THE
BIOLOGICAL BULLETIN
PUBLISHED BY
THE MARINE BIOLOGICAL LABORATORY
Editorial Board
DANIEL L. ALKON, National Institutes of Health and MICHAEL G. O'RAND, Laboratories for Cell Biology,
Marine Biological Laboratory University of North Carolina at Chapel Hill
ROBERT B. BARLOW, JR., Syracuse University
RALPH S. QUATRANO, Oregon State University at
Corvallis
WALLIS H. CLARK, JR., University of California at LlQNEL L REBHUN( Univeisity Of Virginia
Davis
DAVID H. EVANS, University of Florida
HARLYN O. HALVORSON, Brandeis University
RONALD R. HOY, Cornell University
SAMUEL S. KOIDE, The Population Council,
Rockefeller University
FRANK J. LONGO, University of Iowa
JOEL L. ROSENBAUM, Yale University
DOROTHY M. SKINNER, Oak Ridge National
Laboratory
JOHN D. STRANDBERG, Johns Hopkins University
JOHN M. TEAL, Woods Hole Oceanographic
Institution
J. RICHARD WHITTAKER, Boston University
Marine Program and Marine Biological Laboratory
CHARLOTTE P. MANGUM, The College of GEORGE M. WOODWELL, Ecosystems Center, Marine
William and Mary Biological Laboratory
Editor: CHARLES B. METZ, University of Miami
DECEMBER, 1983
Printed and Issued by
LANCASTER PRESS, Inc.
PRINCE & LEMON STS.
LANCASTER, PA.
The BIOLOGICAL BULLETIN is issued six times a year at the
Lancaster Press, Inc., Prince and Lemon Streets, Lancaster, Penn-
sylvania.
Subscriptions and similar matter should be addressed to The
Biological Bulletin, Marine Biological Laboratory, Woods Hole,
Massachusetts. Single numbers, $13.00. Subscription per volume
(three issues), $32.50 ($65.00 per year for six issues).
Communications relative to manuscripts should be sent to Dr.
Charles B. Metz, Marine Biological Laboratory, Woods Hole, Mas-
sachusetts 02543 between May 1 and October 1, and to Dr. Charles
B. Metz, Institute For Molecular and Cellular Evolution, University
of Miami, 521 Anastasia, Coral Gables, Florida 33134 during the
remainder of the year.
THE BIOLOGICAL BULLETIN (ISSN 0006-3185)
Second-class-postage paid at Woods Hole, Mass., and additional mailing offices.
LANCASTER PRESS. INC.. LANCASTER. PA.
11
CONTENTS
No. l, AUGUST 1983
ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY 1
AGUDELO, MARIA I., KENNETH KUSTIN, GUY C. MCLEOD, WILLIAM E.
ROBINSON, AND ROBERT T. WANG
Iron accumulation in tunicate blood cells. I. Distribution and oxidation
state of iron in the blood of Boltenia ovifera, Styela clava, and Molgula
manhattensis 100
ANDERSON, WINSTON A., AND WILLIAM R. ECKBERG
A cytological analysis of fertilization in Chaetopterus pergamentaceus . . 110
BlCKELL, LOUISE R., AND STEPHEN C. KEMPF
Larval and metamorphic morphogenesis in the nudibranch Melibe leonina
(Mollusca: Opisthobranchia) 119
CRONIN, THOMAS W., AND RICHARD B. FORWARD, JR.
Vertical migration rhythms of newly hatched larvae of the estuarine crab,
Rhithropanopeus harrisii 139
FORWARD, RICHARD B., JR., AND KENNETH J. LOHMANN
Control of egg hatching in the crab Rhithropanopeus harrisii (Gould) . . 154
HAND, STEVEN C., AND GEORGE N. SOMERO
Energy metabolism pathways of hydrothermal vent animals: adaptations
to a food-rich and sulnde-rich deep-sea environment 167
HILLER-ADAMS, PAGE, AND JAMES J. CHILDRESS
Effects of feeding, feeding history, and food deprivation on respiration and
excretion rates of the bathypelagic mysid Gnathophausia ingens 182
INCZE, LEWIS S., AND A. J. PAUL
Grazing and predation as related to energy needs of stage I zoeae of the
tanner crab Chionoecetes bairdi (Brachyura, Majidae) 197
MACKIE, G. O., AND C. L. SINGLA
Coordination of compound ascidians by epithelial conduction in the co-
lonial blood vessels 209
OLSON, RICHARD RANDOLPH
Ascidian-Prochloron symbiosis: the role of larval photoadaptations in mid-
day larval release and settlement 221
READ, LAURIE K., LYNN MARGULIS, JOHN STOLZ, ROBERT OBAR, AND
THOMAS K. SAWYER
A new strain of Paratetramitus jugosus from Laguna Figueroa, Baja Cal-
ifornia, Mexico 241
REED-MILLER, CHARLENE
The initial calcification process in shell-regenerating Tegula (Archaeogas-
tropoda) 265
RUTOWSKI, RONALD L.
Mating and egg mass production in the aeolid nudibranch Hermissenda
crassicornis (Gastropoda: Opisthobranchia) 276
SEBENS, KENNETH P.
Settlement and metamorphosis of a temperate soft-coral larva (Alcyonium
siderium Verrill): induction by crustose algae 286
SLOBODKIN, L. B., AND KENNETH DUNN
On the evolutionary constraint surface of hydra 305
iii
CONTENTS
SOUMOFF, CYNTHIA, AND DOROTHY M. SKINNER
Ecdysteroid titers during the molt cycle of the blue crab resemble those
of other Crustacea 32 1
WETHEY, DAVID S.
Geographic limits and local zonation: the barnacles Semibalanus (Balanus)
and Chthamalus in New England 330
No. 2, OCTOBER 1983
AYLING, AVRIL L.
Growth and regeneration rates in thinly encrusting demospongiae from
temperate waters 343
BENAYAHU, Y., AND Y. LOYA
Surface brooding in the Red Sea soft coral Parerythropodium fulvum fulvum
(Forskal, 1775) 353
DUNCAN, THOMAS K.
Sexual dimorphism and reproductive behavior in Almyracuma proximoculi
(Crustacea: Cumacea): the effect of habitat 370
ECKELBARGER, KEVIN J., AND JUDITH P. GRASSLE
Ultrastructural differences in the eggs and ovarian follicle cells of Capitella
(Polychaeta) sibling species 379
EYSTER, LINDA S.
Ultrastructure of early embryonic shell formation in the opisthobranch
gastropod Aeolidia papillosa 394
FREEMAN, JOHN A., TERRY L. WEST, AND JOHN D. COSTLOW
Postlarval growth in juvenile Rhithropanopeus harrisii 409
KAPLAN, SAUL W.
Intrasexual aggression in Metridium senile 416
MILLER, RICHARD L., AND KENNETH R. KING
Sperm chemotaxis in Oikopleura dioica Fol, 1872 (Urochordata: Larvacea)
419
RAMOS-FLORES, TALIA
Lower marine fungus associated with black line disease in star corals (Mon-
tastrea annularis, E. & S.) 429
SUGITA, HlROAKI, AND KOICHI SEKIGUCHI
The developmental appearance of paternal forms of lactate dehydrogenase
and malate dehydrogenase in hybrid horseshoe crabs 436
TSUJI, FREDERICK I., AND ELIZABETH HILL
Repetitive cycles of bioluminescence and spawning in the polychaete,
Odontosyllis phosphorea 444
VITTURI, R., A. MAIORCA, AND E. CATALANO
The karyology of Teredo utriculus (Gmelin) (Mollusca, Pelecypoda) . . . 450
WEDI, STEVEN E., AND DAPHNE FAUTIN DUNN
Gametogenesis and reproductive periodicity of the subtidal sea anemone
Urticina lofotensis (Coelenterata: Actiniaria) in California 458
Yui, MARY A., AND CHRISTOPHER J. BAYNE
Echinoderm immunology: bacterial clearance by the sea urchin Strongy-
locentrotus purpuratus 473
ABSTRACTS OF PAPERS PRESENTED AT THE GENERAL SCIENTIFIC MEETINGS
OF THE MARINE BIOLOGICAL LABORATORY
Cellular, molecular biology, etc 487
Developmental biology 499
Ecology, evolution, plant sciences 504
Gametes and fertilization 512
IV
CONTENTS
Microbiology 520
Neurobiology, learning, behavior 523
Parasitology, pathology, aging 534
Photoreceptors, vision, rhythms 539
No. 3, DECEMBER 1983
BRENCHLEY, G. A., AND J. T. CARLTON
Competitive displacement of native mud snails by introduced periwinkles
in the New England intertidal zone 543
BRETOS, MARTA, ITALO TESORIERI, AND Luis ALVAREZ
The biology of Fissurella maxima Sowerby (Mollusca: Archaeogastropoda)
in northern Chile. 2. Notes on its reproduction 559
CHORNESKY, ELIZABETH A.
Induced development of sweeper tentacles on the reef coral Agaricia agar-
icites: a response to direct competition 569
DEFUR, PETER L., BRIAN R. MCMAHON, AND CHARLES E. BOOTH
Analysis of hemolymph oxygen levels and acid-base status during emersion
'in situ' in the red rock crab, Cancer productus 582
FREEMAN, GARY
Experimental studies on embryogenesis in hydrozoans (Trachylina and
Siphonophora) with direct development 591
GLADFELTER, ELIZABETH H.
