, occurring in
May 1999 on the horizontal site and in August 1999 on the vertical site
(indicated by arrows).

extreme maximum monthly temperatures recorded on the
horizontal site were 6.75 °C hotter than those on the vertical
site; the difference in extreme high monthly temperatures
between the horizontal and vertical sites ranged from a high
of 13.5 °C in April 1999 to a low of 1.9 °C in June 2000
(Fig. 3). Average daily maxima calculated for each month
were also higher on the horizontal site, with an average
difference of 3.6 °C, ranging from 1.7 °C in October 2000
to 6.8 "C in April 1999 (Fig. 4).

Temperatures recorded in the tidepool were not as high as
those on the exposed horizontal microsite. but in general ex-
ceeded those recorded on the aerially exposed vertical face
(Fig. 5). Both the yearly extreme high temperature maximum
(29.8 °C on 2 August 1999) and the highest average daily
maximum (24.5 °C) in the tidepool were recorded in August
1999 (Fig. 5). Water temperatures (recorded by the loggers at
high tide) were highest in August through October 1999 and
July through September 2000 ( - 15 °C), and displayed a pat-
tern that was markedly different from any of those recorded
during aerial exposure (Fig. 6).

Heat shock protein analysis

To compare the physiological status of mussels from the
different microsites, the cellular levels of isoforms of the

TEMPERATURE AND HSP70 IN MUSSELS

379

^35n

| 30

4-)

u
< 5

Horizontal microsite
North-facing microsite

< 2

Figure 4. Comparison of average daily maxima recorded at the two
aerially exposed sites. Again, levels of "chronic" high temperature expo-
sure were highest on the horizontal substratum. In both cases, highest
yearly levels in 1999 occurred in August.

70-kDa Hsp gene family were measured in gill tissue of
mussels collected in each season of the year — in July and
September 1999, and January and May 2000. Figures 7 and
8 show relative endogenous levels of isoforms of Hsp70 in
two groups that separate on SDS-PAGE, a 72-kDa band
(Fig. 7) and a 75-kDa band (Fig. 8). Although the precise
identity of the separate proteins that compose the two sets is
unknown and cannot be determined using one-dimensional
electrophoresis, the two isoforms display changes that to
some degree correspond to the temperature exposure of
Mytiliis. In previous studies, the 72-kDa band varied signif-
icantly with the thermal history of the mussel, with higher
levels in summer than in winter; in contrast, the higher
molecular mass band varied less as a function of season
(Hofmann and Somero. 1995; Roberts et ai. 1997). There-
fore, we have expressed the data using the two sets of
isoforms as separate indicators, where the 75-kDa band
(hereafter Hsc75) reflects constitutive levels of Hsp expres-
sion and the 72-kDa band (hereafter Hsp72) reflects a stress-
inducible subset of the 70-kDa Hsps.

Overall, levels of the 70-kDa molecular chaperones in
mussel gill varied significantly as a function of microsite
(Table 1; Figs. 7, 8). Regardless of season, levels of Hsp72
were always significantly greater in mussels on the horizon-
tal substratum than in mussels attached to the north-facing
surfaces of rocks (ANOVA; P = 0.0001; Fig. 7; Table 1 ).
However, there was no consistent pattern for Hsc75 (Fig. 8).

Extreme Maximum

--•--Average Daily Maximum
— e — Average
— * — Average Daily Minimum
Extreme Minimum

2
8.

§ f

Figure 5. Temperature recorded in a small tidepool. As expected,
temperature extremes were buffered relative to the aerially exposed hori-
zontal substratum. However, high temperature extremes were higher than
those on the aerially exposed, north-facing site, with yearly extremes
reaching nearly 30 °C.

Compared to north-facing mussels, the horizontal mussels
had significantly higher levels of Hsc75 only in January
(P = 0.012). In July, Hsc75 levels in the two groups were

October98

3ecember98

January99

March99

(71

o>
o

August99

October99

\lovember99

JanuaryOO
FebruaryOO
AprilOO

C
3
— )

JulyOO
JeptemberOO

Figure 6. Patterns in water temperature recorded during high tide. The
seasonal pattern in water temperature is markedly different in bi>'!) 111:11;-
nitude and timing from those recorded in any of the microsite !

exposure.

380

2.5n

B. S. T. HELMUTH AND G. E. HOFMANN
Inducible (72 kDa) form

Horizontal microsite
North-facing microsite
Tidepool

6-Jul

24-Sep

21 -Jan

Figure 7. "Inducible" (72-kDa isoform) levels of heat shock protein
from mussels collected at each of the three sites. See Table 1 for statistical
results, and Table 2 for temperature conditions experienced by mussels
prior to each collection. In general, inducible forms were significantly
higher in mussels from the horizontal site than in mussels from the
north-facing site. Differences between the aerially exposed mussels and
mussels from the tidepool were less consistent.

equivalent, and in the other two months the levels were
significantly lower in the horizontal mussels than in the
north-facing mussels (September, P = 0.0001; May. P =
0.0001; Fig. 8).

Constitutive (75 kDa) Form

Horizontal microsite
North-facing microsite
Tidepool

6-Jul

24-Sep

21 -Jan

Figure 8. "Constitutive" (75-kDa isoform) levels of stress proteins
from mussels at each site. Constitutive forms are thought to be affected by
multiple physiological parameters and do not necessarily change with
thermal stress. See Table 1 for results of statistical analysis.

Table 1

Results of statistical analyses of the 72-kDa form of W.v/' 70

Effect of Site within Collection Date*

6 July 1999
24 September 1999
2 1 January 2000
8 May 2000

H = TP > N (F = 4.12. P = 0.0400)
H > TP = N (F - 12.14. P = 0.0013)
H = TP > N (F = 9.64. P = 0.0030)
H > TP = N (F = 19.85, P = 0.0002)

Effect of Collection Date within Site

Horizontal

May

> July = Sept.

= Jan. (F

= 33.0.

= 0.0001)

North-facing

July

= Sept. =

May

> Jan. (F

= 4.0,

= 0.0270)

Tidepool

July

= Sept. =

Jan.

= May (F

= 2.4.

= N.S.)

Overall analysis

using a two-factor ANOVA indicated a

significant

effect of collection date (F = 10.0. P = 0.0001 ). Site (F = 28.2. P =
0.0001) and a significant interaction term (F = 9.23. P = 0.0001). A
series of one-way ANOVAs with Fisher's PLSD post hoc tests were used
to discern the effects of site within collection date, and of collection date
within site.

* H. horizontal; TP. tidepool; N. north-facing.

In a comparison of tidally exposed and constantly sub-
mersed individuals, there were no obvious differences or
trends in either Hsp72 or Hsc75 levels between the hori-
zontal and north-facing mussels and mussels that were
permanently immersed in a tidepool (Figs. 7, 8). Hsp72
levels were greater in horizontal mussels in September and
May as compared to tidepool mussels, but these levels were
equal in July and January (Fig. 7). Hsp72 levels in tidepool
mussels were equivalent to north-facing levels in May and
September, but tidepool mussels had significantly greater
levels in July and January than did their north-facing coun-
terparts. For Hsc75, levels in mussels from the tidepool
were greater than those in horizontal mussels in September.
July, and January, and significantly lower than in mussels
from horizontal surfaces in May. Levels of Hsc75 from
tidepool mussels were significantly greater than in north-
facing mussels in July and January, and significantly lower
than in north-facing mussels in May and September.

Finally, the three microsites displayed variation in levels
of the 70-kDa Hsp bioindicators as a function of time of
collection (Figs. 7, 8). For the horizontal mussels, Hsp72
levels were higher in May than in any other month; how-
ever, all three other months (July. September, and January)
were not significantly different from each other (Fig. 7). In
contrast, the mussels from the north-facing substratum had
their lowest levels of Hsp72 in January; the difference
between January and the other months was statistically
significant. Hsp72 levels in gill from north-facing mussels
were not significantly different amongst the July, Septem-
ber, and May collections. With respect to Hsc75, horizontal
mussels in September and July had equivalent but lower

TEMPERATURE AND HSP70 IN MUSSELS

381

Tank 2

1 int'iiMirt'tnent.\ conducted Jiti
in cm h »»nu7 ciillccrii'ii

the urrA prior

Horizontal

North-lacing

Tidepool

30 June-n Julv 1999

19.8(25.1)

20.5(26.3)

22.0(26.3)

18-24 September 1999

21.1 (26.3)

17.1 (19.1)

20.7(25.3)

1 5-2 1 January 2000

13.0(15.3)

N.R.

I4.l)(lh.l)

2-8 May 2000

20.9(30.0*)

14.2 (22.0)

I1). 0(26.6)

Both the average daily maximum temperature and the extreme temper-
ature (m parentheses) recorded during that week are given (°C).

* The 30 °C temperature recorded on May 8 was for a very brief period
of time (<20 min.).

levels than in January and May; north-facing mussels dis-
played the highest values in May as compared to all other
months, which were not significantly different from each
other. Interestingly, the tidepool mussels exhibited no sea-
sonal effect on Hsp72 levels (Fig. 7), but they did show
some variation in Hsc75 levels (Fig. 8). Specifically, the
levels of Hsc75 in May and September were equivalent to
each other but significantly lower than in the months of July
and January (P = 0.001 ); July and January levels were not
significantly different from each other.

A comparison between the maximum temperature expo-
sure in the week prior to collection (Table 2) and the levels
of Hsp72 (Fig. 7), shows that inducible Hsp levels generally
increased with maximum temperature exposure, but the
correlation was not as good as might be expected (Fig. 9). A
regression of Hsp72 with maximum temperature indicated a
significance level of P = 0.03 (Statview; F = 7.66) when
both north-facing and horizontal mussels were considered
(note that the temperature datum for the January north-
facing site was assumed to be no higher than that on the
horizontal site). Tidepool data (not shown) generally fell
along the same trend line, but reduced the significance level
to P = 0.059 (F ---- 4.55).

Discussion

Intertidal organisms live at the margins of the marine and
terrestrial environments and must contend with the chang-
ing physical conditions of both regimes. Recently, much
attention has been paid to the influence of seawater temper-
ature, and in particular to changes in seawater temperature
as a result of climate, on changes in intertidal communities
(e.g., Barry etal., 1995; Sagarin et «/., 1999). However, few
studies have investigated the importance of aerial exposure
to intertidal organisms in a changing thermal environment
(but see Denny and Paine. 1998). Clearly, extremes in body
temperature (both high and low) experienced during expo-
sure to air far exceed those occurring during high tide.
Depending on the zonational height of the organism, the
duration of exposure to air can be as long as or even longer
than the submersion time.

An important question that remains to be answered is.
how important to an organism's physiological performance
is thermal stress during low tide as opposed to the effects of
water temperature during submersion? Previous evidence
suggests that some intertidal organisms slow their metabolic
rates during aerial exposure, and in some cases resort to
anaerobic metabolism (e.g., Bayne et nL, 1976). Work by
Sanford (1999) has suggested that the rate of predation by
the sea star Pisaster is driven by water temperature and
appears to be unrelated to air temperature. In contrast,
measurements of Hsp production show that the tempera-
tures at which Hsps are induced occur almost exclusively
during low tide (e.g.. Roberts et ul.. 1997; Tomanek and
Somero. 1999). and that the deficit to the protein pool can
have a significant effect on the animal's scope for growth
(Roberts et ul.. 1997). Mass mortalities due to thermal stress
also have been reported primarily as a result of extremes in
temperature experienced during exposure to air (e.g., Glynn.
1968; Suchanek. 1978; Tsuchiya. 1983: Liu and Morton,
1994; Williams and Morritt, 1995). Understanding the rel-
ative importance of thermal stress during submersion versus
during aerial exposure is therefore key if we are to decipher

• Horizontal microsite

1.8
|1.6

North-facing microsite m

i 1-4

(\j

^ 1.2
u

1 ]

•

« 0.8

"5 0.6

C£

0.4

0.2

00 20.00 25.00 30.00 35.

Maximum Weekly Temperature (°C)

Figure 9. Comparison of Hsp72 ("inducible form") levels r\. the
maximum temperature recorded in the week prior to collection. As a result
of data logger failure, no temperature data were collected on the north-
facing site prior to the January collection. For the purposes of this figure,
we thus assume that the north-facing site was no hotter (15.3 °C) than the
horizontal site where temperatures were recorded. A simple regression
reveals a significant relationship between Hsp72 and maximum tempera-
ture (P = 0.03; F ---- 7.66). although it should be noted that the
relationship is significant primarily because of the large spike in Hsp72
production observed in May.

382

B. S. T. HELMUTH AND G. E. HOFMANN

and predict the effects of climate, and of climate change, on
intertidal communities.

Our results show that patterns in body temperature expe-
rienced during low tide cannot be predicted on the basis of
measurements of nearshore water temperature. Similarly,
preliminary evidence (Helmuth, unpubl. data) suggests that
air temperature is also an ineffective proxy for body tem-
perature. This observation is pertinent because air and water
temperatures are frequently the dominant metrics used to
estimate patterns in thermal stress in the intertidal zone
(e.g., Barry er al, 1995; Menge el a/., 1997; Sagarin et til.,
1999; Denny and Paine, 1998; Sanford, 1999; Thompson et
al., 2000). Furthermore, as our data indicate, high spatial
variability due to substratum angle can lead to large differ-
ences in body temperatures. Single measurements of tem-
perature, and particularly those based on water or air tem-
perature, cannot be used to define thermal stress at an
intertidal site or to compare multi-year trends in community
structure as a function of climate change.

Our study also points out gaps in our understanding of
what aspect of the thermal environment drives organismal
stress and of how organisms respond to temporally varying
environmental signals. Widdows (1976), for example,
showed that, when acclimated to cyclic temperatures, Myti-
lus edulis decreased its amplitude of response of filtration
rate and oxygen consumption to changing temperatures.
More relevant to our study, previous research has shown
that the threshold induction temperature and the total cel-
lular pools of Hsps in mussels changed as a function of
season and thermal acclimation in the laboratory (Roberts et
ul.. 1997; Buckley et al.. 2001). Although these studies
clearly demonstrate an effect of thermal history on the
physiology and regulation of the heat shock response, the
mechanism that couples variation in environmental temper-
ature with the physiological response is unknown. Surpris-
ingly, even in the heat shock biology of model cells, there is
no consensus about how the thermal signal is transduced
from the membrane, through protein kinase cascades to the
nucleus (e.g., Lin et ul., 1997; Ng and Bogoyevitch, 2000;
Han et ul.. 2001 ). As ecological physiologists, if we are ever
to determine the pathway of signal transduction of temper-
ature in an organism in nature, we must first understand the
physiologically important aspect of temperature.

Thus, one of the goals of this study was to bridge the gap
between temperature exposure in nature and a predictable
molecular response, the heat shock response. Our results
highlight the complexity of examining an environmentally
induced gene expression event in organisms in a natural
population. Although there are some instances in which the
Hsp levels "match" the predicted result (Fig. 9), there are
others in which the correspondence is poor. For example, as
expected, mussels living on horizontal substrata consis-
tently had higher levels of Hsp72. the inducible isoform of
the 70-kDa Hsp gene family, than did mussels on north-

facing substrata (Fig. 7, Tables 1. 2). In contrast, seasonal
differences in Hsp production (Fig. 7) were less easily
interpreted and did not always display the pattern observed
in other studies of intertidal mussels. For example, levels of
Hsp72 in mussels from the horizontal microsite were nearly
as high (albeit more variable) in January as in July (Fig. 7),
even though recorded body temperatures were considerably
higher in July than in January (Table 2). On the other hand,
higher temperatures recorded in May (Table 2) were re-
flected in Hsp72 production during this time period (Figs. 7,
9). and appear to be most closely related to differences in
extreme temperature (30 vs. 25 °C; Table 2). These patterns,
and in particular the patterns observed in Hsp75 production,
suggest again that there are numerous factors at work in the
control of chaperone levels and that thermal stress may not
be the only factor driving variation in Hsp expression.
Specifically, other physiological stressors such as hypoxia
and desiccation may contribute to temperature's influence
on Hsp induction (see Feder and Hofmann. 1999). Further-
more, our study shows that the seasonal timings of potential
stressors do not always act in concert, and that the timing of
"acute" and "chronic" high temperature exposures varies
with substratum angle. The thermal landscape is highly
variable, and conclusions drawn from any given study could
depend on the sampling regime (e.g.. effects of substratum
angle). Extreme caution must be exercised when collecting
samples over limited spatial and temporal scales as a means
of defining thermal stress at a site.

Our data also address the inherent complexity of using
Hsps as biomarkers in the environment. In general, the heat
shock response is subject to complex regulation in the cell
(Kline and Morimoto, 1997; Ali et al., 1998; Morimoto,
1998;Zhonger«/., 1998). The nature of Hsp gene activation
can change with the length and severity of the thermal stress
(see Lindquist, 1986, for a review; see also Yost et al.,
1990), and Hsp70 mRNA stability varies as a function of
temperature (Petersen and Lindquist, 1988, 1990). In fact.
Hsps are thought to control their own expression via a
negative feedback loop, making the cellular pools and the
induction points interrelated (e.g., DiDominico et al.. 1982;
Craig and Gross, 1991; Shi et al., 1998). Furthermore, once
the Hsps are synthesized, they are also subject to decay just
like any other protein, and their half-life is influenced by the
thermal conditions of the cell. In combination, all the mech-
anistic and complex regulatory aspects of the heat shock
response make for a system that not only is sensitive to
temperature but also is directly influenced by temperature,
just like any other biomolecular process in a cell. Thus, for
example, mussels exposed to lower chronic levels of high
temperature may produce inducible forms of Hsp at a lower
acute temperature level than will mussels that were accli-
mated to high average daily maxima. Thus, a hot day that is
preceded by a week of relatively mild days may elicit a very
different physiological response than an extreme tempera-

TEMPERATURE AND HSP70 IN MUSSELS

383

ture exposure that follows several days of gradually increas-
ing daily maxima. In summary, the effects of both extreme
temperature events (acute temperature exposure) and of the
thermal history (e.g.. chronic temperature exposure) are
likely to be important, but \ve do not yet sufficiently under-
stand the molecular consequences of temperature variation
or how variation in signal transduction and in gene expres-
sion would alter the pools of Hsps.

In some \\ays our study raises more questions than it
answers. Defining "thermal stress" at any given site is likely
to be complex. Substratum angle can have an enormous
effect on the magnitude, timing, and thermal history' of
temperature. Because ectothermic organisms influence their
body temperatures at least partially through their size and
morphology, two organisms at one site might experience
different patterns in the thermal signal, particularly if they
are mobile U'.s-.. Orton. 1929a). Thermal stress may there-
fore be organism-specific, rather than site-specific (Menge
and Olson. 1990). Finally, care must be taken to account for
the thermal conditions occurring during the collection pe-
riod. Thermal stress experienced during low tide results
from the interaction between terrestrial climate and the
timing of low tides as set by the tidal series (Orton. 1929a:
Helmuth. 1999). For example, sites separated by tens of
kilometers have been predicted to experience temperature
maxima that differ by several degrees due to the timing of
low tide during the hottest times of the year: organisms at
sites where low tide occurs at noon may experience much
higher temperatures than those at sites where low tide
occurs in the morning (Helmuth. 1998. 1999). Inter-annual
and decadal-scale variations in tidal exposure have also
been predicted to occur (Denny and Paine. 1998). The
coupling of biochemical indicators of stress with detailed
measurements of temperature may be effective in predicting
the role of climate in driving the ecology of rocky intertidal
communities, and in predicting the effects of climate change
on these ecosystems. However, for ecologists. the tempta-
tion to base large-scale comparisons of the role of thermal
stress on limited measurements of stress proteins must be
balanced by a knowledge of the role of the organism's
"cellular thermostat" in driving its physiological response to
temperature change. Conversely, physiologists must have a
better grasp of how temperatures change in nature if we are
to extrapolate from controlled laboratory experiments to
conditions in the field. Thus, while there is no simple
mechanism for linking patterns in temperature to patterns in
physiological stress, the merger of these levels of approach
promises to be fruitful for understanding the effects of
climate on the rocky intertidal zone.

Acknowledgments

The authors thank Matthew Wright. Sean P. Place, and
Megan Dueck for technical assistance with sample prepa-

ration and western blot analysis. Michael O'Donnell for
assistance in the field, and Morgan Timmerman-Helmuth
for her help in editing the manuscript. In addition, we thank
Hopkins Marine Station of Stanford University and its Di-
rector. Dr. George N. Somero. for access to the study field
sites, and Dr. Mark Denny for his advice and insight. This
research was supported in part by National Science Foun-
dation grants IBN 9985878 to BSTH. IBN 0096100 to
GEH. and NSF 0083369 to GEH and BSTH.

Literature Cited

Ali. A.. S. Bharadwaj, R. O'Carroll. and N. Ovsenek. 1998. Hsp90
interacts with and regulates the activity of Heat Shock Factor 1 in
Xenopus oocytes. Mol. Cell Biol. 18: 4949-4960.

Barry, J. P., C. H. Baxter, R. D. Sagarin, and S. E. Gilman. 1995.
Climate-related, long-term faunal changes in a California rocky inter-
tidal community. Science 267: 672-675.

Bayne, B. L., C. J. Bayne. T. C. Carefoot. and R. J. Thompson. 1976.
The physiological ecology of Mytilus californiamts Conrad. 2. Adap-
tation to low oxygen tension and air exposure. Oecologia 22: 229-250.

Bell. E. C. 1995. Environmental and morphological influences on thallus
temperature and desiccation of the intertidal alga Mastocarpus papil-
latus Kiitzing. J. Exp. Mar. Biol. Ecol. 191: 29-55.

Bertness, M. D. 1981. Predation. physical stress, and the organization of
a tropical rocky intertidal hermit crab community. Ecology 62: 411-
425.

Buckley. B. A., M. E. Owen, and G. E. Hofmann. 2001. Adjusting the
thermostat: changes in the threshold induction temperature for heat
shock protein genes in mussels from the genus Mytilus. J. Exp. Biol.
204 (In press).

Carefoot. T. 1977. Pacific Seashores: a Guide to Intertidal Ecology.
University of Washington Press. Seattle. WA.

Chappie, J. P.. G. R. Smerdon. R. J. Berry, and A. J. S. Hawkins. 1998.
Seasonal changes in stress-70 protein levels reflect thermal tolerance in
the marine bivalve Mytilus edulis L. J. Exp. Mar. Biol. Ecol. 229:
53-68.

Craig, E. A., and C. A. Gross. 1991. N Hsp70 the cellular thermometer0
Trends Biochem. Set. 16: 135-140.

Dahlhoff. E. P.. B. A. Buckley, and B. A. Menge. 2001. Feeding of the
rocky intertidal predator Nucella ostrina along an environmental stress
gradient. Ecology 82: 2S 1 6-2829.

Denny, M. W., and R. T. Paine. 1998. Celestial mechanics, sea-level
changes, and intenidal ecology. Biol. Bull. 194: 108-115.

DiDominico, B. J.. G. E. Bugaisky . and S. Lindquist. 1982. Heat shock
and recovery are mediated by different translational mechanisms. Proc.
\atl. Acad. Sci. USA 79: 6181-6185.

Doty, M. S. 1946. Critical tide factors that are correlated w ith the \ erticul
distribution of marine algae and other organisms along the Pacific
Coast. Ecology 27: 315-328.

Elvin. D. \V.. and J. J. Gonor. 1979. The thermal regime of an intenidal
A/U//M5 californianus Conrad population on the central Oregon coast. J.
Exp. Mar. Biol. Ecol. 39: 265-279.

Feder, M. E., and G. E. Hofmann. 1999. Heat-shock proteins, molec-
ular chaperones. and the stress response. Anna. Rev. Physiol. 61:
243-282.

Glynn, P. \V. 1968. Mass mortalities of echinoids and other reef flat
organisms coincident with midday, low water exposures in Puerto
Rico. Mar. Biol. 1: 226-243

Han, S. I., S. Y. Oh. S. H. Woo, K. H. Kim, J.-H. Kim. H. D. Kim. and
H. S. Kang. 2001. Implication of a small GTPase Rac 1 in the
activation of c-Jun N-terminal kinase and heat shock factor in response
to heat shock. J. Biol. Client. 276: 1889-1895.

384

B. S. T. HELMUTH AND G. E. HOFMANN

Helmut!), B. S. T. 1998. Inteilidal mussel microclimates: Predicting the
body temperature of a sessile invertebrate. Ecol. Monogr. 68: 29-52.

Ht'lmuth, B. 1999. Thermal biology of rocky intertidal mussels: quan-
tifying body temperatures using climatological data. Ecology 80: 15-
34.

Helmulh, B., and M. W. Denny. 1999. Measuring scales of physical
stress in the rocky intertidal. Am. Zoo/. 39: 1 14A (abstract).

Hofmann, G. E., and G. N. Somero. 1995. Evidence for protein damage
at environmental temperature: seasonal changes in levels of ubiquitin
conjugates and Hsp70 in the intertidal mussel Mytilus tmssnliis. J. Exp.
Biol. 198: 1509-1 5 18.

Hofmann, G. E., and G. N. Somero. 1996a. Interspecific variation in
thermal denaturation of proteins in the congeneric mussels Mytilus
trossulns and M. galloprovincialis: evidence from the heat-shock re-
sponse and protein ubiquitinution. Mar. Biol. 126: 65-75.

Hofmann, G. E., and G. N. Somero. 1996b. Protein ubiquitination and
stress protein synthesis in Mytiliix trossulus occurs during recovery
from tidal emersion. Mol. Mar. Biol. Biotechnoi. 5: 175-184.

Hutchins, L. W. 1947. The bases for temperature zonation in geograph-
ical distribution. Ecol. Monogr. 17: 325-335.

Johnson, S. E. II. 1975. Microclimate and energy flow in the marine
rocky intertidal. Pp. 559-587 in Perspectives of Biophysical Ecology.
D. M. Gates and R. B. Schmerl. eds. Springer- Verlag. New York.

Kline, M. P., and R. I. Morimoto. 1997. Repression of the heat shock
factor 1 transcriptional activation domain is modulated by constitutive
phosphorylation. Mol. Cell. Biol. 17: 2107-2115.

Lin, L. Z., Z.-W. Hu, J. H. Chin, and B. B. Hoffman. 1997. Heat shock
activates c-Src tyrosine kinases and phosphatidylinositol 3-kinase in
N1H3T3 fibroblasts. ./. Biol. Client. 272: 31 196-31202.

I ni(li|iiisi. S. 1986. The heat shock response. Annu. Re\: Biochem. 55:
1151-1191.

Liu, J. H., and B. Morton. 1994. The temperature tolerances of Tetra-
clita sijuamosu (Crustacea: Cirripedia) and Septifer virgatus (Bivalvia:
Mytilidae) on a sub-tropical rocky shore in Hong Kong. /. Zoo/. Loml.
234: 325-339.

Menconi. M., L. Benedetti-Cccchi, and F. Cinelli. 1999. Spatial and
temporal variability in the distribution of algae and invertebrates on
rocky shores in the northwest Mediterranean. J. Exp. Mar. Biol. Ecol.
233: 1-23.

Menge, B. A., and A. M. Olson. 1990. Role of scale and environmental
factors in regulation of community structure. Trends Eeol. Evol. 5:
52-57.

Menge, B. A., B. A. Daley, P. A. Wheeler, E. Dahlhoff, E. Sanford, and
P. T. Strub. 1997. Benthic-pelagic links and rocky intertidal com-
munities: bottom-up effects on top-down control'1 Proc. Natl. Acini.
Sci. 94: 14,530-14.535.

Morimoto, R. I. 1998. Regulation of the heat shock transcriptional
response: cross talk between a family of heat shock factors, molecular
chaperones. and negative regulators. Genes Dev. 12: 3788-3796.

Ng, D. C. H., and M. A. Bogoyevitdi. 2(100. The mechanism of heat
shock activation of ERK niitogcn-uctivated protein kmiises m the
interleukin 3-dependent ProB cell line BaF3. ./. Biol. Client. 275:
40,856-40.866.

Orton. J. H. 1929a. Observations on Patella vttlgata Part III. Habitat and
habits. ./. Mar. Biol. As. we. UK 16: 277-288.

Orton, J. H. 1929b. On the occurrence of Echinus e.iculetitiis on the
foreshore in the British Isles. J. Mar. Biol. Assoc. UK 16: 289-29h.

Petersen, R. B., and S. Lindquist. 1988. The Dro.wphila Hsp70 mes-
sage is rapidly degraded at normal temperatures and stabilized by heat
shock. G, - 72: 161-168.

Petersen, R. B., and S. Lindquist. 1990. Differential mRNA stability: A
regulatory strategy tor Hsp70 synthesis. Pp. 83-91 in Posttranscrip-
tional Control of Gene Expression. J. E. G. McCarthy and M. F. Tuite,
eds. Springer-Verlag. New York.

Porter, \V. P., and D. M. Gates. 1969. Thermodynamic equilibria of
animals with environment. Ecol. Monogr. 39: 245-270.

Porter, W. P., J. W. Mitchell, W. A. Beckman, and C. B. DeVVitt. 1973.
Behavioral implications of mechanistic ecology. Thermal and behav-
ioral modeling of desert ectotherms and their microenvironment.
Oecologia 13: 1-54.

Roberts, D. A., G. E. Hofmann, and G. N. Somero. 1997. Heat-shock
protein expression in Mytiliis californiamis: acclimatization (seasonal
and tidal-height comparisons) and acclimation effects. Biol. Bull. 192:
309-320.

Sagarin, R. D., J. P. Barry, S. E. Gilman, and C. H. Baxter. 1999.
Climate-related change in an intertidal community over short and long
time scales. Ecol. Monogr. 69: 465-490.

Sanford, E. 1999. Regulation of keystone predation by small changes in
ocean temperature. Science 283: 2095-2097.

Shi. Y., D. D. Mosser, and R. I. Morimoto. 1998. Molecular chaper-
ones as HSFl-specific transcriptional repressors. Genes Dev. 12: 654-
666.

Snyder, M. J., E. Girvetz, and E. P. Mulder. 2001. Induction of marine
mollusc stress proteins by chemical or physical stress. Arch. Environ.
Contain. To.\icol. 41: 22-29.

Stillman. J. H., and G. N. Somero. 1996. Adaptation to temperature
stress and aerial exposure in congeneric species of intertidal porcelain
crabs (genus Petrolisthes): correlation of physiology, biochemistry and
morphology with vertical distribution. J. E\p. Biol. 199: 1845-1855.

Suchanek, T. H. 1978. The ecology of Mytiliis edulis L. in exposed
rocky intertidal communities. J. E.\p, Mar. Biol. Ecol. 31: 105-120.

Thompson, R. C., M. F. Roberts, T. A. Norton, and S. J. Hawkins.
2000. Feast or famine for intertidal grazing molluscs: a mis-match
between seasonal variations in grazing intensity and the abundance of
microbial resources. Hytlrohiologia 440: 357-367.

Tomanek, L., and G. N. Somero. 1999. Evolutionary and acclimation-
induced variation in the heat-shock responses of congeneric marine
snails (genus Tegula) from different thermal habitats: implications for
limits of thermotolerance and biogeography. J. Exp. Biol. 202: 2925-
2936.

Tsuchiya, M. 1983. Mass mortality in a population of the mussel Mytilus
edulis L. caused by high temperature on rocky shores. J. Exp. Mar.
Biol. Ecol. 66: 101-111.

\Vethey, I). S. 1983. Geographic limits and local zonation: the barnacles
Semihalanus (Balanus) and Chrhainaliis in New England. Biol. Bull.
165: 330-341.

\Vethey, I). S. 1984. Sun and shade mediate competition in the barnacles
Chlhamahis and Semihalanus: a field experiment. Biol. Bull. 167:
176-185.

\\iddo\\s, J. 1976. Physiological adaptation of Mytilus cJulis to cyclic
temperatures. ./. Comp. Phvsinl. 105: 115-128.

Williams, G. A., and I). Morritt. 1995. Habitat partitioning and thermal
tolerance in a tropical limpet. Ce/lana grata. Mar. Ecol. Prog. Ser. 124:
89-103.

Yost, H., R. B. Petersen, and S. Lindquist. 1990. RNA metabolism:
strategies for regulation in the heat shock response. Trends Genet. 6:
223-227.

Zhong. M., A. Orosz, and C. Wu. 1998. Direct sensing of heat and
oxidation by DrosophiUi heat shock transcription factor. Mol. Cell 2:
101-108.

Reference: Bio/. Bull. 201: 385-393. (December 2001)

Competition for Space Among Sessile Marine

Invertebrates: Changes in HSP70 Expression

in Two Pacific Cnidarians

SERGI ROSSI* AND MARK J. SNYDERt

University of California, Davis, and Bodega Marine Laboratory,
P.O. Box 247. Bodega Bay, California 94923

Abstract. The role of stress proteins — either constitutive
(HSC) or inducible (HSP)— of the HSP70 family in intra-
and interspecific competition for space was examined in two
sessile Pacific cnidarians. Anthopleura elegantissima, an
intertidal anemone, and Corynactis califomica. a subtidal
corallimorpharian, express HSP70 in the absence of appar-
ent physical stress. HSP70 protein expression is concen-
trated in the tentacles of A. elegantissima when the animal
is exposed to contact with other benthic organisms. Under
the same conditions, however, HSP concentrations are sim-
ilar in the body and tentacles of C. califomica. When two
different clones of A. elegantissima interact in the field, the
outside polyps (warriors) express more HSP70 than the
inside ones (2.4 versus 0.6 ng HSP70//xg Protein). When
different C. califomica clones interact, HSP70 expression in
the outside and inside polyps is similar ( 1 .5 versus 1 .8 ng
HSP70/ju.g P) and is fairly constant in the corallimorpharian
in the different interspecific encounters. HSP70 expression
is related to the different kinds of aggression encountered by
both cnidarians. HSP70 expression may be involved in the
recovery of tissues damaged by the allelochemical, cyto-
toxical, or corrosive substances produced by different ene-
mies. C. califomica clones appear prepared for war, as
evidenced by the high constant expression of HSP70 in the
polyps. A. elegantissima exhibits differential HSP70 expres-
sion depending on the identity of each neighboring intra- or
interspecific sessile competitor. We propose that stress pro-

Received 26 January 2001; accepted 22 May 2001.

* Current address: Institut de Ciences del Mar. Passeig Nacional, s/n.
08003 Barcelona. Spain.

t To whom correspondence should be addressed. E-mail: mjsnyder@
ucdavis.edu

teins can be used to quantify space competition or aggres-
sion among sessile marine invertebrates.

Introduction

Space on which to live is often the most limiting resource
in marine hard-substratum environments, and patchiness
has evolved under the influence of intense competition for
living space (Connell. 1961; Pequegnat. 1964; Paine, 1971;
Dayton, 1971; Jackson, 1977). Once established, organisms
can show aggressive behavior (Chadwick, 1987) that may
be especially intense in cryptic environments where free
space is almost nonexistent.

In benthic environments, sponges, ectoprocts, cnidarians,
and ascidians can produce biologically active substances
that may be destructive to enemies during space competition
(Whittaker and Feeny, 1973; Uriz et «/.. 1991). These
organisms aggregate in patches that can dominate hard-
bottom substrates (Sutherland. 1978; Chornesky, 1983;
Chadwick. 1987, 1991; Chadwick and Adams, 1991; Lang-
mead and Chadwick, 1999a, b, among others). Growth is
often slow in such organisms, and interactions between
competitors are often nonevident. It is difficult to quantify
competitive interactions in situ, and the manipulation of
organisms is frequently essential to demonstrate the poten-
tial effects of space competition (Schoener. 1983). For
example, investigators have rarely observed agonistic inter-
actions in wild anemones (A. xanthogrammicd), although
these organisms frequently exhibit such behavior in forced
situations (Sebens, 1984). The quantification of damage
from encounters between such organisms and the identifi-
cation of potential mechanisms used to counter the effect of
such aggression have proved difficult. Most studies have
dealt with the organismal responses to the attack and the

385

386

S. ROSSI AND M. J. SNYDER

consequent aggressive behavior displayed by individuals.
Few workers have focused on the capacities, and implied
mechanisms, for tissue recovery following aggressive inter-
actions. We hypothesize that components of the stress re-
sponse such as HSPs may provide evidence of the intensity
of competitive interactions and are one of the mechanisms
by which cnidarians recover from or prepare their tissues for
the effects of competitive or aggressive interactions.

HSPs enhance cell survival by reducing the accumulation
of damaged or abnormal polypeptides within cells (Feder
and Hofmann, 1999). However, whether all wild organisms
routinely, occasionally, or seldom express inducible HSPs is
unknown. For marine invertebrates, most investigators have
examined the effects of thermal variations on constitutive
(HSC70) and inducible (HSP70) responses (Feder and Hof-
mann, 1999). Competitive interactions between sessile or-
ganisms can elicit HSP responses due to protein damage
following the excretion of harmful substances by one or
both competitors (Uriz et ai, 1991; Turon et al., 1996:
Wiens et al.. 1998). One index of tolerance to aggressive
sessile organisms could be the presence and abundance of
mechanisms (such as HSPs) that would resist or ameliorate
the damage inflicted on cellular components by the potential
space competitor. Furthermore, once HSP can be related to
space competition, no manipulation will be necessary to test
such hypotheses. HSP expression could then be a quantita-
tive tool to examine competitive interactions in the field
without human interference.

To determine whether HSP expression patterns could be
related to competitive interactions in marine hard-bottom
sessile invertebrates, two Pacific cnidarians were chosen for
study: the intertidal anemone Anthoplenra elegantissima
and the subtidal corallimorpharian Coiynactis californica.
A. elegantissima forms contiguous aggregations composed
of individuals of a single clone, the products of asexual
reproduction (Francis, 1973b: Sebens, 1982a. b). Free zones
are created where competition between clones occurs
through the outside polyps of the aggregation (called "war-
riors," Francis, 1973a). Compared with polyps in the center
of the clone, the warriors have larger and more abundant
acrorhagi (specialized nonfeeding tentacles) and lack ma-
ture gonads (Francis, 1973b, 1976). The aggressive re-
sponse is not directly involved in either defense against
predators or capture of prey (Francis, 1973b), but functions
in the competition for space. We hypothesize that A. elegan-
tissima warriors may exhibit higher HSP levels than interior
clonemates because they interact more frequently with com-
petitors.

In the subtidally distributed C. californica, the polyps
have no distinctive roles within each clone (Chadwick,
1987). Although the physiology of this group is not as well
understood as that of anemones, several studies have de-
scribed the competition for space and the specific responses
to aggression in corallimorpharians (Chadwick, 1987. 1991 ;

Chadwick and Adams, 1991: Langmead and Chadwick,
1999a. b). Space competition experiments demonstrate that
C. californica influences the abundance and population
structure of other cnidarians by means of its aggressive
behavior (Chadwick. 1987. 1991; Chadwick and Adams,
1991 ). We sought to determine whether the high aggression
in this species is related to elevated HSP levels as prepara-
tion for possible damage resulting from such interspecies
encounters.

We tested two main hypotheses in this work: first, that
stress produced by space competition can induce HSP ex-
pression to counter the effects of aggressive neighbors;
second, that HSP expression can provide a quantitative
assay for space competition in sessile invertebrates.