Circulation of fluids in the gastro vascular system of the reef coral Acropora
cenicornis 619
HANLON, ROGER T., RAYMOND F. HIXON, AND WILLIAM H. HULET
Survival, growth, and behavior of the loliginid squids Loligo plei, Loligo
pealei, and Lolliguncula brevis (Mollusca: Cephalopoda) in closed sea water
systems 637
LEVINTON, JEFFREY S.
The latitudinal compensation hypothesis: growth data and a model of
latitudinal growth differentiation based upon energy budgets. I. Interspecific
comparison of Ophryotrocha (Polychaeta: Dorvilleidae) 686
LEVINTON, JEFFREY S., AND ROSEMARY K. MONAHAN
The latitudinal compensation hypothesis: growth data and a model of
latitudinal growth differentiation based upon energy budgets. II. Intraspecific
comparisons between subspecies of Ophryotrocha puerilis (Polychaeta:
Dorvilleidae) 699
NlCCHITTA, C. V., AND W. R. ELLINGTON
Energy metabolism during air exposure and recovery in the high intertidal
bivalve mollusc Geukensia dernissa granosissima and the subtidal bivalve
mollusc Modiolus squamosus 708
REED-MILLER, CHARLENE
Scanning electron microscopy of the regenerated shell of the marine ar-
chaeogastropod, Tegula 723
SCOFIELD, VIRGINIA L., AND LAUREN S. NAGASHIMA
Morphology and genetics of rejection reactions between oozooids from
the tunicate Botryllus schlosseri 733
TELFORD, MALCOLM, ANTHONY S. HAROLD, AND RICH Mooi
Feeding structures, behavior, and microhabitat of Echinocyamus pusillus
(Echinoidea: Clypeasteroida) 745
VACCA, LINDA L., AND MILTON FINGERMAN
The roles of hemocytes in tanning during the molting cycle: a histochemical
study of the fiddler crab, Uca pugilator 758
CONTENTS
WAHLE, CHARLES M.
Regeneration of injuries among Jamaican gorgonians: the roles of colony
physiology and environment 778
WIDDER, EDITH A., MICHAEL I. LATZ, AND JAMES F. CASE
Marine bioluminescence spectra measured with an optical multichannel
detection system 791
Short Report
GLADFELTER, ELIZABETH H.
Spatial and temporal patterns of mitosis in the cells of the axial polyp of
the reef coral Acropora cervicornis 811
INDEX TO VOLUME 165 816
VI
Volume 165 Number 1
u
THE
BIOLOGICAL BULLETIN
PUBLISHED BY
THE MARINE BIOLOGICAL LABORATORY
Editorial Board
DANIEL L. ALKON, National Institutes of Health and MICHAEL G. O'RAND, Laboratories for Cell Biology,
Marine Biological Laboratory University of North Carolina at Chapel Hill
ROBERTS. BARLOW, JR., Syr? use University RALPH s- QUATRANO, Oregon State University, at
Corvah,
WALLIS H. CLARK, JR., University of California at
Davis LlONEL *• REBHUN, University of Virginia
JOEL L. ROSENBAUM, Yale University
DAVID H. EVANS, University of Florida
DOROTHY_M. SKINNER, Oak Ridge National
HARLYN O. HALVORSON, Brandeis University Laboratory
RONALD R. HOY, Cornell University JOHN D- STRANDBERG, Johns Hopkins University
JOHN M. TEAL, Woods Hole Oceanographic
SAMUEL S. KOIDE, The Population Council,
_ , r ,, T , . Institution
Rockefeller University
J. RICHARD WHITTAKER, Boston University
FRANK J. LONGO, University of Iowa Marine Program and Marine Biological Laboratory
CHARLOTTE P. MANGUM, The College of GEORGE M. WOODWELL, Ecosystems Center, Marine
William and Mary Biological Laboratory
Editor: CHARLES B. METZ, University of Miami
AUGUST, 1983
Printed and Issued by
LANCASTER PRESS, Inc.
PRINCE & LEMON STS.
LANCASTER, PA.
THE BIOLOGICAL BULLETIN
THE BIOLOGICAL BULLETIN is published six times a year by the Marine Biological Laboratory, MBL
Street, Woods Hole, Massachusetts 02543.
Subscriptions and similar matter should be addressed to THE BIOLOGICAL BULLETIN, Marine Bio-
logical Laboratory, Woods Hole, Massachusetts. Single numbers, $10.00. Subscription per volume (three
issues), $27.00 (this is $54.00 per year for six issues).
j
Communications relative to manuscripts should be sent to Dr. Charles B. Metz, Editor, or Pamela
Clapp, Assistant Editor, at 'the Marine Biological Laboratory, Woods Hole, Massachusetts 02543 between
May 1 and October 1, and at the Institute For Molecular and Cellular Evolution, University of Miami,
521 Anastasia, Coral Gables, Florida 33134 during the remainder of the year.
Copyright © 1983, by the Marine Biological Laboratory
Second-class postage paid at Woods Hole, Mass., 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). Very short
papers (less than 10 manuscript pages including tables, figures, and bibliography) will be published in
a separate section entitled "Short Reports." A limited number of solicited review papers may be accepted
after formal review. A paper will usually appear within four months after its acceptance.
The Editorial Board requests that manuscripts conform to the requirements sef below; those manu-
scripts which do not conform will be returned to authors for correction before review.
1. Manuscripts. Manuscripts, including figures, should be submitted in triplicate. (Xerox copies
of photographs are not acceptable for review purposes.) The original manuscript must be typed in double
spacing (including figure legends, footnotes, bibliography, etc.) on one side of 16- or 20-lb. bond paper,
8'/2 by 1 1 inches. 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). Manu-
scripts should conform to the Council of Biology Editors Style Manual, 4th Edition (Council of Biology
Editors, 1 978) 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. Figures. Figures should be no larger than -8'/2 by 1 1 inches. The dimensions of the printed page,
5 by 73/4 inches, should be kept in mind in preparing figures for publication. We recommend that figures
be about l'/2 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 illus-
tration 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 coordin *^ paper. Those to be repro-
d'ji ,d as halftones should be mounted on board, with both designating numbe j or letters and scale bars
affi d directly to the figures. All figures should be numbered in consecutive order, with no distinction
between text and plate figures. The author's name and an arrow indicating orier'ation should appear on
the reverse side of all figures.
3. 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 con-
Continued on Cover Three
THE MARINE BIOLOGICAL LABORATORY
EIGHTY-FIFTH REPORT, FOR THE YEAR 1982— NINETY-FIFTH YEAR
L TRUSTEES AND STANDING COMMITTEES 1
II. MEMBERS OF THE CORPORATION 5
1. LIFE MEMBERS 5
2. REGULAR MEMBERS 7
3. ASSOCIATE MEMBERS 25
III. CERTIFICATE OF ORGANIZATION 28
IV. ARTICLES OF AMENDMENT 29
V. BYLAWS 30
VI. REPORT OF THE DIRECTOR 34
VII. REPORT OF THE TREASURER 43
VIII. REPORT OF THE CONTROLLER 45
IX. REPORT OF THE LIBRARIAN 58
X. EDUCATIONAL PROGRAMS 58
1 . SUMMER 58
2. JANUARY 67
3. SHORT COURSES 71
XI. RESEARCH AND TRAINING PROGRAMS 76
1 . SUMMER 76
2. YEAR-ROUND 84
XII. HONORS 89
XIII. INSTITUTIONS REPRESENTED 92
XIV. LABORATORY SUPPORT STAFF 97
I. TRUSTEES
Including Action of the 1982 Annual Meeting
OFFICERS
PROSSERGIFFORD, Chairman of the Board of Trustees, Woodrow Wilson International Center
for Scholars, Smithsonian Building, Washington, D. C. 20560
DENIS M. ROBINSON, Honorary Chairman of the Board of Trustees, High Voltage Engineering
Corporation, Burlington, Massachusetts 01803
ROBERT MAINER, Treasurer, The Boston Company, One Boston Place, Boston, Massachu-
setts 02 106
PAUL R. GROSS, President of the Corporation and Director of the Laboratory, Marine Bio-
logical Laboratory, Woods Hole, Massachusetts 02543
1
Copyright (& 1983, by the Marine Biological Laboratory
Library of Congress Card No. A38-518
(ISSN 0006-3 185)
2 MARINE BIOLOGICAL LABORATORY
EMERITI
FRANK A. BROWN, JR., Marine Biological Laboratory (Deceased May 1983)
JOHN B. BUCK, National Institutes of Health
AURIN CHASE, Princeton University
ANTHONY C. CLEMENT, Emory University
KENNETH S. COLE, San Diego, California
ARTHUR L. COLWIN, University of Miami
LAURA COLWIN, University of Miami
D. EUGENE COPELAND, Marine Biological Laboratory
SEARS CROWELL, Indiana University
HARRY GRUNDFEST, Columbia University
TERU HAYASHI, Miami, Florida
HOPE HIBBARD, Oberlin College
LEWIS KLEINHOLZ, Reed College
MAURICE KRAHL, Tucson, Arizona
DOUGLAS MARSLAND, Cockeysville, Maryland
CHARLES B. METZ, University of Miami
HAROLD H. PLOUGH, Amherst, Massachusetts
C. LADD PROSSER, University of Illinois
JOHN S. RANKIN, Ashford, Connecticut
A. C. REDFIELD, Woods Hole, Massachusetts (deceased March 1983)
MERYL ROSE, Waquoit, Massachusetts
MARY SEARS, Woods Hole, Massachusetts
CARL C. SPEIDEL, University of Virginia (no mailings)
ALBERT SZENT-GYORGYI, Marine Biological Laboratory
W. RANDOLPH TAYLOR, University of Michigan
GEORGE WALD, Harvard University
CLASS OF 1986
GEORGE H. A. CLOWES, JR., Cancer Research Institute, Boston, Massachusetts
GERALD FISCHBACH, Washington University
JOHN E. HOBBIE, Ecosystems Center
EDWARD A. KRAVITZ, Harvard Medical School
THOMAS REESE, National Institutes of Health
MARJORIE R. STETTEN, National Institutes of Health (Deceased May 1983)
D. THOMAS TRIGG, Wellesley, Massachusetts
J. RICHARD WHITTAKER, Marine Biological Laboratory
CLASS OF 1985
ROBERT W. ASHTON, Gaston Snow Beekman and Bogue, New York, New York
HARLYN O. HALVORSON, Brandeis University
JOHN G. HILDEBRAND, Columbia University
THOMAS J. HYNES, JR., Meredith & Grew, Inc., Boston, Massachusetts
SHINYA INOUE, Marine Biological Laboratory
RICHARD P. MELLON, Richard King Mellon Foundation. Laughlintown, Pennsylvania
JOHN W. MOORE, Duke University
W. D. RUSSELL-HUNTER, Syracuse University
EVELYN SPIEGEL, Dartmouth College
CLASS OF 1984
CLAY ARMSTRONG, University of Pennsylvania
ROBERT B. BARLOW, JR., Syracuse University
JUDITH GRASSLE, Marine Biological Laboratory
HOLGCR JANNASCH, Woods Hole Oceanographic Institution
TRUSTEES AND STANDING COMMITTEES
BENJAMIN KAMINER, Boston University
BRIAN SALZBERG, University of Pennsylvania
W. NICHOLAS THORNDIKE, Boston, Massachusetts
RICHARD W. YOUNG, Houghton Mifflin Company, Boston, Massachusetts
CLASS OF 1983
NINA ALLEN, Dartmouth College
HAYS CLARK, Avon Products, Incorporated
DENNIS FLANAGAN, Scientific American, New York, New York
WILLIAM T. GOLDEN, New York, New York
PHILIP GRANT, University of Oregon
JOEL ROSENBAUM, Yale University
ANN STLIART, University of North Carolina
ANDREW SZENT-GYORGYI, Brandeis University
KENSAL VAN HOLDE, Oregon State University
STANDING COMMITTEES
EXECUTIVE COMMITTEE OF THE BOARD OF TRUSTEES
PROSSER GIFFORD*
PAUL GROSS*
ROBERT MAINER*
HARLYN O. HALVORSON, 1985
J. RICHARD WHITTAKER, 1985
JOHN HILDEBRAND, 1984
BENJAMIN KAMINER, 1984
NINA ALLEN, 1983
SHINYA INOUE, 1983
BUILDINGS AND GROUNDS COMMITTEE
FRANCIS HOSKINS, Chairman
LAWRENCE B. COHEN
A. FARMANFARMAIAN
ALAN FEIN
DANIEL GILBERT
ROBERT GUNNING*
CLIFFORD HARDING, JR.
PHILIP PERSON
ROBERT PRUSCH
THOMAS REESE
EVELYN SPIEGEL
JAY WELLS
HAYS CLARK
DENNIS FLANAGAN
PROSSER GIFFORD*
WILLIAM T. GOLDEN
CAPITAL DEVELOPMENT COMMITTEE
HARLYN O. HALVORSON
RICHARD P. MELLON
RICHARD YOUNG
COMPUTER COMMITTEE
JOHN HOBBIE, Chairman
WILLIAM J. ADELMAN
FRANCIS P. BOWLES
A. FARMANFARMAIAN
WILLIAM S. LITTLE
E. F. MACNICHOL, JR.
CONSTANTINE TOLLIOS
EMPLOYEE RELATIONS COMMITTEE
CATHERINE NORTON, Chairman
WILLIAM EVANS
JOHN HELFRICH
LEE ANNE CAMPBELL
LEWIS LAWDAY
DONALD LEHY
4 MARINE BIOLOGICAL LABORATORY
FINANCIAL POLICY AND PLANNING COMMITTEE
GEORGE H. A. CLOWES, Chairman ROBERT MAINER
ROBERT ASHTON NICHOLAS THORNDIKE
ELLEN GRASS RICHARD WHITTAKER
THOMAS HYNES
HOUSING, FOOD SERVICE AND DAY CARE COMMITTEE
ANN STUART, Chairman JOAN HOWARD
DANIEL ALKON RONALD JOYNER
NINA ALLEN AIMLEE LADERMAN
ROBERT BARLOW BRIAN SALZBERG
MONA GROSS HOMER P. SMITH*
INSTRUCTION COMMITTEE
SHELDON SEGAL, Chairman ROBERT JOSEPHSON
DANIEL ALKON MORTON MASER*
ROBERT ALLEN MERLE MIZELL
JOHN DOWLING GEORGE PAPPAS
JOHN HOBBIE J. RICHARD WHITTAKER
RONALD HOY
INVESTMENT COMMITTEE
W. NICHOLAS THORNDIKE, Chairman WILLIAM T. GOLDEN
JOHN ARNOLD MAURICE LAZARUS
PROSSER GIFFORD* ROBERT MAINER*
JOINT USERS COMMITTEE FOR THE LIBRARY
EDWARD ADELBERG, Chairman ROBERT GAGOSIAN
WILFRED BRYAN FREDERICK GRASSLE
JOHN DOWLING SHINYA INOUE
LIBRARY JOINT MANAGEMENT COMMITTEE
EDWARD ADELBERG, Chairman DEREK SPENCER
PAUL R. GROSS JOHN STEELE
JOE KIEBALA
MARINE RESOURCES COMMITTEE
SEARS CROWELL, Chairman ROBERT PRENDERGAST
CARL J. BERG ROBERT D. PRUSCH
JUNE HARRIGAN JOHN S. RANKIN
TOM HUMPHREYS JOHN VALOIS*
JACK LEVIN JONATHAN WITTENBERG
CYRUS LEVINTHAL
RADIATION COMMITTEE
WALTER S. VINCENT, Chairman JOHN HOBBIE
EUGENE BELL ANTHONY LIUZZI
FRANCIS P. BOWLES E. F. MACNICHOL, JR.
RICHARD L. CHAPPELL MORTON MASER*
PAUL DEWEER HARRIS RIPPS
TRUSTEES AND STANDING COMMITTEES
NINA S. ALLEN, Chairman
JELLE ATEMA
ROBERT BARLOW, JR.