Materials and Methods

Animals and treatments

Anthoplenra elegantissima and Coiynactis californica
were collected from the Bodega Bay area and held in the
running seawater system of the Bodega Marine Laboratory.
All animals were held in ambient seawater (13-15 °C) and
fed adult brine shrimp or frozen seafood. The seawater from
the Bodega Bay area is considered clean, and the animals
used in these experiments are considered to have had min-
imal contact with anthropogenic chemicals that are known
to induce HSP expression (McCain et al.. 1988). All exper-
iments (aquarium and field) were done in September-
October 1998 and 1999 to avoid seasonal differences in
cnidarian behavior. Each experiment, whether forced inter-
actions in an aquarium or in situ interaction, was designed
to assess the effects of neighboring competition for space on
HSP70 expression.

Forced aquarium experiments

The first experiment examined HSP70 protein expression
in A. elegantissima and C. californica in a forced situation.
Six isolated polyps of each species (attached to stones, no
physical stress induced) were moved into contact with each
other (i.e., one polyp of A. elegantissima against one polyp
of C. californica). After 24 h, tentacle samples from three
individuals of each species were removed and frozen in
liquid nitrogen. To quantify the differences between tenta-
cles and body, the other three polyps of each species were
sampled 48 h later, frozen in liquid nitrogen, and then
assayed for HSP70 level by methods detailed below. As
controls, isolated polyp tentacles (n = 5-6, no interacting
species) of A. elegantissima and C. californica were like-
wise sampled in the aquarium.

In situ intraspeciftc competition

We assessed HSP70 expression related to competition for
space in a natural environmental situation (i.e., in natural

COMPETITION AND HSP70 IN CNIDAR1ANS

387

clones in the field). Because collection and transport of
animals to artificial holding conditions can stimulate a stress
response (Sharp e' ai. 1994; Roberts et <//., 1997), clones of
A. elegantissima and C. californica were located and sam-
pled from the Bodega Bay Jetty from a minimum 2 in below
the 0 tide level (permanently submerged). This avoided
significant desiccation, changes in temperature, fluctuations
in salinity and pH, and other effects that are typical of the
environment for the intertidal A. elegantissima but not for
the subtidal C. californica.

For the A. elegantissima intraspecific competition exper-
iments, clones were located by scuba and photographed
(Nikonos V camera, 35-mm lens with macro 1 : 1 or close-up
lens). Polyps of each clone were sampled (;; = 3, tentacles)
from the outside (touching the competitor) and the inside
(touching only the same clone. 10-20 cm from the outside
polyps). Samples were dissected, kept in 13 °C seawater for
no longer than 30 min before freezing in liquid nitrogen, and
stored at — 70 °C. As a control to assess whether HSP70
levels were affected by the extra 30-min tissue incubation in
ambient seawater before freezing, the following experiment
was performed. Individual tentacle samples were obtained
from three individuals of two clones exposed to elevated
temperatures in the intertidal zone (elevated HSP70 is found
in these conditions, Snyder and Rossi, unpubl. obs.). Each
sample was divided into three parts, of which two were
immediately frozen in liquid nitrogen and the third was
submerged in ambient seawater for 40 min prior to freezing
as above.

For the C. californica intraspecific competition experi-
ments, six clones were located and sampled as above. Color
varies greatly between different clonal aggregations, which
is useful in distinguishing clones that show potential in-
traspecific competition. Outside and inside polyps (tentacle
crowns) of each clone were sampled to compare interacting
(<2.5 mm apart) and non-interacting individuals (5-10 cm
apart from the outside ones).

Interspecific competition

To examine the effects that different space competitors in
the benthic substrata have on HSP70 protein levels, we
chose two genera of algae that compete for space with A.
elegantissima and C. californica and two intertidal and two
subtidal invertebrates for A. elegantissima and C. califor-
nica, respectively. The sampled and photographed anemone
clones were always submerged (as described before).

Four clones of A. elegantissima and three of C. califor-
nica that were interacting with a calcareous red alga
(Lithothamnium sp.) were dissected (outside and inside
clone tentacles). Another alga interacting with both cnidar-
ians was a fleshy green alga (Viva sp.), and six clones of
each cnidarian were sampled as above.

In the high subtidal, common space competitors of A.

elegantissima are the anemone A. xanthogrammica and the
cirriped Balanns ampliitrite. Five A. elegantissima clones
interacting with A. xanthogrammica were sampled in the
outside and inside parts of the clones. For B. ampliitrite.
three clones competing for space were likewise sampled.
For C. californica. the subtidal organisms chosen (sponge
Haliclona permollis; ascidian Synoicum parfiistis) were
considered potentially more aggressive than the fleshy al-
gae. Six C. californica clones were chosen for their clear
interactions with H. permollis, and polyps of the outside and
inside part of each clone were dissected. For S. parfustis, the
interaction of the clones was observed in four populations in
the dive area, and outside and inside polyps were sampled.

HSP70 measurements

The western immunoblotting for HSP70 expression was
done as follows. Frozen tentacle samples (stored at —70 °C)
were individually homogenized in 0.2 ml of buffer K con-
taining 5 mA/ NaHPO4. 40 mM HEPES (pH 7.4). 5 mM
MgCK. 70 mA/ potassium gluconate. 150 mM sorbitol, and
1% SDS. Homogenates were centrifuged 10 min at
10,000 X g. and the supernatants were combined with equal
volumes of SDS sample buffer (Laemmli, 1970) and boiled
for 5 min. Supernatant protein levels were determined by
BioRad DC assay, and 20 |Ug of tentacle protein was loaded
in each gel lane. For each blot, 50 ng of standard HSP70
protein (human, StressGen) was included. Discontinuous
SDS gels (1 mm) were 6.2% for the stacking gel and 12%
for the resolving gel. After running for 2 h at 150 V, SDS
gels were electroblotted onto PDVF membranes (for 1 h at
100 V). The protein bands in each western blot were visu-
alized by staining with Ponceau S. HSP70 protein was
detected with mouse monoclonal anti-HSP70 (SPA-822,
StressGen, Victoria, BC); the secondary antibody was goat-
anti-mouse IgG, conjugated to peroxidase (Sigma), and was
visualized with ECL reagents (Amersham) and exposure of
blots to X-ray film.

Blot band intensities were compared by scanning the
X-ray films and analyzing the scans with the NIH Image
software package. For each blot, the scanned intensity of the
HSP was normalized against the intensities of the HSP70
protein standard from that blot; that is, the NIH Image
datum point was divided by the intensity of the HSP70
standard.

Results

Anthopleura elegantissima and Corynactis californica
express a single HSC70 or HSP70 protein (Fig. 1 ). In other
eukaryotes, the HSP70-DnaK protein family comprises
multiple proteins, more than one of which may be detected
by the antibody. For the sake of convenience, we will
collectively term these as "HSP70." The inclusion of pro-
tease inhibitors did not affect HSP70 levels (Fig. 1A,

388

S, ROSSI AND M. J. SNYDER

Anthopleura 1 Anthopleura 2
a b c a b c

•c 3-5

Anthopleura

Corynactis

[•^ !••«••••• —

Tl T2 T3 B1 82 B3 T1 T2 T3 B1 B2 B3

Figure 1. Western blots of HSP70 levels in Anthopleura elegantissima
and Corynactis californica: comparison of tentacles under different sam-
pling conditions and body without tentacles. In (A), triplicate tentacle
samples were taken from two A. elegantissima individuals: (a) the first of
the triplicate samples was immediately frozen in liquid nitrogen, (b)
duplicate of (a) with the addition of protease inhibitors prior to homoge-
mzation, and (c) third sample from each anemone kept in a sample bag
submerged at 13 °C for 40 min prior to freezing in liquid N:. In (B), three
individuals from each species were divided into tentacles only or body
minus tentacles prior to homogenization.

Anthopleura 1 and 2. a versus b); therefore they were
omitted from our studies during the homogenization steps.
The 30-min ambient seawater submersion of subtidal ten-
tacle samples prior to freezing had no effect compared with
immediate freezing (Fig. \ A. Anthopleura 1 and 2, c versus
a and b). In comparing tentacles of the same polyp 24 h after
the first forced interaction between the two cnidarian spe-
cies in the laboratory, no differences were observed (F(3,
8) = 2.0. P < 0.1929) (Fig. 2). Two days later, HSP70
levels in A. elegantissima tentacle were 4 times greater than
before (4.0 ± 0.5 ng HSP70//Lig P in the tentacles; 0.0 ± 0.1
ng HSP70/jug P in the body, power of test = 0.87), but no
differences were detected in C. californica tentacles (1.7 ±
0.9 ng HSP70//j,g P in the tentacles; 0.8 ± 0.9 ng HSP70/fig
P in the body) (Fig. 2). Differences between tentacles and
body were found in A. elegantissima but not in C. califor-

Figure 2. HSP70 expression in tentacles of Anthopleura elegantissima
and Connacris californica at time 0, 24 h after the first contact of the
cnidarians (A. eleg. clones 1 and 2. black and stippled; C. calif. Clones 1
and 2. white and stippled), and 48 h later in both tentacles and body
(without tentacles, stippled) of the same polyps in A. elegantissima and C
californica. The bars are + 1 standard deviation of 3-6 samples. Asterisks
indicate significant differences between groups (P £ 0.05); ns indicates a
lack of significant differences between groups.

Z.5
2

0> 15-

Q_
(0

0.5
0

A. eleg. Out A. eleg. In

C. calif. Out

C. calif. In

Figure 3. Intraspecific competition. HSP70 expression between tenta-
cles of the inside and outside polyps in Anthopleura elegantissima and
Corynactis californica in intraspecific conditions. The bars are + 1 standard
deviation of 4-5 clones. Asterisks indicate significant differences between
groups (P s 0.05); ns indicates a lack of significant differences between
groups.

nica (Fig. 1; F(3, 8) == 18.55. P < 0.0006, power of
test = 0.98). Algal symbionts are at the highest concentra-
tion in A. elegantissima oral disk (Fitt et al., 1982; Weis and
Levine, 1996); these data imply that we are measuring
HSP70 responses in animal tissue. No such differences were
found in the corallimorpharian, which lacks algal symbi-
onts.

HSP70 levels in isolated polyps were also examined
under the same conditions (no contact with any other inver-
tebrate). A. elegantissima tentacles had very low expression
(0.2 ± 0.3 ng HSP70/jLtg P) compared with the previous
contact experiments. C. californica had high expression
(2.1 ± 1.3 ng HSP70/ju,g P) even when there was no direct
(contact) aggression present. Comparing this analysis with
the anemone-corallimorpharian experiments, no differences
were found between HSP70 expressions in C. californica.
There were differences in the HSP70 expression of polyps
between the two cnidarians when they were compared to-
gether (F(\. 9) == 10.81, P < 0.0094).

The mean distance between competitors in field studies as
determined from the photographs was 2.4 ± 0.9 mm (// =
17). This distance is clearly within the range that A. elegan-
tissima tentacle crowns sway during seawater movements
(Francis, 1973a). The results of intraspecific competition in
selected patches of both cnidarians are shown in Figure 3.
There were clear differences in A. elegantissima HSP70
expression between the outside warrior polyps and the in-
side ones (in contact, 2.4 ± 0.5 ng HSP70//Lig P; no contact,
0.6 ± 0.7 ng HSP70//xg P; F(3. 20) = 3.93. P < 0.0234,
power of test = 0.82) when two clones of the same species
interacted. Interestingly. C. californica had similar HSP70
amounts in polyps of different clones (outside 1.5 ± 1.1 ng
HSP70Vg P; inside 1.8 ± 1.3 ng HSP70/Mg P).

The regular cnidarian HSP70 expression in both outside
and inside polyps of the clone in different competition-for-
space situations is illustrated in Figure 4. A. elegantissima

COMPETITION AND HSP70 IN CNIDARIANS

389

Corynactis
vs vs
Ulva Mali.
10 i o

Anthopleura
vs A. xantho vs Ulva

A. xanthogrammica

B a Ian us

i o i o HSP70

Figure 4. Western blot of HSP70 levels in Anthopleura elegantissima

and Con-worm culitornica tentaeles from inside not interacting (i) and
outside interacting (o) analyzed with competitors in the field. C. culifornica
competitors were Ulva sp. and H. perinollis. A. elegantissima competitors
were .A. xanthogrammica and Ulva sp.

had more HSP70 in the warriors than in the inside clone
polyps in general, depending on the competing species (Fig.
4). In Figure 5A, B we show HSP70 levels when both
cnidurians interacted with the same competitors in the Held:
crustose red (Lithothamnium sp.) and fleshy green (Ulva
sp.) algae. Contact with Lithothamnium (Fig. 5 A) resulted
in higher HSP70 expression in the outside A. elegantissima
clone polyps (warriors, 2.4 ± 1.2 ng HSP//o.g P; inside ones
0.5 ± 0.4 ng HSP/jug P. F(3. 10) = 4.82. P < 0.025.
power of test = 0.80). No differences were found between
the inside and outside C. californica polyps in interactions
with either algal species (outside 1.2 ± 0.4 ng HSP/jug P;
inside 1.5 ± 0.5 ng HSP/jug P).

Neither cnidarian showed any significant difference in
HSP70 between inside and outside polyps (Fig. 5B). In A.
elegantissima, the inside polyps ( 1.0 ± 0.8 ng HSP70//ig P)
were similar to the outside ones (0.6 ± 0.6 ng HSP70/jng P).
The expression was also similar for both clone polyps in C.
californica (outside 1.2 ± 0.6 ng HSP70//j.g P: inside 1.3 ±

A. eleg. Out A. eleg. Out C. calif. Out C. calif. In

A. eteg. Out

A. eleg. In

Figure 5. Interspecific competition I. HSP70 expression between ten-
tacles of the inside and outside polyps in Anthopleura elegantissima anil
Corynactis californica in contact with calcareous red (Lithothamnium sp.)
(A) and fleshy green (Ulva sp.) (B) algae. The bars are +1 standard
deviation of 4-6 clones. Asterisks indicate significant differences between
groups (P £ 0.05); ns indicates a lack of significant differences between
groups.

1.5-
1.2'

09'
.E °-6'

* c.

T ' T

S® |

I
0) 51

A. eleg Out A. eleg. In A. eleg. Out A. eleg.

Haliclona Synoicum

C. calif. Out C. calif. In

C. calif. Out C. calif, in

Figure 6. Interspecific competition II. HSP70 expression between
tentacles of the inside and outside polyps in Anthopleura elegantissima and
Corynaclis californica with different competitors. (A) .4. elegantissima
against .4. xanthogrammica and Ba/anu.\; (B) C. californica against Hali-
clona perinollis and Synoicum parfustis. The bars are + 1 standard devia-
tion of 3-5 clones. Asterisks indicate significant differences between
groups (P £ 0.05); ns indicates a lack of significant differences between
groups.

0.8 ng HSP70//Lig P). C. californica HSP70 expression was
always the same in the outside and inside polyps (1-1.8 ng
HSP70//u,g P) in encounters with either A. elegantissima.
other C. californica clones, or either algal species.

For A. elegantissima, two intertidal competitors were
tested in submersed conditions: A. xanthogrammica and
Balanus amphitrite (Fig. 6A). Encounters with A. .\antlio-
grammica resulted in higher HSP70 in A. elegantissima
outside polyps (0.6 ± 0.2 ng HSP70/jug P; inside ones
0.1 ± 0.1 ng HSP70/|Ug P, F(3. 12) = 2.88, P < 0.048,
power of test = 0.99). However. HSP70 levels were low
compared with other situations (interactions with calcareous
algae or other A. elegantissima clones). No differences in
HSP70 level were found with the B. amphitrite interactions
(outside 0.5 ± 0.6 ng HSP70/jug P: inside 0.4 ± 0.4 ng
HSP70//xg P).

Differences in C. californica HSP70 levels occurred
when potential encounters and fights for space were against
the sponge Haliclona perinollis or the ascidian Synoicum
paifustis (Fig. 6B). HSP70 expression was the same in the
outside and inside polyps, but was slightly higher than with
other competitors. Both sponge and ascidian appear to ac-
tivate higher HSP70 expression (H. perinollis outside 3. 1 ±
0.5 ng HSP70/jug P: inside 2.5 ± 0.5 ng HSP70/jug P: S.
parfustis outside 2.4 ± 1.0 ng HSP70/jug P; inside 1.8 ± 0.6
ng HSP70/jug P). Again, no significant differences were
found between inside and outside polyps. When comparing
the response of this cnidarian against the sponge and the
ascidian with all the other encounters, significant HSP70
differences were found (F(5. 79) 18.58. P

0.00001). HSP70 expression in the sponge and ascidian

390

S. ROSSI AND M. J. SNYDER

encounters was 2.2 ± 0.7 ng HSP70/yu,g P, and in all the
other encounters (A. elegantixsiimi and C. californica. cal-
careous and fleshy algae) the HSP70 level was 1.3 ± 0.6 ng
HSP70/^g P.

Discussion

Anthoplcura elegantissima and Corynactis californica
express HSP70 without physical stress (e.g., from temper-
ature, desiccation, changes in pH) or pollution stress (e.g..
due to heavy metals, organochlorines). There are few ex-
amples of cnidarian HSP expression patterns, and all arc
directly (Bosch et ai. 1988; Bosch and Praetzel. 1991;
Sharp et til.. 1994) or indirectly (Hayes and King, 1995;
Sharp et ai, 1997; coral bleaching) related to temperature
stress. This is the first set of observations relating aquatic
invertebrate HSP levels to biological stress and relating
cnidarian HSP expression to parameters other than temper-
ature.

There were significant differences in HSP70 levels be-
tween the two cnidarians, and these depended on the par-
ticular competing species. Perhaps the aggressive behavior
of C. californica (Chadwick, 1987, 1991; Chadwick and
Adams. 1991) causes cellular damage, thereby increasing
HSP70 expression levels in A. elegantissima tentacles (Fig.
2) in the first aquarium experiments. C. culifornica extrudes
mesentarial filaments upon contact with nonfood species,
suggesting that this behavior is used in interspecies aggres-
sive encounters (Chadwick, 1987; Chadwick and Adams.
1991). Prolonged contact with C. californica mesentarial
filaments kills the competitor. In this forced situation, no
stresses other than contact between polyps appear to affect
the tentacles of both cnidarians. In comparison with isolated
(non-interacting) A. elegantissima polyps (Fig. 2). the ex-
pression of HSP70 is nearly 20 times greater after 48 h of
interspecific interactions. The differences shown between
tentacle crown and whole body in -4. elegantissima were not
found in C. californica.

The more striking result is the lack of differences be-
tween the solitary and interacting C. californica polyps in
the aquarium experiences (in Fig. 2, compare 24 and 48 h).
The expression of HSP70 is high and very constant in the
three interspecific encounters (1.3-2.1 ng HSP70//iig P).
One explanation could be that the aggressive behavior of
some corallimorpharians requires cellular protection to
counter the effect of the competing species' response
(Chadwick, 1987; Langmead and Chadwick. 1999a. b).
After a period of contact with C. californica. A. elegan-
tissima moved away via pedal locomotion, suggesting that
the specialized aggressive structures ot the anemone were
ineffective against the corallimorpharian (Francis, 1973a, b;
Chadwick, 1987).

Strong intraspecific competition has been clearly demon-
strated between clones of A. elegantissima (Francis. 1973a,

b; Ayre and Grossberg, 1995, 1996). Contact between ge-
netically different individuals of this species initiates elab-
orate behaviors involving acrorhagial contact (leaving
patches of tissue containing high numbers of nematocysts)
and results in damage to one or both competitors. In addi-
tion, anemones of the genus Anthopleura, including A.
Minthogrammica (discussed below), produce cytolytic and
sodium-channel toxins that presumably damage cellular
constituents such as proteins following contact (Bernheimer
and Lai, 1985; Cline and Wolowyk, 1997; Kelso and Blu-
menthal, 1998). These toxic mechanisms could explain the
high HSP70 levels found in the examined clones (Fig. 3).
The outside warrior polyps bordering neighboring clones
have more HSP70 than the inside ones. Sessile organisms
discontinuously fight for space, depending on growth and
reproductive cycles, the age of competitors, or the nature of
the enemies (Council, 1961; Jackson. 1977; Chadwick,
1991 ). Perhaps when warrior polyps encounter a "known"
competitor (i.e.. in this case a different clone of the same
species), they become "prepared for war." producing
HSP70 levels high enough to avoid serious cellular damage
when real interactions begin. Alternatively, some interac-
tions have already caused some tissue damage, resulting in
higher HSP70.

No differences in HSP70 expression were expected in
interactions between A. elegantissima and a fleshy green
alga (Ulva sp.. Fig. 5B). This algal type escapes from direct
competition for space by growing as rapidly as nutrients and
light levels permit (Lewis, 1964; Paine, 1971). No direct
interactions were evident, and the low HSP70 levels found
in the outside interacting polyps of these clones seem to
confirm their absence, although algae in this genus are
capable of producing harmful secondary compounds (Paine,
1990; Whitfield et <//.. 1999). In the case of Litholhtinmiiini
sp. (Fig. 5 A), it is known that coralline algae grow slowly
(Steneck, 1986: Garrabou and Ballesteros. 2000) and can
synthesixe allelochemicals (as do some other red algae) to
compete for space (Whitfield et <//., 1999). Perhaps the
anemone better detects or is more affected by these Litlwth-
aiiiiiiiiiii chemicals than by those produced by Ulvn.

A. xanthogrammica is a common intertidal competitor
with A. eleganiissima for space (Francis. 1973b; Sebens.
1984). This solitary anemone elicits aggression in A. el-
egantissinui (Francis. 1973b) but does not display the same
mechanisms of defense. Observations made by Sebens
( 1984) support the idea that aggression is common between
these two species, which explains the higher levels of
HSP70 in the outside A. elegantissima polyps in these
interactions (Fig. 6A). Balanits amphitrite, another common
space competitor, seems to have no effect on HSP70 ex-
pression (Fig. 6A). It is possible that the lack of effect was
due to exposure to small individual cirripeds, and it would
be interesting to examine A. elegantissima clones that are in
competition for space with larger clumps of barnacles.

COMPETITION AND HSP70 IN CNIDARIANS

391

In C. californica, HSP70 levels are similar in outside and
inside clone polyps. Therefore the corallimorpharian does
not distinguish between the exposed (outside polyps) and
nonexposed (inside polyps) areas of the clone. More impor-
tantly, even without apparent interactions (Fig. 2). C. cali-
fornica expresses HSP70 at constant levels (1-2 ng
HSP70/ju.g P). In this species, intraspecific competition re-
sults in HSP70 levels that are within the "normal" range
(Fig. 3). and there is no aggressive behavior in intraspecific
contacts (Chadwick. 1987). Perhaps the key to inteipreting
HSP70 expression as a mechanism of competence in C.
californica is the rinding that the highest HSP70 levels were
found in polyps interacting with Haliclona or S\iioiciiiii
(Fig. 6B). Also of importance is that these differences
between interacting and non-interacting polyps were signif-
icant. It is known that sponges and ascidians use chemical
substances to defend themselves or attack potential foes
competing for substrata (Green, 1977; Suchanek et al..
1985; Thompson et al., 1985; Turon et al.. 1996, 1998;
Becerro et til.. 1997). We suggest that HSP70 expression
differences found when the encounter involves ascidians or
sponges may reflect the aggressive toxic substances used by
these enemies (Uriz et ai. 1991).

C. californica appears to be always ''prepared for war" by
its aggressive behavior (Chadwick. 1991). Another organ-
ism that exhibits this strategic use of stress proteins (by
maintaining a basal level of HSP expression) is the desert-
dwelling ant Cataglyphys. This ant presynthesizes HSPs at
relatively low nest temperatures to limit damage from heat
shock on the desert floor. Coupled with continued HSP
production at higher temperatures, this protects the ant from
the high temperatures it experiences when foraging in day-
time (Gehring and Wehner, 1995). Perhaps the presynthesis
of HSP70 in C. californica provides protection from neigh-
bors that intermittently excrete harmful substances. Alter-
natively, the constant HSP70 levels might protect the cor-
allimorpharian against its own aggressive substances, which
it uses to catch prey and to fight for space (Chadwick.
1987). The aggressive behavior of C. californica includes
the extrusion of mesenteric filaments containing gland cells
that secrete strong proteolytic enzymes and nematocysts
that may inject cytolytic toxins into prey or enemies (Van-
Praet, 1985).

Because of the high cost of the HSP expression and its
occasional harmful effect if constantly highly expressed
(Feder et ai. 1992; Krebs and Feder. 1997). we suggest that
expression varies depending on the kind of neighboring
competitor or enemy. Furthermore. A. elegantissima also
expresses high levels of HSP70 in response to physical
factors, especially temperature (Rossi and Snyder. unpubl.
obs.). The anemone has to "share" HSP70 expression be-
tween biological (e.g.. competition for space) and physical
(e.g.. temperature) factors.

It is also possible that other stress proteins contribute to

the responses against biological phenomena such as com-
petitive interactions for space in the benthic environment.
For example, unexpected low-molecular-weight HSP70 ho-
mologs have been found in other cnidarians (Sharp el al..
1994). HSP60 has known roles in thermal acclimation of the
cnidarians Hydra nilgaris and Acropora grandis (Bosch et
al.. 1988; Fang et al.. 1997). The use of SPA-822 HSP70
antiserum can possibly underestimate the number of HSP70
isoforms, and consequently may explain the finding of sin-
gle HSP70 proteins by our methods. However, we have
successfully used the same antiserum and measured two and
three to four different HSP70 isoforms in larval lobsters,
(Homanis amehcaints). and juvenile abalone (Haliotis nife-
scens) and adult mussels (Mytiliis galloprovincialis) respec-
tively (Snyder and Mulder, 2001: Snyder et al.. 2001 ).

Many questions remain unanswered, such as the identity
of the harmful substances or aggressive behaviors that ac-
tivate HSP70 expression in competitive interactions among
sessile marine invertebrates. Among the likely candidates
for cellular damaging allelochemicals are cnidarian sodium-
channel toxins (Kelso and Blumenthal, 1998). cytotoxic and
cytolytic factors (Bernheimer and Lai. 1985; Cline and
Wolowyk, 1997), and an array of toxic alkaloids found in
cnidarians and sponges (e.g., Djura and Faulkner, 1980;
Koh and Sweatman, 2000). Such chemicals can diffuse and
act at some distance from the source or can be deposited on
neighboring organisms by direct contact (e.g., Schmitt et
al.. 1995: Slattery et at.. 1997). Further studies of HSP
proteins may provide important information about the con-
sequent distribution and hierarchy of species in the rocky
benthos.

With this work we propose HSP70 expression as a tool
for evaluating space competition among sessile marine in-
vertebrates, without manipulative experiments. From our
results, it is clear that the expression of the stress proteins
depends on both the particular competing species and the
interacting life stages of each competitor. The energy re-
quired to repair tissue damage cannot be used for other
processes such as reproduction and growth. It will be inter-
esting to measure how the amount of energy an organism
devotes to growth and reproduction varies with the level of
HSP produced during prolonged competition for space.

Acknowledgments

The manuscript was improved by the comments of Drs.
Josep-Marfa Gili, Cadet Hand, and several anonymous re-
viewers. This work was supported by the National Sea
Grant College Program, National Oceanic and Atmospheric
Administration. U.S. Department of Commerce, under grant
number NA66RG0477, project number R/A-108. through
the California Sea Grant College Program and an F.P.I.
fellowship from "Ministerio de Educacion y Ciencia" to
S.R. through the DGICYT grants PB94-0014-C02-OI and

392

S. ROSSI AND M. J. SNYDER

PB98-0496-C03-01. The views expressed herein are those
of the authors and do not necessarily reflect the views of
NOAA or any of its sub-agencies. The U.S. Government is
authorized to reproduce and distribute this publication for
governmental purposes. Contribution 2136 from the Bodega
Marine Laboratory, University of California at Davis.

Literature Cited

Ayre, D. J., and R. K. Grosbcrg. 1995. Aggression, habituation, and
clonal coexistence in the sea anemone Anthopleura elegantissima. Am.
Nat, 146: 427-453.

Ayre, D. J., and R. K. Grosberg. 1996. Effects of social organization on
inter-clonal dominance relationships in the sea anemone Anlhopleiira
elegantissima. Anim. Behav. 51: 1233-1245.

Becerro, M. A., X. Turon, and M. J. Uriz. 1997. Multiple functions for
the secondary metabolites in encrusting marine invertebrates. J. Cliem.
Ecol. 23: 1527-1547.
Bernheimer, A. W.. and C. Y. Lai. 1985. Properties of a cytolytic toxin

from the sea anemone. Stoichacyis kt-nli. Toxicon 23: 791-800.
Bosch, T. C. G., and G. Praetzel. 1991. The heat shock response in
Hydra: immunological relationships of hsp60. the major heat shock
protein of Hydra vulguris. to the ubiquitous hsp70 family. Hydrobio-
logia 216/217: 513-517.

Bosch, T. C. G., S. M. Krylow, H. R. Bode, and R. E. Steele. 1988.
Thermotolerance and synthesis of heat shock proteins; these responses
are present in Hvdra attenuata but absent in Hydra oligactis. Proc.
Nail. Acad. Sci. USA 85: 7927-7931.

Chadwick, N. E. 1987. Interspecific aggressive behavior of the coralli-
morpharian Corynactis ca/ifornicn (Cnidaria: Anthozoa): effects on
sympatric corals and sea anemones. Biol. Bull. 173: I 10-125.
Chadwick, N. E. 1991. Spatial distribution and the effects of competi-
tion on some temperate Scleractinia and Corallimorpharia. Mar. Ecol.
Prog. Ser. 70: 39-48.

Chadwick, N. E., and C. Adams. 1991. Locomotion, asexual reproduc-
tion and the killing of corals by the coral limorpharian Corynaclis
californica. Hydrobioloxia 216/217: 263-269.

Chnrnesky, E. A. 1983. Induced development of sweeper tentacles on
the reef coral Agaricia agaricites: a response to direct competition.
Biol. Bull. 165: 569-581.

dine, E. I., and M. W. Wolowyk. 1997. Cardiac stimulatory, cytotoxic
and cytolytic activity of extracts of sea anemones. Int. J. Pharmacogn.
35: 91-98.

Connell, J. H. 1961. The influence of interspecific competition and other
factors on the distribution of the barnacle Chthamalus stcllatus. Ecol-
ogy 42: 315-328.

Dayton, P. K. 1971. Competition, disturbance, and community organi-
zation: the provision and subsequent utilization of space in a rocky
intertidal community. Ecol. Monogr. 41: 351-389.
Djura, P., and D. J. Faulkner. 1980. Metabolites of the marine sponge

Dercims sp. J. Org. Chum. 45: 735-737.

Fang, L.-s., S.-p. Huang, and K.-l I. in. 1997. High temperature induces
the synthesis of heat-shock proteins and the elevation of intracellular
calcium in the coral Acroporu gruitdi*. Coral Reels 16: 127-131.
Feder, M. E., and G. E. Hot'manii. 1999. Heat-shock proteins, molec-
ular chaperones. and the stress response: evolutionary and ecological
physiology. Aniw. Rev. Physiol. 61: 243-282.

Feder, M. E., J. M. Rossi, J. Solomon, N. Solomon, and S. Lindquist.
1992. The consequences of expressing hsp70 in Drosophila cells at
normal temperatures. Genes Dev. 6: 1402-1413.

Fitt, W. K.. R. L. Pardy, and M. M. Littler. 1982. Photosynthesis,
respiration, and contribution to community productivity of the symbi-

otic sea anemone Anthopleura elegantissima (Brandt. 1835). J. Exp.

Mar. Biol. Ecol. 61: 213-232.
Francis. L. 1973a. Clone specific segregation in the sea anemone An-

thopleura elegantissima. Biol. Bull. 144: 64-72.

Francis. L. 1973b. Intraspecific aggression and its effect on the distri-
bution of Anthopleura elegantissima and some related sea anemones.

Biol. Bull. 144: 73-92.
Francis, L. 1976. Social organization within clones of the sea anemone

Anthopleura elegantissima. Biol. Bull. 150: 361-376.
Garrabou, J., and J. Ballesteros. 2000. Growth of Mesophyllum alter-

nan* and Lithophvlliun frondosum (Corallines. Rhodophyta) in the

northwestern Mediterranean. Eur. J. Phycol. 35: 1-1(1.
Gehring, W. J.. and R. Wehner. 1995. Heat shock protein synthesis and

thermotolerance in Cataglyphis, an ant from the Sahara desert. Proc.

Null. Acad. Sci. USA 92: 2994-2998.
Green, G. 1977. Ecology of to.xicity in marine sponges. Mar. Biol. 40:

207-215.
Hayes, R. L., and C. M. King. 1995. Induction of 70-kD heat shock

protein in scleractinian corals by elevated temperature: significance for

coral bleaching. Mol. Mar. Biol. Biolechnol. 4: 36-42.
Ireland. C. M.. D. M. Roll, T. F. Molinski, T. C. McKee, T. M.

Zahriskie. and J. C. Swersey. 1988. Uniqueness of the marine

chemical environment: categories of marine natural products from

invertebrates. Biomedical importance of marine organisms. Mem.

Calif. Acad. Sci. 13: 41-57.
Jackson, J. B. C. 1977. Competition on marine hard substrata: the

adaptive significance of solitary and colonial strategies. Am. Nat. Ill:

743-767.

Kelso, G. J., and K. M. Blumenthal. 1998. Identification and charac-
terization of novel sodium channel toxins from the sea anemone An-

tli{>plcura xaiithogramtnica. Toxicon 36: 41-51.
Koh, E. G. L., and H. Sweatman. 2000. Chemical warfare among

scleractinians: bioactive natural products from Tubastraea faulkneri

Wells kill larvae of potential competitors. J. Exp. Mar. Biol. Ecol 251:

141-160.
Krebs, R.. and M. E. Feder. 1997. Deleterious consequences of Hsp70

overexpression in Droxophila melanogaster larvae. Cell Stress Cliap-

ITI-HO 2: 60-71.

Laenimli. U. K. 1970. The cleavage of structural proteins during assem-
bly of the head of bacteriophage T4. Nature 227: 680-685.
Langmead. O., and N. E. Chadwick. 1999a. Marginal tentacles of the

corallimorpharian Rhodactis rhodostoma. 1. Role in competition for

space. Mar. Biol. 134: 479-489.
Langmead, O., and N. E. Chadwick. 1999b. Marginal tentacles of the

corallimorpharian Rhoilactix rhodostoma. 2. Induced development and

long-term effects on coral competitors. Mar. Biol. 134: 491-500.
Lewis, J. R. 1964. The Ecology of Rocky Shores, English Universities

Press. London.
McCain, B. B.. D. W. Brown, M. M. Krahn, M. S. Myers, R. C. Clark,

S.-L. Chan, and D. C. Malins. 1988. Marine pollution problems.

North American West Coast. Aauat. Toxicol. 11: 143-162.
Paine, R. T. 1971. A short-term experimental investigation of resource

partitioning in a New Zealand rocky intertidal habitat. Ecology 52:
1096-1106.
Paine. R. T. 1990. Benthic macroalgal competition: complications and

consequences. J. Phycol. 26: 12-17.
Pequegnat, W. E. 1964. The epifauna of a California siltstone reef.

Ecology 45: 272-283.
Roberts, D. A., G. E. Hofmann, and G. N. Somero. 1997. Heat shock

protein expression in A-/V/I/H.V califomianus: acclimatization (seasonal

and tidal-height comparisons) and acclimation effects. Biol. Bull. 192:

309-320.
Schmitt. T. M., M. E. Hay, and N. Lindquist. 1995. Constraints on

COMPETITION AND HSP70 IN CNIDARIANS

393

chemically mediated coevolution: multiple functions for seaweed sec-
ondary metabolites. Ecology 76: 107-123.
Schoener, T. VV. 1983. Field experiments on interspecific competition.

Am. Nat. 122: 240-285.
Sebens, K. P. 1982a. The limits to indeterminate growth: an optimal size

model applied to passive suspension feeders. Eco/ogv 63: 209-222.
Sebens, K. P. 1982b. Asexual reproduction in Anthop/eum elegan-

rissima (Anthozoa:Actinaria): seasonality and spatial extent of clones.

Ecology 63: 434-444.
Sebens, K. P. 1984. Agonistic behavior in the intertidal sea anemone

Anthopleum xanthogrammica, Biol. Bull. 166: 457 — 172.
Sharp, V. A.. D. Miller, J. C. Bythell, and B. E. Brown. 1994. Ex

pression of low molecular weight HSP 70 related polypeptides from the

symbiotic sea anemone Anemonia viridis Forskall in response to heat

shock. J. Exp. Mar. Biol. Ecol. 179: 179-193.
Sharp, V. A., B. E. Brown, and D. Miller. 1997. Heat shock protein

(HSP70) expression in the tropical reef coral Goniopora djiboutiensis.

J. Therm. Biol. 22: 11-19.
Slallery, M., M. T. Hamann, J. B. McClintock, T. L. Perry, M. P.

Puglisi, and \V. V. Yoshida. 1997. Ecological roles for water-borne

metabolites from Antarctic soft corals. Mar. Ecol. Prog. Ser. 161:

133-144.
Snyder, M. J., and E. P. Mulder. 2001. Environmental endocrine

disruption in decapod crustacean larvae: Hormone liters, cytochrome

P450. and stress proteins. Aquat. Toxicol. 55: 177-190.
Snyder, M. J., E. Girvetz, and E. P. Mulder. 2001. Stress protein

induction by chemical exposures in molluscs. Arch. Environ. Contain.

Toxicol. 41: 22-29.
Steneck, R. S. 1986. The ecology of coralline algal crusts: convergent

patterns and adaptative strategies. Annu. Rev. Ecol. Svst. 17: 273-303.
Suchanek, T. H., R. C. Carpenter, J. D. Witman, and C. D. Harvell.

1985. Sponges as important space competitors in deep Caribbean

coral reef communities. Pp. 55-59 in The Ecology of Deep and Shallow

Coral Reefs, M. L. Reaka. ed. Symposia Series for Undersea Research

3( 1 ). NOAA/NURP. Rockville. MD.

Sutherland. J. P. 1978. Functional roles of Sc hizoporella and Sryela in
the fouling community at Beaufort, North Carolina. Ecology 59: 257-
264.

Thompson, J. E., R. P. Walker, and D. J. Faulkner. 1985. Screen-
ing and bioassays for biologically-active substances from forty
marine sponge species from San Diego, California USA. Mar. Biol.
88: 11-21.

Turon, X., M. A. Becerro, M. J. Uriz. and J. Llopis. 1996. Small-scale
associations measures in epibenthic communities as a clue for allelo-
chemical interactions. Oecologia 108: 351-360.

Turon, X., I. Tarjuelo, and M. J. Uriz. 1998. Growth dynamics and
mortality of the encrusting sponge Crambe crambe (Poecilosclerida) in
contrasting habitats: correlation with population structure and invest-
ment in defense. Funct. Ecol. 12: 631-639.

Uriz, M. J., D. Martin, X. Turon, E. Ballesteros, R. Hughes, and C.
Acebal. 1991. An approach to the ecological significance of chemi-
cally mediated bioactivity in Mediterranean benthic communities. Mar.
Ecol. Prog. Ser. 70: 175-188.

V'an-Praet, M. 1985. Nutrition of sea anemones. Adv. Mar. Biol. 22:
65-99.

Weis, V. M.. and R. P. Levine. 1996. Differential protein profiles reflect
the different lifestyles of symbiotic and aposymbiotic Anthopleura
elegantissima. a sea anemone from temperate waters. / Exp. Biol. 199:
883-892.

Whitfield, F. B., F. Helidoniotis, K. J. Shaw, and D. Svoronos. 1999.
Distribution of bromophenols in species of marine algae from eastern
Australia. J. Agric. Food Chem. 47: 2367-2373.