ROBERT GOLDMAN
SAMUEL S. KOIDE
RAYMOND LASER
RESEARCH SERVICES COMMITTEE
MORTON MASER*
BRYAN NOE
BRUCE PETERSON
BIRGIT ROSE
SIDNEY TAMM
JAY WELLS
RESEARCH SPACE COMMITTEE
GERALD FISCHBACH, Chairman
CLAY ARMSTRONG
JOHN ARNOLD
ARTHUR DuBois
GEORGE LANGFORD
HANS LAUFER
EDUARDO MACAGNO
MORTON MASER*
JERRY MELILLO
ALAN PEARLMAN
JOEL ROSENBAUM
JOAN RUDERMAN
BRIAN SALZBERG
ANN STUART
SAFETY COMMITTEE
A. ROBERT GUNNING, Chairman*
DANIEL ALKON
Louis KERR
LEWIS LAWDAY
DONALD LEHY
JANE LEIGHTON
* ex officio
E. F. MAcNiCHOL, JR.
MORTON MASER*
MARK SILVA
RAYMOND STEPHENS
PAUL STEUDLER
FREDERICK THRASHER
JAY WELLS
II. MEMBERS OF THE CORPORATION
Including Action of the 1982 Annual Meeting
LIFE MEMBERS
ABBOTT, MARIE, 259 High Street, R.D. 2, Coventry, CT 06238
ADOLPH, EDWARD F., University of Rochester, School of Medicine and Dentistry, Rochester,
NY 14642
BEAMS, HAROLD W., Department of Zoology, University of Iowa, Iowa City, IA 53342
BEHRE, ELLINOR, Black Mountain, NC 2871 1
BERTHOLF, LLOYD M., Westminster Village #2114, 2025 E. Lincoln Street, Bloomington,
IL 61701
BISHOP, DAVID W., Department of Physiology, Medical College of Ohio, C.S. 10008, Toledo,
OH 43699
BOLD, HAROLD C, Department of Botany, University of Texas, Austin, TX 78712
BRIDGMAN, A. JOSEPHINE, 715 Kirk Road, Decatur, GA 30030
BURBANCK, MADELINE P., Box 15134, Atlanta, GA 30333
BURBANCK, WILLIAM D., Box 15134, Atlanta, GA 30333
BURDICK, C. LALOR, 900 Barley Drive, Barley Mill Court, Wilmington, DE 19807
CARPENTER, RUSSELL L., 60 Lake Street, Winchester, MA 01890
CHASE, AURIN, Professor of Biology Emeritus, Princeton University, Princeton, NJ 08540
CHENEY, RALPH H., 45 Coleridge Drive, Falmouth, MA 02540
CLARKE, GEORGE L., 44 Juniper Road, Belmont, MA 02178
CLEMENT, ANTHONY C., Department of Biology, Emory University, Atlanta, GA 30322
COLE, KENNETH S. 2404 Loring Street, San Diego, CA 92109
6 MARINE BIOLOGICAL LABORATORY
COLWIN, ARTHUR, 320 Woodcrest Road, Key Biscayne, FL 33149
COLWIN, LAURA, 320 Woodcrest Road, Key Biscayne, FL 33149
COPELAND, D. E., 41 Fern Lane, Woods Hole, MA 02543
COSTELLO, HELEN M., 507 Monroe Street, Chapel Hill, NC 27514
GROUSE, HELEN, Institute of Molecular Biophysics, Florida State University, Tallahassee, FL
32306
DILLER, IRENE C, 2417 Fairhill Avenue, Glenside, PA 19038
DILLER, WILLIAM F., 2417 Fairhill Avenue, Glenside, PA 10038
ELLIOTT, ALFRED M., 2345 Tarpon Road, Naples, FL 33992
FERGUSON, JAMES K. W., 56 Clarkehaven Street, Thornhill, Ontario L4J 2B4 Canada
FRAENKEL, GOTTFRIED S., Department of Entomology, University of Illinois, 320 Morrill
Hall, Urbana, IL 61801
FRIES, ERIK F. B., 3870 Leafy Way, Miami, FL 33133
OILMAN, LAUREN C., Department of Biology, University of Miami, PO Box 24918, Coral
Gables, FL33124
GREEN, JAMES W., Department of Physiology, Rutgers University, Piscataway, NJ 08854
GRUNDFEST, HARRY, Department of Neurology, College of Physicians and Surgeons, Co-
lumbia University, New York, NY 10032
GUTTMAN, RITA, 75 Henry Street, Brooklyn, NY 1 1210
HAMBURGER, VIKTOR, Professor Emeritus, Washington University, St. Louis, MO 63130
HAMILTON, HOWARD L., Department of Biology, University of Virginia, Charlottesville, VA
22901
HARTLINE, H. KEFFER, The Rockefeller University, New York, NY 10021 (Deceased March
1983)
HIBBARD, HOPE, 143 East College Street, Apt. 309, Oberlin, Ohio 44074
HISAW, F. L., 5925 SW Plymouth Drive, Corvallis, OR 97330
HOLLAENDER, ALEXANDER, Associated Universities, Inc., 1717 Massachusetts Avenue, NW,
Washington, DC 20036
HUMES, ARTHUR, Marine Biological Laboratory, Woods Hole, MA 02543
JOHSON, FRANK H., Department of Biology, Princeton University, Princeton, NJ 08540
KAAN, HELEN, 62 Locust Street, Falmouth, MA 02540
KAHLER, ROBERT, P.O. Box 423, Woods Hole, MA 02543
KILLE, FRANK R., 500 Osceola Avenue, Winter Park, FL 32789
KLEINHOLZ, LEWIS, Department of Biology, Reed College, Portland, OR 97202
LEVINE, RACHMIEL, 2024 Canyon Road, Arcadia, CA 91006
LOCHHEAD, JOHN H., 49 Woodlawn Road, London SW 6 6PS, England, U. K.
LYNN, W. GARDNER, Department of Biology, Catholic University of America, Washington,
DC 200 17
MAGRUDER, SAMUEL R., 270 Cedar Lane, Paducah, KY 42001
MANWELL, REGINALD, D., Syracuse University, Lyman Hall, Syracuse, NY 13210
MARSLAND, DOUGLAS, Broadmead N12, 13801 York Road, Cockeysville, MD 21030
MILLER, JAMES A., 307 Shorewood Drive, E. Falmouth, MA 02536
MILNE, LORUS J., Department of Zoology, University of New Hampshire, Durham, NH
03824
MOORE, JOHN A., Department of Biology, University of California, Riverside, CA 92521
MOUL, E. T., 43 F. R. Lillie Road, Woods Hole, MA 02543
NACHMANSHON, DAVID, Department of Neurology, College of Physicians and Surgeons,
Columbia University, New York, NY 10032
PAGE, IRVING H., Box 516, Hyannisport, MA 02647
PLOUGH, HAROLD H., 31 Middle Street, Amherst, MA 01002
POLLISTER, A. W., Box 23, Dixfield, ME 04224
POND, SAMUEL E., P.O. Box 63, E. Winthrop, ME 04343
PROSSER, C. LADD, Department of Physiology and Biophysics, University of Illinois, Urbana,
IL 61801
PRYTZ, MARGARET MCDONALD, 21 McCouns Lane, Oyster Bay, NY 1 1771
RANKIN, JOHN S., JR., Box 97, Ashford, CT 06278
RENN, CHARLES E., Route 2, Hempstead, MD 21074
MEMBERS OF THE CORPORATION
REZNIKOFF, PAUL, 1 1 Brooks Road, Woods Hole, MA 02543
RICHARDS, A. GLENN, Department of Entomology, Fisheries and Wildlife, University of
Minnesota, St. Paul, MN 55101
RICHARDS, OSCAR W., Pacific University, Forest Grove, OR 97462
SCHARRER, BERTA, Department of Anatomy, Albert Einstein College of Medicine, 1300
Morris Park Avenue, Bronx, NY 10461
SCHMITT, F. O. Room 1 6-5 1 2, Massachusetts Institute of Technology, Cambridge, MA 02 1 39
SHEMIN, DAVID, Department of Biochemistry and Molecular Biology, Northwestern Uni-
versity, Evanston, IL 60201
SICHEL, ELSA, 4 Whitman Road, Woods Hole, MA 02543
SONNENBLICK, B. P., Department of Zoology and Physiology, Rutgers University, 195 Uni-
versity Avenue, Newark, NJ 07102
SPEIDEL, CARL C, 1873 Field Road, Charlottesville, VA 22903
STEINHARDT, JACINTO, 1508 Spruce Street, Berkeley, CA 94709
STUNKARD, HORACE W., American Museum of Natural History, Central Park West at 79th
Street, New York, NY 10024
TAYLOR, W. RANDOLPH, Department of Biology, University of Michigan, Ann Arbor, MI
48109-
TEWINKEL, Lois E., 4 Sanderson Avenue, Northampton, MA 01060
TRACER, WILLIAM, The Rockefeller University, 1230 York Avenue, New York, NY 10021
TRAVIS, DOROTHY F., 35 Coleridge Drive, Falmouth, MA 02540
WALD, GEORGE, Higgins Professor of Biology Emeritus, Harvard University, Cambridge,
MA 02138
WICHTERMAN, RALPH, 31 Buzzards Bay Avenue, Woods Hole, MA 02543
YOUNG, D. B., 1 137 Main Street, N. Hanover, MA 02357
ZINN, DONALD J., P.O. Box 589, Falmouth, MA 02541
ZORZOLI, ANITA, Department of Botany, Vassar College, Poughkeepsie, NY 12601
ZWEIFACH, BENJAMIN W., c/o Ames, University of California, La Jolla, CA 92037
REGULAR MEMBERS
ACHE, BARRY W., Whitney Marine Laboratory, University of Florida, Rt. 1 Box 121, St.
Augustine, FL 32084
ACHESON, GEORGE H., 25 Quissett Avenue, Woods Hole, MA 02543
ADEJUWON, CHRISTOPHER A., Chemical Pathology Department, University of Ibadan, Iba-
dan, Nigeria
ADELBERG, EDWARD A., Department of Human Genetics, Yale University Medical School,
New Haven, CT 065 11
AFZELIUS, BJORN, Wenner-Gren Institute, University of Stockholm, Stockholm, Sweden
ALBERTE, RANDALL S., University of Chicago, Barnes Laboratory, 5630 S. Ingleside Avenue,
Chicago, IL 60637
ALKON, DANIEL, Section on Neural Systems, Laboratory of Biophysics, NIH, Marine Bio-
logical Laboratory, Woods Hole, MA 02543
ALLEN, GARLAND E., Department of Biology, Washington University, St. Louis, MO 63130
ALLEN, NINA S., Department of Biology, Dartmouth College, Hanover, NH 03755
ALLEN, ROBERT D., Department of Biology, Dartmouth College, Hanover, NH 03755
ALSCHER, RUTH, Department of Biology, Manhattanville College, Purchase, NY 10577
AMATNIEK, ERNEST, 4797 Boston Post Road, Pelham Manor, NY 10803
ANDERSON, EVERETT, Department of Anatomy, LHRRB, Harvard Medical School, Boston,
MA 021 15
ANDERSON, J. M., Cornell University, Emerson Hall, Ithaca, NY 14850
ARMSTRONG, CLAY M., Department of Physiology, University of Pennsylvania Medical
School, Philadelphia, PA 19174
ARMSTRONG, PETER B., Department of Zoology, University of California, Davis, CA 95616
ARNOLD, JOHN M., Pacific Biomedical Research Center, University of Hawaii, 42 Ahui
Street, Honolulu, HI 96813
ARNOLD, WILLIAM A., 102 Balsam Road, Oak Ridge, TN 37830
8 MARINE BIOLOGICAL LABORATORY
ASHTON, ROBERT W., Gaston Snow Beekman and Bogue, 14 Wall Street, New York, NY
10005
ATEMA, JELLE, Marine Biological Laboratory, Woods Hole, MA 02543
ATWOOD, KJMBALL C, 100 Haven Avenue, Apt. 21-E, New York, NY 10032
AUGUSTINE, GEORGE JR., Department of Physiology and Anatomy, University of California,
Los Angeles, CA 94720
AUSTIN, MARY L., 506'/2 N. Indiana Avenue, Bloomington, IN 47401
BACON, ROBERT, P.O. Box 723, Woods Hole, MA 02543
BALDWIN, THOMAS O., Department of Biochemistry and Biophysics, Texas A & M Uni-
versity, College Station, TX 77843
BANG, BETSY, 76 F. R. Lillie Road, Woods Hole, MA 02543
BARKER, JEFFERY L., NIH Bldg. 36 Room 2002, Bethesda, MD 20205
BARLOW, ROBERT B., JR., Institute for Sensory Research, Syracuse University, Merrill Lane,
Syracuse, NY 13210
BARTELL, CLELMER K., 2000 Lake Shore Drive, New Orleans, LA 70122
BARTH, LUCENA J., 26 Quissett Avenue, Woods Hole, MA 02543
BARTLETT, JAMES H., Department of Physics, Box 1921, University of Alabama, University,
AL 35486
BATTELLE, BARBARA-ANNE, National Eye Institute, Bethesda, MD 20205
BAUER, G. ERIC, Department of Anatomy, University of Minnesota, Minneapolis, MN 55414
BEAUGE, Luis ALBERTO, Institute de Investigacion Medica, Casilla de Correo 389, 5000
Cordoba, Argentina
BECK, L. V., Department of Pharmacology, School of Experimental Medicine, Indiana Uni-
versity, Bloomington, IN 47401
BEGG, DAVID A., LHRRB, Harvard Medical School, Boston, MA 021 15
BELL, EUGENE, Department of Biology, Massachusetts Institute of Technology, 77 Massa-
chusetts Avenue, Cambridge, MA 02139
BENNETT, M. V. L., Department of Neuroscience, Albert Einstein College of Medicine, 1300
Morris Park Avenue, New York, NY 10461
BENNETT, MIRIAM F., Department of Biology, Colby College, Waterville, ME 04901
BERG, CARL J., JR., Marine Biological Laboratory, Woods Hole, MA 02543
BERMAN, MONES, NIH, Theoretical Biology NCI, Bldg. 10 Room 4B56, Bethesda, MD 20205
(Deceased August 1982)
BERNE, ROBERT W., University of Virginia, School of Medicine, Charlottesville, VA 22908
BERNHEIMER, ALAN W., New York University, School of Medicine, New York, NY 10016
BEZANILLA, FRANCISCO, Department of Physiology, University of California, Los Angeles,
CA 90052
BIGGERS, JOHN D., Department of Physiology, Harvard Medical School, Boston, MA 021 15
BISHOP, STEPHEN H., Department of Zoology, Iowa State University, Ames, IA 50010
BLAUSTEIN, MORDECAI P., Department of Physiology, School of Medicine, University of
Maryland, 655 W. Baltimore Street, Baltimore, MD 21201
BODIAN, DAVID, Department of Otolaryngology, Johns Hopkins University, Baltimore, MD
21205
BOETTIGER, EDWARD G., 29 Juniper Point, Woods Hole, MA 02543
BOGORAD, LAWRENCE, The Biological Laboratories, Harvard University, Cambridge, MA
02138
BOOLOOTIAN, RICHARD A., Science Software Systems, Inc., 11899 W. Pico Blvd., W. Los
Angeles, CA 90064
BOREI, HANS G., Department of Zoology, University of Pennsylvania, Philadelphia, PA
19174
BORGESE, THOMAS A., Department of Biology, Lehman College, Bronx, NY 10468
BORISY, GARY G., Laboratory of Molecular Biology, University of Wisconsin, Madison, WI
53715
BOSCH, HERMAN F., Whipple Hill, Richmond, NH 03470
BOTKJN, DANIEL, Department of Biology, University of California, Santa Barbara, CA 93106
BOWEK VAUGHN T., Woods Hole Oceanographic Institution, Redfield Bldg. 3-32, Woods
Hoi,, MA 02543
MEMBERS OF THE CORPORATION 9
BOYER, BARBARA C., Department of Biology, Union College, Schenectady, NY 12308
BOWLES, FRANCIS P., Box 674, Woods Hole, MA 02543
BRINLEY, F. J., Neurological Disorders Program, NINCDS, 716 Federal Building, Bethesda,
MD 20205
BROWN, FRANK A., JR., Marine Biological Laboratory, Woods Hole, MA 02543 (Deceased
May 1983)
BROWN, JAY C., Department of Neurobiology, University of Virginia, Charlottesville, VA
22908
BROWN, JOEL E., Department of Physiology and Biophysics, Health Sciences Center, SUNY,
Stony Brook, NY 1 1 794
BROWN, STEPHEN C., Department of Biological Sciences, SUNY, Albany, NY 12222
BUCK, JOHN B., National Institutes of Health, Laboratory of Physical Biology, Bethesda, MD
20205
BURDICK, CAROLYN J., Department of Biology, Brooklyn College, Brooklyn, NY 1 1210
BURGER, MAX, Department of Biochemistry, Biocenter of the University of Basel, Klingel-
bergstrasse 70, CH-4056 Basel, Switzerland
BURKY, ALBERT, Department of Biology, University of Dayton, Dayton, OH 45469
BUSH, LOUISE, 7 Snapper Lane, Falmouth, MA 02540
CANDELAS, GRACIELA C., Department of Biology, University of Puerto Rico, Rio Piedras,
PR 00931
CARLSON, FRANCIS D., Department of Biophysics, Johns Hopkins University, Baltimore,
MD21218
CASE, JAMES, Department of Biological Sciences, University of California, Santa Barbara,
CA93106
CASSIDY, REV. J. D., O.P., Department of Biological Sciences, University of Illinois at Chicago
Circle, Box 4348, Chicago, IL 60680
CEBRA, JOHN J., Department of Biology, Leidy Labs, G-6, University of Pennsylvania, Phil-
adelphia, PA 19174
CHAET, ALFRED B., University of West Florida, Pensacola, FL 32504
CHAMBERS, EDWARD L., Department of Physiology and Biophysics, University of Miami,
School of Medicine, Miami, FL 33152
CHAPPELL, RICHARD L., Department of Biological Sciences, Hunter College, Box 201, New
York, NY 10021
CHAUNCEY, HOWARD H., 30 Falmouth Street, Wellesley Hills, MA 02181
CHILDS, FRANK M., Department of Biology, Trinity College, Hartford, CT 06106
CITKOWITZ, ELENA, 410 Livingston Street, New Haven, CT 0651 1
CLARK, A. M., Department of Biological Sciences, University of Delaware, Newark, DE
19711
CLARK, ELOISE E., National Science Foundation, 1800 G Street, NW, Washington, DC 20550
CLARK, HAYS, 26 Deer Park Drive, Greenwich, CT 06830
CLARK, WALLIS H., JR., Aquaculture Program, Department of Animal Science, University
of California, Davis, CA 95616
CLAUDE, PHILIPPA, Primate Center, Capitol Court, Madison, WI 53706
CLAYTON, RODERICK K., Cornell University, Section of Genetics, Development and Phys-
iology, Ithaca, NY 14850
CLOWES, GEORGE H. A., JR., The Cancer Research Institute, 194 Pilgrim Road, Boston, MA
02215
CLUTTER, MARY, Cellular and Physiological Biosciences Section, National Science Foun-
dation, 1800 G Street, Washington, DC 20550
COBB, JEWELL P., President, California State University, Fullerton, CA 92634
COHEN, ADOLPH I., Department of Opthamology, School of Medicine, Washington Uni-
versity, 660 S. Euclid Avenue, St. Louis, MO 631 10
COHEN, CAROLYN, Rosenstiel Basic Medical Sciences Research Center, Brandeis University,
Waltham, MA 02 1 54
COHEN, LAWRENCE B., Department of Physiology, Yale University, 333 Cedar Street, New
Haven, CT 065 10
10 MARINE BIOLOGICAL LABORATORY
COHEN, SEYMOUR S., Department of Pharmacological Science, SUNY, Stony Brook, NY
11790
COHEN, WILLIAM D., Department of Biological Sciences, Hunter College, New York, NY
10021
COLLIER, JACK R., Department of Biology, Brooklyn College, Brooklyn, NY 1 1210
COLLIER, MARJORIE McCANN, Biology Department, St. Peter's College, Jersey City, NJ
07306
COOK, JOSEPH A., The Edna McConnell Clark Foundation, 250 Park Avenue, New York,
NY 10017
COOPERSTEIN, S. J., University of Connecticut, School of Medicine, Farmington Avenue,