Whittaker, R. H., and P. P. Feeny. 1973. Allelochemics: chemical
interaction between species. Science 171: 757-768.

Wiens, M., C. Koziol, H. M. A. Hassanein, R. Batel, H. C. Schroder,
and W. E. G. Miiller. 1998. Induction of gene expression of the
chaperones 14-3-3 and HSP70 by PCB 118 (2.3',4,4',5-pentachloro-
biphenyl) in the marine sponge Geodia cydonium: novel biomarkers for
polychlorinated biphenyls. Mar. Ecol. Prog. Ser. 165: 247-257.

Reference: Biol. Bull. 201: 394-404. (December 2001)

NO/cGMP Signaling and HSP90 Activity Represses
Metamorphosis in the Sea Urchin Lytechinus pictus

CORY D. BISHOP AND BRUCE P. BRANDHORST*

Department of Molecular Biology ami Biochemistry, Simon Fraser University;
Burnaby, British Columbia V5A 1S6. Canada

Abstract. Nitric oxide (NO) signaling repressively regu-
lates metamorphosis in two solitary ascidians and a gastro-
pod. We present evidence for a similar role in the sea urchin
Lytechinus pictus. NO commonly signals via soluble gua-
nylyl cyclase (sGC). Nitric oxide synthase (NOS) activity in
some mammalian cells, including neurons, depends on the
molecular chaperone heat shock protein 90 (HSP90); this
may be so in echinoid larvae as well. Pluteus larvae con-
taining juvenile rudiments were treated with either radicicol
L- or D-nitroarginine-methyl-ester (L-NAME and o-NAME),
or IH-[l,2,4]oxadiazolo[4,3-a]quinoxalin-l-one (ODQ). in-
hibitors of HSP90, NOS, and sGC. respectively. In all
instances, drug treatment significantly increased the fre-
quency of metamorphosis. SNAP, a NO donor, suppressed
the inductive properties of L-NAME and biofilm, a natural
inducer of metamorphosis. NADPH diaphorase histochem-
istry indicated NOS activity in cells in the lower lip of the
larval mouth, the preoral hood, the gut, and in the tube feet
of the echinus rudiment. Histochemical staining coincided
with NOS immunostaining. Microsurgical removal of the
oral hood or the pre-oral hood did not induce metamorpho-
sis, but larvae lacking these structures retained the capacity
to metamorphose in response to ODQ. We propose that the
production of NO repressively regulates the initiation of
metamorphosis and that a sensory response to environmen-
tal cues reduces the production of NO, and consequently
cGMP, to initiate metamorphosis.

Received 17 November 2000; accepted 7 September 2001.
* To whom correspondence should be addressed. Email: brandhor@sfu.ca
Abbreviations: D-NAME, D-nitroarginine-methyl-ester; GA, geldanamy-
cin; GBD, geldanamycin binding domain; HSP90. heat shock protein 90;
L-NAME, L-nitroarginine-methyl-ester; NO, nitric oxide; NOS, nitric oxide
synthase; ODQ, lH-[l,2,4]oxadiazolo[4.3-a]quinoxalin-l-one; RD, radici-
col; sGC, soluble guanyl cyclase; SNAP, S-nitroso-N-acetylpenicillamine.

Introduction

Many species of sea urchin undergo maximal indirect
development (Davidson, 1991). Embryonic development
generates a bilaterally symmetrical feeding pluteus larva
that bears no resemblance to an adult sea urchin. After a
period of growth in the plankton, an adult rudiment forms
on the left side of the larva, within the vestibule. Once
competence is reached, and in response to appropriate cues,
the pluteus larva settles and undergoes a radical transfor-
mation into a pentaradially symmetrical juvenile sea urchin.
The initial events of this transformation, as described for
Lytechinus pictus, are completed within an hour (Cameron
and Hinegardner, 1974, 1978; Pearse and Cameron, 1991).
Briefly, the larval arms bend away from the vestibule, from
which the tube feet of the rudiment extend, allowing podial
attachment to the substratum. The rudiment becomes ex-
posed to the exterior and then everts. The larval epithelium,
including that of the arms, contracts and collapses onto the
aboral surface of the juvenile, involving drastic changes in
cell shape. The vestibular epithelium extends to cover the
aboral surface, forming the epithelium of the juvenile and
enclosing degenerating larval cells. Extensive remodeling
of internal structures such as the digestive tract continue for
several days as the juvenile begins the reproductive stage of
its life cycle as a benthic feeder.

Competent echinoid larvae will settle and initiate meta-
morphosis if provided with a hard surface covered with an
appropriate organic film, particularly a microbial film (re-
viewed by Strathmann. 1987; Pearse and Cameron. 1991;
also see Discussion). In the absence of such cues, some
species delay metamorphosis (Caldwell, 1972; Cameron
and Hinegardner. 1974). When placed in clean glass or
plastic dishes with fresh seawater, L. pictiis larvae rarely
metamorphose. This allows experimental investigation ot
the induction of metamorphosis. The mechanism by which

394

NO REPRESSES ECHINOID METAMORPHOSIS

395

external cues are detected and transduced into the initiation
of metamorphosis remains poorly understood, but appar-
ently involves a neurosensory response. Further, it is not
clear whether larval or juvenile sensory perception (or both)
is responsible for transducing external signals under natural
conditions. Evidence for the involvement of neural re-
sponses from both the larva and the juvenile has been
reported. Electrical stimulation of the oral ganglion or the
apical neuropile of Dendraster excentricus larvae induced
metamorphosis (Burke. 1983). In contrast, observation of
settling behaviors and the prevention of settling (and con-
sequently of metamorphosis) in the presence of inducers
clearly demonstrates a role for the juvenile sensory appara-
tus in L. pictus (Cameron and Hinegardner. 1974; Burke.
1980: our observations). Investigations of the molecular and
anatomical basis of signaling events that regulate echinoid
metamorphosis can thus be placed in this historical context.

Nitric oxide synthase (NOS) catalyzes the conversion of
L-arginine to L-citrulline with the production of the gas nitric
oxide (NO). NOS expression and NO function have been
documented in both nervous and non-nervous tissues alike
across a range of eukaryotic organisms, indicating their
antiquity and importance in regulating many cellular pro-
cesses (Schulte el ai. 1998; Cueto el al, 1996: Kuzin el al.
1996; Czar et al.. 1997). That NO is diffusible through
biological membranes suggests that it may have served as a
primitive signaling system between cells before more elab-
orate mechanisms of cell adhesion and receptor-based sig-
naling evolved. In mammalian cells, NOS activity in vivo
requires interaction with heat shock protein 90 (HSP90)
(Garcia-Cardena et al.. 1998: Bender et al., 1999). We
recently reported that metamorphosis of two species of
ascidian tadpole larvae is induced by drugs that inhibit the
activity of the protein chaperone HSP90, NOS. or soluble
guanylyl cyclase (sGC) (Bishop et al.. 2001 ). Among larval
tissues, NOS activity is concentrated in the tail muscle cells
of the ascidian tadpole Cnemidocarpa finmarkiensis. Re-
moval of the tail stimulates metamorphosis of the head,
consistent with there being a signal, probably NO. from the
tail that represses metamorphosis. NOS produces NO. a
gaseous signaling molecule whose most common effector is
sGC (Garthwaite et al., 1995; Salter et al.. 1996: Hebeiss
and Kilbinger. 1998). Thus, inhibition of NOS often results
in a corresponding reduction of cGMP (McDonald and
Murad. 1996. for review). Metamorphosis of the marine
gastropod llyanassa obsolete! was also reported to be in-
duced by inhibition of NOS activity (Froggett and Leise,
1999). indicating that NO may repress metamorphosis in a
variety of animals.

To further test the idea that NO-mediated repression of
metamorphosis occurs widely within the bilaterian clade.
we used the sea urchin L. pictus. We report that NO/cGMP
signaling is an important regulator of the events surrounding
the transition of form from the larva to the juvenile in L

pictus, that it is downstream from a natural inductive cue,
and that this regulation may be dependent upon HSP90
function. NOS was detected in both larval and juvenile
organs; such organs may be involved in sensing or trans-
ducing the response to natural inductive cues.

Materials and Methods

Obtaining and culturing lan-ae

Specimens of Lytechinus pictus were purchased from
Marinus (Long Beach, CA) and held in recirculating sea-
water tanks. Eggs were induced to shed by intracoelomic
injection of 0.5 mol KC1, then washed and fertilized. Em-
bryos in Millipore (0.45 /im) filtered seawater (MFSW) at
16 °C containing 50 /ng/ml penicillin and streptomycin were
continuously stirred at 20 or 60 rpm using plastic paddles.
After hatching, the embryos were collected by filtration on
93-;um-mesh Nitex and resuspended in fresh MFSW; this
washing procedure was repeated frequently throughout lar-
val growth, and the concentration of larvae was gradually
reduced to I/ml or less. Algae were obtained from the
Northeast Pacific Culture Collection (NEPCC) at the Uni-
versity of British Columbia. Vancouver. Either a mix of
Pyrenomonas salina (NEPCC strain 076; Center for Culture
of Marine Phytoplankton (CCMP) strain 3C) and Dunaliella
tertiolecta (NEPCC strain 001: CCMP strain 1320) or only
the latter were fed to plutei every 2-3 days in quantities such
that most algae had been ingested as of the next feeding.

Pharmacological inhibition

L-NAME (L-nitroarginine-methyl-ester) and its enantio-
mer o-NAME, radicicol (RD), and ODQ ( lH-[l,2.4]oxadia-
zolo[4,3-a]quinoxalin-l-one) were obtained from Sigma
Chemical Corp. (St. Louis. MO). Because there is variation
in the rate of development of the juvenile rudiment, indi-
v idual L. pictus plutei were selected by examination under
a stereomicroscope and transferred to wells of 24-well plas-
tic culture dishes (Flow Labs, McLean. VA). Larvae were
selected for experiments based on the presence of a large,
pigmented rudiment having well developed spines and tube
feet. Each well contained about 10 larvae in 2 ml MFSW or
experimental solutions in MFSW. To quadruplicate sets of
these selected larvae were added L-NAME, o-NAME. RD.
ODQ. or MFSW in 1- or 2-ml final volumes. Metamorpho-
sis was monitored using a stereomicroscope: it was scored if
the larval epithelium had collapsed on top of an everted
juvenile. The activity of tube feet was used as an indicator
of larval vitality. The concentrations of L-NAME. RD, and
ODQ used in the experiments reported here were chosen
because they elicited a metamorphic response in ascidian
larvae (Bishop et al.. 2001). L-NAME and D-NAME were
prepared as 1 M stocks in water and diluted to a final
concentration of 1-10 mM with MFSW. ODQ was prepared

396

C. D. BISHOP AND B. P. BRANDHORST

as a 100 mM stock in DMSO and diluted into MFSW to 50
ju,M. RD was prepared as a 5 mM stock in DMSO and
diluted into MFSW to 5/iAf. SNAP (S-nitroso-N-
acetylpenicillamine) was prepared as a 100 mM stock in
DMSO and diluted to 0.1 mM in MFSW. For RD, ODQ.
and SNAP treatments, experimental and control wells all
contained a final concentration of 0.1% DMSO; this con-
centration of DMSO did not have any inductive properties.
Unless significant metamorphosis was observed sooner, ex-
periments were scored at 24, 48, and sometimes 72 h. A low
frequency of spontaneous metamorphosis was observed for
larvae placed in plastic dishes; this response tends to occur
shortly after the assessment of a larvae and its transfer into
a well. If such a response was observed before the addition
of drugs, juveniles were removed.

To create a natural inductive cue, glass Syracuse dishes
were submerged for several days in recirculating tanks
containing natural seawater. Ten larvae were exposed to the
substrate in MFSW either in the presence or absence of 0.1
mM SNAP. Results shown are from a single experiment.

Microsurgical removal of oral hoods and pre-oral hoods
was accomplished using a fine-edged stainless steel pin
(Fine Science Tools, Vancouver, BC) fused to a glass pi-
pette. Dissected oral and pre-oral hoods retained their ca-
pacity to swim. Quadruplicate sets of 5 larvae or hoods per
well (a total of 20 operations) were used for each experi-
ment.

All experiments were tested for statistical significance by
performing a one-tailed Student's / test with the assumption
of hoinoscedastic variance. In all graphs (made using Mi-
crosoft Excel 97), the asterisks denote statistical signifi-
cance; P values are provided in the figure legends. Specific
statistical comparisons are described in the figure legends.

NADPHd histochemistry and NOS immunohistochemistry

The NADPH diaphorase staining protocol described by
Weinherg et nl. ( 1996) was used with modifications. Larvae
were fixed in 2% glutaraldehyde and 1% formaldehyde in
sodium phosphate buffer for 1 h at room temperature. Form-
aldehyde was freshly prepared by dissolving paraformalde-
hyde (EM grade, Ted Pella, CA) in MFSW, adjusting the
pH to 7.4, and then diluting in PB to 1%. After rinsing with
PB, fixed larvae were incubated in 0.4 mg/ml nitrotetrazo-
lium blue substrate with 2 mg/ml NADPH from 2 to 16 h at
37 °C. As a negative control, specimens were incubated in
50% ethanol for 2 h and then incubated in nitrotetrazolium
blue in the absence of NADPH; no staining was observed
under these conditions. Under the fixation conditions used,
NOS is the only diaphorase expected to be active ( Weinberg
c/ «/.. 1996). Stained larvae were examined as whole
mounts b\ microscopy or were dehydrated in a graded
ethanol series, embedded in polyester wax (BDH Labora-
tory Supplies. Poole, England), and sectioned at 8 /xm.

Sectioned larvae were examined using an Olympus Vanox
microscope, and images were captured using a Sony DXC-
950 3CCD camera.

Universal anti-NOS (Affinity Bioreagents, Golden, CO)
polyclonal rabbit antibody was used to detect NOS in grow-
ing and mature larvae. Larvae were fixed for 2 h at room
temperature in 4% formaldehyde (prepared as outlined
above). Fixed larvae were blocked with PB saline contain-
ing 5% bovine serum albumin and 0.1% Triton-X-100 and
then incubated in 1:100 anti-NOS overnight at 4 °C. Larvae
were incubated in secondary antibody (goat anti-rabbit-
Alexa 568, Molecular Probes, Eugene OR) for 2 h at room
temperature and then rinsed, mounted, and viewed on a
Zeiss LSM 410 confocal microscope. Images were pro-
cessed using Adobe Photoshop 5.5 or 6.0.

Results

Inhibitors of nitric oxide synthase, guanylyl cyclase, and
HSP90 induce metamorphosis

Treatment of larvae with the NOS inhibitor L-NAME in-
duced a significant increase in the frequency of metamorphosis
in comparison with larvae treated with seawater (Fig. 1A) or
D-NAME (Fig. !B). In a time-course experiment, the fre-
quency of metamorphosis was scored every hour for 6 h (Fig.
IB). Some larvae responded rapidly (within 2 h) to either L- or
D-NAME but others required several hours. Because D-NAME
is used as an inactive enantiomer of L-NAME, the observed
inductive property of D-NAME was unexpected and substan-
tial, although less so than for L-NAME (Fig. 1 ). To confirm that
the inductive properties of L-NAME or D-NAME were due to
a reduction in NO levels, larvae were co-incubated with L-
NAME or D-NAME and the NO donor SNAP. At a 10-fold
lower concentration than i.-NAME or D-NAME, SNAP com-
pletely suppressed their inductive properties (Fig. I A). In a
variation of that experiment. SNAP was added 4 h after L-
NAME had been added. This also resulted in the suppression
of metamorphosis (Fig. 2).

Marine biotilms consisting of bacteria and other micro-
organisms have previously been shown to induce metamor-
phosis in L. pictiis larvae (Cameron and Hindgardner,
1974). This is considered to be a cue that approximates that
of a natural benthic environment. We exposed larvae to a
biofilm grown in recirculating tanks (containing natural
seawater from local sources) in the presence or absence of
SNAP to test whether NO signaling was downstream of a
sensory pathway that is responsive to a natural cue. SNAP
suppressed the inductive properties of the biofilm in a
reversible manner (Fig. 3). SNAP was effective at suppress-
ing metamorphosis among larvae that had been exposed to
biofilm for several hours, but had not yet metamorphosed
(Fig. 3).

Soluble guanylyl cyclase (sGC) is the most common
downstream effector of NO signaling (Salter ct ai, 1996;

NO REPRESSES ECHINOID METAMORPHOSIS

397

(A)

DO hours post-SNAP D7 hours post-SNAP 124 hours post -SNAP

1.2-

- 1
|

o 0.8

E 0.6

£0.4

f 0.2H

D8 D24

(B)

« 0.8 •

QL-NAME

DO-NAME

pJ-.

E 0.6 •

'o
J0.4-

- 0.2 •

-1-

0 -

2345

Hours after treatment

Figure 1. i -NAME and n-NAME treatments induce metamorpho
a time-dependent fashion; SNAP suppresses their inductive properties. (A)
Larvae were incubated in 1 mM L-NAME or D-NAME or co-incubated
with O.I mM SNAP. The frequency of metamorphosis was scored alter S
and 24 h. Asterisks indicate a significant difference between larvae treated
with i -NAME or D-NAME and seawater controls (PLS < 0.004: /', . ,
6.9 X 10~7; PK4 < 3.2 X 10 5: H = 4). Asterisks in parentheses indicate
a significant dilference between L-NAME and L-NAME + SNAP or
D-NAME and D-NAME + SNAP (PLS < 0.002; PL,24 < 3.0 X 10 "; n =
4). The value from a statistical comparison between L,4 and L + S,4
cannot be calculated, since the respective means are 1 and (I with no
variation. (B) The frequency of metamorphosis \\as monitored on the hour,
for 6 h. Asterisks indicate a significant difference in the frequency of
metamorphosis between larvae treated with 1 mA/ L-NAME or D-NAME.
(P, < 0.03: /', < 0.004; />, < 0.01: />„ < 0.002: n = 4).

Heheiss and Kilbinger. 1998). To test the involvement of
cGMP signaling in /.. pictus metamorphosis, we incubated
larvae in ODQ. There was a significant increase in the
frequency of metamorphosis in comparison with controls in

L-NAME

L-NAME+SNAP

Figure 2. SNAP can suppress metamorphosis alter the addition of
L-NAME. Eight wells (10 larvae /well) were incubated with i.-NAME.
After 4 h. the frequency of metamorphosis reached approximately 0.5. then
0.1 mM SNAP was added to four of the wells. The frequency of meta-
morphosis was scored 7 and 24 h thereafter. The asterisk indicates a
significant difference between lime points m the frequency of metamor-
phosis among larvae treated with L-NAME (P,,_7h < 0.02; fY-,4h < 0.0005;
» = 4). The asterisk in parentheses indicates a significant difference in the
frequency of metamorphosis between larvae treated with L-NAME and
L-NAME + SNAP after 24 h (A>,4 < 0.03).

MFSW (Fig. 4). In another experiment, larvae were treated
with radicicol. an inhibitor of HSP90 function. Radicicol
and geldanamycin frequently lead to a decrease in the
activity or abundance of HSP9()'s client proteins (Yen ft a/..
1994; Schulte ct til.. 1998). Therefore, based on a hypoth-
esized interaction of HSP90 and NOS in urchins, and the
observation that inhibiting NOS activity induces metamor-
phosis, we expected that treatment with RD would increase

D 4 hours D 2 hours • 18 hours

(+SNAP/-SNAP) (-SNAP/+SNAP)

Figure 3. SNAP suppresses the inductive properties of biotilm in a
reversible manner. Larvae were exposed to a biolilm in the presence or
absence of O.I mM SNAP After 4 h the conditions were reversed such that
SNAP was washed out of the ilish that contained it and added to the dish
that lacked it. The frequency of metamorphosis was seined at 2 and IS h
thereafter (i. e.. 6 and 22 h. respectively after initial exposure to biotilm).
This experiment was not amenable to statistical analysis.

398

C. D. BISHOP AND B, P. BRANDHORST

o
I °'6

°0.4

0.2-

D3 hours
D24 hours
• 48 hours

s.w.

ODQ

Figure 4. Inhibitors of HSP90 or sGC induce metamorphosis. I ai\ae
were treated with 0.1% DMSO (control). 5 pM RD. or 50 fiM ODQ The
frequency of metamorphosis was monitored after 3. 24. and 48 h. Asterisks
indicate a significant difference in the frequency of metamorphosis of
larvae treated with RD or ODQ compared to controls. (PRD24 — 0.01;
PRD4S s 0.02: n = 4). ODQ caused a significant increase within 3 h
(POD(j, £ 0.0004; n = 4), with no further significant increase.

the frequency of metamorphosis over the untreated controls.
and it did (Fig. 4). We have not measured directly whether
RD reduces the activity of NOS.

NOS activity is present in neurons of liin'al tissues anil
tube feet of the rudiment

The NADPH diaphorase histochemical assay was used
under conditions specific for vertebrate NOS enzymes.
Whole larvae were stained and observed; some were then
fixed and sectioned before examination by microscopy.
Feeding larvae stained for diaphorase activity in the lower
lip of the mouth, mid- and hindgut, at the tips of postural
arms, and in cells within the lobe between the anterolateral
arms. These sites of NADPH diaphorase activity most likely
represent sites of NOS activity. Larvae having large rudi-
ments resembling those used for the inhibitor assays were
also sectioned and stained. As shown in Figure 5, diapho-
rase activity was found in a variety of structures. Stained
cells were found within the larval gut epithelium (Fig.
5A-C). The basioepithelial nerve plexus of juvenile tube
feet was intensely stained and appeared to extend processes
to the outer surfaces of the tube foot (Fig. 5D, E). Stained
cells were observed at the tips of larval arms (Fig. 5F| and
in the pre-oral hood (Fig. 5 A). Stained cells, often having a
neuronal appearance, were observed in epaulettes (Fig. 5G)
and the lower lip of the larval mouth (Fig. 5H-J).

We stained larvae with anti-NOS antibodies to see
whether sites of NADPH activity were coincident with the
location of NOS. Prominent staining was observed in the
lower lip of the mouth (Fig. 5K) and in some cells situated
in the pre-oral hood (not shown). Stained cells in the lower
lip are roughly symmetrically arranged around the pharyn-

geal lumen. The number of immunoreactive cells in the
lower lip was variable from larva to larva. It is not clear if
this perceived variation was due to variation in the actual
number of NOS-positive cells in this region or in the sen-
sitivity of immunostaining. Since the variation was ob-
served among larvae in individual immunostaining experi-
ments, the former possibility is more likely. To gain a
three-dimensional perspective on the organization of NOS-
positive cells in the lower lip of the mouth, serial optical
sections were captured and projected as three-dimensional
images (Fig. 5L, M). In agreement with histochemically
stained sections, these cells extend processes toward the
apical surface of the oral epithelium. We also saw a general
correspondence between histochemical and immunohisto-
chemical staining in other tissues (data not shown), indicat-
ing that sites of NADPH diaphorase activity correspond to
sites of NOS expression.

Dendraster excentricus pluteus larvae contain cells that
express catecholamines in the lower lip of the mouth; this
region was thus termed an oral ganglion (Burke, 1983).
Removal of the oral hood (OH), which includes the oral
ganglion, induced metamorphosis (Burke 1983). That ob-
servation and the expression of NOS in the oral ganglion
cells of L. pictus larvae led to the hypothesis that these cells
repress metamorphosis via their production of NO. To test
this idea, we microsurgically removed either the entire OH
or the pre-oral hood (PH) from mature larvae and scored the
frequency of metamorphosis. This operation did not induce
metamorphosis of L. /rictus after 6 h (not shown), so L-
N AME was added to see if larvae lacking the OH or the PH
had retained their capacity to undergo metamorphosis. Nei-
ther postoperative larvae nor the OH and PH were respon-
sive to i -NAME at concentrations that induced metamor-
phosis in intact control larvae (Fig. 6A). To further test if
the postoperative larvae and the dissected tissues had re-
tained the capacity to metamorphose, we added 50 ^M
ODQ after 14 h of incubation in L-NAME. This resulted in
a very rapid metamorphic response (Fig. 6A). The OH and
PH did not initially undergo epithelial collapse typical of
intact metamorphosing larvae, although they did so within
24 h (data not shown). Microscopic analysis indicated that
the OH and PH were not necrotic, but rather they had
undergone genuine cellular rearrangements characteristic of
the epithelium of metamorphosing larvae. Therefore, micro-
surgical removal of the OH or the PH did not lead to
metamorphosis of postoperative larvae, and apparently de-
creased their capacity to respond to NOS inhibition but not
sGC inhibition. Dissected OH and PH tissues underwent
rearrangements typical of metamorphosing larvae, but only
after a protracted period in drug.

Discussion

Independent pharmacological inhibition of NOS, HSP90,
and sGC led to a significant increase in the frequency of

NO REPRESSES ECH1NOID METAMORPHOSIS

399

W '<

\«-&*

-»*—

»*&

/ /:'"'~

Figure 5. NOS expression in larvae was analyzed by NADPH diaphorase histochemistry and NOS immuno-
histochemistry. (A) Section of a 26-day larva showing dark blue staining in the fore and mid-gut (M) and in a cell in
the pre-oral hood (arrowhead). (B) Oblique longitudinal section showing the arrangement of stained mid-gut epithelial
cells. (C) Higher magnification cross section showing staining in the basal portion of columnar epithelial cells lining
the mid-gut. The inset shows a low magnification whole-mount view of a stained larval mid-gut. (D) Longitudinal
section of a tube foot from a juvenile rudiment contained in a larva. Stained cells of the basioepithelial nerve plexus
are tightly apposed to the ectodermal epithelium. (E) Slightly oblique transverse section of a tube foot showing stained
nerve plexus with possible projections to the outer surface of the epithelium. (F) Whole-mount staining of a post-oral
larval arm from a 26-day-old larva. (G) Section of an epaulette showing staining in cells at the distal tip. (H) Frontal
section of the larval mouth showing stained cells. (I) Lateral section of the oral hood and mouth. (J) Higher
magnification view of the oral epithelium showing basal position of stained cell bodies. O = outside and I = inside.
A \on-like projections having bulbous termini (arrowhead) extend to the ciliated apical surface. Scale bars: 40 /urn in
A inset of C; 20 jim in B. F. H. and I: 8 /im in C-E. G. J. (K-M) NOS immunostaining. (K) Mouth of a larva. Only
cells in the lower lip are immunoreactive. (L) 3-D projection of NOS-positive oral cells. (M) Same projection as (L)
but rotated to show cell polarity. Apical side of the oral epithelium (outside of larva) is to the left. Sharp line at the
right demarcates the end of the confocal stack. Scale bars are not available for K-M.

metamorphosis of L. pictus larvae. These results are con-
sistent with our model for the signaling system that regu-
lates the initiation of metamorphosis in ascidians: cells
having NOS activity (probably dependent upon interaction
with HSP90) release NO that stimulates the activity of

guanylyl cyclase to produce cGMP that inhibits metamor-
phosis (Bishop et ai, 2001 ). This proposal is also based, in
part, on evidence that NO represses metamorphosis in a
gastropod (Froggett and Leise. 1999; Leise et al. 2001).
Natural inductive cues may be operating via receptor-based

400

C. D. BISHOP AND B. P. BRANDHORST

(A)

(B)

Dctrls Dlarva-OH llarvae-PH

L-NAME

ODQ

Figure 6. (A) Removal of the entire oral hood or the pre-oral hood does not induce metamorphosis and
diminishes the response to L-NAME. but not ODQ. L-NAME was added to postoperative larvae (POL) and the
dissected fragments 6 h after surgery. After 14 h in l.-NAME, the metamorphic response was significantly reduced in
comparison with intact control larvae in L-NAME. Asterisks indicate significant differences between the frequency of
metamorphosis of intact and postoperative larvae (POL) (PLjrl4<, m,nu.. OH < 0.02; ^ur>-ac mmu, PH < 0.002). ODQ was
added to both POL and dissected fragments, and the frequency of metamorphosis was scored. After 2 h. ODQ rapidly
induced metamorphosis. Asterisks indicate a significant difference between POL before and after ODQ treatment
(PL ^ ^ ^ ^<0.006:/>lanlon,lnu,thepreol_l|holx)<0.0003).(B)Aschematicdrawingofai,yrer/»'H»ip/fn« larva
indicating the position of NOS-positive cells found in the lower lip of the mouth (arrow) and the point at which
portions of the larvae were surgically removed. Dashed lines indicate the plane of the cuts. NOS-positive oral cells
are removed along with the entire oral but not with the pre-oral hood. Arms are virtually absent in well-fed, stirred
larvae. PH = pre-oral hood; OH = oral hood; M = mouth; G = gut; E = epaulettes.

sensory perception that is upstream of NO/cGMP signaling.
A low-molecular-weight, water-soluble compound isolated
from biofilm has inductive properties in L pictus (Cameron
and Hinegardner. 1974).

Frequently, some of the treated larvae did not respond by
initiating metamorphosis, even after longer incubations (72
h in some cases). In similar experiments, we have found that
a fraction of selected larvae do not respond to dishes coated
with microbial film. The fraction of resistant larvae in such
experiments was variable (not shown), but similar to the
fraction resistant to potent drug treatments such as ODQ
(Fig. 4). Variation in response to inducers. whether natural
or otherwise, may represent variation in sensitivity of sen-
sory perception, levels of NO repression, or response to a
reduction of NO signaling (or a combination thereof) among
larvae of a clutch and among clutches. Perhaps the resistant
larvae had not achieved competence to undergo metamor-
phosis, despite the morphological similarity of their rudi-
ments to those that did metamorphose. In fact, in some
cases, larvae containing less well developed rudiments were
responsive to drugs, whereas those with large, highly pig-
mented rudiments were not. It is clear that competence does
not strictly correspond to the presence of a fully formed
rudiment within the larva. Assessing competence is prob-
lematic in that one does not know whether a lack of re-
sponse is due to lack of competence or failure to respond to
an inductive cue.

To our knowledge, the concept of competence does not

describe a specific biological state in any marine inverte-
brate having planktotrophic larvae and benthic adults. Lar-
vae with no rudiments or abnormal rudiments do not re-
spond to inducers of metamorphosis (Cameron and
Hinegardner, 1978; CDB, unpubl. obs.), so competence in
urchins represents a discrete change in the physiological
state of the larva that is related to the growth and develop-
ment of the juvenile. Competence is a phenomenon that
requires further investigation and should be considered in
all studies on the regulation of metamorphosis. The acqui-
sition of competence coincides with the initiation of rneta-
morphosis in some animals, but not in others (Birkeland el
nl.. 1971; Degnan el til., 1997; Bishop el ai. 2001). This
indicates that the fitness consequences associated with the
timing of, and substrate choice during, settling and meta-
morphosis vary. What other signaling systems may be con-
tributing to the timing events surrounding life cycle trans-
formations? Studies on thyroxine in echinoids suggest its
involvement in the evolutionary loss of larval feeding. The
addition of exogenous thyroxine leads to a reduction of
larval structures and the time to metamorphosis in D. ex-
cent ricus (}. Hodin, Friday Harbor Laboratories, and A.
Heyland, University of Florida; pers. comm.l. It will be
interesting to know if and how NO/cGMP and hormonal
signals interact to regulate the timing of life cycle transfor-
mations in echinoids.

Between different clutches, we have observed a striking
difference in the response of larvae to NOS inhibition.

NO REPRESSES ECHINOID METAMORPHOSIS

401

Increased sensitivity is manifested as a more rapid response
given that identical concentrations of L-NAME and
o-NAME were used. We cannot rule out other variations in
culturing conditions, such as larval densities and food. The
data on NO signaling presented here are from the clutch that
was the most sensitive to NOS inhibition. This clutch often
responded to NOS inhibition within 2 h. whereas another
clutch often took 24-48 h to show a significant effect. The
source of this variation is not clear.

We have shown that D-NAME has inductive properties
that are suppressed by SNAP, indicating that application of
D-NAME also leads to a decrease in NO. Although D-
NAME is often used as an inactive negative control for
L-NAME treatment, we propose that it does inhibit NOS,
but less effectively than L-NAME; others have also noted
this activity (Babal et ui, 2000). Therefore, D-NAME
should be used as a less active enantiomer of L-NAME, not
an inactive enantiomer. The extent to which D-NAME is
useful as a control for L-NAME treatment will depend on
the sensitivity of the experimental system to NO reduction
and the concentration of drug used.

There was a lag in the response to radicicol after the
beginning of treatment. Radicicol competes with ATP for
binding to HSP90, thereby inhibiting its function in binding
and folding proteins (Schulte et al., 1998; Sharma et ui,
1998). As a protein chaperone. HSP90 interacts with mem-
bers of several signal transduction pathways (reviewed by
Pratt, 1998). In concert with accessory proteins, HSP90
promotes the folding and maintenance of the active state of
several known client proteins (Aligue et al., 1994; Whitesell
et al.. 1994: Nathan and Lindquist, 1995; reviewed by
Caplan, 1999). NOS activity in some mammalian cells,
including neurons, requires an interaction with HSP90; all
three vertebrate isoforms of NOS are degraded in the pres-
ence of geldanamycin (GA), another HSP90 inhibitor (Joly
et al.. 1997; Garcia-Cardena et al., 1998; Bender et al..
1999). Like RD, this agent inhibits HSP90 function by
competing with ATP for binding (Promrodou et al.. 1997).
When the folding function of HSP90 is impaired by inhib-
itory drugs such as RD and GA, its client proteins (which
are often in complexes including HSP90) may be caught in
a partially folded state that is then recognized by the ubi-
quitin-proteasome protein degradation machinery (reviewed
by Pratt, 1998; Caplan, 1999). Thus, some client proteins
are expected to be degraded or lose activity after HSP90
activity is inhibited. In this circumstance, a response to
inhibition of HSP90 would not be expected until its activity
had become limiting and its critical client proteins had lost
activity or decayed. Such a lag in response was observed for
three HSP90 inhibitors that induced metamorphosis when
applied to ascidian larvae (Bishop et al., 2001). Thus, we
consider this lag to be a consequence of the mechanism by
which RD probably leads to a decline in NOS activity.

However, a direct demonstration of interaction between
HSP90 and NOS in urchins is warranted.

All of the biochemical characterizations concerning the
inhibitory properties of anti-HSP90 drugs have been con-
ducted with vertebrate cells. It is relevant to assess whether
RD is likely to have the same effect on L. pictits HSP90 as
it does on vertebrate HSP90. The crystal structure of a
geldanamycin-HSP90 complex has been determined (Steb-
bins et al., 1997). The geldanamycin binding domain
(GBD) is 43% conserved at the amino acid level between
vertebrates and E. coli; the aspartic acid residue (Asp93) is
absolutely conserved among all HSP90 homologs from 35
species. A hydrogen bond network between Asp93 and the
carbamate group of GA has been suggested by structural
and functional studies to play the most critical role in the
binding of HSP90 to GA (Schnur et al.. 1995; Stebbins et
al.. 1997). Thus, it is probable that GA has similar inhibi-
tory properties on HSP90 from all organisms. RD and GA
share no structural similarities, but RD can compete with
GA for binding to the N-terminal portion of HSP90 that
contains the GBD (Schulte et al., 1998). Moreover, like GA
treatment, RD depletes cells of known HSP90 client pro-
teins (Schulte et al., 1998). It is reasonable then to expect a
set of highly conserved intermolecular interactions between
the GBD of HSP90 of different organisms and RD and
hence, a highly conserved mechanism of inhibition of
HSP90 by RD. Consistent with this conclusion, GA and RD
had similar effects on the initiation of ascidian metamor-
phosis (Bishop et al., 2001 ) and morphogenetic movements
during sea urchin embryonic development (CB. unpubl.
obs.).

Under natural circumstances, the initiation of metamor-
phosis by competent L. pictus larvae results from a sensory
response to appropriate environmental cues. Minimally, this
is a biochemical cue, although a hard surface is usually
required (Cameron and Hinegardner, 1974). It is not clear
what cells or organs are involved in transducing this chemo-
and mechanosensory perception into a metamorphic re-
sponse. The rate of biphasic potentials recorded from the
larval body or near the rudiment increases more in response
to a substrate "conditioned" with a microbial film than to an
unconditioned substrate (Satterlie and Cameron, 1985). This
suggests that both the larval and juvenile neural systems are
responsive to environmental stimuli. We have not tested
whether the drugs used herein can induce metamorphosis in
the absence of contact with a hard surface, but the suppres-
sion by SNAP of the inductive properties of biofilm dem-
onstrate that NO signaling is downstream of sensory per-
ception leading to metamorphic events.

Various experiments have attempted to address how
metamorphosis is initiated and coordinated. The results can
differ among echinoid species. Although many species re-
quire a hard surface for settlement before metamorphosis,
larvae of the sand dollar D. excentricus suspended in sea-

402

C. D. BISHOP AND B. P. BRANDHORST

water can be induced to metamorphose by a heat-labile,
low-molecular-weight compound extracted from the sand of
a bed of adults (Highsmith. 1982; Burke. 1983, 1984).
Low-voltage electrical stimulation of the oral ganglion on
the lower lip of the larval mouth or the apical neuropile
between the preoral and anterolateral arms on the preoral
hood region of the D. excentricus larva induced metamor-
phosis (Burke. 1983). These sensitive larval areas have
axonal connections (Burke, 1983). and there is a ciliary
patch on the pre-oral hood that may have a sensory function
(Nakajima. 1986). Electrical stimulation of the oral gan-
glion has been reported to induce metamorphosis in several
echinoids, including L. pictus (Burke and Gibson, 1986),
although Cameron and Hinegardner (1974) reported other-
wise for L. pictus. The difference in these results may be
methodological. Recently, Beer et al. (2001) reported that
cells in the lower lip of the larval mouth of the sea urchin
Psammechinus miliaris develop immunoreactivity to a se-
rotonin antibody. We found staining for NOS protein and
NOS activity in cells that appear to be neurons in the lower
lip of the mouth, corresponding to the region of the oral
ganglion (Burke, 1983), and in cells of the preoral hood,
perhaps corresponding to the apical neuropile (Burke,
1983). When Burke (1983) excised the oral hood of D.
excentricus — including the oral ganglion and apical neuro-
pile— both fragments of the larva rapidly began metamor-
phosis, but this did not occur when only the preoral hood —
lacking the oral ganglion — or larval arms — lacking both
sites — were excised. The excised preoral hood and remain-
ing larva were able to respond to a chemical cue for meta-
morphosis, but isolated larval arms did not (Burke. 1983).
Isolated larval arms of some species, including D. excentri-
cus, can be induced to contract by treatment with divalent
ionophores or the neural transmitters adrenalin, noradrena-
lin. and dopamine (Burke, 1982. 1983). Dopamine induced
only a few whole D. excentricus larvae to initiate metamor-
phosis, suggesting the local response of arms can be inhib-
ited centrally.

On the basis of his experiments. Burke (1983) proposed
that there is a mutually inhibitory control of metamorphosis
between the oral hood and remainder of the D. excentricus
larva that is switched off in response to an appropriate cue
(or electrical stimulation). The inhibitory region of the oral
hood appears to be localized to the larval mouth (Burke.
1983), while the preoral and remaining regions of the larva
must have sensory receptors for the chemical cue that in-
duces metamorphosis. Data from histological sectioning
and optical reconstructions of the L. pictus oral epithelium
stained for NOS suggest that nitrergic neurons may reside
within this epithelium, possibly performing a sensory role
related to feeding or metamorphosis. These NOS-express-
ing cells were considered as candidate NO-signaling cen-
ters. We removed the pre-oral hood or the entire oral hood.
In the former case, most of the oral cells remained with the

larva; in the latter, they were removed (see Fig. 6B). In
direct contrast to D. excentricus, L. pictus did not metamor-
phose in response to the removal of the oral hood, a basic
distinction between these two species. Moreover, both
classes of L. pictus postoperative larvae were less sensitive
to NOS inhibition than were the intact controls, but they
apparently retained their sensitivity to inhibition of sGC. In
Figure 6A, ODQ was added directly to wells containing
postoperative larvae that had been treated with L-NAME for
14 h; there may have been an additive effect of the two
drugs. Accordingly, when tested separately, a five-fold ex-
cess of t-NAME or ODQ is required to induce metamor-
phosis of larvae lacking the oral hood over concentrations
that induce metamorphosis of control larvae (CDB, unpubl.
obs.). These experiments are difficult to interpret with re-
spect to Burke' s model of mutual inhibition, but they do
suggest the involvement of cells in the oral hood of L. pictus
in a pathway that regulates metamorphosis by NO/cGMP
signaling.

The regulatory role of these and other NOS-expressing
cells in larvae or juveniles may be additive. In L. pictus, the
tube feet of the rudiment appear to have sensory receptors
that may be involved in inducing metamorphosis (Burke.
1980). We found intense staining for NOS activity in the
nerve plexus lining the outer epithelial layer of the tube feet
of the rudiment. NO has been implicated in the relaxation of
adult tube feet (Billack et al., 1998). The presence of NOS
in cells associated with other structures that may have a
sensory role (the pre-oral hood, the tips of the anterolateral
arms and epaulettes) suggests that the drugs we used act on
one or more of these groups of cells to inhibit their produc-
tion of NO. Indeed, microsurgical and expression data in-
dicate that multiple larval structures and perhaps juvenile
structures transduce sensory information, by NO/cGMP sig-
naling, which leads to the initiation of metamorphosis. The
frequency of metamorphosis of larvae of L. variegatus was
increased by excess potassium or calcium ions (Cameron et
til., 1989). Metamorphosis of Strongylocentrotus purpura-
tus larvae was induced by treatment with calcium ionophore
A23187 or quercetin, an inhibitor of a [Ca.Mgj-ATPase
(Klein et al., 1985). Ionic fluxes may play a role, perhaps in
coordinating cellular responses (Burke. 1983; Pearse and
Cameron. 1991). Some mammalian isoforms of NOS (en-
dothelial and neuronal) are dependent on Ca:+ for their
activation (Mayer et al., 1998). The inductive properties of
Ca2+ flux may relate to the role of Ca2+ in the regulation of
NOS activity.

With this report, there is now evidence that NO plays a
repressive role in regulating the initiation of metamorphosis
in a protostome (llvanassa) and three deuterostomes (two
ascidians and an echinoid) (Froggett and Leise, 1999;
Bishop et al., 2001). NO is involved in metamorphosis of
larvae that do not grow before metamorphosis and retain
much of the larval tissue (ascidians), larvae that grow as

NO REPRESSES ECHINOID METAMORPHOSIS

403

swimming veliger larvae but do not undergo profound
changes upon metamorphosis (Ilyunussa), and larvae that
undergo extensive growth and catastrophic metamorphosis
in which most larval tissues are degraded and replaced by a
radically different juvenile (echinoids). NO, a universal and
ancient signaling molecule in eukaryotes. may have a role in
regulating metamorphosis in a wide diversity of animals.

Sea urchin larvae are optically clear and can easily be
cultured in large numbers. This fact, and a rich experimental
literature on settling and metamorphosis, make echinoids a
useful system with which to investigate the neuroanatomical
basis for the regulation of metamorphosis. These features
and our findings provide a basis for a more focused exper-
imental effort to identify which cells or organs repress
metamorphosis by NO production in L. piclits.

Acknowledgments

This research was funded by a Research Grant from the
Natural Sciences and Engineering Research Council of Can-
ada. We thank Robert Burke. Andrew Cameron, and Victor
Vacquier for helpful discussions, and two anonymous re-
viewers for useful comments.

Literature Cited

Aligue, R., H. Akhavan-Niak. and P. Russell. 1994. A role for Hsp90
in cell cycle control: Wee 1 tyrosine kinase activity requires interaction
with Hsp90. EMBO J. 13: 6099-6106.

Babal. P., O. Pechanova, and I. Bernatova. 2(1(10. Long-term admin-
istration of D-NAME induces hemodynamic and structural changes in
the cardiovascular system. Physiol. Res. 49: 47-54.

Beer. A-J., C. Moss, and M. Thorndyke. 2001. Development of sero-
tonin-like and SALMFamide-like immunoreactivity in the nervous
system of the sea urchin Psammechinus mi/iaris. Biol. Bull. 200:
268-280.

Bender, A. T., A. M. Silverstein, D. R. Demady. K. C. Kanelakis, S.
Noguchi, VV. C. Pratt, and V. Osawa. 1999. Neuronal nitric oxide
synthase is regulated by the hsp90-based chaperone system in r/ro.
J. Biol. Chem. 274: 1472-1478.

Billack, B., J. D. Laskin, P. T. Heck. W. Troll, M. A. Gallo, and D. E.
Heck. 1998. Alterations in cholinergic signaling modulate contrac-
tion of isolated sea urchin tube feet: potential role of nitric oxide. Biol.
Bull. 195: 196-197.

Birkeland, C., F. S. Chia, and R. R. Strathmann. 1971. Development,
substrate selection, delay of metamorphosis and growth in the sea star.
Mediaster aequalis Stimpson. Biol. Bull. 141: 99-108.

Bishop, C. D.. W. R. Bates, and B. P. Brandhorst. 2001. Regulation of
metamorphosis in ascidians involves NO/cGMP signaling and HSP90.
J. Exp. Zoo/. 289: 374-384.

Burke, R. 1980. Podial sensory receptors and the induction of metamor-
phosis in echinoids. J. Exp. Mar. Biol. Ecol. 47: 223-234.

Burke. R. 1982. Echmoid metamorphosis: retraction and resorption of
larval tissues. Pp. 513-518 in Echinoderms: Proceedings of the Inter-
national Echinodenns Conference. Tampa Ba\. 14 — 17 September
1981. J. M. Lawrence, ed. A. A. Balkema. Rotterdam.

Burke. R. 1983. Neural control of metamorphosis in Dendraster excen-
tricm. Biol. Bull. 164: 176-188.

Burke, R. 1984. Pheromonal control of metamorphosis in the Pacific
sand dollar. Dendraster excentricus. Science 255: 442-443.

Burke, R. D., and A. \V. Gibson. 1986. Cytological studies for the
studies of larval echinoids with notes on methods for inducing meta-
morphosis. Methods Cell Biol. 27: 295-308.

Caldwell. J. \V. 1972. Development, metamorphosis, and substrate se-
lection of the larvae of the sand dollar. Mellila quinques perforate.
M.Sc. Thesis. University of Florida. Gainesville.

Cameron. R. A., and R. T. Hinegardner. 1974. Initiation of metamor-
phosis in laboratory-cultured sea urchins. Biol. Bull. 146: 335-342.

Cameron, R. A., and R. T. Hinegardner. 1978. Early events of meta-
morphosis in sea urchins; description and analysis. J. Morphol. 157:
21-32.

Cameron, R. A., T. R. Tosteson, and V. Hensley. 1989. The control of
sea urchin metamorphosis: ionic effects. Dev. Growth. Differ. 31:
589-594.

Caplan, A. J. 1999. Hsp90's secrets unfold: new insights from structural
and functional studies. Trends Cell Biol. 9: 262-268.

Cueto. M., O. Hernandez-Perera, R. Martin. M. L. Bentura, J. Ro-
drigo, S. Lamas, and M. P. Golvano. 1996. Presence of nitric oxide
synthase activity in roots and nodules of Lupinus albus. FEBS Lett.
398: 59-164.

Czar, M. J., M. D. Galigniana, A. M. Silverstein, and W. B. Pratt. 1997.
Geldanamycin, heat shock protein 90-binding benzoquinone ansamy-
cin. inhibits steroid-dependent translocation of the glucocorticoid re-
ceptor from the cytoplasm to the nucleus. Biochemistry 36: 7776-
7785.

Davidson, E. H. 1991. Spatial mechanisms of gene regulation in meta-
zoan embryos. Development 113: 1-26.

Degnan, B. M., D. Souter, S. M. Degnan, and S. C. Long. 1997.
Induction of metamorphosis with potassium ions requires the develop-
ment of competence and an anterior signaling centre in the ascidian
Herdmania momiis. Dev. Genes Evol. 206: 370-376.

Froggett, S. J., and E. M. Leise. 1999. Metamorphosis in the marine
snail llyanassa obsolete!, yes or NO? Biol. Bull. 196: 57-62.

Garcia-Cardena, G., R. Fan, V. Shah, R. Sorrentino, G. Cirino, A.
Papapetropoulos, and W. C. Sessa. 1998. Dynamic activation of
endothelial nitric oxide synthase by Hsp90. Nature 392: 821-824.

Garthwaite. J.. E. Southam, C. L. Boulton, E. B. Neilsen, K. Schmidt,
and B. Mayer. 1995. Potent and selective inhibition of nitric oxide-
sensitive guanylyl cyclase by lH-[1.2.4|oxadiazolo[4,3-a]quinoxalin-
1-one. Mol. Phannacoi 48: 184-188.

Hebeiss, K., and H. Kilbinger. 1998. Nitric oxide-sensitive guanylyl
cyclase inhibits acetylcholine release and excitatory motor transmission
in the guinea-pig ileum. Neuroscience 82: 623—629.

Highsmith, R. C. 1982. Induced settlement and metamorphosis of sand
dollar (Dendraster excentricus} larvae in predator-free sites: adult sand
dollar beds. Ecology 63: 320-337.

Joly, G. A., M. Ayres, and R. G. Kilbourn. 1997. Potent inhibition of
inducible nitric oxide synthase by geldanamycin, a tyrosine kinase
inhibitor, in endothelial, smooth muscle cells, and in rat aorta. FEBS
Lett. 403: 40-44.

Klein, W. H., C. D. Carpenter. L. Philpotts, and B. P. Brandhorst.
1985. The sea urchin Spec family of calcium binding proteins: char-
acterization and consideration of possible roles in larval development.
Pp. 272-295 in The Cellular and Molecular Biology of Invertebrate
Development. R. Sawyer and R. Showman, eds. University of South
Carolina Press. Columbia. SC.

Kuzin, B., I. Roberts, N. Peunova, and G. Enikolopov. 1996. Nitric
oxide regulates cell proliferation during Droxuphila development. Cell
87: 639-649.

Leise, E. M., K. Thavaradhara, N. R. Durham, and B. E. Turner. 2001.
Serotonin and nitric oxide regulate metamorphosis in the marine snail
Ilvunassa obsoleta. Am. Zoo/. 41: 258-267.

Mayer, B., S. Pfeiffer, A. Schranimel, D. Koesling.. K. Schmidt, and F.

404

C. D. BISHOP AND B. P. BRANDHORST

Brunner. 1998. A new pathway of nitric oxide/cGMP signaling

involving S-nitroglutathione. J. Bid. Chem. 273: 3264-3270.
McDonald, L. J.. and F. Murad. 1996. Nitric oxide and cyclic GMP

signaling. Proc. Soc. Exp. Bio/. Med. 211: 1-6.
Nakajima, Y. 1986. Presence of a ciliary patch in preoral epithelium of

sea urchin plutei. De\: Growth Differ. 28: 243-249.
Nathan, D. F., and S. Lindquist. 1995. Mutational analysis of Hsp90

function: interactions with a steroid receptor and a protein kinase. Mol.

Cell. Biol. 15: 3917-392?.
Pearse, J. S., and R. A. Cameron. 1991. Echinoderrr.ata: Echinoidea.

Pp. 513-662 in Reproduction of Marine Invertebrates. Vol. VI. A. C.

Giese. J. S. Pearse. and V. B. Pearse, eds. The Boxwood Press. Pacific

Grove. CA.
Pratt, W. B. 1998. The hsp90-based chaperone system: involvement in

signal transduction from a variety of hormone and growth factor

receptors. Proc. Soc. Exp. Biol. Med. 217: 420-434.
Promrodou, C., S. M. Roe, R. O'Brien, J. E. Ladbury, P. VV. Piper, and

L. H. Pearl. 1997. Identification and structural characterization of the

ATP/ADP-binding site in the HSP90 molecular chaperone. Cell 90:

65-75.
Salter, M., P. J. Strijbos, S. Neale. C. Duffy, R. L. Follenfant, and J.

Garthvvaite. 1996. The nitric oxide-cyclic GMP pathway is required

for nociceptive signalling at specific loci within the somatosensory

pathway. Neuroscience 73: 649-655.

Satterlie, R. A., and R. A. Cameron. 1985. Electrical activity at meta-
morphosis in larvae of the sea urchin Lytechinus pictiis (Echinoidea:

Echinodermata). J. Exp. Zool. 235: 197-204.
Schnur, R. C., M. L. Corman, R. J. Gallaschun, B. A. Cooper, M. F,

Dee, J. L. Doty, M. L. Muzzi, J. D. Moyer, C. I. DiOrio, E. G.

Barbacci, P. E. Miller, A. T. O'Brien, M. J. Morin, B. A. Foster,

V. A. Pollack. D. M. Savage, D. E. Sloan, L. R. Pustilnik, and M. P.

Moyer. 1995. Inhibition of the oncogene product p!85(erbB-2) in
vitro and in vivo by geldanamycin and dihydrogeldanamycin deriva-
tives. J. Med. Chem. 38: 3806-3812.

Schulte. T. W., S. Akinaga, S. Soga, W. Sullivan, B. Stensgard. D. Toft,
and L. M. Neckers. 1998. Antibiotic radicicol binds to the N-
terminal domain of Hsp90 and shares important biologic activities with
geldanamycin. Cell Stress Chaperones 3: 100-108.

Sharma, S. V., T. Agatsuma, and H. Nakano. 1998. Targeting of the
protein chaperone HSP90 by the transformation suppressing agent
radicicol. Oncogene 16: 2639-2645.

Stebbins, C. E., A. A. Russo, C. Schneider, N. Rosen, F. U. Hartl, and
N. P. Pavletich. 1997. Crystal structure of an Hsp90-geldanamycin
complex: targeting of a protein chaperone by an antitumor agent. Cell
89: 239-250.

Strathmann, M. F. 1987. Reproduction and Development of Marine
Invertebrates of the Northern Pacific Coast. University of Washington
Press, Seattle.

Weinberg, R. J., J. G. Valtschanoff, and H. H. W. Schmidt. 1996. The
NADPH diaphorase histochemical stain. Pp. 237-248 in Methods in
Nitric Oxide Research, M. Feelisch and J. S. Stamler. eds. Wiley and
Son. Chichester, United Kingdom.

Whitesell, L., E. G. Mimnaugh, B. D. Costa, C. E. Myers, and L. M.
Neckers. 1994. Inhibition of heat shock protein Hsp90-pp60v"MX~ het-
eroprotein complex formation by benzoquinone ansamycins: essential
role for stress proteins in oncogenic transformation. P roc. Natl. Acad.
Sci. USA 91: 8324-8328.

Yen, A., S. Soong, H. J. Kwon, M. Y'oshida, T. Btppu, and S. Varvaya-
nis. 1994. Enhanced cell differentiation when RB is phosphorylated
and down-regulated by radicicol. a src-kinase inhibitor. Exp. Cell Res.
214: 163-171.

Reference: Biol. Bull. 201: 405-416. (December 2001)

Developmental Patterns and Cell Lineages of
Vermiform Embryos in Dicyemid Mesozoans

HIDETAKA FURUYA1. F. G. HOCHBERG2, AND KAZUHIKO TSUNEKI1

Department of Biology, Graduate School of Science, Osaka University, Toyonaka, Osaka 560-0043,

Japan: and 2Department of Invertebrate Zoology. Santa Barbara Museum of Natural History,

2559 Puesta del Sol Road. Santa Barbara, California 93105-2936

Abstract. Patterns of cell division and cell lineages of the
vermiform embryos of dicyemid mesozoans were studied in
four species belonging to four genera: Conocyema polymor-
plui, Dicyema apalachiensis, Microcyema vespa, and
Pseudicyema nakaoi. During early development, the follow-
ing common features were apparent: ( 1 ) the first cell divi-
sion produces prospective cells that generate the anterior
peripheral region of the embryo; (2) the second cell division
produces prospective cells that generate the posterior pe-
ripheral region plus the internal cells of the embryo: (3) in
the lineage of prospective internal cells, several divisions
ultimately result in cell death of one of the daughter cells.
Early developmental processes are almost identical in the
vermiform embryos of all four dicyemid genera. The cell
lineages appear to be invariant among embryos and are
highly conserved among species. Species-specific differ-
ences appear during later stages of embryogenesis. The
number of terminal divisions determines variations in pe-
ripheral cell numbers among genera and species. Thus, the
numbers of peripheral cells are fixed and hence species-
specific.

Introduction

All members of the phylum Dicyemida are found in the
renal sacs of benthic cephalopod molluscs (Nouvel, 1947:
McConnaughey. 1951; Hochberg. 19901. Their bodies
consist of the smallest number of cells among multicel-
lular animals (usually 10 to 40) and are organized in a
very simple fashion. Although recent studies have re-
vealed that they might not be truly primitive animals
deserving the name of mesozoans (Katayama et al., 1995:
Kobayashi et al.. 1999). they are still one of the most

Received 14 May 2001; accepted 17 August 2001.

interesting groups of lower invertebrates. Each species is
characterized by a fixed number of cells. The somatic
cells therefore undergo a limited number of species-
specific divisions during embryogenesis. The analysis of
embryonic cell lineages in dicyemids is intriguing, since
it may provide clues towards an understanding of the
simplest patterns of cell differentiation in multicellular
animals. A comparative study of cell lineage and devel-
opmental processes among related species of dicyemids
is also relevant to advance our understanding of morpho-
logical evolution in these simple animals.

Dicyemids produce two distinct types of embryos: ver-
miform embryos from an asexual agamete and infusoriform
embryos from fertilized eggs (Furuya et al.. 1996). From the
standpoint of morphological evolution, the vermiform em-
bryo is the more pertinent target for study because its shape
is similar to that of an adult. The cell lineage of vermiform
embryos has been fully documented in only two dicyemids,
Dicyema acitticephalum and D. japonicum (Furuya et al.,
1994). Among other species, cell lineages have been de-
scribed only to a limited extent in Microcyema vespa and
Pseudicyema truncatiim (Lameere, 1919; McConnaughey,
1938; Schartau, 1940: Nouvel. 1947: Bogomolov. 1970;
Lapan and Morowitz. 1975). Details of cell lineage in the
phylum as a whole remain to be determined.

In this paper we describe the pattern of cell divisions and
cell lineages in the embryogenesis of vermiform embryos in
dicyemids belonging to four genera: Conocyema polymor-
pha. Dicvema apalachiensis. Microcyema vespa. and
Pseudicyema nakaoi. Our data reveal that cell lineages in
vermiform embryos are highly conserved among species;
but species-characteristic features appear in the later embry-
ogenesis, and these are related to morphological evolution
and speciation.

405

406

Materials and Methods

H. FURUYA ET AL

Micmcyenia vespa

Specimens of Cotwcyenta polymorpha van Beneden,
1882, Dicyema apalachiensis Short, 1962, Microcyema
vespa van Beneden, 1882, and Pseudicyema tninciinim
(Whitman, 1883) were examined in the collections of the
Department of Invertebrate Zoology, Santa Barbara Mu-
seum of Natural History, Santa Barbara, California. Cono-
cveina polymorpha, found in Octopus rulgaris, was col-
lected by Henri Nouvel in the Mediterranean Sea (Nouvel,
1947). Microcvema vt'.v/w and P. tninciitiini, found in Sepia
offcinalis, were also collected by Nouvel in the Mediterra-
nean Sea (Nouvel, 1947). Dic\eimi apalachiensis, found in
Octopus joubini, was collected by Robert B. Short in the
Gulf of Mexico off Florida (Short, 1962).

Specimens of Pseudicveimi nukaoi Furuya, 1999, were
prepared for this study. A total of 57 host cuttlefishes. Sepia
esculenta, were purchased in the western part of Japan.
When dicyemids were detected in the kidney of the host
cuttlefish, small pieces of renal appendages with attached
dicyemids were removed and smeared on slide glasses. The
smears were fixed immediately in Bouin's fluid for 24 h and
then stored in 70% ethyl alcohol. The fixed smears were
stained in Ehrlich's hematoxylin and counterstained in eo-
sin. Stained smears were mounted using Entellan (Merck).

Embryos within the axial cell of parent nematogens were
observed with the aid of a light microscope under an oil-
immersion objective at a magnification of 2000 diameters.
Cells were identified by their position within the embryo,
their size, and the intensity of stain taken up by the nucleus
and cytoplasm. By careful examination, we were able to
identify each swollen nucleus that was about to divide and
each metaphase figure in terms of the cell that was about to
divide into two daughter cells. Each developing embryo was
sketched at three optical depths, and three-dimensional di-
agrams were reconstructed from these sketches. Measure-
ments and drawings were made with the aid of an ocular
micrometer and a drawing tube (Olympus U-DA), respec-
tively. Fully formed embryos consisted at most of 23 cells,
and special techniques such as injection of a tracer and
videoscopy were not required for determination of the cell
lineage. Early divisions of the vermiform embryos exam-
ined in this study were the same as those of Dicyema
(iciiticep/uiliini and D. japoniciim (see Furuya et «/.. 1994).
The terminology of cells used by Furuya et al. ( 1994) was
adopted in designating the cells in the present paper.

Results

In the Dicyemida. two adult forms, the nematogen and
the rhombogen, develop asexually from an agamete (axo-
blast) through a vermiform embryo within the axial cell of
parent nematogens (Fig. I).

Agamete diameter is about 6 /J,m. The first division is
meridional and unequal, producing two daughter cells. A
and B. Cell B becomes the mother cell of the peripheral
cells of the embryo's head. The second division involves
only cell B. This division is occasionally skipped. It is
extremely unequal, producing two daughter cells that are
quite different in size. The smaller of these two cells
ultimately degenerates without contributing to embryo-
genesis. The third division, involving cell A, is latitudinal
and equal, producing two daughter cells, 2A and 2a. Cell
2A is the mother cell of peripheral cells in the tail, and
cell 2a is the prospective axial cell. In the 2a lineage,
extremely unequal divisions occur at around the 5- and
7-cell stages. The resultant much smaller daughter cells
remain attached to the larger daughter cells until they
ultimately degenerate without contributing to embryo-
genesis. The fourth division, involving cell 2B, is merid-
ional and equal, producing two daughter cells, 3B and 3B.
At this 4-cell stage, two pairs of cells, 2A-2a and 3B-3B.
are arranged crosswise with respect to one another. The
furrow of the fourth division coincides with the plane of
bilateral symmetry of the embryo. The pattern of division
and the cell lineage are the same for the descendants of cell
3B and those of cell 3_B. The fifth division, involving cell
2A, is latitudinal and equal; resulting in the 5-cell embryo.
The plane of cell division coincides with the plane of
bilateral symmetry, and it separates the right cell 3 A from
the left cell 3A. These cells do not divide further but become
the two peripheral cells of the tail region, known as the
uropolar cells (Figs. 2a-c. 3a).

The pattern of cell division beyond the 5-cell stage
changes from spiral to bilateral. After the 5-cell stage,
divisions occur not one by one but in pairs, and the divisions
become almost synchronized. Subsequent developmental
stages thus proceed with odd numbers of cells, yielding, for
example, a 7-cell embryo, and so on.

The sixth division is extremely unequal. Both the 3B and 3B
cells divide, and they produce a pair of large cells and a pair of
much smaller daughter cells. The smaller cells degenerate and
eventually disappear. At around the 5-cell stage, cell 2a, the
prospective axial cell, undergoes an extremely unequal divi-
sion. The resultant smaller cell degenerates and eventually
disappears. The seventh division is equal, and results in the
7-cell embryo. Cells 4B and 4B divide and produce two pairs
of daughter cells, 5B ' and 5B2 plus 5B1 and 5B2. respectively.
The future anterior-posterior axis of the embryo corresponds
almost exactly to the 5B'-3A axis at the 7-cell stage. About the
7-cell stage, cell 3a, the prospective axial cell, undergoes an
extremely unequal division. The resulting smaller cells degen-
erate and eventually disappear.

After the seventh division, the order of division is not
necessarily identical among developing embryos. The

DEVELOPMENT OF VERMIFORM EMBRYOS

407

AX"

IP-DV

I^L

Figure 1. Light micrographs of nematogens in four species of dicyemids. Scale bars represent 10 jim.
Abbreviations: AG, agamete: AX, axial cell; DC, degenerating cell; DP. diapolar cell; DV. developing
vermiform embryo; MP, metapolar cell; PP. parapolar cell; PR; propolar cell; S. syncytium; UP. uropolar cell;
V. vermiform embryo. Microcycimi n:\pu: (a) whole body of a young individual; (b) a vermiform embryo in the
axial cell of the nematogen. Conocyema polymorpha'. (c) whole body of a nematogen; (d) developing vermiform
embryos in the axial cell of the nematogen. Dicyema upalachiensis: (e) whole body of the nematogen.
Pi>eudic\eina nakaoi: (f) whole body of a nematogen; (g) developing vermiform embryos in the axial cell of the
nematogen.

5B1 cell pair divides equally and produces two pairs of
daughter cells, 6B" and 6B12 plus 6B" and 6B12. The
5B2 cell pair divides equally and produces two pairs of
daughter cells, 6B21 and 6B22 plus 6B21 and 6B22. Nei-

ther pair divides further, and they form the anterior part
of the embryo (Fig. 2a, b). The 6B22 cell pair develops
into the parapolar cells, while the 6B", 6B12, and 6B2i
cell pairs eventually form a syncytium, which is more

408

H. FURLIYA ET AL.

a6B,§Bi:.

6B;;

. 5B"2 4B'2 SBi1 6A'i:

• _ ^- , C A 1 2 '

5B'1: SB-' 5B:2 5B

Figure!. The late-stage vermiform embryos of four species of dicyemids. Scale bar represents 10 ^m. Cilia
are omitted. See text for explanations of cell division notations. Other abbreviations: AG. agamete: AX, axial
cell; DP. diapolar cell; MP. metapolar cell; PR, propolar cell; PP. parapolar cell; S. syncytium; UP. uropolar cell.
Microcvema vespa: (a) a late-stage embryo (sagittal optical section) — note an agamete (5a~) in the cytoplasm of
an axial cell (5a'l; (b) a late-stage embryo (sagittal optical section); (c) formed embryo (sagittal optical
section) — pairs of 6B", 6B'2. and 6B:i cells form a syncytium (S) that is more conspicuously stained with
hematoxylin. Conocyema polymorpha; (d) a late-stage embryo (sagittal optical section) — note an agamete (6a~)
incorporated in the cytoplasm of an axial cell (6a'): (e) a late-stage embryo (lateral view); (f) formed embryo
(lateral view) — pairs of 4B" and 5B21 cells form propolar cells (PR) that are more conspicuously stained with
hematoxylin. Dicyema apalachiensis: (g) 13-cell stage — note an anaphase figure of 4B': cell and a metaphase
figure of 4B12 cell; (h) 15-cell stage — the 5B1" and 5B':i pairs form the propolar cells (PR), while the 5B"2
and 5B'22 pairs form another type of polar cell, the metapolar cell (MP); (I) formed embryo (lateral view) —
propolar cells and metapolar cells are more conspicuously stained with hematoxylin. Pseudicyema nakaoi: (j)
22-cell stage — note a metaphase figure in 4B12 cell — the plane of this division, in contrast to the divisions of
4B|: pair (Fig. 2g), are oblique to the anterior-posterior axis, and as the result, cells of the propolar tier alternate
with cells of the metapolar tier (see Fig. 2k, 1 ); (k) a late-stage embryo (lateral view); (I) formed embryo (lateral
view) — propolar cells and metapolar cells are more conspicuously stained with hematoxylin.

conspicuously stained with hematoxylin than the other
cells (Fig. Ib). The two parapolar cells cover more than
half of the syncytium (Fig. 2c). As peripheral cells are
formed, the prospective axial cell, 4a, divides unequally.
The large anterior cell, 5a', undergoes no further divi-
sions and becomes the axial cell, while the smaller pos-
terior cell, 5a2, is incorporated into the axial cell and

becomes an agamete (Fig. 2a). At this stage, cilia are first
evident on the peripheral cells. Cilia on the external
surface of the syncytium are directed anteriorly and are
more densely distributed than on other peripheral cells.
The fully formed embryo consists of three types of
cells: peripheral cells, one syncytial cell, and an axial
cell, which contains two agametes (Figs. Ib, 2c). The

DEVELOPMENT OF VERMIFORM EMBRYOS

409

H2a— i-3a — i-4a — f.

••• rtA

.- Af^

i-x >-x loa

,QA ..

2Al3A

Ur
... I ID

\5B'^

r3B r:4B \ 11-

L CD* J

16B"

PP
s

o o * r ^

[dA

1 ID

{/1A-

4A
r4A'

... up
no

_ . 1

••• Ur

1 IP

rSB'"

F3B— I' r5B»
2B ^4B-HfB,a

PR
MP

.. DD

JL3 "I™ 4LJ

[— 1, rsf7

2Bi 4B • — H — —

MP
PR

1 1 n _— ji n •

•• Mr
DD

OL5 T" 4L5

•B —

2a r-3a ^4a — r

A A ' .

c rfia
•5a— fo

x ^6a-

-AX
-AG
.. np

r4A
r3A-| ,CA."

.. no

14A- [D^;,

1 ID

4A' ••

'•• Ur

• • DP

.. RD

4A: f — 7,

-4R'1 ..

r36'— ««..

.. DD

)

.. DD

l3B:i— 4B:

— [ 2,

.. pp

J3B [4jp'""

— ]

., DD

13B;T— 4B:

** rK

«-x

IDB
r6a'

-AX

rfiA"

632

-AG
DP

[5A'-[

. np

r4AH

. fiA'21 ...

... i in

1 5 A'«

3AJ

LcA'22

0 1

L 4/\

^2G_^1 £>»n2

rfiA':'

U r

IfiA122

L A A:

* 4A •'

r4B"

r5B'"

-PR

3B 1

~-l5B'«-

_r5B'-

-MP
PR

hAO

•2B'

LOD

r otj

r i

L ^D2'

[5B

-PR

r3B_ 1 "

5B"-
_r5B':i •

-MP
-PR

•2B-

— LsBf-

-MP

CR."

OR: A p :

P'igure 3. Cell lineages of vermiform embryos. A cross (x) indicates that a cell, formed as the result of an
unequal division, degenerates and does not contribute to the formation of the embryo. See text for explanations
of cell division notations. Other abbreviations: AG, agamete: AX, axial cell; DP. diapolar cell: MP. metapolar
cell; PP, parapolar cell; PR. propolar cell; S. syncytium; UP. uropolar cell, (a) Microcyema vespa: (b)
Conocyema polymorpha: (cl Dicyema apalachiensis', (d) Pseudicyema nakaoi.

peripheral cells are composed of two parapolar cells and
two uropolar cells. The swollen cephalic head region
consists of a calotte and two parapolar cells. The calotte
is a syncytium. The trunk is composed of two uropolar
cells. Further development involves growth and en-
largement of the syncytium and branching of the axial

cell (Fig. la). Total length, excluding cilia, of the fully
formed vermiform embryo is about 50 /xm. and the width
is about 20 ^m. The cell lineage of the vermiform
embryo is summarized in Figure 3a. No variations
in cell lineage were found in more than 50 embryos
examined.

410

H. FURUYA ET AL

Conocyema polymorpha

Agamete diameter is about 7 fxm. The first division is
meridional and unequal, producing two daughter cells that
are slightly different in size. A and B. The larger cell B
becomes the mother cell of the peripheral cells of the
embryo's head (propolar and parapolar cells). The second
division involves only the smaller cell A. This division is
latitudinal and equal, producing two daughter cells, 2A and
2a. Cell 2A is the mother cell of the peripheral cells of the
posterior trunk and tail, and cell 2a is the prospective axial
cell (Fig. 3b). In the 2a lineage, extremely unequal divisions
occur at around the 5-, 1 1-. and 13-cell stages (Fig. 3b). and
the resultant much smaller daughter cells remain attached to
the larger daughter cells until they ultimately degenerate
without contributing to embryogenesis. The third division,
involving cell B, is meridional and equal, producing two
daughter cells, 2B and 2B. At this 4-cell stage, two pairs of
cells, 2A-2a and 2B-2B, are arranged crosswise with respect
to one another. The furrow of the third division coincides
with the plane of bilateral symmetry of the embryo. The
pattern of division and the cell lineage are the same for the
descendants of cell 2B and those of cell 2B (Fig. 3b). The
fourth division, involving cell 2A. is meridional and equal,
resulting in the 5-cell embryo. The plane of cell division
again coincides with the plane of bilateral symmetry, and it
separates the right cell 3 A from the left cell 3A. The pattern
of cell division and the cell lineage are the same for de-
scendants of cell 3A and for those of cell 3 A (Fig. 3b).

The pattern of cell division beyond the 5-cell stage
changes from spiral to bilateral. Beyond the 5-cell stage,
divisions occur not one by one but in pairs, and they become
almost synchronized. Subsequent developmental stages thus
proceed with odd numbers of cells.

At around the 5-cell stage, cell 2a, the prospective axial
cell, undergoes an extremely unequal division. The resultant
smaller cell degenerates and finally disappears. The fifth cell
division is equal and results in the 7-cell embryo. Thus, cells
2B and 2B divide and produce two pairs of daughter cells,
3B' and 3B2 plus 3B1 and 3B2. respectively. The future
anterior-posterior axis of the embryo corresponds almost
exactly to the 3B'-3A axis at the 7-cell stage. The sixth
division is slightly unequal. Cells 3A and 3 A divide into two
pairs of daughter cells, 4A1 and 4A2 plus 4A1 and 4 A2.
Cells 4A: and 4A2 are the smallest cells at this stage. In
addition to the 2a lineage, cells 3B: and 3B2 undergo
unequal divisions, each generating a pair of cells, one large
and one much smaller. The smaller cells degenerate and
finally disappear, although they remain in place on the
developing embryo until later stages.

The 3B1 pair divide equally into 4B11 and 4B12 pairs.
These cells undergo no further divisions. The 4B" pair
become the propolar cells, and the 4B12 pair become the
parapolar cells (Figs. 2e, f). The cell in the 2a lineage (cell

3a) is incorporated into the inside of the embryo, and the
3B1 pair divide and rearrange their descendants. At around
the 11 -cell stage, the 3a cell undergoes an extremely un-
equal division.

The 13-cell stage is achieved by equal divisions of cells
4A2 and 4A2. The resultant 5 A21 and 5 A22 pairs undergo no
further divisions and become diapolar cells and uropolar
cells, respectively. Soon after these divisions, the 4B2 pair
divide equally into two pairs of daughter cells, 5B21 and
5B22 plus 5B21 and 5B22. The 5B21 and 5B22 pair undergo
no further divisions and become the propolar cells and
parapolar cells, respectively (Fig. 2e, f). At around the
13-cell stage, the prospective axial cell 4a undergoes an
extremely unequal division.

At the final stage of embryogenesis, the prospective axial
cell, 5a. divides unequally. The large anterior cell, 6a',
undergoes no further divisions and becomes the axial cell,
while the smaller posterior cell, 6a2, is incorporated into the
axial cell and becomes an agamete (Fig. 2d).

The fully formed vermiform embryo of Conocyema poly-
iiiorpha consists of 14 peripheral cells and one axial cell,
which contains one or two agametes (Fig. 2f). The head
peripheral cells are composed of four propolar cells and
parapolar cells. The propolar cells have short, dense cilia
and form the calotte, which is more conspicuously stained
with hematoxylin than the other cells. Four diapolar cells
make up the trunk peripheral cells. The caudal peripheral
cells are uropolar cells. The length, excluding cilia, of the
fully formed embryo is about 25 /urn, and the width is about
10 jum. The cell lineage of the vermiform embryo is sum-
marized in Figure 3b. No variations in cell lineage were
found in more than 80 embryos examined.

Dic\ciim (iptilcicliiensis

Agamete diameter is about 5.5 jum. The first division is
meridional and equal, producing two daughter cells of equal
size. The subsequent patterns of development and cell lin-
eage up to the 7-cell stage are the same as those described
for CO/U>C\CIIHI polymorpha.

At the 7-cell stage, cells 3B2 and 3B2 undergo unequal
divisions, each generating a pair of cells, one large and one
much smaller cell. The 9-cell stage is achieved by unequal
divisions of cells 3B1 and 3B1. The resultant small cells,
4B ' ' and 4B". divide again into two pairs of daughter cells,
5B " ' and 5B " 2 plus 5B1" and 5B1'2. in the anterior part of
the embryo (Fig. 2g-i). Almost simultaneously, the resultant
large cells, 4B12 and 4B12, divide again into two pairs of
daughter cells, 5B121 and~5B122 plus 5B12' and 5B122. in the
anterior part of the embryo (Fig. 2h. i). These cells undergo
no further divisions, and the 5Bm and 5B121 pairs become
the propolar cells, while the 5B112 and 5B122 pairs become
the metapolar cells. The cell in the 2a lineage (cell 4a) is

DEVELOPMENT OF VERMIFORM EMBRYOS

411

incorporated into the inside of the embryo, and the other
cells, 4B" and 4B". divide and rearrange their descendants.

At the 9-cell stage, cells 3A and 3A divide equally into
two pairs of daughter cells, 4A1 and 4A2 plus 4A1 and 4A:.
They do not divide further; the 4A ' pair become the uropolar
cells, and the 4A2 pair become the diapolar cells (Fig. 3c). At
the final stage of embryogenesis, the prospective axial cell. 4a.
divides unequally. The large anterior cell. 5a'. becomes the
axial cell, and the smaller posterior cell. 5a2, is incorporated
into the axial cell and becomes an agamete.

The vermiform embryo of Dicyema apalachiensis con-
sists of 14 peripheral cells and one axial cell, which contains
one or two agametes. The peripheral cells of the head region
are composed of four propolar cells, four metapolar cells,
and two parapolar cells. The propolar and metapolar cells
have short, dense cilia and form the calotte. Two diapolar
cells make up the trunk peripheral cells. Two caudal periph-
eral cells are uropolar cells. The length, excluding cilia, of
the fully formed embryo is about 30 /nm. and the width is
about 10 ij.m. The cell lineage of the vermiform embryo is
summarized in Figure 3c. No variations in cell lineage were
found in more than 50 embryos examined.

Pseudicyema nakuoi

Agamete diameter is about 6.5 /urn. The first division is
equal, producing two daughter cells of equal size. The
subsequent patterns of development and cell lineage up to
the 9-cell stage are the same as those described for Dicvema
apalachiensis (see Fig. 3c).

At the 9-cell stage, cells 3B2 and 3B2 undergo extremely
unequal divisions, each generating a pair of cells, one large
and one much smaller cell. The smaller cells resulting from
this division degenerate and finally disappear. In the 2a line,
unequal divisions occur at around the 5-. 9-. and 1 1 -cell
stages (Fig. 3d). The pattern of development and cell lin-
eages up to the 9-cell stage are the same in both P. trunca-
twn and P. nakaoi. Further development was not studied in
P. truncation, because adequate material was not available.