Farmington, CT 06032
CORLISS, JOHN O., Department of Zoology, University of Maryland, College Park, MD 20742
CORNELL, NEAL W., 6428 Bannockburn Drive, Bethesda, MD 20817
CORNMAN, IVOR, 10A Orchard Street, Woods Hole, MA 02543
COSTELLO, WALTER J., College of Medicine, Ohio University, Athens, OH 45701
COUCH, ERNEST F., Department of Biology, Texas Christian University, Fort Worth, TX
76129
CREMER-BARTELS, GERTRUD, Universitats Augenklinik, 44 Munster, West Germany
CRIPPA, MARCO, Faculte de Scientces, Univeristats de Geneve, 20 quai Ernest-Ansermet,
Geneve 4, Switzerland
CROW, TERRY J., Department of Physiology, University of Pittsburgh, School of Medicine,
Pittsburgh, PA 1 526 1
CROWELL, SEARS, Department of Biology, Indiana University, Bloomington, IN 47401
DAIGNAULT, ALEXANDER T., W. R. Grace Company, 114 Avenue of the Americas, New
York, NY 10036
DAN, KATSUMA, Professor Emeritus, Tokyo Metropolitan Union, Meguro-ku, Tokyo, Japan
DANEILLI, JAMES F., 185 Highland Street, Worcester, MA 01609
DAVID, JOHN R., Seeley G. Mudd Bldg., Room 504, 250 Longwood Avenue, Boston, MA
02115
DAVID, ROBERTA A., Seeley G. Mudd Bldg., Room 504, 250 Longwood Avenue, Boston,
MA 02115
DAVIS, BERNARD D., Bacterial Physiology Unit, Harvard Medical School, Boston, MA 02 1 1 5
DAVIS, JOEL P., Seapuit, Inc., P.O. Box G, Osterville, MA 02655
DAW, NIGEL W., 78 Aberdeen Place, Clayton, MO 63105
DEGROOF, ROBERT C., 511 Carpenter Lane, Philadelphia, PA 19119
DEHAAN, ROBERT L., Department of Anatomy, Emory University, Atlanta, GA 30322
DELANNEY, Louis E., Institute for Medical Research, 751 Bascom Avenue, San Jose, CA
95128
DEPHILLIPS, HENRY A., JR., Department of Chemistry, Trinity College, Hartford, CT 06106
DETERRA, NOEL, Marine Biological Laboratory, Woods Hole, MA 02543
DETTBARN, WOLF-DIETRICH, Department of Pharmacology, School of Medicine, Vanderbilt
University, Nashville, TN 37127
DEWEER, PAUL J., Department of Physiology, School of Medicine, Washington University,
St. Louis, MO 63110
DISCH, ZACHARIAS, College of Physicians and Surgeons, Columbia University Eye Institute,
630 W. 165th Street, New York, NY 10032
DIXON, KEITH E., School of Biological Sciences, Flinders University, Bedford Park, South
Australia
DOWDALL, MICHAEL J., Department of Biochemistry, University Hospital and Medical
School, Nottingham N672 UH, U. K.
DOWLING, JOHN E., The Biological Laboratories, Harvard University, 16 Divinity Street,
Cambridge. MA 02 1 38
DRESDEN, MARC H., Department of Biochemistry, Baylor College of Medicine, Houston,
TX 77025
DUDLEY, PATRICIA L., Department of Biological Sciences, Barnard College, Columbia Uni-
versitv. New York, NY 10027
MEMBERS OF THE CORPORATION 1 1
DUNHAM, PHILIP B., Department of Biology, Syracuse University, Syracuse, NY 13210
EBERT, JAMES D., Office of the President, Carnegie Institution of Washington, 1530 P Street,
NW, Washington, DC 20008
ECK.BERG, WILLIAM R., Department of Zoology, Howard University, Washington, DC 20059
ECKERT, ROGER O., Department of Zoology, University of California, Los Angeles, CA
90024
EDDS, KENNETH T., Department of Anatomical Sciences, SUNY, Buffalo, NY 14214
EDDS, LOUISE, College of Osteopathic Medicine, Grosvenor Hall, Ohio University, Athens,
OH 45701
EDER, HOWARD A., Albert Einstein College of Medicine, 1300 Morris Park Avenue, Bronx,
NY 10461
EDWARDS, CHARLES, Department of Biological Sciences, SUNY, Albany, NY 12222
EGYUD, LASZLO G., P.O. Box 342, Woods Hole, MA 02543
EHRENSTEIN, GERALD, NIH, Bethesda, MD 20205
EHRLICH, BARBARA E., Department of Physiology, Albert Einstein College of Medicine, 1300
Morris Park Avenue, New York, NY 10461
EICHEL, HERBERT J., 226 W. Rattinghouse Square, Philadelphia, PA 19174
EISEN, ARTHUR Z., Chief of Division of Dermatology, Washington University, St. Louis,
MO 63110
ELDER, HUGH YOUNG, Institute of Physiology, University of Glasgow, Glasgow, Scotland
ELLIOTT, GERALD F., The Open University Research Unit, Foxcombe Hall, Berkeley Road,
Boars Hill, Oxford, England, U. K.
EPEL, DAVID, Hopkins Marine Station, Pacific Grove, CA 93950
EPSTEIN, HERMAN T., Department of Biology, Brandeis University, Waltham, MA 02154
ERULKAR, SOLOMON D., 318 Kent Road, Bala Cynwyd, PA 19004
ESSNER, EDWARD S., Kresege Eye Institute, Wayne State University, Detroit, MI 48201
ETTIENE, EARL M., Department of Anatomy, Harvard Medical School, Boston, MA 02115
FAILLA, PATRICIA M., Argonne National Laboratory, Office of the Director, Argonne, IL
60439
FARMANFARMAIAN, A., Department of Physiology and Biochemistry, Rutgers University,
New Brunswick, NJ 08903
FAUST, ROBERT G., Department of Physiology, Medical School, University of North Carolina,
Chapel Hill, NC27514
FEIN, ALAN, Laboratory of Sensory Physiology, Marine Biological Laboratory, Woods Hole,
MA 02543
FERGUSON, F. P., National Institute of General Medical Sciences, NIH, Bethesda, MD 20205
FESSENDEN, JANE, Marine Biological Laboratory, Woods Hole, MA 02543
FINKELSTEIN, ALAN, Albert Einstein College of Medicine, 1 300 Morris Park Avenue, New
York, NY 10461
FISCHBACH, GERALD, Department of Anatomy and Neurobiology, Washington University,
School of Medicine, St. Louis, MO 631 10
FISCHMAN, DONALD A., Cornell University Medical College, Department of Anatomy and
Cell Biology, 1300 York Avenue, New York, NY 10021
FISHER, J. MANNERV, Department of Biochemistry, University of Toronto, Toronto, Ontario,
Canada M5S 1AB
FISHMAN, HARVEY M., Department of Physiology, University of Texas, Medical Branch,
Galveston, TX 77550
FLANAGAN, DENNIS, Editor, Scientific American, 415 Madison Avenue, New York, NY
10017
Fox, MAURICE S., Department of Biology, Massachusetts Institute of Technology, Cam-
bridge, MA 02 1 39
FRANZINI, CLARA, Department of Biology G-5, School of Medicine, University of Pennsyl-
vania, Philadelphia, PA 19174
FRAZIER, DONALD T., Department of Physiology and Biophysics, School of Medicine, Uni-
versity of Kentucky Medical Center, Lexington, KY 40536
FREEMAN, GARY L., Department of Zoology, University of Texas, Austin, TX 78172
12 MARINE BIOLOGICAL LABORATORY
FRENCH, ROBERT!., Department of Biophysics, University of Maryland, School of Medicine,
Baltimore, MD 21201
FREYGANG, WALTER J., JR., 6247 29th Street, NW, Washington, DC 20015
FULTON, CHANDLER M., Department of Biology, Brandeis University, Waltham, MA 02154
FURSHPAN, EDWIN J., Department of Neurophysiology, Harvard Medical School, Boston,
MA 021 15
FUSELER, JOHN W., Department of Cell Biology, University of Texas, Medical Branch, 53233
Harry Hines Blvd., Dallas, TX 75235
FUTRELLE, ROBERT P., Department of Genetics and Development, 515 Morrill Hall, Uni-
versity of Illinois, Urbana, IL 68101
FYE, PAUL, Woods Hole Oceanographic Institution, Woods Hole, MA 02543
GABRIEL, MORDECAI, Department of Biology, Brooklyn College, Brooklyn, NY 11210
GAINER, HAROLD, Section of Functional Neurochemistry, NIH, Bldg. 36 Room 2A21, Be-
thesda, MD 20205
GALATZER-LEVY, ROBERT M., Suite 1813, 55 East Washington Street, Chicago, IL 60602
GALL, JOSEPH G., Department of Biology, Yale University, New Haven, CT 06520
GASCOYNE, PETER, Marine Biological Laboratory, Woods Hole, MA 02543
GELFANT, SEYMOUR, Department of Dermatology, Medical College of Georgia, Augusta,
GA 30904
GELPERIN, ALAN, Department of Biology, Princeton University, Princeton, NJ 08540
GERMAN, JAMES L., Ill, The New York Blood Center, 310 East 67th Street, New York, NY
10021
GIBBS, MARTIN, Institute for Photobiology of Cells and Organelles, Brandeis University,
Waltham, MA 02154
GIBSON, A. JANE, Wing Hall, Cornell University, Ithaca, NY 14850
GIFFORD, PROSSER, Woodrow Wilson International Center for Scholars, Smithsonian Build-
ing, Washington, DC 20560
GILBERT, DANIEL L., NIH, Laboratory of Biophysics, NINCDS, Bldg. 36 Room 2A-29,
Bethesda, MD 20205
GIUDICE, GIOVANNI, Via Archirafi, 22 Palermo, Italy
GLUSMAN, MURRAY, Department of Clinical Psychiatry, Columbia University, 722 W. 168th
Street, New York, NY 10032
GOLDEN, WILLIAM T., 40 Wall Street, New York, NY 10005
GOLDMAN, DAVID E., 63 Loop Road, Falmouth, MA 02540
GOLDMAN, ROBERT D., Department of Cell Biology and Anatomy, Northwestern University,
303 E. Chicago Avenue, Chicago, IL 6061 1
GOLDSMITH, MARY H. M., Department of Biology, Yale University, New Haven, CT 06520
GOLDSMITH, PAUL K., 551 1 Oakmont Avenue, Bethesda, MD 20034
GOLDSMITH, TIMOTHY H., Department of Biology, Yale University, New Haven, CT 06520
GOLDSTEIN, MOISE H., JR., Johns Hopkins University School of Medicine, 720 Rutland
Avenue, Baltimore, MD 21205
GOODMAN, LESLEY JEAN, Department of Zoology and Comparative Physiology, Queen Mary
College, Mile End Road, London El 4NS England, U. K.