After the unequal divisions of the 3B2 pair, cell pairs 4A1
and 3B1 undergo equal divisions almost simultaneously and
produce two pairs of daughter cells. 5A11 and 5A12 plus
4B" and 4B12. to form the 13-cell-stage embryo. At around
the 13-cell stage, the prospective axial cell. 4a. undergoes an
unequal cell division. The 5A11 pair divide equally and
produce two pairs of daughter cells, 6A1" and 6A112 plus
6A" ' and 6A1 12. The plane of this division is parallel to the
anterior-posterior axis, in contrast to the previous division,
which occurs parallel to the perpendicular axis. As a result,
cells 6A1 " and 6A1" are situated on the left and right sides
of the embryo, respectively.

At the 15-cell stage, cell pairs 4B2 and 5A12 undergo
equal divisions almost simultaneously, and produce two
pairs of daughter cells, 5B21 and 5B22 plus 6A121 and6A122,

to form the 19-cell-stage embryo. These pairs undergo no
further divisions. Cell pair 5B2' become the parapolar cells,
and the 5B22 pair become the anterior diapolar cells (Fig.
3d). Cell pair 6A121 become the uropolar cells, and the
6A122 pair become the posterior diapolar cells (Fig. 3d).

At the 1 9-cell stage, the 4B" and 4B12 pairs divide
equally into two pairs of daughter cells, 5B111 and 5B"
plus 5B121 and 5B122, in the anterior part of the embryo
(Fig. 2k. 1). These divisions proceed cell by cell. The daugh-
ter cells undergo no further divisions, and the 5B1" and
5B12' pairs become the propolar cells, while the 5B112 and
5B122 pairs become the metapolar cells (Fig. 2j). The planes
of these divisions, in contrast to the previous division, are
oblique to the anterior-posterior axis. As the result, cells of
the propolar tier alternate with cells of the metapolar tier
(Fig. 2k, 1).

At the final stage of embryogenesis, the prospective axial
cell, 5a. divides unequally. The large anterior cell. 6a'.
becomes the axial cell, and the smaller posterior cell, 6a2, is
incorporated into the axial cell and becomes an agamete.

The fully formed vermiform embryo of P. nakaoi con-
sists of 22 peripheral cells and one axial cell, which contains
one agamete. The peripheral cells of the head region are
composed of four propolar cells, four metapolar cells, and
two parapolar cells. The propolar and metapolar cells have
dense short cilia and form the calotte. Ten diapolar cells
make up the trunk peripheral cells. Two caudal peripheral
cells are uropolar cells. The body length, excluding cilia, of
the fully formed embryo is about 70 ju.m. and the body width
is about 16 /am. The cell lineage of the vermiform embryo
is summarized in Figure 3d. No variations in cell lineage
were found in more than 50 embryos examined.

Discussion

Patterns of development of the vermiform embryos of
four species of dicyemids belonging to four genera, namely
Microcyema vespa, Coiwcveimi polymorpha, Dicvema
apalachiensis, and Pseudicyema nakaoi, were studied in
detail. In the embryogenesis of each species, cell divisions
proceed without variation and result in fully formed em-
bryos with a definite number and arrangement of cells. The
process of development of vermiform embryos is very sim-
ple and seems to be programmed similarly to that of infu-
soriform embryos and infusorigens (Furuya el al.. 1992b,
1993. 1995). Seven different cell lineages including those of
two previously described species, D. acuticephalum and D.
japonicum (Figs. 3. 4; see also Furuya el ai, 1994), could be
compared. Early developmental processes up to the 7-cell
stage are almost identical in vermiform embryos examined
in this study and those of D. acuticephalum and D. japoni-
cum (Figs. 5. 6).

Our results are different from earlier reports with respect
to the timing of cell-fate specification (Lameere. 1919;

412

2a-r3a-r4a —

-r5a— r6a-i;V

A r^

LX Lx

LX ux fa ••

C A"

•• Mo

A- r4A-

1 IP

r 3 A I

5A •

Pip

2A-

C A11

- Ur

14A:-

f OM ""

...i IP

5A_

r3B— p

r2B-

4B1 fiR12,

x ^5B--{fB::

••PR
-MP
-PR
-MP
PP

I-3B-'

— p4B: — [

PiD

•B—

^D D
— On"1
i DD

4& r6Br-

X L ggizj" —

-• Ur

-PR
-MP
-PR

-MP
PP

l3B2

— .-4B' J

n on:

LX UOD •'

/ID2 ..

... PP

H. FURUYA ET AL.

2aT3a-r4a 1- 5a r
LX Lx *-x *•

.£A"'

c r /d
6a— I ,
x L 7a2 ••••

'• AVA

••AG
RD

r5A" -C

•• Ur
HP

A. [4A

A A121

3A| 15A-

L RA"2

-- Ur
r»D

L A A2 ...

; *4A

•• DP

p\D

?A c A u r^A

^A r5A — ITTT;

r4A'J 6A

•• UP

r— 1

-KA'2' .

.. i ID

IC-AI:

3AJ L5A_

L R A122 ..

p»Q

—

* DA

1 A A'

L 4A ••

_CP" ...

•• DP
.. DP

• TR' .— rir"

_J5B

LSR'2 ...

.. MD

Lx
r2B> |1Q5,

r562"-

-PR

UB —

l5B;'2-

CD^^1

-MP
PP

L ,1 n "

... ^j^j

LB—

LSB222-

_CD11 ...

-DP

... PR

r5B

T.CQ12

L2B< 1C"

r5B2"-

-PR

.R^_f —

LSB^-

-MP

3B^^-1
L -i rv?

r5B22'-

... pp

I-5B*-

-DP

Figure 4. Cell lineages of vermiform embryos in Dicyema acuticephalum and D. japonicum (modified from
Furuya el al, 1994). A cross (x) indicates that a cell resulting from an unequal division degenerates and does
not contribute to the formation of the embryo. Abbreviations: AG, agamete: AX, axial cell: DP, diapolar cell;
MP. metapolar cell; PP, parapolar cell; PR, propolar cell; UP. uropolar cell, (a) D acuticephalum with 18
peripheral cells; (b) D. acuticephalum with 16 peripheral cells; (c) D. japimicum.

Gersch, 1938; McConnaughey. 1951). According to
Lameere (1919). in M. vespa and P. tnincanim the first
division is unequal, and as a result two daughter cells of
different sizes are produced. One of the daughter cells
(usually the larger one) is described as a prospective axial
cell, and the other is regarded as the mother cell of the
peripheral cells. However, in all species examined, we
found that the prospective axial cell was produced at the
second division, not at the first division. Gersch (1938)
and McConnaughey ( 195 1 ) also claimed that the prospec-
tive axial cell is produced at the first division in D. typus,
D. hahinnuhi, Dicyemennea abelis, and Dicyentennea
californica. Although the possibility that two types of
first division exist cannot be excluded in this study, the
results of those early observations remain to be con-
firmed.

Early developmental pattern and cell lineage

Comparisons of developmental processes and cell lin-
eages among various species of dicyemids reveal conser-

vative features in the early development. Although dicy-
emids from different host species and geographically
different distributions were compared, the developmental
processes and cell lineages are almost identical from an
agamete to the 7-cell stage (Fig. 5). Cell-fate segregation
appears in the very early stages of embryogenesis. Three
types of prospective cells that form the body of embryos,
such as the agamete, the axial cell, and the peripheral
cells, can be identified as early as the 3-cell stage. In the
development of vermiform embryos, cell fates may be
initially segregated. This conserved feature among spe-
cies may represent the basic plan in forming bodies of
vermiform embryos (Fig. 6). These features in cell lin-
eage suggest that the early developmental processes have
persisted through the evolution of dicyemids. Vermiform
embryos develop in the confined space of an axial cell
located within the parent nematogen. This peculiar hab-
itat thus may constrain the developmental process, as
well as limit the size and number of cells that compose
the body. As a result, development may appear to be
conserved.

DEVELOPMENT OF VERMIFORM EMBRYOS

413

AG 2 3 4 5 7

agamete (axoblast)
a-lineage
A-lineage
B-lineage

9 11 13 15 17 19

Figure 5. Developmental processes of vermiform embryos in several species of dicyemids. The develop-
mental patterns and cell lineages from (he agamete (AG) to 7-cell stage are identical among the species. The
numerals in the bottom row represent cell number stages in the development. Arrows in the developing embryos
indicate daughter cells that were produced by the proceeding division, (a) Microcyema vespa', (b) Conocyema
polymorpha: (cl Dicyema apalachiensis', (d) D. acuticephalum with 16 peripheral cells; (e) D. acuticephaluin
with 18 peripheral cells; (fl D. japonicum; (g) Pseudicveina nakaoi.

Variations of terminal divisions in cell lineages

Species-specific patterns of development and cell lin-
eages appear in the later stages of embryogenesis. The most
striking difference is seen in terminal divisions in the cell
lineage that give rise to variations in peripheral cell num-
bers. For instance, species-specific differences in the periph-
eral cell number between Dicyema acuticephalum and D.
japonicum can be attributed to the number of divisions of
the 4A1 pair (Fig. 4; Furuya et ai, 1994). In other species,
additional terminal divisions occur toward the end of the
establishment of another cell lineage as well. The various
numbers of terminal divisions, which are genetically deter-
mined, clearly play a significant role in the morphogenesis
of vermiform embryos and may be correlated with specia-
tion in the dicyemids.

In most species of dicyemids, vermiform embryos have a
constant number of peripheral cells. However, some species
of dicyemids, such as Dicvema acuticephalum, D. hilohum,
D. benthoctopi, D. erythrum, D. lycidoceum, and D. rhadi-
nuni. have a variable number of peripheral cells (Nouvel,
1947; Couch and Short, 1964; Hochberg and Short, 1970;

Furuya et ai, 1992a; 1994; Furuya, 1999). Such intraspe-
cific variation in peripheral cell numbers could be attributed
to minor differences in numbers of terminal divisions in
certain cell lineages (compare Fig. 4a and b).

In the developmental patterns of vermiform embryos, the
cell lineages do not vary, and the terminal divisions usually
occur bilaterally. Thus, several even numbers of peripheral
cells are formed as the result of a pair of terminal divisions
in both the 2A- and B-cell lineages. In species that have a
variable number of peripheral cells, such as Dicyema eryth-
rum, D. Ivcidoceum, and D. rhadinum, some peaks are
evident in even numbers of peripheral cells (see tables in
Furuya, 1999). The number of terminal divisions may not be
strictly programmed in these exceptional species.

Later development and lan'al morphologv

In the evolution of dicyemids, various types of vermiform
embryos must have been produced as deviations from a
common developmental pattern. Unusual species, such as
Microcyema vespa and Conocyema polymorpha, not only
differ morphologically from other dicyemids but are distinct

414

H. FURUYA ET AL

2a-

r3A
L 1A

{agamete (axoblast)
axial cell

•] peripheral cells of trunk
. J (diapolars) & tail (uropolars)

B—

2BJ

1-5R-

[3BJ-
3B-

peripheral cells of head
(propolars & metapolars)
& trunk (diapolars)

Figure 6. A common cell lineage in all the vermiform embryos ex-
amined. At the first division, an agamete (AG) divides to produce two
daughter cells. A and B. Cell A divides into two daughter cells. Cell 2a is
a mother cell for both an axial cell and agamete. Descendants of cell 2A
form the peripheral cells of both trunk and tail. Descendants of cell B form
the peripheral cells of both the head and anterior trunk. A cross (x)
indicates that a cell formed by unequal division degenerates and does not
contribute to the formation of the embryo.

in the later stages of development. As shown in the cla-
dogram (Fig. 7), these two species of dicyemids are clearly
distinct and separate when compared with the clade com-
posed of the genera Dicyemu and Pseudicyema. Some
changes that occur in cell lineages certainly are reflected in
morphological features.

The genus Pseudicyema. as diagnosed by Nouvel ( 1933),
is morphologically very similar to Dicyema. As a result, it
occasionally has been treated as a subgenus of Dicyema
(Hochberg. 1990). The difference between Pseudicyema
and Dicvema depends on whether cells of the propolar tier
are alternate or opposite with respect to the cells in the
metapolar tier. The developmental processes in these genera
are different only in the terminal cell lineage and the pattern
of cell divisions at the final stage of embryogenesis. On the
basis of cell lineages, differences between Dicyema and
Pseudicvema are within the range of inter-species differ-
ences in Dicvema, as shown in the cladogram. However, as
far as calotte configuration and the process of calotte for-
mation are concerned, Dicyemu and Pseudicyema can be
clearly identified as separate groups. Although cell lineage
is an important character, it may not necessarily help to
determine the definition of genera. Detailed comparative
studies on cell lineages and organization of infusoriform
embryos are also indispensable in separating dicyemid taxa.

In recent years, it has been argued that the evolution of
morphological features requires alterations in developmen-
tal processes. In dicyemids, the cell lineage of Microcyema
vespa is closer to a conservative lineage than in other genera
in the phylum, but vermiform embryos of M. vespa show a
distinctive form not seen in other genera. It is possible that

in M. vespa the developmental process may be truncated,
resulting in a simple cell lineage and a body organization
with a very small number of peripheral cells. However,
changes in cell lineage may not always contribute to mor-
phological characters. For example, there are some differ-
ences in the later cell lineage between Dicyema acuticeph-
tilitm and D. apalachiensis. but these dicyemids are very
similar in general body shape.

Cell death

McConnaughey (1951) described chromatin elimination
from the prospective axial cell. In Dicyema acuticephalum
and D. japonicum, what appears to be a mass of eliminated
chromatin is actually a small cell that is produced as the
result of an extremely unequal division (Furuya et al.,
1994). In Pseudicvenui tnincatnm and Microcyema vespa,
Lameere (1919) noted that the prospective axial cell under-
went an unequal division and that the smaller daughter cell
itself divided once or twice to produce two or four small

• Conocyerrndae -

- Dicyemidae •

Figure 7. Cladogram of six species of dicyemids based on cell lin-
eages of the vermiform embryos. These dicyemids might have been de-
rived from an ancestor that had a basic cell lineage as shown in Figure 6.
Modifications in cell lineages might result in diversity of morphology
giving rise to two separate families, namely. Conocyemidae and Dicyemi-
dae. Sketches at top of cladogram indicate the size and shape of the whole
bodies of adult stages of each species. Bars represent modifications of the
different cell lineages as follows: (1) Early development as shown in
Figure 6. (2A) Calotte is formed with a tier of polar cells. (2B) Calotte is
formed with two tiers of polar cells: propolars and metapolars present. (3A)
Calotte form-, a syiicytium; diapolars absent. (3B) Calotte is cellular;
diapolars present. (4A) Calotte is formed from both 3B1- and 3B:-cell
lineages. (4B) Calotte is formed only from 3B'-cell lineage. (5A) Propolars
are located perpendicularly above metapolars. (5B) Propolars are obliquely
oriented to metapolars. (6A) Cell death occurs both in 3B1- and 3B"-cell
lineages. (oB) Cell death occurs only in 3B--cell lineages; both 4A'- and
4A2-cells undergo no further divisions.

DEVELOPMENT OF VERMIFORM EMBRYOS

415

Table 1

The number of cell divisions in each cell lineage

Dicvemid

A-cell lineage

Peripheral cell number

A-cell lineage

a-cell lineage

B-cell lineage

Total cell divisions

Micracyema vespa

3(2)

10(3*)

15(5)

Conocvenui polyrnorpha

4(3)

9(2)

19(5)

Dicyema apalachiensis

3(2)

11 (2)

18(4)

Dicyema aciiticepha/imi

16"

5(4)

13(4)

24(8)

Dicyema acuticephalum

18"

5(4)

15(4)

26(8)

Dicvciim jupuniciim

22"

5(4)

13(2)

28(6)

Pseudicyema nakaoi

4(3)

13(2)

27(5)

The numbers in parentheses represent the number of extremely unequal cell divisions.

* One cell division was not consistently observed.

# From Furuya et al. ( 1994).

cells that do not degenerate. We were able to examine the
details of these small cells in several dicyemids, including
the species studied by Lameere. In contrast to Lameere's
observation, the small cell does not undergo further divi-
sions. In his report, the small cells were exclusively derived
from a prospective axial cell (a-cell lineage), but we recog-
nized they are formed in both the a-cell and B-cell lineages,
as recognized in the previous study of D. acuticephalum and
D. japonicuin.

The small cells eventually die and are eliminated without
contributing to the embryogenesis. This is considered to be
a programmed cell death as described in the development of
infusoriform embryos (Furuya et al., 1992b). In the dicy-
emids examined, extremely unequal divisions take place
four to eight times during embryogenesis (Table 1). The
number of such divisions is as definite according to species
as the number of peripheral cells. In the a-cell lineage, much
programmed cell death appears frequently in dicyemids that
consist of a large number of peripheral cells. It seems
possible that successive, extremely unequal divisions in the
a-cell lineage may be required to maintain an increased
amount of cytoplasm in the large axial cell. The axial cell
retains most of the cytoplasm of the mother cell and en-
larges after each cell division. In most dicyemids, the axial
cell elongates as peripheral cell numbers increase. Thus,
peripheral cell number appears to be correlated to the num-
ber of programmed cell deaths.

The B-cell lineage gives rise to the head region, in which
cell death occurs in all dicyemids examined. In contrast, no
cell death was observed in the A-cell lineage. The A-cell
lineage gives rise to the trunk and tail region, which are
composed of standard peripheral cells. Programmed cell
death in dicyemids appears in cell lineages associated with
remarkably differentiated cells, e.g.. the axial cell and cal-
otte cells. Thus, cell death may be intimately involved in the
advanced characteristic differentiation of cells.

Several features in developmental pattern and cell
lineages among species

The early development of dicyemids is conservative and
may be summarized as follows: (1) the first cell division
produces prospective cells that generate the anterior periph-
eral region of the embryo; (2) the second cell division
produces prospective cells that generate the posterior pe-
ripheral region plus the internal cells within the embryo; (3)
in the lineage of prospective internal cells, several divisions
ultimately result in the death of one of the daughter cells.
Developmental processes to the 7-cell stage are almost
identical in the vermiform embryos of the four genera
examined (Figs. 5, 6).

In contrast, distinct species-specific differences appear in
the order and number of terminal divisions of peripheral
cells. Most of the changes in terminal divisions can be
correlated with individual body length. Generic differences
appear in the number of cells that contribute to the calotte
during the final stage of embryogenesis. Distinct morpho-
logical features typically emerge following a final cell di-
vision or after the embryo escapes from the axial cell of the
adult. Subsequent processes, proceeding without cell divi-
sions, are cell differentiation in the head region and cell
elongation in the trunk region.

On the basis of cell lineage, a simple cladogram was
constructed (Fig. 7). Cell lineages from an agamete to the
7-cell stage were almost identical among species (bar 1 ).
The terminal of B-cell lineage indicates some variation
among species. In the family Conocyemidae, a calotte is
formed with a tier of polar cells (bar 2A), whereas in the
Dicyemidae a calotte consists of two tiers of polar cells,
propolars and metapolars (bar 2B). Thus, the tree indicates
that two clusters initially separate to form two families. In
Microcvema, a calotte and peripheral cells form a syncytium
(bar 3A), but in Conocvenui a calotte is cellular and diapo-
lars are present (bar 3B). In Dicyema japonicum, the calotte

416

H. FURUYA ET AL

is formed in 3B1- and 3B2-cell lineages (bar 4A), but in D.
acuticephalum. D. apalachiensis, and Pseudicyema nakaoi
the calotte is formed only in 3B'-cell lineage (bar 4B). The
orientation of propolars to metapolars separates Pseudicy-
ema from Dicyenui. In Pseudicyema, propolars are ob-
liquely oriented to metapolars (bar 5B). In Dicyema, propo-
lars are located perpendicularly above metapolars (bar 5A).
In D. acuticephalum, cell death occurs both in 3B1- and in
3B:-cell lineages (bar 6 A), but in D. apalachiensis it occurs
only in 3B2-cell lineage (bar 6B). Based on the above
criteria, separation of the dicyemids into two families may
be justified; however, the generic state of Pseudicyema
apparently warrants further study.

Acknowledgments

We wish to express our gratitude to the late Dr. Yutaka
Koshida, Professor Emeritus of Osaka University, for his
continual advice and valuable suggestions on the biology of
dicyemids. This study was supported in part by research
grants from the Nakayama Foundation for Human Science.
Japan Society for the Promotion of Science (no. 12740468).
and the Santa Barbara Museum of Natural History.

Literature Cited

Bogomolov, S. I. 197(1. On the question of the type of cleavage in the
dicyemids. Pp. 22-33 in Questions of Evolutionary Morphology ami
Biocenology, Kazan University Press. Kazan, Russia.

Couch, J. A., and R. B. Short. 1964. Dic\ema bilohum sp. n. (Mesozoa:
Dicyemidae) from the northern Gulf of Mexico. J. Parasitoi 50:
641-645.

Furuya, H. 1999. Fourteen new species of dicyemid mesozoans from six
Japanese cephalopods, with comments on host specificity. Species
Diversity 4: 257-319.

Furuya, H., K. Tsuneki, and Y. Koshida. 1992a. Two new species of
the genus Dicyi'inu (Mesozoa) from octopuses of Japan with notes on
D. misakiense and D. acuticephalum. Zoo/. Sci. 9: 423-437.

Furuya, H., K. Tsuneki, and Y. Koshida. 1992b. Development of the

intusoriform embryo of Dicyema japonicum (Mesozoa: Dicyemidae).

Bint. Bull. 183: 248-257.
Furuya, H., K. Tsuneki. and Y. Koshida. 1993. The development of

the hermaphroditic gonad in four species of dicyemid mesozoans. Zoo/.

Sci. 10: 455-466.
Furuva. H., K. Tsuneki, and Y. Koshida. 1994. The development of

the vermiform embryos of two mesozoans. Dicyema acuticephalum

and Dicyema japonicitm. Zoo/. Sci. 11: 235-246.
Furuya, H., K. Tsuneki, and Y. Koshida. 1996. The cell lineages of

two types of embryo and a hermaphroditic gonad in dicyemid meso-
zoans. Dev. Growth Differ. 38: 453-463.
Gersch. J. 1938. Der Entwicklungszyklus der Dicyemiden. Z. wiss. Zoo/.

151: 515-605.
Hochherg, F. G. 1990. Diseases caused by protistans and metazoans. Pp.

47-202 in Diseases of Marine Animals. Vol. III. O. Kinne, ed. Biolo-

gische Anstalt Helgoland. Hamburg.
Hochberg, F. G., and R. B. Short. 1970. Dicyemennea littlei sp. n. and

Dicyema benthoctopi sp. n.: dicyemid Mesozoa from Benlhoctopus

mcgellanicus. Trans. Am. Microsc. Soc. 89: 216-224.
Katayama, T., H. Wada, H. Furuya, N. Satoh, and M. Yamamoto.

1995. Phylogenetic position of the dicyemid mesozoa inferred from

18S rDNA sequences. Bi,>l. Bull. 189: 81-90.
Kobayashi. M., H. Furuya, and \V. H. Holland. 1999. Dicyemids are

higher animals. Nature 401: 762.
Lameere, A. 1919. Contributions a la connaissance des Dicyemides.

Bull. Biol. Fr Belt;. 53: 234-275.
Lapan, E. A., and H. J. Morowitz. 1975. The dicyemid Mesozoa as an

integrated system for morphogenetic studies. 1. Description, isolation

and maintenance. J. Exp. Zoo/. 193: 147-160.
McConnaughey, B. H. 1938. The dicyemid Mesozoa. J. Entomol. Zoo/.

30: 1-12.
McConnaughey, B. H. 1951. The life cycle of the dicyemid Mesozoa.

Univ. Calif. Publ. Zoo/. 55: 295-336.

Nouvel, H. 1933. Recherches sur la cytologie. la physiologic et la biolo-
gic des Dicyemides. Ann. Inst. Oceanogr. Monaco 13: 165-255.
Nouvel, H. 1947. Les Dicyemides. lre partie: systematique. generations,

vermiformes, infusorigene et sexualite. Arch. Biol. Paris 58: 59-220.
Schartau, O. 1940. Der Entwicklungszyklus von Microcvema vespa van

Beneden (Heterocyemidae). Pubbl. Stn. Zoo]. Ncipo/i 18: 118-128.
Short, R. B. 1962. Two new dicyemid mesozoans from the Gulf of

Mexico. Tulane Stud. Zoo/. 9: 101-111.

Reference: Biol. Bull. 201: 417-423. (December 2001)

Shaping of Colony Elements in Laomedea flexuosa

Hinks (Hydrozoa, Thecaphora) Includes a Temporal

and Spatial Control of Skeleton Hardening

IGOR A. KOSSEVITCH'. KLAUS HERRMANN2*, AND STEFAN BERKING2

^Department of Invertebrate Zoology, Biology Faculty, Moscow State University, Moscow 1 19899.
Russia; and 'Zoologisches Institut, Universitdt zu Koln, Weyertal 119, 50923 Koln. Germany

Abstract. The colonies of thecate hydroids are covered
with a chitinous tubelike outer skeleton, the perisarc. The
perisarc shows a species-specific pattern of annuli, curva-
tures, and smooth parts. This pattern is exclusively formed
at the growing tips at which the soft perisarc material is
expelled by the underlying epithelium. Just behind the apex
of the tip, this material hardens. We treated growing cul-
tures of Laomedea flexuosa with substances we suspected
would interfere with the hardening of the perisarc (L-cys-
teine, phenylthiourea) and those we expected would stimu-
late it (dopamine. A'-acetyldopamine). We found that the
former caused a widening of and the latter a reduction in the
diameter of the perisarc tube. At the same time, the length
of the structure elements changed so that the volume re-
mained almost constant. We propose that normal develop-
ment involves a spatial and temporal regulation of the
hardening process. When the hardening occurs close to the
apex, the diameter of the tube decreases. When it takes
place farther from the apex, the innate tendency of the tip
tissue to expand causes a widening of the skeleton tube. An
oscillation of the position at which hardening takes place
causes the formation of annuli.

Introduction

The fragile, almost beautiful pattern of hydrozoa colonies
attracts every observer's interest. There are many variations
reminiscent of plumes, plants, or minute trees. At closer
examination the pattern of the colonies is fixed by a rigid
outer skeleton, the chinitous perisarc. We are interested in
how this perisarc is shaped.

Received 5 February 2001; accepted 26 July 2001.
* To whom correspondence should be addressed. E-mail: k.herrmannfs'
uni-koeln.de

Inside the perisarc is a hollow tube of soft tissue com-
posed of two cell layers separated by an extracellular ma-
trix, the mesogloea. This matrix is flexible. In general, a
colony comprises two pans: a net of tubes (stolons or
hydrorhiza) generally fixed to a substratum, and shoots
(hydrocauli) emerging vertically from these stolons in a
more or less regular pattern. The shoots bear polyps (hy-
dranths) with which the animals catch their prey. All parts
of the colony are covered with the perisarc. In thecates, the
polyp expands out of the tubelike endings of the perisarc
covering (Fig. 1 ).

The perisarc of the stolons is an almost uniform tube that
is flattened at the site of tight contact to the substratum. The
perisarc of the shoots in Laomedea flexuosa Hinks, used in
this research, forms a repetitive pattern (Fig. 1). One ele-
ment of the shoot — the internode — consists of two se-
quences of annuli separated by a smooth, slightly bent tube
and followed by the finely structured housing (hydrotheca)
of the polyp. The sequence, the number, and the size of the
pattern elements are almost invariant and species specific
(Kosevich, 1990). The exact composition of the perisarc is
unknown, but it appears to contain up to 30% of chitin
(Jeuniaux, 1963: Holl et ai, 1992).

Both the stolon and the shoot tubes increase in length
exclusively at their tips (Kiihn, 1914:Hyman, 1940). Hence,
the pattern of the perisarc emerges exclusively at that site.
Close to the apex of the elongating tube, this material is
rather soft and flexible. Its shape is exactly that of the
underlying tissue. The perisarc material "hardens" some
dozens of micrometers proximal to the apex, and from that
time onward it has a fixed shape. The pattern of the perisarc
is a time recording of the activity of the tissue in the tip.

Our interest is to learn how the perisarc is shaped. One
possibility is that in the course of growth, cell-cell interactions

417

418

I. A. KOSSEVITCH ET AL

hydranlh

shoot tip

distal
annulation zone
proximal

hydrotheca

regressed
hydranth

1 mm

Figure 1. A colony of Laomedea flexuosa illustrating the perisarc and
two hydramhs.

cause a differential curving of the tip surface, and that this
pattern is simply fixed by the perisarc. In this case the time at
which the perisarc hardens has no influence on the shape of the
perisarc. But shaping could also involve a differential pattern
in time and space of perisarc hardening. When the perisarc
hardens closer to the apex, the diameter of the growing tube
should decrease thereafter. But when hardening takes place
more distantly, the diameter can increase due to the tendency
of the tissue in the apex to expand.

To test whether a differential hardening of the perisarc could
play a role in the process of shaping, we treated the colonies
with substances that could be expected to either support or to
antagonize the hardening process. The effects we observed
indicate that both a spatial and temporal pattern of perisarc
hardening is involved in the shaping of the perisarc.

Materials and Methods

Animals

Colonies of Laomedea flexnosa Hincks (Thecaphora,
Campanulariidae) were cultured on glass microscope slides
in artificial seawater (Tropic Marine, 1000 mOsmol, pH
8.2-8.3) in a 5-1 aquarium at 18 °C. The animals were fed
daily with Anemia sulinii nauplii.

Test system

Shoots with newly emerged tips were used as test sys-
tems. Shoots including 4 to 6 distal internodes were isolated
from the colony 2-4 h after feeding. The pieces were used
immediately.

The treatment was performed in 4-ml petri dishes. Nor-
mally, the treatment lasted for 14 to 36 h. Under such
conditions the shoot tip completed the formation of the
internode in about 20 to 24 h. The animals were not fed
during the course of the experiment. However, the tips of
fed and unfed specimens grew with the same speed (Ko-
sevich, 1991). The medium was not changed. The results
were scored at different times, starting 14 h after the begin-
ning of the treatment. Measurements were made by means
of an ocular micrometer. The proximal internodes of the
isolated shoots that had completed their development before
the start of the experiment served as the control, and were
termed untreated.

Chemicals used for treatments

The stock solutions of the following chemicals were
prepared in distilled water: 10 mM/1 dopamine (Sigma). 10
mM/1 W-acetyldopamine (Sigma). 5 mMl phenylthiourea
(Sigma), 0.1% Calcofluor white (Fluorescent Brightener 28
[Sigma]). The following stock solutions were prepared in
seawater: 10 mM/1 L-cysteine (Sigma), adjusted to pH 8.2-
8.3, each time freshly prepared: staining solution for phenol
compounds with fast red salt (Chroma, Stuttgart) according
to Romeis (Clara, after Romeis, 1968), treatment for 3 to 5
min under visual control; 0.001% Congo red (Merck) and
Evans blue (Merck), treatment for 5 to 15 min under visual
control; 4% formaldehyde (Merck), treatment for 24 h.

Statistics

The significance of differences between data obtained
following the various treatments was calculated by means of
the F-test and the one-tailed t test.

Results

Architecture of the stolon tip anil the shoot ti/>

The stolon and the shoot tip differ in size. During forma-
tion of the smooth part of a shoot, the tissue tube in the tip
is about 160-250 ju.ni in diameter and is in tight contact with
the perisarc over a length of 250-350 jam (the perisarc is
translucent). The tissue tube in the stolon tip is 200-300 ^m
in diameter and is in tight contact with the perisarc over a
length of 300-500 /urn. In both cases, adjacent to that region
the tissue tube is much smaller in diameter and has tight
contact with the perisarc at only a few positions.

The mode of perisarc formation

The composition of the perisarc is not well known but
includes chitin and proteins (Jeuniaux, 1963; Chapman,
1973). The proteins of related species were found to contain
a high concentration of disulfide bonds (Chapman, 1937;
Bouillon and Levi, 1971). Phenol compounds are expected

PERISARC FORMATION IN LAOMEDEA

419

Figure 2. Laome dea flexuosa stained with various compounds, (a) A shoot stained with Calcofluor white.
Note that the new tip is strongly stained, but the staining in the previously formed internode occurs in patches,
(b) A stolon tip stained with formaldehyde. Note the decrease in the brightness of the fluorescence in the
proximal direction, (c) Shoot perisarc stained with lm/M/1 ot dopamine. The proximal part of the intemode and
of the hydrotheca are well stained, but the distal annulation zone is almost unstained. The scale bar represents
100 turn.

to play a role in the hardening by causing a crosslinking
between the proteins and the chitin (Knight. 1970).

Figure 2a shows the result of staining with Calcofluor
white, which stains various carbohydrate fibrils, including
amorphous chitin (Compere, 1996). The treatment stains the
perisarc of the tip and in particular the outer surface of all
ectodermal cells in the tip, that is. in the region in which all
ectodermal cells contact the perisarc. Proximal to the tip, the
ectoderm is not in close contact with the perisarc. In this
region, the surface of the ectoderm, staining is observed to
be in the shape of patches. The diameter of such a patch
corresponds to the diameter of one or several ectodermal
cells. The perisarc in the proximal part shows very little
staining. No correlation between the spatial pattern of
stained cells and the perisarc pattern could be detected. It
appears that perisarc material is almost continuously se-
creted by the epithelial cells along the whole shoot, with the
cells in the tip being the most active ones. That correlates
with the rinding that in old parts of the colony the perisarc
is thicker than in younger parts. For example, the thickness
of the perisarc wall in the smooth part of the internodes was
found to change from proximal (the eldest) towards distal
(the youngest) as follows (in /nm): 1 1.95-8.34-7.56-6.10-
4.39-3.17. Note that the distal part is stained but the prox-
imal is not.

After formaldehyde was applied, a fluorescent stain ap-
peared in cells of the ectoderm or at their surface. The
stained cells were more numerous within the tip. but were

also found in smaller numbers along the whole tissue prox-
imal to the tip (Fig. 2b). This result may indicate the
presence of phenol compounds, which are known to play a
role in the hardening process or sclerotization of the
chitin-containing exoskeletons of various animals, includ-
ing cnidarians (Knight, 1968. 1970; Holl et al. 1992).

Although the perisarc looks almost uniform within an
internode, it is not. Treatment with dyes including dopa-
mine. fast red salt, Evans blue, and Congo red revealed a
distinct pattern of staining of the perisarc. The most intense
and spatially different staining was obtained with dopamine
(Fig. 2c). The staining intensity decreases gradually from
the most proximal position to the distal end of the smooth
part. The distal annulated zone is not stained, whereas in the
hydrotheca the staining is intense again. In elder internodes
the pattern is identical, but the staining is deeper. Thus, the
pattern of staining does not correspond simply to the thick-
ness of the perisarc wall. Because of the chemical nature of
the various agents and their binding specificity, we argue
that these substances bind to phenol compounds, which may
have played a role in cross-linking the proteins and the
chitin in the skeleton (cf. Holl et al., 1992).

The influence of L-cvsteine on shoot patterning

L-cysteine is able to interfere with the formation of di-
sulride bonds between and within proteins. Thus, the appli-
cation of L-cysteine may antagonize perisarc hardening if

420

I. A. KOSSEVITCH ET AL.

Figure 3. Alteration of the shoot perisarc shape due to treatment with L-cysteine and phenylthiourea.
Treatment with 1 mAffl of L-cysteine. (a); 2 mMI\ of L-cysteine (b): 0.25 mM/1 of phenylthiourea (c). Note that
the proximal annulation zone is smoothened, the smooth part is crumbled, and the distal annulation zone is
smoothened and widened. Compare the normal pattern elements on the left side of each graph. The scale bar
represents 100 ju.m.

the formation of disulfide bonds is involved in this process.
In addition, L-cysteine impedes the formation of diphenols
(Horowitz et al., 1970). Diphenolic compounds including
dopamine and A'-acetyldopamine were shown to be in-
volved in the sclerotization of the cuticle of insects (Kramer
et ai, 1987; Sugumaran, 1987).

Treatment of shoots with L-cysteine greatly altered the
shape of the perisarc. The perisarc tube widened, crumbled,
and displayed folds at unusual positions (Fig. 3a, b). The
smooth part and the distal annulated zone were especially
affected. Most important, the annuli of the distal part, which
form after the onset of the treatment, were not separated by
the usual deep indentations, but displayed a much smoother
pattern. (Compare as internal control the old pattern ele-
ments that formed before the start of treatment [Fig. 3]). The
effect was observed following application of up to 1-2
mA//l of L-cysteine. Concentrations ten times higher caused
the tissue to disintegrate.

Although the shape of the perisarc was altered to a great
degree, the sequence of the pattern elements — such as the
proximal annulated zone, the smooth part, the distal annu-
lated zone, and the hydrotheca — was laid down as usual. It
appears that even the volume of these elements was not
significantly changed. Thus, the applied concentrations of
L-cysteine did not strongly affect the pattern-forming pro-
cesses in the tissue, but rather adversely affected the normal
perisarc hardening. Due to the L-cysteine treatment, the
perisarc remained soft for a longer period of time, allowing

external and internal mechanical forces to produce the ob-
served malformations.

The influence of phenylthiourea on stolon and shoot
patterning

Phenylthiourea, due to its sulfhydryl moiety, was also
expected to interfere with the hardening of the perisarc. As
was found for L-cysteine, phenylthiourea hinders the forma-
tion of diphenols by interaction with the monophenol mono-
oxygenases (Lerch, 1983). Treatment of shoots by applica-
tion of 0.25-0.5 mM/1 of phenylthiourea resulted in the
formation of bent and crumpled pattern elements. In partic-
ular, the distal annulated zone and the smooth part of the
shoot were affected (Fig. 3c). Following treatment with
L-cysteine, the annuli were not separated by the usual deep
indentations but displayed a much smoother pattern. The
sequence of pattern elements was unchanged.

Dopamine

The diphenol dopamine is an intermediate on the way to
those diphenols that are involved in cross-linking of com-
ponents of the cuticle in insects. In L. flexuosa Knight
(1970) found dopamine and a phenoloxidase. He suggested
that both substances generate quinones that react to cross-
link structural proteins. We found that 0. 1 mM/1 of dopa-
mine reduced the maximal diameter of both the smooth part
and the distal annulated zone. At the same time, the length

PER1SARC FORMATION IN LAOMEDEA

421

Figure 4. Alteration of the shoot pensarc shape due to treatment with dopamine and .V-ucetyldopamine. (a I
Treatment with 0.1 m/W/1 of dopamine (b) Untreated control (c) Treatment with 0.1 mM/1 of iV-acetyldopamine.
Note that the smooth part and the distal annulation zone are increased in length, and the proximal annulation zone
is irregular in shape due to treatment with W-acetyldopamine. Compare Fig. 4b as control. The scale bar
represents 100 /xm.

of these pattern elements increased (Fig. 4a, Fig. 5). In the
annulated zone, the ratio between the maximal outer diam-
eter of the annuli and the diameter of the furrow between
adjacent annuli remained almost unchanged (not shown). In
the proximal annulated zone, the effect was less pro-
nounced. One reason may be the short interval between the
onset of treatment and the formation of the proximal annuli.
Further, the composition of the perisarc may play a role.
The resultant staining of the perisarc was strong in the

proximal annulated zone and almost absent in the distal
annulated zone (Fig. 2c).

N-acetyldopatnine

In insects, A/-acetyldopamine is thought to be an interme-
diate between dopamine and the diphenols used for cross-
linking of the cuticle (Kramer et ni, 1987: Sugumaran.
1987). Knight (1970). however, suggested that the mecha-

0)
10
TO

-2

I I

. Distal
annulation

-4
-6

Smooth part

-8

increase in length (%)
0.1mM/l N-acetyldopamine

-2

-10

Proximal
annulation

Smooth part

Distal
annulation

increase in length (%)
0.1mM/l Dopamine

Figure 5. Treatments of growing tips with dopamine and /V-acetyldopamine change the diameter-to-length
ratio of the structural elements of a shoot internode. A concentration of 0. 1 mMI\ of the compounds was applied.
The graphs show the changes in the size (dimensions) of the intemode parts. The data are given as percent of
increase as compared to the respective control, the previously formed internode. The bars indicate the standard
deviation of the mean, dopamine. two experiments, n = 9. 10. and iV-acetyldopaniniL- n =

422

I. A. KOSSEVITCH ET AL.

nism of sclerotization of the hydroid perisarc differs from
that of insects, because he failed to detect /V-acetyldopamine
and phenolic-/3-glucosides in hydroids. In L. flexuosa, a
concentration of 0. 1 mA//l of /V-acetyldopamine caused the
smooth part and the annuli of the distal annulated zone to
become narrower and longer (Figs. 4a, b; Fig. 5). Further, as
observed for the treatment with dopamine, the ratio between
the maximal outer diameter of the annuli and the diameter
of the furrow between adjacent annuli remained almost
unchanged. Unlike dopamine, /V-acetyldopamine strongly
affected the proximal annulated zone, eliminating its regular
annulation pattern (Fig. 4c). /V-acetyldopamine may act
faster than dopamine.

There is no indication of an unspecitic, cytotoxic action
of the chemicals. One can see in the figures that the older
colony elements are unaffected by the treatment: polyps
stretch out of their hydrotheca, and they are able to catch
their prey. The hydrothecae formed during treatment with
the chemicals are well shaped, and living polyps formed
with tentacles.

Discussion

The delicate species-specific pattern of a thecate colony is
laid down exclusively at the growing tip. At this site, the
tissue has permanent contact with the expelled soft material
from which the outer skeleton, the perisarc, is formed. Some
dozen micrometers proximal to the apex of the tip. the
perisarc hardens, which fixes the pattern of the perisarc.

It is obvious that the soft material is molded by the outer
shape of the underlying tissue. This outer shape is deter-
mined by the property and activity of the cells that built the
tissue tube, particularly those cells that produce the growing
tip. In the tip, the tissue moves back and forth rhythmically.
This phenomenon, termed growth pulsation, has been stud-
ied extensively (Beloussov et ai, 1992).

The staining with Calcofluor white suggests that the
amorphous perisarc material that eventually forms fibrils,
including chitin fibrils, is secreted by almost all the ecto-
dermal epithelial cells of the growing tip, as well as by some
epithelial cells along the body axis. The phenolic com-
pounds, which Knight contended to be involved in the
cross-linking of the perisarc, appear to be contained in
so-called tanning cells (Knight. 1970). These cells are con-
centrated in the tip and also exist in lower density in the
proximal parts. They have no broad contact with the outer
surface of the epithelial sheet of the growing tip and are
embedded between the epithelial cells (Knight. 1970).

Our data suggest that a differential hardening of the
perisarc is involved in the shaping of the perisarc tube. We
treated a growing culture with substances that we expected,
from their chemical nature, to affect the hardening process.
Phenylthiourea and L-cysteine were expected to impede the
hardening; dopamine and /V-acetyldopamine were expected

to support it. The putative "softeners" caused a crumbling
and a widening of the perisarc. Of particular importance is
that the constrictions between the annuli were smoothed out
in the distal annulated zone. The putative "hardeners"
caused the perisarc tube in all internode parts to become
narrower and longer. The applied concentration of the var-
ious chemicals was apparently not toxic to the animals: in
the presence of the chemicals the polyp and the hydrotheca
of the internode formed normally and the polyps behaved
normally — for example, in stretching out to catch their prey.

We suggest that the composition of the soft perisarc
material surrounding the apex changes with time. The na-
ture of the compounds is largely unknown. In insects, low-
molecular-weight catechols such as /V-acetyldopamine and
/V-/3-alanyldopamine are involved in sclerotization. These
are converted to quinones, which react in cross-linking
proteins (for general review see Waite, 1990). Knight
(1970) proposed a different mechanism of action for
Laomedea flexuosa: failing to detect the mentioned sub-
stances, he detected dopamine instead, and suggested that it
was active in sclerotization. Waite (1990) stated that "this
should be taken with caution since the entire animal was
methanol-extracted." In organisms other than insects, dopa-
containing proteins are thought to cause the sclerotization
through a process of "autotanning" (Smyth, 1954; Brown,
1952; Pryor, 1962) in which the dopa moieties are converted
to quinones. Additional molecules — of chitin, collagen, fi-
broin, or cellulose, for example — are necessary as "fillers."
This mode of sclerotization is well-distributed throughout
the animal kingdom, and Waite and coworkers ( 1990) found
dopa-containing proteins in the cnidarian Pachycerianthus
fimbriatus. Our results do not help resolve the question of
which mode of sclerotization acts in L. flexuosa. We know,
however, that the concentration of one or several of the
components changes rhythmically during the growth of the
shoot internode. These rhythms are much slower than those
of the aforementioned growth pulsations. If the hardening
occurs closer to the apex, the diameter of the ring-shaped
border between the hard and the soft perisarc decreases,
forcing the tissue to squeeze through this opening. Under
these conditions, the perisarc tube elongates with a reduced
diameter. A widening of the diameter needs at least two
prerequisites: the hardening has to happen more distally
from the apex than before, and the tissue of the apex must
form a bulb. Evidence for bulb formation may be that the
tissue tube in both the tip of the stolon and the tip of the
shoot has a tight contact to the perisarc, while in proximal
regions the tissue tube is much smaller than the inner lumen
of the perisarc tube. Further, the shoot and the stolon occa-
sionally form a bulb at the wound after cutting (Kossevitch,
unpubl. obs.).

In the process of annulus formation, the zone of harden-
ing may move rhythmically closer to and then farther away
from the apex. This may occur in either a continuous or a

PER1SARC FORMATION IN LAOMEDEA

423

stepwise manner. When a hydrotheca starts to form, the
zone of hardening lags behind in relation to the apex of the
protruding tissue. That causes a widening of the tissue tube
and subsequently of the perisarc tube as well.

In L. ficxHosa, the observed bending of the tube in the
smooth part of the internode (cf. Fig. 1 ) may be the result of
an asymmetry in the hardening of the perisarc along the
circumference of the tip. It may occur closer to the tip apex
at the side that faces the shoot axis, imposing a spatial
control of hardening in addition to the temporal control.

In other animals with an exoskeleton, such as arthropods,
the integument may be shaped by changes in the hardening
process together with changes in the pressure of the tissue
against the forming integument. In arthropods other than
thecate hydrozoa. the hardening can start at various posi-
tions and can spread at different speeds from those posi-
tions. The resultant shape of the integument can thus be
more complex than in hydrozoa.

The various treatments we applied caused the perisarc to
bend, to fold, and to crumble. However, the sequence of the
pattern elements up to hydrotheca formation was as normal
as possible. The volume of the tissue responsible for the
formation of the corresponding element was largely un-
changed. The decrease in the diameter of the perisarc tube
was compensated for by the elongation of the tube. This
indicates that the very tip determines the sequence of pattern
elements. The respective decisions of the tip were not in-
fluenced by ( 1 ) the chemicals applied in the concentrations
noted, (2) the disturbance of the shape and movements of
the tissue in the tip, (3) the shape of the tissue tube in a more
proximal region, nor (4) the altered tension and pressure of
the proximal tissue on the tissue in the very tip. These four
points are in agreement with the observation that the exper-
imentally isolated shoot tip continued the patterning pro-
gram of the perisarc tube up to the formation of the polyp's
housing. The tissue itself was transformed into only the
apical part of the polyp; the proximal part of the perisarc
tube was free of tissue (Kosevich, 1991 ).

Acknowledgments

This work was in part supported by the Deutscher Aka-
demischer Austauschdienst (PKZ A/98/40374) for I.A.K.

Literature Cited

Beloussov, L. V., J. A. Labas, and N. I. Kazakova. 1992. Growth
pulsations in hydroid polyps: kinematics, biological role and cytophysi-
ology. Pp. 183-193 in Oscillations and Morphogenesis. L. Rensing. ed.
Marcel Dekker, New York.

Bouillon. J., and C. Levi. 1971. Structure et ultrastructure des attaches

hydranthes-hydrotheques chez les polypes Thecata. Z. Zellforsch. 121:

218-231.
Brown, C. H. 1952. Some structural proteins of Mylilus edulis. Q. J.

Microsc. Sci. 93: 487-502.
Chapman. G. 1973. A note on the composition of some coelenterate

exoskeletal materials. Camp. Biochern. Physiol. 45B: 279-282.
Chapman. S. S. 1937. Localization of -SH and -S-S- in Obelia genicu-

lara. Growth 1: 299.
Compere, P. 1996. Cytochemical labelling of chitin. Pp. 66-87 in

Chitin in Life Sciences. M. M. Giraud-Guille, ed. Jacques Andre

Publisher, Lyon, France.
Holl, S. M., J. Schaefer, W. M. Goldberg. K. J. Kramer, T. D. Morgan,

and T. L. Hopkins. 1992. Comparison of black coral skeleton and

insect cuticle by a combination of carbon- 13 NMR and chemical

analyses. Arch. Biochem. Biophys. 292: 107-111.
Horowitz, N. H., M. Fling, and G. Horn. 1970. Tyrosinase (Neuro-

sphora crassa). Methods En-ymol. 17A: 615-620.
Hyman, H. 1940. The Invertebrates: Protozoa through Ctenophom.

McGraw-Hill. New York. Pp. 400-497.
Jeuniaux, C. 1963. Chitine et Chitinolyse. Un chapitre de la biologie

moleculaire. P. Masson. Paris.
Knight, D. P. 1968. Cellular basis for quinone tanning of the perisarc in

the thecate hydroid Campanularia (= Obelia) flexuosa Hinks. Nature

218: 584-586.
Knight, D. P. 1970. Sclerotization of the perisarc of the caliptoblastic

hydroid, Laomedea flexuosa. 1. The identification and localization of

dopamine in the hydroid. Tissue Cell 2: 467-477.
Kosevich [Kossevitch], I. A. 1990. Development of stolon's and stem's

internodes in hydroid genera Obelia (Campanulariidae). Vestn. Mask.

Univ. Biol.. N 3: 26-32. [In Russian; English summary.]
Kosevich [Kossevitch]. I. A. 1991. Comparison of upright's and stolon's

tips function in hydroid colony Obelia loveni (Allm.) (Campanulari-
idae). Vestn. Mosk. Univ. Biol., N 2: 44-52. [In Russian; English

summary.]
Kramer, K. J., T. L. Hopkins. J. Schaefer, T. D. Morgan, J. R.

Garbow. G. S. Jacob, E. O. Stejskal, and R. D. Speirs. 1987.

Mechanisms of insect cuticle stabilization. How do tobacco hornworms

do it? Pp. 331-355 in Molecular Entomology, J. H. Law, ed. Alan R.

Liss. New York.
Kiilin. A. 1914. Entwicklungsgeschichte und Verwandtschaftsbeziehun-

gen der Hydrozoen. I. Teil; Die Hydroiden. Ergebnisse Fortschr. Zool

4: 1-284.
Lerch, K. 1983. Neurospora tyrosinase: structural, spectroscopic and

catalytic properties. Mol. Cell Biochem. 52: 125-138.
Pryor, M. G. M. 1962. Sclerotization. Pp. 371-396 in Comparative

Biochemistry. Vol. IV, M. Florkin und H. S. Mason, eds. Academic

Press, New York.
Romeis, B. 1968. Mikroskopische Technik. R. Oldenbourg, Munich.

Germany.
Smyth, J. D. 1954. A technique for the histochemical demonstration of

polyphenoloxidase and its application to eggshell formation in hel-
minths and byssus formation in Mytilus. O. J. Microsc. Sci. 95: 139-
152.
Sugumaran. M. 1987. Quinone methide Sclerotization. Pp. 357-367 in

Molecular Entomology. J. H. Law, ed. Alan R. Liss, New York.
Waite, J. H. 1990. The phylogeny and chemical diversity of quinone-

tanned glues and varnishes. Camp. Biochem. Physiol. 97B: 19-29.

Reference: Biol. Bull. 201: 424-434. (December 21)01)

Detection of Salinity by the Lobster,
Homarus americanus

CHRISTOPHER G. DUFORT, STEVEN H. JURY1, JAMES M. NEWCOMB2,
DANIEL F. O'GRADY III, AND WINSOR H. WATSON III3

Zoologv Department and Center for Marine Biologv, University of New Hampshire,
Durham, New Hampshire 03824

Abstract. Changes in the heart rates of lobsters (Homarus
americanus) were used as an indicator that the animals were
capable of sensing a reduction in the salinity of the ambient
seawater. The typical response to a gradual ( 1 to 2 ppt/min)
reduction in salinity consisted of a rapid increase in heart
rate at a mean threshold of 26.6 ± 0.7 ppt, followed by a
reduction in heart rate when the salinity reached 22.1 ± 0.5
ppt. Animals with lesioned cardioregulatory nerves did not
exhibit a cardiac response to changes in salinity. A cardiac
response was elicited from lobsters exposed to isotonic
chloride-free salines but not to isotonic sodium-, magne-
sium- or calcium-free salines. There was little change in the
blood osmolarity of lobsters when bradycardia occurred,
suggesting that the receptors involved are external. Further-
more, lobsters without antennae, antennules, or legs showed
typical cardiac responses to low salinity, indicating the
receptors are not located in these areas. Lobsters exposed to
reductions in the salinity of the ambient seawater while both
branchial chambers were perfused with full-strength seawa-
ter did not display a cardiac response until the external
salinity reached 21.6 ± 1.8 ppt. In contrast, when their
branchial chambers were exposed to reductions in salinity
while the external salinity was maintained at normal levels,
changes in heart rate were rapidly elicited in response to
very small reductions in salinity (down to 29.5 ± 0.9 ppt in
the branchial chamber and 31.5 ± 0.3 ppt externally). We
conclude that the primary receptors responsible for detect-
ing reductions in salinity in H. americanus are located
within or near the branchial chambers and are primarily
sensitive to chloride ions.

Received 27 April 2000; accepted 21 July 2001.
1 Current address: SUNY-New Paltz, New Paltz. NY 12561.
- Current address: Georgia State University. Atlanta. GA 30303.
3 To whom correspondence should he addressed. E-mail:
whw@cisunix.unh.edu

Introduction

Several studies have provided evidence for osmolarity or
salinity receptors in crustaceans, but the location of such
receptors and the precise ionic stimuli that activate them are
not fully understood. In a study designed to localize the
salinity receptors of the green crab Carcinus inaenas. Hume
and Berlind (1976) selectively exposed different parts of
crabs to seawater with a salinity of 15 ppt. They concluded
that the salinity receptors were located in or near the excur-
rent opening of the branchial chambers. In the crayfish
Procamhams simulans, the branchial chamber also appears
to be the location of receptors that mediate cardiovascular
responses to changes in salinity (Larimer, 1964). Although
the antennae and antennules of lobsters are exquisitely
sensitive to a wide range of chemicals (Tierney et ai, 1988;
Johnsons «/., 1989; see review by Atema and Voigt, 1995),
it is unclear whether they play a role in sensing salinity.
During Hume and Berlind's (1976) investigation of salinity
detection in C. inaenas, removal of the antennae and anten-
nules had no effect. In contrast, Lagerspetz and Mattila
(1961). demonstrated that the antennules and antennae
played an important role in the detection of low salinity in
the isopod Asellns sp. and the amphipod Gammants oce-
anicns. and Tazaki and Tanino (1973) concluded that the
antennae of the spiny lobster Panuliris japonicus have
mechanoreceptors that also function as osmoreceptors.
There is also evidence that the legs of crustaceans have
receptors that provide important information about salinity.
The porcelain crab Porcellana platycheles is able to dis-
criminate between water of different salinities by using its
walking legs (Davenport and Wankowski. 1973). and
Schmidt (1989) recorded electrophysiological responses to
changes in salinity from sensilla on the legs of C. inaenas.
Thus, there is some limited evidence for receptors capable
of sensing salinity changes in crustaceans, but the locations

424

SALINITY DETECTION IN LOBSTERS

425

of these receptors and the transduction mechanisms in-
volved are poorly understood.

Little is known about how marine invertebrates detect
changes in salinity. The bivalves Mytilus editlis and Scro-
bicnlaria pltinti close their shells in response to salinity
reductions, and the receptors controlling these shell closures
are primarily sensitive to Na+, Mg ++, Ca+ + , and possibly
Cl~. rather than to osmolarity (Davenport, 1981; Akberali
and Davenport, 1982). However, there is also evidence for
osmoreceptors in both marine molluscs and crustaceans
(Davenport. 1972; Davenport and Wankowski, 1973;
Tazaki, 1975; Schmidt, 1989). One goal of this study was to
determine whether lobsters detect reductions in salinity by
using a transduction mechanism that is sensitive to changes
in the concentration of certain ions, or one that responds to
alterations in ambient osmolarity.

Hoiminis americaiuis, the American lobster, is an excel-
lent organism in which to investigate responses to changing
salinity, both in terms of the sensory processes involved and
how this behavior is adaptive in certain habitats. Although
the American lobster is generally considered to be stenoha-
line and intolerant to salinities below 25 ppt (Dall, 1970),
adult and juvenile lobsters are known to inhabit salt
marshes, bays, and estuaries that are characterized by fre-
quent fluctuations in salinity (Thomas and White, 1969;
Munro and Therriault, 1983; Able et ai. 1988; Jury et til..
1995;Howellm;/., 1999; Watson et til.. 1999; reviewed by
Charmantier et ai. 2001). For example, lobsters are regu-
larly found in the Great Bay Estuary, New Hampshire,
where the salinity is normally between 22 and 28 ppt in the
summer but routinely drops below 15 ppt each spring
(Loderetai., 1983; Short. 1992; Watson et til.. 1999). After
heavy storms, the salinity can also fall to less than 5 ppt
(Jury et ai. 1995), a value below the lower lethal limit for
adult lobsters, which is from 8 to 14 ppt, depending on
temperature, oxygen and acclimation conditions (McLeese,
1956). Moreover, even if lobsters are able to survive short-
term exposure to low salinity, the resulting physiological
stress may have deleterious long-term effects on growth or
reproduction (Jury et ai, 1994a; Houchens. 1996).

Field studies have shown that lobster movements in es-
tuaries tend to be correlated with seasonal changes in tem-
perature and salinity (Munro and Therriault, 1983; Watson
et til.. 1999), or with storms that cause substantial decreases
in salinity (Jury et til., 1995). Laboratory studies have also
demonstrated that both adult lobsters (Jury et til., 1994b)
and larval lobsters (Scarratt and Raine, 1967) avoid low-
salinity water. For example, when given a choice between
two passageways, one containing water held at a low salin-
ity (10-15 ppt) and one at a higher salinity (20-25 ppt),
93% of the lobsters tested moved through the high-salinity
passageway. Lobsters also moved out of their shelters if the
salinity in those shelters was lowered below 12.5 ppt (Jury
et ul.. 1994b). This avoidance response to low salinity
strongly suggests that lobsters possess the ability to detect

decreases in either osmolarity or the concentrations of spe-
cific ions.

It is well known that many crustaceans will exhibit a drop
in heart rate (bradycardia) or ventilation rate (apnea) in
response to novel stimuli (Maynard, 1960; McMahon,
1999). Therefore, as was pointed out by Florey and Kreibel
( 1974), heart rate "can serve as a most sensitive indicator of
sensory perception and it could well be used in studies on
perceptual physiology." For example. Larimer (1964)
showed that crayfish exhibited changes in heart rate when
exposed to ( 1 ) solutions low in oxygen; (2) different con-
centrations of NaCI and L-glutamic acid; and (3) sudden
changes in temperature. A cardiac assay was also used by
Offutt (1970) to measure the ability of H. americaiuis to
detect sounds of different frequencies, and by Jury and
Watson (2000) to measure the thermosensitivity of H.
americanus.

In the present study, we employed a cardiac assay to
demonstrate that lobsters are able to sense drops in salinity
of greater than 4 ppt. Although removal of the legs, anten-
nae, and antennules had little impact on their responsive-
ness, selective perfusion of the branchial chamber revealed
that this is the most likely location of receptors sensitive to
changes in salinity. Finally, by exposing lobsters to seawa-
ter deficient in certain ions, we determined that lobsters
probably detect changes in salinity by monitoring the con-
centration of chloride rather than by sensing changes in
osmolarity.

Materials and Methods

Adult (>82 mm carapace length), intermolt lobsters of
both sexes were captured in research traps in the Gulf of
Maine, near New Castle, New Hampshire. They were held
in recirculating tanks at a salinity of 32 ppt and a temper-
ature of 1 2 to 14 °C for up to 2 weeks prior to use. Through-
out this paper, "normal" seawater refers to full-strength
(along the NH coast), 32 ppt seawater.

Cardiac assay

Changes in heart and ventilation rates were used as indi-
cators that lobsters sensed an alteration in their environ-
ment. Two pairs of wire electrodes, insulated except at the
tips, were inserted through the carapace of each lobster and
then fastened to the shell with tape and cyanoacrylate glue.
Typically, one pair was implanted through the dorsal cara-
pace on either side of the heart, and the second pair was
inserted through the lateral carapace over the scapho-
gnathites (gill bailers). The electrodes were connected to an
impedance converter (UFI, Morro Bay, CA) that produced
analog signals proportional to the magnitude of the move-
ments of the heart or gill bailer. The impedance converter
output was then digitized using a MacLab analog-to-digital
interface (AD Instruments, Grand Junction, CO), and this
digitized signal was recorded on a Macintosh computer. In

426

C. G. DUFORT ET AL.

some cases, data were also recorded on an AstroMed Dash
4 polygraph (Grass Instruments, Quincy, MA).

Lobsters were placed individually in a 3-1 light-tight
acrylic plastic chamber that was continuously perfused with
cooled (12 to 14 °C) normal seawater taken from a large
holding aquarium. The experimental chamber was con-
nected by tubing to two 1-1 stimulus bottles, one containing
experimental (0 ppt) water, and the other containing control
seawater (32 ppt). Valves were used to control whether the
lobster received an experimental or a control stimulus. Dur-
ing the ion-sensitivity experiments (see subsequent section
of Materials and Methods), the experimental water con-
sisted of solutions that were isotonic to the seawater in the
recording chamber (950-1050 mOsm) but deficient in one
or more specific ions. The temperature of the water in the
stimulus bottles was held constant by placing them in a 25-1
water bath that was maintained at the same temperature as
the chamber holding the lobster.

Lobsters were secured in the chamber with elastic bands
fastened loosely across their dorsal carapace and left over-
night to adjust to this new environment. Previous studies
have indicated that cardiac responses are more pronounced
and are elicited with smaller stimulus intensities when lob-
sters are left in the chamber overnight instead of being
tested shortly after electrode implantation (Offutt, 1970;
Florey and Kriebel, 1974; Jury and Watson, 2000).

All lobsters were first tested to determine whether their
heart rates altered in response to tactile stimulation, 10
betaine (Atema and Voigt, 1995). or shadows (Larimer,
1964). Only lobsters that exhibited a cardiac response to
these stimuli were used in subsequent experiments. A
Quicktime video showing a lobster cardiac response to a
low-salinity stimulus can be viewed at the following web-
site: http://zoology.unh.edu/faculty/win/winvideos.htm.

Snlinir\- detection threshold

In this experiment, 30 lobsters ( 15 male. 15 female) were
tested for their ability to sense changes in salinity. For each
animal, the salinity in the recording chamber was gradually
lowered from a starting value of 32 ppt to less than 20 ppt,
at a rate of 1-2 ppt/min, while heart and scaphognathite
rates were continuously recorded. To monitor salinity, a
piece of tubing was placed in the experimental chamber
near the inhalent opening to the lobster's branchial chamber.
Throughout the experiment water from this area of the
chamber continuously dripped out of this tubing. At 1-min
intervals, the salinity of the water flowing from this tubing
was determined, in parts per thousand, using a refractome-
ter. It took 10 s for water to flow from the chamber to the
end of the tubing, and data were adjusted for this time lag.

During these experiments, under these controlled condi-
tions, the heart rates were very stable, deviating less than
4% from one minute to the next. Thus, a sudden, stimulus-
induced increase or decrease in heart rate was very obvious

4 \ Heart

4 j Scaphognathite

1 35

50 100

Time (s)

Figure 1. Impedance recordings (in volts) of heart and scaphognathite
activity during reductions in salinity for a typical test animal. The salinity
was decreased at a rate of 1-2 ppt/min. (A) The initial response was
typically a rapid increase in heart rate, or tachycardia, which took place in
this experiment at a salinity of 31 ppt and was accompanied by an increase
in ventilation rate, as can be seen in the scaphognathite recording. Initiation
of tachycardia is indicated by the open arrow. (B) As the salinity was
decreased further, to 25 ppt. the lobster responded with a rapid decrease in
heart rate, or bradycardia. Initiation of bradycardia is indicated by the
closed arrow. Bradycardia was usually accompanied by a decrease in
ventilation rate (i.e., apnea).

(see Fig. 1 for example). However, even under the most
stable conditions, occasionally lobsters will spontaneously
skip a heartbeat, or ventilate their gill bailers in the reverse
direction, which is often accompanied by a small change in
heart rate. Therefore, to avoid counting these events as
responses to salinity drops, we set a more obvious and
conservative criterion for designating a change in rate as
either a tachycardia or bradycardia response. This criterion
was either an increase or a decrease of at least 25% from the
baseline heart rate that lasted for more than 10 s. When a
cardiac response occurred, the salinity measured at the
beginning of the response (taking into account the time lag)
was considered to be the salinity detection threshold (SDT)
for that animal. All results are reported as a mean ± SEM.

Ion sensitivity assay

To determine which ions were used to detect differences
in salinity, cardiac responses were measured while exposing
lobsters (n == 37) to artificial saline solutions that were
deficient in one or more specific ions. Most lobsters were

SALINITY DETECTION IN LOBSTERS

427

exposed to at least two different salines, yielding a total of
61 trials. All artificial saline solutions were isotonic with
seawater, so the osmolarity did not change as they were
introduced into the experimental chamber, but the concen-
trations of certain ions did change.

Each artificial saline solution was deficient in one. or a
combination, of the following ions: sodium, chloride, mag-
nesium, and calcium. The solutions tested were: 550 mM
sodium bicarbonate. 530 mM sodium acetate, 530 mM
sodium phosphate. 590 mM choline chloride, and 530 mM
sodium chloride, as well as artificial seawater (423 mM
NaCl. 9 mM KC1. 9.27 mM CaCK. 22.94 mM MgCU 25.50
mM MgSO4, 2.14 mM NaHCO,). sodium-free seawater
(9.40 mM KC1. 9.00 mM CaCU. 22.10 mM MgCK. 25.60
mM MgSO4. 455 mM choline chloride, and 2.10 mM
KHCO,) and chloride-free seawater (25.50 mM MgSO4.
2.14 mM NaHCO,. 422.30 mM NaNO,. 9.69 mM KNO3.
and 9.27 mM Ca(NO3)2). The pH of most solutions was
adjusted to 7.6-7.7 with hydrogen chloride, acetic acid,
sodium hydroxide, or potassium hydroxide, depending on
the ions being tested. A few solutions, such as sodium
acetate and choline chloride, were allowed a larger pH range
(6.7-8. 1 ). because adjusting the pH would alter the concen-
tration of either Na+, K+. or CT ions, as well as the overall
osmolarity. In separate tests, lobsters did not exhibit cardiac
responses when only the pH of natural seawater was
changed over the range 6.2 to 8.1.

Blood osmolarity experiments

These experiments were carried out to determine whether
significant changes in hemolymph osmolarity take place
during the type of salinity reduction protocol used in the
salinity detection studies. Individual lobsters (/; = 8) were
placed in the experimental chamber, and the salinity was
decreased at a rate of 1.5 ppt/min. Before the salinity was
decreased, and every 2 min during the study, the lobster was
quickly removed from the chamber and 0.3 ml of hemo-
lymph was removed from the base of one of the walking
legs using a 1 -ml tuberculin syringe and a 26-gauge needle.
Because all SDTs in previous experiments occurred less
than 16 min after exposure to low salinity, these experi-
ments were conducted for 16 min. Hemolymph samples
were placed in 1-ml eppendorf tubes on ice. Seawater sam-
ples were also taken from the experimental chamber every
2 min and placed in tubes on ice. Control lobsters (n = 8)
were subjected to a similar protocol, except that the salinity
of the seawater was kept constant. The osmolarities of all
the hemolymph and water samples were measured using a
Wescor vapor-pressure osmometer. The heart and ventila-
tion rates were not measured from these lobsters because the
repetitive blood sampling caused dramatic changes in heart
rate that were not related to reductions in salinity.

Cardioregulatory nerve lesions

To determine whether changes in heart activity are me-
diated by the cardioregulatory nerves, three groups of lob-
sters were tested for cardiac responses to salinity changes.
The first group of lobsters (n = 5) had their cardioregula-
tory nerves cut (lesion); the second group (;; = 5) had the
same operation as the first group except that their car-
dioregulatory nerves were left intact (sham): and the third
group (n = 5) was handled, but did not undergo an oper-
ation (control). The baseline heart rates of all lobsters were
recorded for more than 1 h before surgery and again at least
4 days after surgery, to determine if lesioning the car-
dioregulatory nerves had any effect on baseline heart rates.
Once baseline heart rates were recorded, all lobsters were
then tested for a cardiac response to reduced salinity. All
recordings were carried out as described above, after lob-
sters had become accustomed to the recording chamber
overnight.

Lesions were performed as described in Guirguis and
Wilkens (1995). Briefly, a small (3-cm2) rectangular piece
of dorsal carapace just above the heart was removed, and
superficial cuts were made with fine scissors through the
connective tissue along the border of the opening. The shell
was then replaced and secured with tape and cyanoacrylate
glue.

Ablations

First, the SDTs of the experimental lobsters (n -- 15)
were determined by the cardiac assay method. Then, after
chilling the animals for 30 min, their antennae (;; = 5),
antennules (n -- 5). or all walking legs (/; = 5) were
removed. After at least 4 weeks of recovery in a flow-
through tank at the UNH Coastal Marine Laboratory, the
lobsters were tested again to determine the salinity reduc-
tion necessary to elicit bradycardia.

Selective perfusion of the branchial chambers

Since the branchial chamber cannot be isolated by lesion-
ing, a different technique was employed to determine
whether this region is receptive to reductions in salinity.
Both branchial chambers of six lobsters were cannulated
with polyethylene (PE) tubing (1.57 mm I.D.). Four lengths
of PE tubing were inserted into each branchial chamber
through small holes drilled in the carapace near the dorsal
edge of the branchial chambers and glued into place. These
four lengths of tubing were connected to a flow divider,
which in turn was connected to a valve that permitted the
perfusion of each branchial chamber with either normal
seawater or reduced-salinity water. A fifth length of PE
tubing (1.19 mm I.D.) was inserted through the shell pos-
terior to the exhalent area of each branchial chamber to
monitor water salinity. As in all previously described ex-
periments, lobsters were left in the experimental chamber.

428

C. G. DUFORT ET AL.

180 -,

160 -

140 -

^ 120 -

g 100 \

^ 80-

& 60 -

40-

20-

fWi

Tachycardia

^; in «

v i ; I * i

>*.

'% Vu; x

r35

-30
-25
-20

- 15

- 10

- 5

210

410 610

Time (s)

810

0
1010

ex
ex

Figure 2. Changes in heart and scaphognathite rates in response to changes in salinity. As in Figure 1,
salinity was decreased at a rate of 1-2 ppt/min. Each heart and scaphognathite data point is an average of 10 s
of data from a digital ratemeter. while each salinity data point comes from a single refractometer measurement
each min. Initial heart and scaphognathite rates were 35 and 50 bpm. respectively. Tachycardia first occurred at
31 ppt (upward arrow) and was accompanied by an increase in ventilation rate. When the salinity reached a value
of 22 ppt. the lobster responded with bradycardia (downward arrow). The bradycardia response was accompa-
nied by apnea. Shortly after the salinity began to increase, both heart and scaphognathite rates rebounded to
levels well above baseline and then slowly recovered towards baseline over time.

with normal seawater flowing through both the branchial
chambers and the experimental chamber (tank), overnight.
The next day, lobsters were exposed to the following treat-
ments. (1) Normal seawater (32 ppt) was perfused through
the tank while the salinity in the branchial chambers was
gradually reduced. (2) The salinity in the tank was reduced
while the branchial chambers were perfused with normal
seawater. (3) The salinity in the tank was reduced and no
solutions were perfused through the branchial chamber, as
in a typical salinity-reduction experiment. Treatment #1 was
always carried out last; the other two treatments were ran-
domized. Animals were given at least 1 h to recover be-
tween treatments. Water from both branchial chambers and
the experimental chamber dripped into a reserve tank
through PE tubing so that salinity could be sampled each
minute using a refractometer.

Results

Control heart rates and cardiac response controls

After overnight acclimation in the experimental chamber,
the lobsters tested before their salinity detection threshold
(SDT) was measured (n = 32) had a mean heart rate of
52.2 ± 3.3 beats/min (bpm). The heart rates of lobsters
under these control conditions were very consistent, and
thus changes in heart rate in response to drops in salinity
were quite evident and easy to identify. For example, in a
separate experiment (cardioregulatory nerve lesion con-
trols), when we averaged the heart rates for 5 consecutive
min, in 10 different lobsters, the mean standard deviation

was only 1.2 bpm, or a 4% deviation from the average heart
rate (48.4 bpm).

In response to a variety of novel stimuli, 30 of the 32
lobsters tested exhibited a transient bradycardia that was
usually, although not always, accompanied by a reduction in
ventilation rate (apnea). Stimuli which were effective in
eliciting bradycardia included 10~4 M betaine, shadows,
and tactile stimulation of the carapace. The abrupt and
transient reductions in heart rate typically lasted 30 to 120 s,
although on one occasion the heart rate stayed below base-
line for 10 min. None of the lobsters showed any response
to control applications of full-strength ambient seawater,
which was true in all subsequent experiments as well. Only
lobsters that exhibited a cardiac response to novel stimuli
were tested for their response to changes in salinity.

Responses to a reduction in salinity

All the lobsters tested for their response to a reduction in
salinity exhibited a dramatic change in heart rate when the
salinity detection threshold was reached. The typical re-
sponse was an increase, followed by a decrease, in heart rate
(Figs. 1, 2). The initial tachycardia lasted for 178.0 ± 13.0
s (H = 22 because not all lobsters tested responded with an
increase in heart rate) and was often, but not always, ac-
companied by a significant increase in ventilation rate. On
average, heart rate increased significantly from 45.0 ± 3.0
to 66.9 ± 2.9 bpm or 48% (paired t test, P < 0.0001 . n =
22). The bradycardia that occurred next was usually accom-
panied by a transient decrease in scaphognathite pumping

SALINITY DETECTION IN LOBSTERS

429

210

410 610

Time (s)

1010

Figure 3. Heart and scaphognathite rates during a drop in salinity for a lobster with lesioned cardioregu-
latory nerves. The baseline heart and scaphognathite rates were 63 and 120 bpm, respectively. When the salinity
reached 3 1 ppt. the scaphognathite rate dropped suddenly and continued a pattern of intermittent stops and starts
until the salinity increased again. Over the course of the salinity drop, heart rate declined slowly to 53 bpm, a
decrease of 16% from baseline, and then slowly increased back to baseline. Similar results with four additional
lobsters indicate that cardiac responses to changes in salinity are mediated by the cardioregulatory nerves.

(Fig. 2). During bradycardia, the heart rate fell significantly
from 50.3 ± 3.1 to 17.0 ± 1.33 bpm (paired t test, P <
0.0001, n = 30, a 66% decrease in rate) and remained
below baseline for 123.0 ± 8.2 s. Following bradycardia,
heart and ventilation rates usually increased above baseline
for several minutes before full recovery (Fig. 2).

Seventy-three percent (22 of 30) of the lobsters exhibiting
cardiac responses to drops in salinity expressed a biphasic
change in heart rate; 27% expressed a bradycardia response
with no tachycardia. Tachycardia, when present, always
preceded the bradycardia and always occurred before the
salinity reached 25 ppt. Although the bradycardia response
was much more reliable, occurring in every lobster tested, it
did not occur until the salinity had dropped to nearly 20 ppt.
It is possible that the lobsters not exhibiting tachycardia
may already have been in an excited state, because their
average baseline heart rate was 64.6 ± 6. 1 bpm and animals
expressing tachycardia in response to reduced salinity in-
creased their heart rate to 66.9 ± 2.9. In contrast, animals
that did display a tachycardia response had a mean initial
heart rate of 45.0 ± 3.0. Due to the more reliable nature of
the bradycardia response, it was used in the ablation and
ion-sensitivity experiments as an indicator that lobsters
sensed changes in salinity.

Salinity detection threshold

Lobsters first expressed a tachycardia response when the
salinity had fallen to 26.6 ± 0.7 (n = 22), representing a
5.4 ppt drop in salinity relative to ambient levels (32.0 ± 0.2
ppt). The salinity at which the tachycardia response oc-
curred did not differ significantly (unpaired t test, P > 0.5)

from females (26.3 ± 0.7 ppt) to males (26.9 ±1.1 ppt).
The bradycardia response in the 30 animals tested occurred
at 22.1 ± 0.5 ppt, which represents an average drop of 9.9
ppt from the ambient salinity. For the lobsters that showed
both bradycardia and tachycardia responses, the salinity at
which bradycardia occurred was significantly lower than
that at which tachycardia occurred (paired t test. P <
0.001). The SDT for bradycardia was also significantly
higher (unpaired t test, P < 0.05) for females (23.1 ± 0.4
ppt) than for males (21.0 ± 1.0 ppt).

Involvement of cardioregulatory nen'es

Under control conditions, prior to treatment, there was no
significant difference (P = 0.9775. one-way ANOVA)
between the baseline heart rates of control (n = 5, 43.7 ±
5.5 bpm). experimental (n = 5. 45.3 ± 4.0 bpm). and
sham-lesioned lobsters (n = 5, 44.9 ± 6.9 bpm). After
recovery from treatment (4-7 days), the heart rates of both
the sham and lesioned groups were elevated in comparison
to the control group, but this difference was not statistically
significant (P = 0.2994. one-way ANOVA: control
40.8 ± 6.7 bpm; experimental 56.6 ± 7.8 bpm; sham-
lesioned 51.3 ± 6.2 bpm). Lobsters in the control and sham
groups (/; = 10) all exhibited bradycardia in response to a
1 to 2 ppt/min reduction in salinity before the salinity in the
experimental chamber reached 20 ppt. There was no signif-
icant difference (P = 0.2362. unpaired t test) in the mean
SDTs of these two groups of lobsters (30.4 ± 1.7 ppt for
controls and 25.8 ± 3.2 ppt for sham lesions). None of the
lesioned animals (n = 5) exhibited bradycardia in response
to salinity reductions down to 20 ppt (Fig. 3). Two of the 5

430

C. G. DUFORT ET AL

lesioned animals showed a slow decrease in heart rate
during the course of the salinity reduction, but the magni-
tude of these rate decreases did not qualify them as a
bradycardia under our criteria (Fig. 3). Interestingly, all of
the lesioned lobsters exhibited reductions in ventilation
rates during the course of the salinity reduction (Fig. 3). The
salinity at which lesioned lobsters reduced their ventilation
rates was not significantly different (P = 0.8930, one-way
ANOVA), from that of control or sham-lesioned lobsters
(control n = 3 [because scaphognathite records were poor
in 2 of the 5 lobsters], 28.3 ± 2.6 ppt; lesion n = 5, 26.6 ±
2.2 ppt; sham-lesion n = 5, 26.4 ± 3.2 ppt), suggesting that
the salinity response elements in the nervous system had
been activated, but the lobsters were unable to modify their
heart rates due to the lesions.