GOTTSCHALL, GERTRUDE Y., 315 E. 68th Street, 9-M, New York, NY 10021
GOUDSMIT, ESTHER M., Department of Biology, Oakland University, Rochester, MI 48063
GOULD, ROBERT MICHAEL, Institute for Basic Research in Developmental Disabilities, 1050
Forest Hill Road, Staten Island, NY 10314
GOULD, STEPHEN J., Museum of Comparative Zoology, Harvard University, Cambridge,
MA 02 138
GRAHAM, HERBERT, 36 Wilson Road, Woods Hole, MA 02543
GRANT, PHILIP, Department of Biology, University of Oregon, Eugene, OR 97403
GRASS, ALBERT, The Grass Foundation, 77 Reservoir Road, Quincy, MA 02170
GRASS, ELLEN R., The Grass Foundation, 77 Reservoir Road, Quincy, MA 02170
GRASSLE, JUDITH, Marine Biological Laboratory, Woods Hole, MA 02543
GREEN, JONATHAN P., Department of Biology, Roosevelt University, 430 S. Michigan Av-
enue, Chicago, IL 60605
MEMBERS OF THE CORPORATION 1 3
GREENBERG, MICHAEL J., Department of Biological Sciences, Florida State University, Tal-
lahassee, FL 32306
GREGG, JAMES H., Department of Zoology, University of Florida, Gainesville, FL 3261 1
GREIF, ROGER L., Department of Physiology, Cornell University, Medical College, New
York, NY 10021
GRIFFIN, DONALD R., The Rockefeller University, 1230 York Avenue, New York, NY 1002 1
GROSCH, DANIEL S., Department of Genetics, Gardner Hall, North Carolina State University,
Raleigh, NC 27607
GROSS, PAUL R., President and Director, Marine Biological Laboratory, Woods Hole, MA
02543
GROSSMAN, ALBERT, New York University Medical School, New York, NY 10016
GUNNING A. ROBERT, 377 Hatchville Road, Hatchville, MA 02536
GWILLIAM, G. P., Department of Biology, Reed College, Portland, OR 97202
HALL, ZACK W., Department of Physiology, University of California, San Francisco, CA
94143
HALVORSON, HARLYN O., Rosenstiel Basic Medical Sciences Research Center, Brandeis
University, Waltham, MA 02154
HAMKALO, BARBARA A., Department of Molecular Biology and Biochemistry, University
of California, Irvine, CA 92717
HANNA, ROBERT B., SUNY, College of Environmental Science and Forestry, Syracuse, NY
13210
HARDING, CLIFFORD V., JR., Kresege Eye Institute, Wayne State University, Detroit, MI
48210
HAROSI, FERENC I., Laboratory of Sensory Physiology, Marine Biological Laboratory, Woods
Hole, MA 02543
HARRIGAN, JUNE F., Laboratory of Biophysics, Marine Biological Laboratory, Woods Hole,
MA 02543
HARRINGTON, GLENN W., Department of Microbiology, School of Dentistry, University of
Missouri, 650 E. 250th Street, Kansas City, MO 64108
HASCHEMEYER, AUDREY E. V., Department of Biological Sciences, Hunter College, New
York, NY 10021
HASTINGS, J. W., The Biological Laboratories, Harvard University, Cambridge, MA 02138
HAYES, RAYMOND L., JR., Department of Anatomy, School of Medicine, Morehouse College,
Atlanta, GA 30314
HAYASHI, TERU, 7105 SW 112 Place, Miami, FL 33173
HENLEY, CATHERINE, 7401 Westlake Terrace, Apt. No. 1516, Bethesda, MD 20034
HERNDON, WALTER R., University of Tennessee, 506 Andy Holt Tower, Knoxville, TN
37916
HESSLER, ANITA Y., 5795 Waverly Avenue, La Jolla, CA 92037
HEUSER, JOHN, Department of Biophysics, Washington University School of Medicine, St.
Louis, MO 63110
HIATT, HOWARD H., Office of the Dean, Harvard School of Public Health, 677 Huntington
Avenue, Boston, MA 021 15
HIGHSTEIN, STEPHEN M., Division of Cellular Neurobiology, Albert Einstein College of
Medicine, 1300 Morris Park Avenue, Bronx, NY 10461
HILDEBRAND, JOHN G., Department of Biological Sciences, Fairchild Center #9 13, Columbia
University, New York, NY 10027
HILL, ROBERT B., Department of Zoology, University of Rhode Island, Kingston, RI 02881
HILLIS-COLINVAUX, LLEWELLYA, Department of Zoology, Ohio State University, Columbus,
OH 43210
HILLMAN PETER, Department of Biology, Hebrew University, Jerusalem, Israel
HINEGARDNER, RALPH T., Division of Natural Sciences, University of California, Santa Cruz,
CA 95064
HINSCH, GERTRUDE W., Department of Biology, University of South Florida, Tampa, FL
33620
14 MARINE BIOLOGICAL LABORATORY
HOBBIE, JOHN E., Ecosystems Center, Marine Biological Laboratory, Woods Hole, MA 02543
HODGE, ALAN J., Marine Biological Laboratory, Woods Hole, MA 02543
HODGE, CHARLES, IV, P.O. Box 4095, Philadelphia, PA 19118
HOFFMAN, JOSEPH, Department of Physiology, School of Medicine, Yale University, New
Haven, CT 06515
HOLLYFIELD, JOE G., Baylor School of Medicine, Texas Medical Center, Houston, TX 77030
HOLTZMAN, ERIC, Department of Biological Sciences, Columbia University, New York, NY
10027
HOLZ, GEORGE G., JR., Department of Microbiology, SUNY, Syracuse, NY 13210
HOSKIN, FRANCIS C. G.. Department of Biology, Illinois Institute of Technology, Chicago,
IL60616
HOUGHTON, RICHARD A., Ill, Ecosystems Center, Marine Biological Laboratory, Woods
Hole, MA 02543
HOUSTON, HOWARD, 2500 Virginia Avenue, NW, Washington, DC 20037
HOWARTH, ROBERT, Ecosystems Center, Marine Biological Laboratory, Woods Hole, MA
02543
HOY, RONALD R., Section of Neurobiology and Behavior, Cornell University, Ithaca, NY
14850
HUBBARD, RUTH, The Biological Laboratories, Harvard University, Cambridge, MA 02138
HUFNAGEL, LINDA A., Department of Microbiology, University of Rhode Island, Kingston,
RI 02881
HUMMON, WILLIAM D., Department of Zoology, Ohio University, Athens, OH 45701
HUMPHREYS, SUSIE H., Gerontology Research Center, NIA, NIH, Baltimore City Hospital,
Baltimore, MD 21224
HUMPHREYS, TOM D., University of Hawaii, PBRC, 41 Ahui Street, Honolulu, Hawaii 968 13
HUNTER, BRUCE W., Box 321, Lincoln Center, MA 01773
HUNTER, ROBERT D., Department of Biological Sciences, Oakland University, Rochester,
NY 48063
HUNZIKER, HERBERT E., Esq., P.O. Box 547, Falmouth, MA 02541
HURWITZ, CHARLES, Basic Science Research Lab, Veterans Administration Hospital, Albany,
NY 12208
HURWITZ, JERARD, Albert Einstein College of Medicine, Department of Molecular Biology,
1300 Morris Park Avenue, Bronx, NY 10461
HUXLEY, HUGH E., Medical Research Council, Laboratory of Molecular Biology, Cambridge,
England, U. K.