Ion sensitivity

Most (21 of 24) of the lobsters exposed to isotonic
solutions lacking chloride showed a typical bradycardia
response (Fig. 4). However, cardiac responses were seen in
only 2 of 19 lobsters exposed to isotonic solutions lacking
other ions, but containing chloride (Fig. 4). Statistically, the
occurrence of a bradycardia was significantly dependent on
the lack of chloride (Fig. 4). For example, lobsters did not
exhibit bradycardia when exposed to an isotonic solution of
choline chloride, but they did upon exposure to solutions of

100 -i

n=7

80-

60-

40-

20-

n=24

n=ll

n=10

Seawater

Na-free
Solutions

NaCl

Cl-free

Solutions

Figure 4. Percentage of trials (n = 61) in which lobsters exhibited
bradycardia responses when exposed to natural and artificial solutions
containing various amounts of certain ions. Lobsters did not usually
respond when exposed to solutions containing chloride, such as seawater
(natural [« = 7, 32 ppt] and artificial (» = 4]). sodium-free solutions
(choline chloride [n = 7] and sodium-free seawater [n = 3]), and NaCl
(n = 9). However, lobsters did usually exhibit bradycardia when exposed
to solutions lacking chloride, such as distilled water (» = 7) and chloride-
free solutions (sodium bicarbonate [n = 8], sodium acetate [n = 7),
sodium phosphate [n = 5], and chloride-free seawater [n = 4]). Statis-
tically, the occurrence of bradycardia was significantly dependent on the
lack of chloride (Fisher's exact test, P < 0.0001). Each lobster ( n = 37)
was usually subjected to one to four different solutions, with sufficient time
between solutions for the heart to recover to its baseline rate (>2 h).
Distilled water, when used, was always the last solution tested.

1000-1

•n 900-

r32

800-

a
in

6 8 10 12

Time (min)

14 16 18

Figure 5. Comparison of the osmolarity of ambient seawater and
lobster hemolymph during a typical salinity-reduction experiment. Blood
samples and water samples were taken every 2 min, from eight lobsters,
and averaged (± SEM). Control hemolymph values are also shown for
eight lobsters held at a constant salinity for 16 min. There was no statis-
tically significant difference between the hemolymph osmolarity of control
and experimental animals after 10 min. However, after 16 min there was a
slight, but statistically significant, difference. Equivalent salinity values, in
parts per thousand, are shown on the right-hand vertical axis for compar-
ison.

sodium phosphate and sodium acetate. Lobsters also ex-
pressed bradycardia in chloride-free but not in sodium-free
artificial seawater. Thus, when only chloride was missing
they detected a change, but when some combination of
sodium, calcium, and magnesium was missing they re-
sponded as if the solution was normal seawater. The only
exceptions were two lobsters that responded when exposed
to NaCl solutions. It is not clear why they responded and
seven other lobsters did not. These experiments indicate that
( 1 ) a change in osmolarity is not required for lobsters to
sense a change in salinity; and (2) as long as chloride is
present at normal concentrations, lobsters do not sense
changes in the concentrations of other ions.

Changes in blood osmolarity as ambient salinity is
reduced

In these experiments, the salinity of the seawater in the
experimental chamber was reduced from 31 to 18 ppt over
16 min, and the osmolarity of lobster hemolymph and the
seawater in the chamber were measured every 2 min. The
control study was identical, except the salinity was not
changed. After 10 min there was no statistically significant
change (2-way ANOVA with replication P > 0.10) in the
blood osmolarity of the test animals (n = 8) when com-
pared to the blood osmolarity of control lobsters (/; = 8)
(Fig. 5). For comparison, in the salinity reduction experi-
ments, the external salinity had dropped almost to 20 ppt
after 10 min, which was usually sufficient to elicit a brady-
cardia response. After 16 min there was a small but statis-
tically significant difference (2-way ANOVA with replica-
tion, P < 0.01) in experimental blood osmolarity

SALINITY DETECTION IN LOBSTERS

431

compared to controls (Fig. 5). Thus, although it is possible
that sensitive internal receptors could detect this slight de-
crease in blood osmolarity, the time course and magnitude
of the change — in comparison to the response of the lob-
sters— make it more likely that external salinity receptors
detect the more robust declines that occur in the ambient
seawater.

Ablation experiments

Lobsters with antennules (n = 5), antennae (n = 5). or
walking legs (n = 5) ablated were responsive to declining
salinity both before and after removal of these putative
receptor sites (Fig. 6). There was no statistically significant
difference between the mean SDTs obtained before and
after removal of these structures (paired t test, P > 0.5 in
all three groups; antennae P = 0.94, antennules P = 0.30,
legs P = 0.80).

Branchial perfusion

Both branchial chambers in six lobsters were cannulated
so that the salinity could be differentially controlled in both
the branchial chambers and the experimental chamber. The
day after cannulation, when the lobsters were exposed to
different treatments, the mean heart rate of the lobsters was
51.8 ± 6.3 bpm. The SDTs were then determined in re-
sponse to ( 1 ) a typical drop in external salinity. (2) a drop in
external salinity with the branchial chambers perfused with
normal salinity seawater; and (3) perfusion of the branchial
chambers with low-salinity seawater while exposing the
animal to normal seawater.

When these cannulated lobsters were exposed to a typical
drop in external salinity, with no seawater perfusion of their
branchial chambers, their SDT was 26.7 ± 1.4 ppt. During
these experiments, the salinities in both branchial chambers
and the experimental chamber were recorded each minute.
These data showed that the salinity in the branchial cham-

C- 30

25-|

Pre-lesion

Post-lesion

Antennae

Antennules Walking Legs

Figure 6. Mean salinity detection thresholds (SDTs) for pre- and
post-lesion lobsters. The salinity level at which bradycardia occurred was
measured prior to removal of antennules, antennae, and legs (pre-lesion
SDT), and then compared to values obtained 4 weeks after ablations
(post-lesion SDT). Five lobsters were tested after ablation of each putative
receptor site. There were no statistically significant differences between
any of the means (paired t test. P > 0.5).

Time (s)

200

400 600

Time (s)

800

1000

Figure 7. Cardiac responses of two lobsters to (A) perfusion of the
branchial chambers with low-salinity water while providing the experi-
mental chamber with normal seawater: and ( B ) perfusion of the branchial
chambers with normal seawater while the salinity in the experimental tank
was lowered. In both experiments, the salinity in the branchial chambers
and the experimental tank was recorded every minute. The heart rates
shown were averaged every 10 s. Compare how fast the lobster responded
when low-salinity water was perfused directly into its branchial chambers
with how long it took the other lobster to respond when the salinity in its
branchial chambers was maintained close to 32 ppt, while the salinity in the
experimental tank was lowered.

bers was always 1-2 ppt higher than the changes in salinity
recorded in the ambient seawater. For example, in one
lobster, the salinity in the branchial chambers was 28.0 ppt
when the SDT was 26.7 ppt.

Perfusing the branchial chambers with normal seawater
while dropping the external salinity caused the cardiac
response to occur at an average external salinity of 2 1 .6 ±
1.8 ppt (Fig. 7), which was lower than the SDT obtained
from these same lobsters during the typical experiment
described above. However, this difference in thresholds was
not statistically significant (Mann- Whitney U test, P =
0.06). At the time the lobsters showed a cardiac response,
the salinity in their branchial chambers was still signifi-
cantly higher than the external salinity (Mann-Whitney U
test, P < 0.05) due to the perfusion with normal seawater.
However, it had decreased to a value of 29.0 ± 1.2 ppt due
to dilution with the lower salinity water in the experimental
tank that was being pumped through the branchial chambers

432

C. G. DUFORT ET AL

by the scaphognathites. In contrast, when the salinity in the
branchial chambers was lowered, while the lobster was
being perfused with normal seawater, a cardiac response
was expressed almost immediately, when the branchial
chamber salinity had only dropped to 29.5 ± 0.9 ppt and the
external salinity had only been reduced to 31.5 ± 0.3 ppt.
Interestingly, in all three experiments, there was no statis-
tically significant difference between the salinity values in
the branchial chambers when a cardiac response took place
(repeated measures ANOVA, P > 0.5 (P = .66)). These
data, taken together, suggest that some, if not all, of the
salinity receptors are located in or near the branchial cham-
bers.

Discussion

The ability of American lobsters to detect changes in
salinity was examined by monitoring heart and ventilation
rates while exposing the animals to a gradual reduction in
salinity, either in the ambient seawater or in water directly
flowing into the branchial chambers. The typical response to
a reduction in salinity consisted of tachycardia followed by
bradycardia. In the first set of experiments, tachycardia
occurred when the salinity had decreased from 32 to 26.6 ±
0.7 ppt, whereas bradycardia was not expressed until the
salinity dropped to 22.1 ± 0.5 ppt. During direct perfusion
of the branchial chamber, bradycardia was elicited in re-
sponse to very small drops in salinity (SDT = 29.5 ± 0.9,
measured in the branchial chamber); when the branchial
chambers were perfused with normal seawater while the
external salinity was dropped, lobsters were less responsive
than during control experiments. These data suggest that the
primary salinity receptors mediating the cardiac responses
investigated in this study are located in or very near the
branchial chambers.

In the behavioral avoidance experiments conducted by
Jury et al. ( 1994b), lobsters first became restless and started
to move out of their shelters when the salinity in their
shelters dropped below 18 ppt. These observations, along
with observations of lobsters during cardiac assays and
more recent electrocardiogram recordings obtained from
freely moving lobsters (D. O'Grady, University of New
Hampshire, unpubl. data), indicate that the bradycardia re-
sponse, and not the more sensitive tachycardia response, is
more often correlated with avoidance behaviors. Therefore,
even though lobsters can detect relatively small reductions
in salinity, which may cause them to become aroused and
may increase their heart rate, they may not exhibit avoid-
ance behaviors until the salinity drops to about 18 ppt, a
level well below that necessary to elicit bradycardia. Thus,
as suggested by McGaw and McMahon ( 1996) and Guirguis
and Wilkens (1995), bradycardia is probably a shock or
startle response, indicating that animals sense a potentially
dangerous stimulus and are initiating an avoidance behav-
ior.

We suggest that tachycardia is one of the earliest indica-
tors that lobsters have sensed a change in salinity, and that
this sensory input leads to arousal and a readiness for a
change in behavior. Most of the lobsters in our study that
did not exhibit tachycardia had elevated heart rates before
the stimulus was applied, so they may already have been in
a relatively excited state. In Callinectes sapidus. the blue
crab, drops in salinity trigger a similar tachycardia, and the
available data suggest this increase in heart rate facilitates
certain behaviors associated with low osmolarity (McGaw
and Reiber, 1998). Lobsters induced to walk on a treadmill
exhibit a very rapid increase in heart rate at the onset of
activity, which is comparable to the changes observed in our
experiments (Guirguis and Wilkens, 1995; Rose et al..
1998; O'Grady et al., 2001). This increase is mediated by
the cardioregulatory nerves, and as in our experiments, the
tachycardia is probably an arousal response that helps pre-
pare the lobster for activity.

Although the physiological role of brief changes in heart
and ventilation rates is not obvious, the physiological role of
long-term changes is clear. Increased oxygen uptake and
enhanced circulation of the hemolymph are necessary to
serve the metabolic demands associated with osmoregula-
tion (Jury <•/<//., 1994a; Houchens, 1996), locomotion (Guir-
guis and Wilkens, 1995; Rose et al.. 1998), and higher
temperatures (S. Schreiber, University of New Hampshire,
unpubl. data). Under these circumstances, the initial and
rapid changes in heart rate appear to be mediated by the
cardioregulatory nerves, whereas circulating hormones ap-
pear to be involved in long-term modulation (Guirguis and
Wilkens, 1995; McMahon. 1999; Jury and Watson. 2000:
O'Grady et al.. 2001).

Marine animals may sense drops in ambient salinity by
detecting a change in osmolarity (Davenport. 1972; Daven-
port and Wankowski, 1973; Tazaki. 1975; Schmidt. 1989),
or they may utilize a sensory mechanism that is sensitive to
the concentration of one or more of the ions present in
seawater (Davenport, 1981; Akberali and Davenport, 1982).
One further possibility is that a change in osmolarity could
alter the responsiveness of another type of receptor. For
example, in Callinectes sapidiis, the sensitivity of olfactory
sensilla decreases in low-salinity water because the osmotic
stress causes the outer dendritic segments to change size
(Gleeson et al.. 1996, 1997). However, in the two species of
molluscs that have been studied in the most detail, Mytiln.i
edulis and Scrobiciilaria plana. and in two crustaceans,
Carcinus inaenus (Hume and Berlind, 1976) and Homarus
americaniis (present study), the salinity detection systems
involved are sensitive to the concentration of certain ions
rather than to overall osmolarity. Both molluscs are primar-
ily sensitive to sodium, magnesium, and calcium, and only
slightly responsive to changes in chloride levels (Daven-
port. 1981; Akberali and Davenport, 1982). Hume and Ber-
lind (1976) were unable to determine if any single ion was
detected during salinity reductions in Cardans maenus. In

SALINITY DETECTION IN LOBSTERS

433

contrast, the lobsters in the present study exhibited the
typical low-salinity response when exposed to saline solu-
tions lacking chloride, even though the osmolarity of the
artificial saline was identical to that of seawater. Moreover,
they did not exhibit that response when exposed to solutions
that lacked other ions but contained appropriate concentra-
tions of chloride ions. Thus, although marine crustaceans
may employ any of several mechanisms to detect changes in
salinity, the American lobster appears to detect drops in
salinity by monitoring changes in the concentration of chlo-
ride ions.

This study provides evidence that at least some salinity
receptors in lobsters are located in or near the branchial
chamber. A similar conclusion was reached by Hume and
Berlind (1976) for Carchnis maenas and by Larimer ( 1964)
for crayfish. Previous studies of various crustaceans have
suggested that osmoreceptors might be located on the an-
tennules or antennae (Lagerspetz and Mattila, 1961: Van
Weel and Christofferson, 1966; Tazaki and Tanino, 1973),
and the dactyls (Case el a!., 1960; Davenport. 1972; Dav-
enport and Wankowski. 1973; Schmidt, 1989). However, in
this investigation we did not find any evidence that the
antennules. antennae, or legs were necessary for the detec-
tion of salinity changes in Hoinams americanus.

There may be internal receptors for salinity or osmolarity
in lobsters, but three lines of evidence strongly implicate
external receptors. First, there was no statistically signifi-
cant change in blood osmolarity during the first 10 min of
our experiments, even though changes in heart rate typically
occur within the first 5 min. when the external salinity had
been reduced by 6 to 8 ppt. Second, when the branchial
chamber was perfused with low-salinity water while the rest
of the animal was exposed to normal seawater. bradycardia
occurred very rapidly in response to very small drops in
salinity. Finally, when the branchial chamber was selec-
tively perfused with normal seawater, animals became less
responsive to changes in external salinity. Their eventual
response was typically correlated with a slight decrease in
the branchial chamber salinity, which was difficult to main-
tain at 32 ppt when the external salinity reached low levels.
Thus, while the available evidence suggests that external
salinity receptors probably exist, further studies are clearly
needed to better localize and characterize these sensory
structures.

Lobsters inhabit estuarine and coastal habitats where
storms and spring runoff often produce large drops in sa-
linity that may last for days or weeks (Charmantier et al..
2001). This puts a tremendous demand upon the limited
ability of the animals to osmoregulate, causing a marked
increase in metabolism and, at salinities less than 10 ppt,
extensive mortality (McLeese, 1956; Thomas and White,
1969; Jury et al., 1994a; Houchens. 1996). The avoidance
responses to drops in salinity that lobsters exhibit in the
laboratory probably serve in their natural habitat to move
them to an area that might have a higher salinity (Jury et al.,

1994b. 1995). Although we have used bradycardia as an
assay for detection of salinity and possibly as an index of an
impending avoidance response, the true adaptive signifi-
cance of this response still needs to be resolved. In the field,
bradycardia would probably be triggered when reductions in
salinity are rapid, long-lasting, or of sufficient magnitude to
cause osmoregulatory stress. During the spring runoff sea-
son in the Great Bay estuary, the salinity typically drops at
a rate of 0.2 ppt/min; the rate of decrease is probably even
greater during a storm with heavy rains (see the UNH/
CICEET IDEMS website: www.ciceet.unh.edu). It is likely
that lobsters would detect such a change, and their reaction
would be twofold. First, they would avoid the low-salinity
water and seek deeper water, closer to the coast, that would
have a higher salinity (Jury et til.. 1994b, 1995). Second,
they would increase their metabolism and heart and venti-
lation rates to help fuel the Na+/K+-ATPases necessary to
keep their blood osmolarity higher than the ambient water
(Jury et al., 1994a; Charmantier et ul., 2001 ). The metabolic
demands of these behavioral and physiological adaptations
are likely to be too large to allow both to occur simulta-
neously. Results from recent studies, in which we measured
locomotion, ventilation, and heart rates while exposing lob-
sters to gradual drops in salinity, indicate that when they are
faced with this dilemma, lobsters will eventually stop walk-
ing and give priority to osmoregulation (D. O'Grady, Uni-
versity of New Hampshire, unpubl. data). Field studies are
necessary to further test this hypothesis and clarify how
lobsters regulate their heart and ventilation rates in response
to naturally occurring changes in their environment.

Acknowledgments

We thank the anonymous reviewers whose comments
greatly improved the manuscript. We also thank Hunt How-
ell for his input on, and assistance with, all aspects of this
work, John Sasner for his advice during the early stages of
this study, Mike Kinnison for his preliminary studies in the
Spaulding Lab, Noel Carlson for his help at the Coastal
Marine Laboratory, and Ed Millman for his meticulous
editing. Special thanks to Glenn Crossin for his aid in fine-
tuning the bradycardia assay and Mary Calhoun for pa-
tience, driving, and support. This project was supported by
USDA (Hatch) and NOAA (Sea Grant) grants to WHW and
Hunt Howell. as well as funds from the UNH Marine
Program and Graduate School. It is contribution number
376 of the Center for Marine Biology/Jackson Estuarine
Laboratory series.

Literature Cited

Able, K. W., K. L. Heck, Jr., M. P. Fahay, and C. T. Roman. 1988.

Use of salt-marsh peat reefs by small juvenile lobsters on Cape Cod.
Massachusetts. Estuaries 11: 83-86.

Akberali. H. B., and J. Davenport. 1982. The detection of salinity
changes by the marine bivalve molluscs Scrobicularia plana (da Costa)
and Mytilus ectulis L. J. Exp. Mar. Bio/. Ecol. 58: 59-71.

434

C. G. DUFORT ET AL.

Atema, J., and R. Voigt. 1995. Behavior and sensory biology. Pp.

313-348 in Biolog\ of the Lobster, Homarus americanus J. R. Factor.

ed. Academic Press. New York.
Case, J., G. F. Gwillian, and F. Hanson. 1960. Dactyl chemoreceptors

of brachyurans. Biol. Bull. 119: 308.
Charmantier, G.. C. Haond. J.-H. Lignot, and M. Charmantier-

Daures. 2001. Ecophysiological adaptation to salinity throughout a

life cycle: a review in homarid lobsters. J. Exp. Biol. 204: 967-977.
Dall, W. 1970. Osmoregulation in the lobster Homarus americanus. J.

Fish. Res. Board Can. 27: 1123-1130.
Davenport, J. 1972. Salinity tolerances and preferences in the porcelain

crab, Porcellana platycheles and P. longicomis. Mar. Behav. Physiol.

1: 123-138.
Davenport, J. 1981. The opening response of mussels (Mytilns edulis)

exposed to rising seawater concentrations. J. Mar. Biol. Assoc. UK 61:

667-678.
Davenport, J., and J. Wankowski. 1973. Pre-immersion salinity-choice

behavior in Porcellana platycheles. Mar. Biol. 22: 313-316.
Florey, E., and M. E. Kriebel. 1974. The effects of temperature, anoxia

and sensory stimulation on the heart rate of unrestrained crabs. Camp.

Biochem. Physiol. 48A: 285-300.
Gleeson, R. A., L. M. McDowell, and H. C. Aldrich. 1996. Structure of

the aesthetasc (olfactory) sensilla of the blue crab. Callinectes sapidus:

transformations as a function of salinity. Cell Tissue Res. 284: 279-

288.
Gleeson, R. A., M. G. Wheatly. and C. L. Reiber. 1997. Penreceptor

mechanisms sustaining olfaction at low salinities: insight from the

euryhaline blue crab Callinectes sapidus. J. Exp. Biol. 200: 445-456.
Guirguis, M. S., and J. L. VVilkens. 1995. The role of the cardioregu-

latory nerves in mediating heart rate responses to locomotion, reduced

stroke volume, and neurohormones in Homarus americanus. Biol. Bull.

188: 179-185.
Houchens, C. R. 1996. A comparison of the osmoregulatory capabilities

of estuarine and coastal populations of the American lobster. Homarus

americanus. Master's thesis. University of New Hampshire, 76 pp.
Howell, W. H., W. H. Watson III, and S. H. Jury. 1999. Skewed sex

ratio in an estuarine lobster (Homarus americanus} population. J.

Shellfish Res. 18: 193-201.
Hume, R. I., and A. Berlind. 1976. Heart and scaphognathite rate

changes in a euryhaline crab, Carcinus maenas. exposed to dilute

environmental medium. Biol. Bull. 150: 241-254.
Johnson, B. R., R. Voigt, and J. Atema. 1989. Response properties of

lobster chemoreceptor cells: response modulation by stimulus mix-
tures. Physiol. Zoo/. 62: 559-579.
Jury, S. H., and W. H. Watson III. 2000. Thermosensitivity of the

lobster, Homarus americanus. as determined by cardiac assay. Biol.

Bull. 199: 257-264.
Jury, S. H., M. T. Kinnison, W. H. Howell, and W. H. Watson III.

1994a. The effects of reduced salinity on lobster (Homarus ameri-
canus Milne Edwards) metabolism: implications for estuanne popula-
tions. J. Exp. Mar. Biol. Ecol. 176: 167-185.
Jury, S. H., M. T. Kinnison, W. H. Howell. and W. H. Watson 111.

1994b. The behavior of lobsters in response to reduced salinity. J.

Exp. Mar. Biol. Ecol. 180: 23-37.
Jury, S. H., W. H. Howell. and W. H. Watson HI. 1995. Lobster

movements in response to a hurricane. Mar. Ecol. Prog. Ser. 119:

305-310.
Lagerspetz, K., and M. Mattila. 1961. Salinity reactions of some fresh-

and brackish-water crustaceans. Biol. Bull. 120: 44-53.

Larimer, J. 1964. Sensory-induced modifications of ventilation and

heart rate in crayfish. Cornp. Biochem. Physiol. 12: 25-36.
Loder. T. C., J. A. Love, C. E. Penniman, and C. D. Neefus. 1983.

Long-term environmental trends in nutrient and hydrographic data

from the Great Bay Estuarine System. New Hampshire-Maine. Uni-
versity of New Hampshire Marine Program, UNH-MP-D/TR-SG-

83-6. 65 pp.
Maynard. I). M. 1960. Circulation and heart function. Pp. 161-226 in

The Physiology of Crustacea. Vol. I., T. H. Waterman, ed. Academic

Press. New York.
McGaw, I. J., and B. R. McMahon. 1996. Cardiovascular responses

resulting from variation in external salinity in the dungeness crab,

Ctmcer miigister. Physio/. Zool 69: 1384-1401.
McGaw, I. J., and C. L. Reiber. 1998. Circulatory modification in the

blue crab Callinectes sapidus. during exposure and acclimation to low

salinity. Comp. Biochem. Physiol. 121A: 67-76.
McLeese, D. W. 1956. Effects of temperature, salinity and oxygen on the

survival of the American lobster. J. Fish. Res. Board Can. 13: 247-

272.
McMahon, B. R. 1999. Intrinsic and extrinsic influences on cardiac

rhythms in crustaceans. Comp. Biochem. Physiol. 124A: 539-547.
Munro, J., and J. Therriault. 1983. Migrations saisonnieres du homard

(Homarus americanus) entre la cote et les lagunes des Iles-de-la-

Madeleine. Can. J. Fish. Aauat. Sci. 40: 905-918.
Oft'utt. G. C. 1970. Acoustic stimulus perception by the American

lobster. Homarus americanus (Decapoda). Experientia 26: 1276-1278.
O'Grady, D. F., S. H. Jury, and W. H. Watson 111. 20(11. The use of

a treadmill to study the relationship between locomotion, ventilation

and heart rate in the lobster, Homarus americanus. Mar. Freshw. Res.

(In press).
Rose, R. A., J. L. Wilkens, and R. L. Walker. 1998. The effects of

walking on heart rate, ventilation rate and acid-base status in the lobster

Homarus iiincricunus. J. Exp. Biol. 201: 2601-2608.
Scarratt, D. J., and G. E. Raine. 1967. Avoidance of low salinity by

newly hatched lobster larvae. J. Fish. Res. Board Can. 24: 1403-1406.
Schmidt. M. 1989. The hair-peg organs of the shore crab. Carcinus

maenas (Crustacea. Decapoda): ultrastructure and functional properties

of sensilla sensitive to changes in seawater concentration. Cell Tissue

Res. 257: 609-621.
Short, F. T., ed. 1992. The Ecology of the Great Bay Estuary. New

Hampshire and Maine: An Estuarine Profile and Bibliography. NOAA

Coastal Ocean Program Publ. 222 pages.
Tazaki, K. 1975. Sensory units responsive to osmotic stimuli in the

antennae of the spiny lobster, Panulirus japonicus. Comp. Biochem.

Physiol. 51A: 647-653.
Tazaki, K., and T. Tanino. 1973. Responses of osmotic concentration

changes in the lobster antenna. Experientia 29: 1090-1091.
Thomas, M. L. H., and G. N. White. 1969. Mass mortality of estuarine

fauna at Biddeford P. E. associated with abnormally low salinities. J.

Fish. Res. Hoard Can. 26: 701-704.
Tierney, A. J.. R. Voigt, and J. Atema. 1988. Response properties of

chemoreceptors from the medial antennular filament of the lobster

Homarus americanus. Biol. Bull. 174: 364-372.
Van Weel, P. B. and J. P. Christofferson. 1966. Electrophysiological

studies on perception in the antennules of certain crabs. Physiol. Zool.

39: 317-325.
Watson, W. H. III. A. Vetrovs, and W. H. Howell. 1999. Lobster

movements in an estuary. Mar. Biol. 134: 65-75.

INDEX

Acetylcholine, muscarinic. 276

Actin. 240. 241

Action potentials occur spontaneously in squid giant axons with moder-
ately alkaline intracellular pH. 186

Age structure of the Pleasant Bay population of Crepidula fomicata: a
possible tool for estimating horseshoe crab age. 296

Aggregation. 175

Aggression. 385

Aging, 296

AKESSON, BERTIL. see Thomas G. Dahlgren. 193

Alga. 34. 121. 126.287

Alkaline intracellular pH. 186

AMENT. SETH A., see Nadav Shashar. 267

Ammonium. 288

An initial study on the effects of signal intermittency on the odor plume
tracking behavior of the American lobster. Homarus americanus, 274

Animal-algal symbiosis, 348. 360

Annelida. 193

Anthopleura, 385

Aquarium, acoustics in. 278

Arbacia eggs. 234

Argopecten, 285

ARMSTRONG. PETER B.. see Mara L. Conrad. 246

ARNOLDS, D. E. W.. see S. J. Zottoli, 277

Array, 263

ASAMOAH, N. O.. see S. J. Zottoli. 277

Ascidian. 52

Asexual reproduction. 45

Asexual reproduction in Pygospio elegans Claparede (Annelida.
Polychaeta) in relation to parasitism by Lepocreadium setiferoides
(Miller and Northup) (Platyhelminthes, Trematoda). 45

Asterias. 95. 175

ATEMA. JELLE. see Corinne Kozlowski. 274

Aurelia. 104

Auditory processing. 280

Axonal transport, 240, 243

Axoplasmic vesicle. 243

BABA. SHOJI A., see Yoshihiro Mogami. 26

BALLARIN. LORIANO. ANTONELLA FRANCHINI, ENZO OTTAVIANI, AND AR-

M \NDO SABBADIN, Morula cells as the major immunomodulatory

hemocytes in ascidians: evidences from the colonial species Botryllus

achlosseri, 59
Ballast water. 297

BARLOW, R., see M. Errigo. 271; S. Meadors, 272
Bay scallop. 285

BECHTEL. DEANNA L.. see Alan M. Kuzirian, 297
Behavior. 6. 271.323. 424
Benthic alga. 287

BERKING, STEFAN, see Igor A. Kossevitch, 417
Bimodal units in the torus semicircularis units of the toadfish (Opsaniis

tau). 280

BINDER. MANFRED, see David S. Hibbett. 319
Biogeography. 95. 104
Biogeography of Asterias: North Atlantic climate change and speciation.

Bioluminescence, 339
Biomass. 292

Biomechanics. 126

BISHOP, CORY D.. AND BRUCE P. BRANDHORST. NO/cGMP signaling and

HSP90 activity repress metamorphosis in the sea urchin Lytechinits

pictus. 394

BLACKSTONE. NEIL W.. see Lawrence M. Ponczek. 76
Blastodisc, 251

BOLAND, WILHELM. see Ingo Maier, 1 2 1
BORST. DOUGLAS, see Nadav Shashar, 267
Boinilus, 59

BOYER. BARBARA C.. see Susan D. Hill. 257
BRANDHORST. BRUCE P.. see Cory D. Bishop. 394
BROWN. JEREMIAH R., KYLE R. SIMONETTA, LESLIE A. SANDBERG, PHILLIP

STAFFORD, AND GEORGE M. LANGFORD, Recombinant globular tail

fragment of myosin-V blocks vesicle transport in squid nerve cell

extracts. 240

BRYAN, BRUCE, see Osamu Shimomura. 339
BUCHSBAUM, ROBERT, see Libby Williams, 287
Buoyancy, sea urchin egg. 234
BURBACH, J. PETER H.. ANITA J. C. G. M. HELLEMONS, MARCO HOEKMAN.

PHILIP GRANT, AND HARISH C. PANT. The stellate ganglion of the squid

Lo/igo pealeii is a model for neuronal development: expression of a

POU Class VI homeodomain gene Rpf-1. 252
BURGER, MAX, see William J. Kuhns. 238
Bythograea. 167

CA1, 236

Calcium, 248, 263

Camouflage. 269, 301

CAMPBELL. A. C., S. COPPARD, C. D'ABREO, AND R. TUDOR-THOMAS.
Escape and aggregation responses of three echinoderms to conspecific
stimuli. 175

Capitella, 257

Cardiac. 424

CARLINI, DAVID B.. see Brad A. Seibel. 1

CARMICHAEL, RUTH, see Laurie Fila, 294; Sara P. Grady, 296

CARROLL, IAN. see Hemant M. Chikarmane. 285

/3-catenin, 255

Catfish, 278

Cell

division. 241
lineage. 405

Centrifuge microscope. 234

Centrifuge polarizing microscope with dual specimen chambers and injec-
tion ports. 234

Cephalopod, 1, 136, 154, 186, 240. 251. 267

cGMP signaling. 394

CHANG, FRED, see P. T. Tran. 235

CHAPPELL, RICHARD L.. AND STEPHEN REDENTI. Endogenous zinc as a
neuromodulator in vertebrate retina: evidence from the retinal slice.
265

CHARMANTIER, GUY. see Anne-Sophie Martinez. 167

CHARMANTIER-DAURES. MIREILLE. see Anne-Sophie Martinez. 167

CHEVEZ, C.. see S. J. Zottoli, 277

CHIAO, CHUAN-CHIN. AND ROGER T. HANLON, Cuttlefish cue visually on
area — not shape or aspect ratio — of light objects in the substrate to
produce disruptive body patterns for camouflage. 269

Chionoeceles opilio. 204

CHIKARMANE. HEMANT M., ALAN M. KUZIRIAN, IAN CARROLL. AND ROBBIN
DENGLER. Development of genetically tagged bay scallops for evalu-
ation of seeding programs. 285

435

436

INDEX TO VOLUME 201

Chloride. 424

Chloroplast. 34

Choice chamber. 1 75

Cholinergic modulation of odor-evoked oscillations in the frog olfactory

bulb. 276
Cichlidae. 258
Clam oocyte extract. 241
CLAY, JOHN R.. AND ALAN M. KUZIRIAN, A novel, kinesin-rich preparation

derived from squid giant axons. 243
CLAY, JOHN R., AND ALVIN SHRIER, Action potentials occur spontaneously

in squid giant axons with moderately alkaline intracellular pH, 186
Climate change, 374
Clonal biology, 76
Cnidaria, 104, 385
Coelenterate, 339
Cohort analysis, 296
Collagen. 136, 154
Colonial animal, 76
A comparison of sounds recorded from a catfish (Orinocodoras eigen-

manni. Doradidae) in an aquarium and in the field, 278
Competition. 385
Competition for space among sessile marine invertebrates: changes in

HSP70 expression in two Pacific cnidarians, 385
CONRAD, MARA L.. R. L. PARDY, AND PETER B. ARMSTRONG, Response of

the blood cell of the American horseshoe crab. Limulus polyphemus,

to a lipopolysacchande-like molecule from the green alga Chlorella.

246

Conspecific stimuli, 175
Contractile ring. 241
COPPARD, S.. see A. C. Campbell, 175
Coral

bleaching. 348, 360
reef, 348, 360
Cortical flow. 241
Corynactis, 385
Crab, 17, 167, 204
CRAWFORD. KAREN, Ooplasm segregation in the squid embryo, Loligo

pealeii. 251

Crepidula fomicata. 296
Cuttlefish. 269
Cuttlefish cue visually on area — not shape or aspect ratio — of light objects

in the substrate to produce disruptive body patterns for camouflage.

269

Cytokinesis, 241

Cytological basis of photoresponsive behavior in a sponge larva. 323
Cytoskeleton. 251

D'ABREO. C.. see A. C. Campbell. 175

DAHLGREN, THOMAS G.. BERTIL AKESSON, CHRISTOFFER SCHANDER, KEN-
NETH M. HALANYCH. AND PER SUNDBERG, Molecular phylogeny of the
model annelid Ophryotrocha, 193

Deep sea, 1

DEGNAN, BERNARD M.. see Sally P. Leys, 323

DELANEY, KERRY, see Benjamin Hall. 276

Delayed insemination results in embryo mortality in a brooding ascidian,
52

DENGLER, ROBBIN. see Hemant M. Chikarmane, 285

DEPINA, ANA S.. see Torsten Wollert, 241

Detection of salinity by the lobster. Homarus americanus, 424

Development. 272, 285, 405

Development of genetically tagged bay scallops for evaluation of seeding
programs, 285

Developmental patterns and cell lineages of vermiform embryos in dicy-
emid mesozoans, 405

Dichroism. 231

Dicyemid, 405

Differentiation of pharyngeal muscles on the basis of enzyme activities in
the cichlid Tramitichromis intermedius, 258

Dinoflagellate. 348, 360

Dissolved nitrogen dynamics in groundwater under a coastal Massachu-
setts forest, 288

Dissolved organic nitrogen, 288

Disturbance. 360

DODGE, F., see M. Errigo. 271; S. Meadors. 272

Dorsal cell. 277

Drag. 126

Drag, drafting, and mechanical interactions in canopies of the red alga
Chnndrits crispiis, 126

DUFORT, CHRISTOPHER G., STEVEN H. JURY. JAMES M. NEWCOMB. DANIEL
F. O'GRADY III, AND WiNSOR H. WATSON III, Detection of salinity by
the lobster, Homarus americanus, 424

Dye coupling. 277

Dye coupling evidence for gap junctions between supramedullary/dorsal
neurons of the cunner. Taiitogolabrus adspersus, 277

Echinoderm. 175

Echinus. 175

EDDS-W ALTON. P. L., see R. R. Fay. 280

Eddy chemotaxis. 274

EEG, 218

Effect of cloning rate on fitness-related traits in two marine hydroids, 76

The effects of salt marsh haying on benthic algal biomass. 287

Egg

Arbaciu, 234
longevity, 84
Egg longevity and time-integrated fertilization in a temperate sea urchin

(Strongylocentrotus droebachiensis), 84
Elastic energy storage, 136
Electrical activity, spontaneous. 186
Electrotonic coupling. 277
Elysia, 34
Embryo loss, 52
Endogenous zinc, 265
Endogenous zinc as a neuromodulator in vertebrate retina: evidence from

the retinal slice, 265
Endosymbiosis. 34
Endotoxin, 246
Epi-fluorescence. 235
Epibiont. 296
ERRIGO. M.. C. McGuiNESS, S. MEADORS, B. MITTMANN, F. DODGE, AND R.

BARLOW. Visually guided behavior of juvenile horseshoe crabs. 271
Escape. 175

jet. 154. 252
Escape and aggregation responses of three echinoderms to conspecific

stimuli, 175
Estuarine crab, 17
Estuary. 290. 292
Evidence for directed mitotic cleavage plane reorientations during retinal

development within the zebrafish, 254
Evolution of marine mushrooms. 319
Exocytosis. 246
Extracts, 240
Eye. 272

FAY. R. R.. AND P. L. EDDS-WALTON. Bimodal units in the torus semicir-
cularis units of the toadfish fOpsanus taut, 280

FELDMAN, ROBERT A., see Kenneth M. Halanych, 65

FERNANDEZ-BUSQUETS. XAVIER. see William J. Kuhns, 238

Fertilization. 52. 84. 234

FILA. LAURIE, RUTH HERROLD CARMICHAEL, ANDREA SHRIVER, AND IVAN
VALIELA, Stable N isotopic signatures in bay scallop tissue, feces, and
pseudofeces in Cape Cod estuaries subject to different N loads, 294

Fingerprinting, 285

FLOOD, PER R.. see Osamu Shimomura. 339

Flow. 126

INDEX TO VOLUME 201

437

Fluorescence

microscopy, 23 1

polarization, 231

Fluorescence polarization ratio of GFP crystals. 23 1
Fractionation, 294

FRANCHINI. ANTONELLA. see Loriano Ballarin. 59
Free radicals, 261
FRET. 231

FULLER. S. N., see S. J. Zottoli. 277
Fungi. 319

FURUYA, HIDETAKA. F. G. HocHBERG. AND KAZUHIKO TsuNEKl. Develop-
mental patterns and cell lineages of vermiform embryos in dicyemid
mesozoans. 405

Gap junction. 277

GARZA. J. M.. see R. V. Hernandez. 236

Genetic tag. 285

GERSHWIN, LISA-ANN. Systematics and biogeography of the jellyfish Aure-

liu lubiiita (Cnidaria: Scyphozoa). 104
GFP. 231
Gill current. 6

GODA, MAKOTO, see Shinya Inoue, 23 1 ; Robert A. Knudson. 234
GOULD. ROBERT M.. see Ryan Smith, 255

GRADY. SARA P., DEBORAH RUTECKI, RUTH CARMICHAEL, AND IVAN VA-
LIELA, Age structure of the Pleasant Bay population of Crepidiila
fornicata: a possible tool for estimating horseshoe crab age. 296
GRANT. PHILIP, see J. Peter H. Burbach, 252
Grass

Fellowships. 218

Foundation, 218

Instrument Company, 218
GRAVES. M. E.. see R. V. Hernandez. 236
Gravity. 26
Grazer inclusion. 292
Green fluorescence protein. 231
Groundwater. 288. 290

Growth, visual field, and resolution in the juvenile Limiilus lateral eye, 272
Gulf of Maine, 45

HALANYCH. KENNETH M.. see Thomas G. Dahlgren. 193

HALANYCH. KENNETH M., ROBERT A. FELDMAN. AND ROBERT C. VRIJEN-
HOEK. Molecular evidence that Sclerolinum brattstromi is closely
related to vestimentiferans, not to frenulate pogonophorans (Sibo-
glinidae, Annelida). 65

HALL. BENJAMIN, AND KERRY DELANEY, Cholinergic modulation of odor-
evoked oscillations in the frog olfactory bulb. 276

HANLON, ROGER T., see Nadav Shashar. 267; Chuan-Chin Chiao, 269;
Allen F. Mensinger. 282

HANTEN, JEFFREY J.. AND SIDNEY K. PIERCE, Synthesis of several light-
harvesting complex I polypeptides is blocked by cycloheximide in
symbiotic chloroplasts in the sea slug. Elysia chlorotica (Gould); a
case for horizontal gene transfer between alga and animal?, 34

Hardening. 417

Harvard Medical School. 218

HAI \\VELL, ALYSON M., CHRISTOPHER NEILL, IVAN VALIELA. AND KEVIN D.
KROEGER. Small-scale heterogeneity of nitrogen concentrations in
groundwater at the seepage face of Edgartown Great Pond. 290

Haying, salt marsh. 287

Hearing, 280

Heat shock protein. 374. 385, 394

HELLEMONS. ANITA J. C. G. M.. see J. Peter H. Burbach. 252

HH.MITH, BRIAN S. T.. AND GRETCHEN E. HOFMANN. Microhabitats. ther-
mal heterogeneity, and patterns of physiological stress in the rocky
intertidal zone. 374

Hemocyte, 59

HERBERHOLZ. JENS, AND BARBARA SCHMITZ. Signaling via water currents in
behavioral interactions of snapping shrimp (Alpheus helerochaelis), 6

Hermaphroditism, 193

HERNANDEZ. R. V., J. M. GARZA, M. E. GRAVES. J. L. MARTINEZ. JR., AND
R. G. LEBARON, The process of reducing CA1 long-term potentiation
by the integrin binding peptide, GRGDSP. occurs within the first few
minutes following theta-burst stimulation, 236

HERRMANN, KLAUS, see Igor A. Kossevitch. 417

HERTWECK, CHRISTIAN, see Ingo Maier, 121

HIBBETT. DAVID S., AND MANFRED BINDER, Evolution of marine mush-
rooms, 319

Hidden in plain sight: the ecology and physiology of organismal transpar-
ency, 301

HILI , SUSAN D.. AND BARBARA C. BOYER, Phalloidin labeling of develop-
ing muscle in embryos of the polychaete Capiiella sp. I. 257

HINCKLEY, EVE-LYN S., CHRISTOPHER NEILL, RICHARD MCHORNEY. AND
ANN LEZBERG, Dissolved nitrogen dynamics in groundwater under a
coastal Massachusetts forest. 288

Hippocampus, 236

HIZA. N. A., see S. J. Zottoli. 277

Ho. MICHAEL, see William J. Kuhns. 238

HOCHBERO. R. G., see Hidetaka Furuya, 405

HOEKMAN, MARCO, see J. Peter H. Burbach, 252

HOFMANN, GRETCHEN E.. see Brian S. T. Helmuth. 374

Homeodomain gene, 252

Host response. 45

HSP70, 385

HSP90. role of. 394

Hytiractinia, 76

Hydrogen peroxide. 297

Hydrogen peroxide: an effective treatment for ballast water. 297

Hydroid, 76

Hydromineral regulation in the hydrotherma! vent crab Bythograea thcr-
myilron. 167

Hydrostatic skeleton, 136. 154

Hydrothermal vent. 167

/lyumissa obsoleta, 292

Immunity, invertebrate, 246

Immunology. 59

Independent contrast. 1

INOUE, SHINYA. AND MAKOTO GODA, Fluorescence polarization ratio of

GFP crystals, 23 1

INOUE, SHINYA. see Robert A. Knudson. 234
INOUYE, SATOSHI, see Osamu Shimomura. 339
Integrins, 236, 238
Invasive species. 297
Iridophore, 267

ISHII, JUNKO, see Yoshihiro Mogami, 26
Isolation and properties of the luciferase stored in the ovary of the scy-

phozoan medusa Periphylla periphylla. 339
Isotope. 294

JAMES, PATRICK I., see Alan M. Kuzirian, 297

Jellyfish, 104

JOHNSEN, SONKE, Hidden in plain sight: the ecology and physiology of

organismal transparency. 301
JOHNSON, AMY S.. Drag, drafting, and mechanical interactions in canopies

of the red alga Chondrus crispus, 1 26
JUNG, SuNG-KwoN. see Gilad Twig. 261
JURY, STEVEN H.. see Christopher G. Dufort, 424

KAATZ, INGRID M.. AND PHILLIP S. LOBEL, A comparison of sounds re-
corded from a catfish (Orinocodoras eigenmanni, Doradidae) in an
aquarium and in the field. 278

KAATZ, INGRID M.. see Aaron N. Rice, 258

KALTENBACH. JANE, see William J. Kuhns. 238

438

INDEX TO VOLUME 201

KAVANAGH. EMMA, see Ryan Smith. 255

KELLER, BRUCE, see Robert B. Silver. 263

KIER, WILLIAM M.. see Joseph T. Thompson. 136, 154

Kinesin. 243

KISHORI. B.. see P. Sreenivasula Reddy. 17

KNOWLTON. N.. see W. W. Toller. 348. 360

KNUDSON. ROBERT A.. SHINYA INDUE, AND MAKOTO GODA, Centrifuge

polarizing microscope with dual specimen chambers and injection

ports. 234
KOSSEVITCH. IGOR A., KLAUS HERRMANN, AND STEFAN BERKING, Shaping

of colony elements in Laomedea flexuosa Hinks (Hydrozoa. Theca-

phora) includes a temporal and spatial control of skeleton hardening.

417
KOZLOWSKI, CORINNE, KARA YOPAK, RAINER VOIGT, AND JELLE ATEMA, An

initial study on the effects of signal intermittency on the odor plume

tracking behavior of the American lobster, Homarus americanus, 274
KRIEBEL, MAHLON E., see Robert B. Silver. 263
KROEGER, KEVIN D.. see Alyson M. Hauxwell, 290
KUHNS, WILLIAM J., DARIO RUSCIANO, JANE KALTENBACH, MICHAEL Ho,

MAX BURGER, AND XAVIER FERNANDEZ-BUSQUETS, Up-regulation of

integrins a, /3, in sultate-starved marine sponge cells: functional

correlates, 238

KUZIRIAN. ALAN M.. see John R. Clay, 243; Hemant M. Chikarmane, 285
KUZIRIAN. ALAN M.. ELEANOR C. S. TERRY. DEANNA L. BECHTEL, AND

PATRICK I. JAMES. Hydrogen peroxide: an effective treatment for

ballast water. 297

Laminariales, 121

LANGFORD, GEORGE M.. see Jeremiah R. Brown. 240; Torsten Wollert, 241

Laomedea, 417

Larva, 323

development of, 394
gastrula. 26
Lateral line. 280

LEBARON. R. G.. see R. V. Hernandez, 236
Leukotriene B4. 248
LEVER, MARK, see Melissa Novak, 292
LEYS, SALLY P.. AND BERNARD M. DEGNAN. Cytological basis of photore-

sponsive behavior in a sponge larva, 323
LEZBERG, ANN, see Eve-Lyn S. Hinckley. 288
Limuliis. 246. 271. 272. 296
LINK, BRIAN A.. Evidence for directed mitotic cleavage plane reonentations

during retinal development within the zebrafish. 254
Lipopolysaccharide. 246

LOBEL, PHILLIP S., see Aaron N. Rice. 258; Ingrid M. Kaatz. 278
Lobster. 274, 424
Local field potential. 276
Locomotion. 136. 154
LtBj evokes the calcium signal that initiates nuclear envelope breakdown

through a multi-enzyme network in sand dollar (Echinaracnius

puniiii) celK. 24S

Macrophage, 261

MAIER. INGO, CHRISTIAN HERTWECK. AND WILHELM BOLAND, Stereochem-
ical specificity of lamoxirene. the sperm-releasing pheromone in kelp
(Laminariales, Phaeophyceae). 121

Mangrove. 319

Mariculture. 282. 283

Manculture of the toadtish Opsuniis ran. 282

Marine Biological Laboratory
and the Grass Foundation. 2 1 8
Annual Report, v. 200( 1 ). R 1
General Scientific Meetings. Short Reports, 227

MARTINEZ, ANNE-SOPHIE. JEAN- YVES TOULLEC. BRUCE SHILLITO, MIREILLE
CHARMANTIER-DAURES. AND GUY CHARMANTIER, Hydromineral regu-
lation in the hydrothermal vent crab Bythograea thennydron 167

MARTINEZ, J. L., JR.. see R. V. Hernandez. 236

Mating success, 204

Maximum likelihood. 193

MBL. see Marine Biological Laboratory

McCuRDY. DEAN G.. Asexual reproduction in Pygospio elegans Claparede
(Annelida. Polychaeta) in relation to parasitism by Lepocreadium
serifemides (Miller and Northup) (Platyhelminthes. Trematoda). 45

McGuiNESS. C.. see M. Errigo, 271; S. Meadors, 272

McHoRNEY. RICHARD, see Eve-Lyn S. Hinckley. 288

MEADORS, S., C. MCGUINESS, F. A. DODGE, AND R. BARLOW, Growth,
visual field, and resolution in the juvenile Limuliis lateral eye, 272

MEADORS, S., see M. Errigo. 271

Mechanoreception, 280

MEIDEL, SUSANNE K., AND PHILIP O. YUND, Egg longevity and time-
integrated fertilization in a temperate sea urchin (Strongylocentrotus
droebachiensis), 84

MENSINGER, ALLEN F., see Leila E. Rieder, 283

MENSINGER. ALLEN F., KATHERINE A. STEPHENSON. SARAH L. POLLEMA,
HAZEL E. RICHMOND, NICHOLE PRICE, AND ROGER T. HANLON, Mari-
culture of the toadfish Opsanus tan, 282

Mesozoa. 405

Messenger RNAs located in spiny dogfish oligodendrocyte processes, 255

MESSERLI. MARK A., see Gilad Twig, 261

Metabolism. 1

Metabolism of pelagic cephalopods as a function of habitat depth: a
reanalysis using phylogenetically independent contrasts, 1

Metamorphosis. 394

Methionine-enkephahn induces hyperglycemia through eyestalk hormones
in ihe estuarine crab Scylla serrata, 17

Microhabitats. thermal heterogeneity, and patterns of physiological stress
in the rocky intertidal zone, 374

Microphytobenthos, 292

Microscope
centrifuge, 234
polarizing. 234

Microscopy. 231, 235

Microsporidian dynactin, 245

Microsporidian spore/sporoplasm dynactin in Spraguea, 245

Midbrain. 280

Midwater. 1

Mitosis, 248

MITTMANN. B.. see M. Errigo. 271

Model. 248

MOGAMI, YOSHIHIRO. JuNKO IsHil. AND SHOJi A. SABA. Theoretical and
experimental dissection of gravity-dependent mechanical orientation
in gravitactic microorganisms. 26

Molecular

evolution, 255
phylogeny, 193
systematics. 319

Molecular evidence that Sderolinum brattstromi is closely related to
vestimentiferaiis. not to frenulate pogonophorans (Siboglinidae, An-
nelida). 65

Molecular phylogeny of the model annelid Ophryotrocha, 193

Montastraea annitlaiis, 348. 360

Morphology. 126. 104

MORRISON. HILARY G., see Ryan Smith. 255

Morula cells as the major immunomodulatory hemocytes in ascidians:
evidences from the colonial species Botryllus schlosseri, 59

Mud snail, 292

Muscle

development, 257
physiology. 258

Mutualism. 348. 360

Mycology. 319

Myosin
"ll, 241
V, 240

Mytilus ctilifornianu*. 374

INDEX TO VOLUME 201

439

NEILL, CHRISTOPHER, see Eve-Lyn S. Hinckley. 288: Alyson M. Hauxwell.
290

Neural adaptation. 265

Neuroepithelium, 254

Neuronal development. 252

Neuromodulator, 265

Neuromuscular junction. 263

Neurophy.siology. 218

NEWCOMB. JAMES M.. see Christopher G. Dut'ort, 424

NIERMAN, J. E., see S. J. Zottoli. 277

Nitrate, 288

Nitric oxide, 394

Nitrogen, 288
loading. 290, 294

NO/cGMP signaling and HSP90 activity repress metamorphosis in the sea
urchin Lytechinus pictus, 394

NOBLITT. G. CARL, IV, see Libby Williams. 287

NOVAK, MELISSA, MARK LEVER, AND IVAN VALIELA, Top down vs. bot-
tom-up controls of microphytobenthic standing crop: role of mud
snails and nitrogen supply in the littoral of Waquoit Bay estuaries. 292

A novel, kinesin-rich preparation derived from squid giant axons. 243

Nuclear envelope breakdown. 248

Nutrient. 292

O'GRADY, DANIEL F., III. see Christopher G. Dufort. 424

Odor tracking behavior. 274

Odor-gated rheotaxis. 274

Olfaction. 276

Oligodendrocyte, 255

Ontogenetic changes in fibrous connective tissue organization in the oval

squid. Sepioteuthis lessoniana Lesson, 1830, 136
Ontogenetic changes in mantle kinematics during escape jet locomotion in

the oval squid. Sepioteuthis lessoniaua Lesson. 1830. 154
Ontogeny. 136. 154

Ooplasm segregation in the squid embryo, Loligo pealeii. 251
Ophryotrocha, 193
Opioid peptide. 17
Optics. 301
Orientation, 26

The origins of The Grass Foundation, 218
Oscillation. 276
Osmolarity. 424
Osmoregulation, 167
OTTAVIANI, ENZO, see Loriano Ballarin. 59

PANT. HARISH C., see J. Peter H. Burbach, 252

PAPPAS, GEORGE D.. see Robert B. Silver. 263

Pdnuneciitm, 26

Parasitism. 45

PARDY. R. L.. see Mara L. Conrad, 246

Pelagic. 1

Peptide. opioid. 17

Perisarc. 4 1 7

Phaeophyceae, 121

Phalloidin labeling of developing muscle in embryos of the polychaete

Capitellti sp. I. 257
Pharyngeal jaw. 258
Pherornone. 1 2 1

PHILLIPPI. AIMEE. see J. Stewart Savage, 52
Photoreceptor, 323

feedback, 265
Phototaxis, 323
Phylogenetics, 65
Phylogeny, 1. 193

PIERCE, SIDNEY K.. see Jeffrey J. Hanten. 34
Plankton. 297

Pleasant Bay. 296

Pleistocene. 95

Plum Island Sound. 287

Podocorynci, 76

Pogonophora. 65

Polarization reflecting iridophores in the arms of the squid Loligo pi'tili'ii.
267

Polarization

microscope, 234
vision, 267

POLLEMA, SARAH L.. see Allen F. Mensinger, 282

Polychaete, 193
larva, 257

Polygamy, 204

PONCZEK, LAWRENCE M., AND NEIL W. BLACKSTONE, Effect of cloning rate
on fitness-related traits in two marine hydroids, 76

Population biology, 104

Porifera. 323

Porocytosis. 263

Porocytosis: quantal synaptic secretion of neuro-transmitter at the neuro-
muscular junction through arrayed vesicles, 263

PORTNOY, DAVID S.. see Aaron N. Rice, 258

Potentiation. long-term. 236

PRICE, NICHOLE. see Allen F. Mensinger, 282

The process of reducing CA 1 long-term potentiation by the integrin bind-
ing peptide, GRGDSP. occurs within the first few minutes following
theta-burst stimulation. 236

Proopiomelanocortin, 255

Pstiiumt'chinits, 175

Quantum, 263

Rana pipiens. 276

16S rDNA. 65

18S rDNA. 65

Real-time detection of reactive oxygen intermediates from single micro-

glial cells. 261
Receptor, 424
Recombinant globular tail fragment of myosin-V blocks vesicle transport

in squid nerve cell extracts, 240
Reconstitution of active pseudo-contractile rings and myosin-II-mediated

vesicle transport in extracts of clam oocytes, 241
REDDY, P. SREENIVASLILA. AND B. KISHORI, Methionine-enkephalin induces

hyperglycemia through eyestalk hormones in the estuarine crab Scylla

serrata. 1 7

REDENTI, STEPHEN, see Richard L. Chappell. 265
Reflection. 267
Repopulation of zooxanthellae in the Caribbean corals Montastraea annu-

laris and M. faveolata following experimental and disease-associated

bleaching, 360
Reproduction. 193
asexual. 45
sexual, 121

Reproductive success, 45
Resolution. 272
Response of the blood cell of the American horseshoe crab. Linmlua

polvphemus. to a lipopolysaccharide-like molecule from the green

alga Chlorella. 246
Retina. 272

development of. in zebrafish, 254
slice. 265

Ribosomal RNA genes. 34X
RICE, AARON N., DAVID S. PORTNOY. INGRID M. KAATZ, AND PHILLIP S.

LOBEL, Differentiation of pharyngeal muscles on the basis of enzyme

activities in the cichlid Tramitichromis intermedius, 258
RICHMOND, HAZEL E.. see Allen F. Mensinger, 282

440

INDEX TO VOLUME 201

RIEDER. LEILA E., AND ALLEN F. MENSINGER. Strategies for increasing

growth of juvenile toadfish, 283
Rocky intertidal zone. 374
RONDEAU. AMELIE, AND BERNARD SAINTE-MARIE, Variable mate-guarding

time and sperm allocation by male snow crabs (C/uonoeceles opiliol

in response to sexual competition, and their impact on the mating

success of females. 204
Rossi, SERGI, AND MARK J. SNYDER. Competition for space among sessile

marine invertebrates: changes in HSP70 expression in two Pacific

cnidarians. 385

ROWAN. R.. see W. W. Toller. 348. 360
RUSCIANO. DARIO, see William J. Kuhns, 238
RUTECKI, DEBORAH, see Sara P. Grady, 296

SABBADIN, ARMANDO, see Loriano Ballarin, 59

SAIDEL, WILLIAM M., see Nadav Shashar. 267

SAINTE-MARIE. BERNARD, see Amelie Rondeau. 204

Salinity, 424

Salt marsh, 287

SANDBERG, LESLIE A., see Jeremiah R. Brown. 240; Torsten Wollert, 241

Scallop. 294

SCHANDER. CHRISTOFFER, see Thomas G. Dahlgren, 193

SCHMITZ. BARBARA, see Jens Herberholz. 6

Sclerolinum, 65

Scyphozoa, 104

ScvlUi. 17

Sea slug, 34

Sea urchin, 26. 84, 394

Secretion. 263

Segmentation, 257

SEIBEL, BRAD A.. AND DAVID B. CARLINI, Metabolism of pelagic cephalo-
pods as a function of habitat depth: a reanulysis using phylogeneti-
cally independent contrasts. 1

Self-referencing. 26 1

Sensory processing, 280

Sex ratio. 204

Sexual

competition. 204
reproduction. 121

Shaping of colony elements in Laumeilea flexuosa Hinks (Hydrozoa,
Thecaphora) includes a temporal and spatial control of skeleton hard-
ening. 417

SHASHAR. NADAV. DOUGLAS BORST, SETH A. AMENT. WILLIAM M. SAIDEL.
ROXANNA M. SMOLOWITZ, AND ROGER T. HANLON, Polarization re-
flecting iridophores in the arms of the squid Loligo pealeii. 267

SHILLITO. BRUCE, see Anne-Sophie Marline/.. 167

SHIMOMURA, AKEMI, see Osamu Shimomura, 339

SHIMOMURA, OSAMU, PER R. FLOOD, SATOSHI INOUYE, BRUCE BRYAN. AND
AKEMI SHIMOMURA, Isolation and properties of the luciferase stored in
the ovary of the scyphozoan medusa Pcriphyllu pcriphylla, 339

SHIRIHAI, ORIAN S., see Gilad Twig, 261

SHRIER, ALVIN, see John R. Clay, 186

SHRIVER, ANDREA, see Laurie Fila. 294

Siboglinidae, 65

Signaling, 6

Signaling via water currents in behavioral interactions of snapping shrimp
(Alpheus heterochaelis), 6

SILVER. ROBERT B., MAHLON E. KRIEBEL, BRUCE KELLER. AND GEORGE D.
PAPPAS, Porocytosis: quantal synaptic secretion of neuro-transmitter
at the neuromuscular junction through arrayed vesicles. 263

SILVER, ROBERT, LtB4 evokes the calcium signal that initiates nuclear
envelope breakdown through a multi-enzyme network in sand dollar
(Echinaracnius punna) cells, 248

SIMONETTA. KYLE R.. see Jeremiah R. Brown. 240

Skate, 265

Small-scale heterogeneity of nitrogen concentrations in groundwater at the
seepage face of Edgartown Great Pond. 290

SMITH. PETER J. S., see Gilad Twig. 261

SMITH, RYAN. EMMA KAVANAGH, HILARY G. MORRISON. AND ROBERT M.

GOULD, Messenger RNAs located in spiny dogfish oligodendrocyte

processes, 255

SMOLOWITZ, ROXANNA M.. see Nadav Shashar. 267
Snow crab, 204

SNYDER, MARK J., see Sergi Rossi. 385
Sound, swimbladder
Speciation, 95
Sperm

depletion, 204
economy, 204
limitation, 204
Sponge, 323

Spontaneous electrical activity, 186
Squid. 240, 251, 252. 267

giant axon, 186, 252
Stable N isotope, 294
Stable N isotopic signatures in bay scallop tissue, feces. and pseudofeces in

Cape Cod estuaries subject to different N loads. 294
STAFFORD. PHILLIP, see Jeremiah R. Brown. 240
The stellate ganglion of the squid Loligo pealeii is a model for neuronal

development: expression of a POU Class VI homeodomain gene

Kpf-1. 252

STEPHENSON. KATHERINE A., see Allen F. Mensinger. 282
Stereochemical specificity of lamoxirene, the sperm-releasing pheromone

in kelp (Laminariales, Phaeophyceael. 121
STEWART-SAVAGE, J., AIMEE PHILLIPPI, AND PHILIP O. YUND. Delayed

insemination results in embryo mortality in a brooding ascidian, 52
Strategies for increasing growth of juvenile toadrish, 283
Stress protein. 374, 385. 394
Subtractive hybridization, 255
SUNDBERG, PER, see Thomas G. Dahlgren, 193
Supramedullary neurons, 277
Symbiodinium, 348, 360
Synaptic

plasticity, 236
vesicle, 263
Synthesis of several light-harvesting complex I polypeptides is blocked by

cycloheximide in symbiotic chloroplasts in the sea slug, Elysia chlo-

nuica (Gouldl: a case for horizontal gene transfer between alga and

animal'?. 34
Systematics. 104
Systematics and biogeography of the jellyfish Aurelia lithium (Cnidaria:

Scyphozoa). 104

TABOADA. L. A., see S. J. Zottoli, 277

Teleost, 2X2

Temporal pattern. 52

TERRY. ELEANOR C. S.. see Alan M. Kuzirian, 297

Theoretical and experimental dissection of gravity-dependent mechanical

orientation in gravitactic microorganisms, 26
Thermal stress, 374
THOMPSON, JOSEPH T., AND WILLIAM M. KIER. Ontogenetic changes in

fibrous connective tissue organization in the oval squid. Sepiuteuthis

/ow'w'tmu Lesson. 1830. 136
THOMPSON, JOSEPH T., AND WILLIAM M. KIER, Ontogenetic changes in

mantle kinematics during escape jet locomotion in the oval squid,

Sepioteuthis lessoniana Lesson, 1830. 154
Toadfish. 282. 283
TOLLER. W. W.. R. ROWAN, AND N. KNOWLTON, Repopulation of zooxan-

ihellae in the Caribbean corals Montastraea unnularis and M.faveo-

latu following experimental and disease-associated bleaching. 360
TOLLER. W. W.. R. ROWAN, AND N. KNOWLTON. Zooxanthellae of the

Minituxiracu unnularis species complex: patterns of distribution of

tour taxa of Swihioiliniiini on different reefs and across depths, 348
Top down r.v. bottom-up controls of microphytobenthic standing crop: role

of mud snails and nitrogen supply in the littoral of Waquoit Bay

estuaries. 292
TOULLEC, JEAN- YVES, see Anne-Sophie Martinez, 167

INDEX TO VOLUME 201

441

TK \\. P. T., AND FRED CHANG, Transmitted light fluorescence microscopy

revisited. 235
Trans-fluorescence, 235

Transmitted light fluorescence microscopy revisited, 235
Transparency, 301

TSUNEKI. KAZUHIKO. see Hidetaka Furuya, 405
TUDOR-THOMAS, R., see A. C. Campbell. 175

TWIG. GlLAD, SUNG-KWON JltNG. MARK A. MESSERLI. PETER J. S. SMITH.

AND ORIAN S. SHIRIHAI. Real-time detection of reactive oxygen inter-
mediates from single microglial cells, 261

Ultrastructure, 301
Up-regulation of integrins a,
functional correlates. 238

in sulfate-starved marine sponge cells:

VOIGT, RAINER, see Corinne Kozlowski. 274
VRIJENHOEK. ROBERT C.. see Kenneth M. Halanych. 65

Waquoit Bay, 292. 294

WARES. J.P., Biogeography of Asterias: North Atlantic climate change and

speciation. 95

WATSON. WINSOR H., III. see Christopher G. Dufort. 424
WEIDNER. EARL. Microsporidian spore/sporoplasm dynactin in Spraguea,

245
WILLIAMS. LIBBY, G. CARL NOBLITT IV, AND ROBERT BUCHSBAUM. The

effects of salt marsh haying on benthic algal biomass, 287

WOLLERT, TORSTEN, ANA S. DEPlNA, LESLIE A. SANDBERG, AND GEORGE M.

LANGFORD. Reconstitution of active pseudo-contractile rings and my-
osin-II-mediated vesicle transport in extracts of clam oocytes. 241

VALIELA. IVAN, see Alyson M. Hauxwell. 290: Melissa Novak, 292; Laurie

Fila. 294; Sara P~ Grady. 296

Variable mate-guarding time and sperm allocation by male snow crabs
l Chionoecetes opiliol in response to sexual competition, and their
impact on the mating success of females. 204
Vaucheria, 34
Ventilation. 424
Vermiform embryo. 405
Vesicle transport, 240. 241
Vision. 271, 301
Visual

cue, 269

field. 272

predation, 301
Visually guided behavior of juvenile horseshoe crabs, 271

YOPAK, KARA, see Corinne Kozlowski. 274

YUND. PHILIP O., see J. Stewart Savage. 52; Susanne K. Meidel. 84

Zebrafish retinal development. 254

Zooxanthella, 348, 360

Zooxanthellae of the Montastraea annularis species complex: patterns of
distribution of four taxa of Symbiodinium on different reefs and across
depths, 348

ZOTTOLI. S. J., D. E. W. ARNOLDS, N. O. ASAMOAH. C. CHEVEZ. S. N.
FULLER, N. A. HIZA. J. E. NIERMAN. AND L. A. TABOADA, Dye
coupling evidence for gap junctions between supramedullary/dorsal
neurons of the cunner. Tautogolabrus adspersus, 211

ZOTTOLI, STEVEN J., The origins of The Grass Foundation, 218

Biomimetic Engineering Conference

March 3-8, 2002

Biological organisms exhibit sophisticated crystal engineering
capabilities that underlie the remarkable material properties of
mineralized tissues such as bone and teeth, and the beautiful and
functional nacre of molluscs and abalone. Increasing interest is
being paid to nature's processing strategies, particularly by
materials scientists looking for bio-inspired methods to engineer
unique ceramics coatings or composites for use in magnetic,
optical, biomedical. and protective coatings applications. In
particular, the engineering of hard tissues may benefit from
biomimetic approaches since the benign conditions allow for the
incorporation of biomolecular compounds into the organic/
inorganic composite during fabrication.

Sandestin, Florida

Each day will open with a keynote address to highlight the day's
topics: each of the sessions will include eight to ten presentations.
with afternoons left free for ad hoc meetings and informal
discussions. An evening poster session will also promote dialogue
among the attendees.

Additional information about this Conference — and a registra-
tion form — can be found at

The United Engineering Foundation is located at

Three Park Avenue, 27th Floor, New York. NY 10016-5902;

Tel: 2I2-S9I-7836. Fax: 212-591-7441. E-mail: engfnd@aol.com

THE

BIOLOGICAL
BULLETIN

2002 Subscription Rates
Volumes 202-203

*Paid Subscriptions include both print and electronic subscriptions at: www.biolbull.org

Institutional* Individual*

One vear subscription (6 issues - 2

volumes)

S260.00

$105.00

Single volume (3 issues)

S130.00

$52.50

Single Issues

S 45.00

$20.00

*Surface delivery included in above prices.
For prompt delivery, we encourage subscribers outside the U.S. to request airmail service.

Airmail Delivery Charge

U.S. and Canada: $ 25.00

Mexico: $ 60.00

All other locations: $100.00

Orders Must Be Prepaid in U.S. Dollars, Check Payable to The Marine Biological Laboratory

About The Biological Bulletin

ISSN: 0006-3185

Frequency: Bimonthly

Number of issues per year: 6

Months of Publication: February, April, June, August, October, December

Subscriptions entered for calendar year

Volume indexes contained in June and December issues

Annual report of the Marine Biological Laboratory contained in August issue

Most back issues available

Claims handled upon receipt

No agency discounts

Internet:
www.biolbull.org

Please address orders to:

Wendy Child

Subscriptions

The Biological Bulletin

Marine Biological Laboratory

7 MBL Street

Woods Hole, MA 02543-1015 U.S.A.
Fax: 508-289-7922 Tel: 508-289-7402 Email: wchild@mbl.edu

Published by the Marine Biological Laboratory
Woods Hole, Massachusetts, 02543 U.S.A.

THE BIOLOGICAL BULLETIN

(www.biolbull.org)

2002 SUBSCRIPTION FORM

(VOLUMES 202-203, 6 ISSUES)

All subscriptions run on the calendar year; price includes both print and online journals.

(please print)

NAME:

INSTITUTION:
ADDRESS:

CITY: STATE:

POSTAL CODE: COUNTRY:

TELEPHONE: FAX:

E-MAIL ADDRESS:

O Please send me a 2002 subscription to The Biological Bulletin at the rate indicated below:

a Individual: $105.00 (6 ISSUES) G Institutional: $260.00 (6 ISSUES)

a Individual: $ 52.50 (3 ISSUES) O Institutional: $130.00 (3 ISSUES)

Check one: D February, April, June or D August, October, December

G Please send me the following back issue(s):

a Individual: at $20.00 (PER ISSUE) O Institutional: at $45.00 (PER ISSUE)

Delivery Options

Surface Delivery (Surface delivery is included in the subscription price.)

Air delivery (Please add the correct amount to your payment.)

a U.S. and Canada: $25.00 d Mexico: $60.00 a All other locations: $100.00

Payment Options

Enclosed is my check or U.S. money order for $ made payable to The Marine Biological Laboratory

Please send me an invoice. (Note: Payment must be received before subscription commences.)

Please charge my O VISA, O MasterCard G Discover Card $

Account No.: Exp. Date:

Signature: Date:

Return this form with your check or credit card information to:

Marine Biological Laboratory
Subscription Office » The Biological Bulletin »7 MBL Street » Woods Hole, MA 02543-1015

ARINE IVESOURCES CENTER

MARINE BIOLOGICAL LABORATORY • WOODS HOLE, MA 02543 • (508)289-7700
WWW.MBL.EDU/SERVICES/MRC/INDEX.HTML

Animal and Tissue Supply for
Education & Research

• 150 aquatic species available for shipment via
online catalog: <http://www.mbl.edu/animals/
index.html>; phone: (508)289-7375; or
e-mail: specimens@mbl.edu

• zebrafish colony containing limited mutant strains

• custom dissection and furnishing of specific organ
and tissue samples

zebrafish facilities

MRC Services Available

• basic water quality analysis

• veterinary services (clinical, histopathologic.
microbial services, health certificates, etc.)

• aquatic systems design (mechanical, biological,
engineering, etc.)

• educational tours and collecting trips aboard
the RA/ Gemma

Using the MRC for Your Research

• capability for advanced animal husbandry (temperature, light control, etc.)

• availability of year-round, developmental life stages

• adaptability of tank system design for live marine animal experimentation

"Chemosensory Neurobiology in the
Marine Environment"

a 3 -week summer course at the

Bermuda Biological Station for Research
June 16-July 5,

Dr. Hank Trapido-Rosenthal, Bermuda Biological

Station forResearch
Dr. Charles Derby, Georgia State University

We will study chemosensory neurobiology in the marine environment at the physio-
logical, biochemical, and molecular levels. Lectures will deal with chemoreception in a
variety of marine organisms. In laboratory exercises and research projects the olfactory
system of the spiny lobster, Pamdints argus, will serve as the main teaching and research
tool. Emphasis is on experimental techniques and approaches to the study of chemo-
sensory biology. Receptor cell electrophysiology, immunocytochemistry, BrdU labeling
of cell proliferation, biochemistry of receptor and perireceptor phenomena, and molecular
biology (PCR, sequencing, and other techniques) will be taught and applied to the study
of novel research questions relating to chemical sensing, including basic function and
applications (e.g. environmental biology).

The course is designed to benefit graduate students and advanced undergraduates with
interests in organismal, systems, cellular, or molecular biology.

Competitive scholarships are available to cover tuition, room, and board.

For more information or applications, see
http://www.bbsr.edu/Education/summercourses/summercourses.html

For questions, contact the instructors:

Charles Derby

Dept. of Biology, Georgia State University, P. O. Box 4010
Atlanta, GA 30302-4010

cderby@gsu.edu

(404) 65 1-3058 (office)

http://www.gsu.edu/~biocdd/

Hank Trapido-Rosenthal

Bermuda Biological Station for Research, 17Biological Lane

Ferry Reach, GE01, Bermuda

hank@sargasso.bbsr.edu

(441)297-1880
http://www.bbsr.edu/ About_BBSR/Faculty_Profiles/hank/hank.html/

18]

2002 SUMMER RESEARCH FELLOWSHIPS

FINDING AVAILABLE FOR SIMMER RESEARCH AT
THE MARINE BIOLOGICAL LABORATORY

The Marine Biological Laboratory is pleased lo announce the availability
of funding for Summer Research Fellowships in 2002 for junior or senior
investigators holding a Ph D , M.D., or equivalent degree. These
prestigious awards provide costs for research and housing, and also enable
Fellows to benefit from the rich intellectual and interactive environment
of the scientific community at the MBL

Proposals for Fellowship support will be considered in, but are not limited
to. the following fields of investigation:

Cellular & Molecular Physiology Parasitology

Developmental Biology Microbiology

Neurobiology Molecular Biology

Ecology

In addition, specific Fellowships also provide state-of-the-art
microscopy support

ADDITIONAL INFORMATION IS AVAILABLE ON OUR WEB-SITE!

http://www.mbl.edu/fellowships

APPLICATION DEADLINE FOR
FELLOWSHIPS is JANUARY 15, 2002

FOR APPLICATION FORMS AND ADDITIONAL INFORMATION, PLEASE CONTACT:

Sandra Kaufmann. Fellowship Coordinator
(508)289-7441; skaufinan@mbl.edu

Applications are encouraged from women and
members ofunderrepresenled minorities.

The MBL is an EEO> Affirmative Action Institution

OGL- OCEAN GENOME
LEGACY

I Team Leader

A new not-for-profit foundation devoted to the
preservation, distribution and evolutionary genomics of
DNA from marine organisms is seeking a Team Leader.

The successful candidate will have a Ph.D. in cell or
molecular biology, 4+ years' experience in molecular
biology, and be highly motivated to work at the bench.
Responsibilities will include the assembly of a small
team of scientists for the long-term preservation of
marine organism DNA, development of tissue culture
techniques for marine invertebrate cells, construction of
genome libraries, and evolutionary genomics of special
marine transition groups.

The OGL laboratory will be temporarily located at New
England Biolabs, Inc., Beverly, MA until a new facility is
constructed in Ipswich, MA.

Applicants should send a resume and three references to:
Carol Ann Zapustas, New England Biolabs, Inc.,
32 Tozer Road, Beverly, MA 01915.

OCEAN GENOME LEGACY

An Equal Opportunity Employer.

Marine Biological Laboratory

2002 Course Offerings

Advances In Genome Technology &

Blolnforrnatlcs
October 6 - November 1

Analytical & Quantitative Light Microscopy
May 9- May 17

Biology of Parasitism: Modern Approaches

June 13 - August 10

Embryology: Concepts & Techniques In

Modern Developmental Biology

June 16- July 27

Frontiers In Reproduction: Molecular &
Cellular Concepts & Applications

May 19 - June 29

Fundamental Issues In Vision Research

August 11 - August 24

Medical Informatics

1st Session: May 26 - June 2
2nd Session September 29 - October 6

Methods In Computational Neurosclence

August 4 - September 1

Mlcroblal Diversity

June 16 - August 2

Mlcrolnjectlon Techniques In Cell Biology

May 21 - May 28

Molecular Biology of Aging

July 30 - August 17

Molecular Mycology: Current Approaches
to Fungal Pathogenesls

August 12 - August 30

Substantial financial assistance is
available for many of our courses!

For more information contact:

Carol Hamel,
Admissions Coordinator

(508) 289-7401

admissions@mbl.edu

http://courses.mbl.edu

The MBL is an EEO/Affirmative Action Institution

Neural Development & Genetics

of Zeb rails h
August 18 - August 31

Neural Systems & Behavior

June 16- August 10

Neurobiology
June 16- August 17

Neurolnformatlcs

August 17 - September 1

Optical Microscopy & Imaging In the
Blomedlcal Sciences

October 9 - October 18

Physiology: The Biochemical & Molecular
Basis of Cell Signaling

June 16- July 27

Rapid Electrochemical Measurements In
Biological Systems

May 9 - May 13

Summer Program In Neurosclence, Ethics,
& Survival (SPINES)

June 15 - July 13

Workshop on Molecular Evolution

July 28 - August 9

Marine Biological Laboratory, 7 MBL Street, Woods Hole, MA 02543

Carl Zeiss Microlmaging, Inc.

One Zeiss Drive
Thornwood, NY 10594

1.800.233.2343

micro@zeiss.com

www.zeiss.com/micro

ZEISS

MBI. WHOI LIBRARY

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "The Biological bulletin"

See other formats