HYNES, THOMAS J., JR., Senior Vice President, Meredith and Grew, Inc., 125 High Street,
Boston, MA 02110
ILAN, JOSEPH. Department of Anatomy, Case Western Reserve University, Cleveland, OH
44106
INGOGLIA, NICHOLAS, Department of Physiology, New Jersey Medical School, Newark, NJ
07103
INOUE, SADUYKJ, Electron Microscopy Laboratory, McGill University Cancer Center, 655
Drummond Street, Montreal, P. A., Canada, HG3 1Y6
INOUE, SHINYA, Marine Biological Laboratory, Woods Hole, MA 02543
ISENBERG, IRVING, Department of Biochemistry and Biophysics, Oregon State University,
Corvallis, OR 97331
ISSELBACHER, KURT J., Massachusetts General Hospital, Boston, MA 021 14
ISSADORIDES, MARIETTA R., Department of Psychiatry, University of Athens, Monis Petraki
8, Athens, 140 Greece
IZZARD, COLIN S., Department of Biological Sciences, SUNY, Albany, NY 12222
JACOBSON, ANTONE G., Department of Zoology, University of Texas, Austin, TX 78712
JAFFE, LIONEL, Department of Biology, Purdue University, Lafayette, IN 47907
JAHAN-PARWAR, BEHRUS, Worcester Foundation for Experimental Biology, 222 Maple Av-
enue, Shrewsbury, MA 01545
JANNASCH, HOLGER W., Woods Hole Oceanographic Institution, Woods Hole, MA 02543
JEFFERY, WILLIAM R., Department of Zoology, University of Texas, Austin, TX 78712
MEMBERS OF THE CORPORATION 1 5
JENNER, CHARLES E., Department of Zoology, University of North Carolina, Chapel Hill,
NC 27514
JENNINGS, JOSEPH B., Department of Zoology, Baines Wing, University of Leeds, Leeds LS
2 9-JT, England, U. K.
JONES, MEREDITH L., Smithsonian Institution, Division of Worms, Washington, DC 20650
JONES, RAYMOND F., Department of Biology, SUNY, Stony Brook, NY 11790 (Deceased
August 1982)
JOSEPHSON, ROBERT K., School of Biological Sciences, University of California, Irvine, CA
92664
JOYNER, RONALD W., Department of Physiology, University of Iowa, Iowa City, IA 52242
KABAT, E. A., Department of Microbiology, College of Physicians and Surgeons, Columbia
University, New York, NY 10032
KAFATOS, FOTIS C, The Biological Laboratories, Harvard University, Cambridge, MA 02138
KALEY, GABOR, Department of Physiology, Basic Sciences Bldg., New York Medical College,
Valahalla, NY 10595
KALTENBACH, JANE, Department of Biological Sciences, Mount Holyoke College, South
Hadley, MA 01075
KAMINER, BENJAMIN, Department of Physiology, Boston University, School of Medicine,
Boston, MA 02 1 1 8
KAMMER, ANN E., Division of Biology, Kansas State University, Manhatten, KS 66506
KANE, ROBERT E., University of Hawaii, PBRC, 41 Ahui Street, Honolulu, HI 96813
KANESHIRO, EDNA S., Department of Biological Sciences, University of Cincinnati, Cincin-
nati, OH 45221
KAPLAN, EHUD, The Rockefeller University, 1230 York Avenue, New York, NY 10021
KARAKASHIAN, STEPHEN J., 165 West 91st Street, Apt. 16-F, New York, NY 10021
KARUSH, FRED, Department of Microbiology, School of Medicine, University of Pennsyl-
vania, Philadelphia, PA 19174
KATZ, GEORGE M., Department of Neurology, College of Physicians and Surgeons, Columbia
University, New York, NY 10032
KEAN, EDWARD L., Case Western Reserve University, Department of Ophthalmology and
Biochemistry, Cleveland, OH 44101
KELLY, ROBERT E., Department of Anatomy, College of Medicine, University of Illinois,
P.O. Box 6998, Chicago, IL 60680
KEMP, NORMAN E., Department of Zoology, University of Michigan, Ann Arbor, MI 48104
KENDALL, JOHN P., Fanueil Hall Associates, One Boston Place, Boston, MA 02108
KEYNAN, ALEXANDER, Vice President, Hebrew University, Jerusalem, Israel
KING, THOMAS J., Division of Cancer Research Resources and Center, NIH, Bldg. 31 Room
10A03, Bethesda, MD 20205
KINGSBURY, JOHN M., Department of Botany, Cornell University, Ithaca, NY 14853
KJRSCHENBAUM, DONALD, Department of Biochemistry, SUNY, 450 Clarkson Avenue,
Brooklyn, NY 11203
KLEIN, MORTON, Department of Microbiology, Temple University, Philadelphia, PA 19122
KLOTZ, I. M., Department of Chemistry, Northwestern University, Evanston, IL 60201
KOIDE, SAMUEL S., Population Council. The Rockefeller University, 1230 York Avenue,
New York, NY 10021
KONIGSBERG, IRWIN, R., Department of Biology, Gilmer Hall, University of Virginia, Char-
lottesville, VA 22903
KORNBERG, SIR HANS, Department of Biochemistry, University of Cambridge, Cambridge
CB2 7QW, England, U. K.
KOSOWER, EDWARD M., Department of Chemistry, Tel Aviv University, Tel Aviv, Israel
KRAHL, M. E., 2783 W. Casas Circle, Tucson, AZ 85741
KRANE, STEPHEN M., Massachusetts General Hospital, Boston, MA 021 14
KRASSNER, STUART M., Department of Developmental and Cell Biology, University of Cal-
ifornia, Irvine, CA 92717
KRAUSS, ROBERT, FASEB, 9650 Rockville Pike, Bethesda, MD 20205
16 MARINE BIOLOGICAL LABORATORY
KRAVITZ, EDWARD A., Department of Neurobiology, Harvard Medical School, Boston, MA
02115
KRIEBEL, MAHLON E., Department of Physiology, B.S.B., Upstate Medical Center, 766 Irving
Avenue, Syracuse, NY 13210
KRIEG, WENDELL J. S., 1236 Hinman, Evanston, IL 60602
KUHNS, WILLIAM J., University of North Carolina, 512 Faculty Lab Office Bldg. 231-H,
Chapel Hill, NC27514
KUSANO, KIYOSHI, Illinois Institute of Technology, Department of Biology, 3300 South
Federal Street, Chicago, IL 60616
LADERMAN, AIMLEE, Box 689, Teaneck, NJ 07666
LAMARCHE, PAUL H., Eastern Maine Medical Center, 489 State Street, Bangor, ME 04401
LANDIS, DENNIS M. D., Department of Neurology, Massachusetts General Hospital, Boston,
MA 02114
LANDIS, STORY C, Department of Neurobiology, Harvard Medical School, Boston, MA
02115
LANDOWNE, DAVID, Department of Physiology, University of Miami, R-430, P.O. Box
016430, Miami, FL 33101
LANGFORD, GEORGE M., Department of Physiology, University of North Carolina, Medical
Sciences Research Wing 206H, Chapel Hill, NC 27514
LASH, JAMES W., Department of Anatomy, School of Medicine, University of Pennsylvania,
Philadelphia, PA 19174
LASTER, LEONARD, President, University of Oregon, Health Sciences Center, Portland, OR
97201
LAUFER, HANS, Biological Sciences Group U-42, University of Connecticut, Storrs, CT 06268
LAUFFER, MAX A., Department of Biophysics, University of Pittsburgh, Pittsburgh, PA 1 5260
LAZAROW, JANE, 221 Woodlawn Avenue, St. Paul, MN 55106
LAZARUS, MAURICE, Federated Department Stores, Inc., 50 Cornhill, Boston, MA 02108
LEADBETTER, EDWARD R., Biological Sciences Group U-42, University of Connecticut,
Storrs, CT 06268
LEAK, LEE VIRN, Department of Anatomy, Howard University, Washington, DC 20001
LEDERBERG, JOSHUA, President, The Rockefeller University, 1230 York Avenue, New York,
NY 10021
LEDERHENDLER, IZJA I., Laboratory of Biophysics, Marine Biological Laboratory, Woods
Hole, MA 02543
LEE, JOHN J., Department of Biology, City College, Convent Avenue and 138th Street, New
York, NY 10031
LEFEVRE, PAUL G., Department of Physiology, Health Sciences Center, East Campus —
SUNY, Stony Brook, NY 1 1 794
LEIGHTON, JOSEPH, 1201 Waverly Road, Gladwyne, PA 19035
LEIGHTON, STEPHEN, NIH, Bldg. 13 Room 3W13, Bethesda, MD 20205
LENHER, SAMUEL, 50-C Cokesbury Village, Hockessin, DE 19707
LERMAN, SIDNEY, Laboratory for Ophthalmic Research, Emory University, Atlanta, GA
30322
LERNER, AARON B., Yale Medical School, New Haven, CT 06510
LEVIN, JACK, Clinical Pathology Service, Veterans Administration Hospital — 113A, 4150
Clement Street, San Francisco, CA 94120
LEVINE, RACHMIEL, 2024 Canyon Road, Arcadia, CA 91006
LEVINTHAL, CYRUS, Department of Biological Sciences, Columbia University, 908 Scher-
merhorn Hall, New York, NY 10027
LEVITAN, HERBERT, Department of Zoology, University of Maryland, College Park, MD
20742
LING, GILBERT, 307 Berkeley Road, Marion, PA 19066
LIPICKY, RAYMOND J., Laboratory of Biophysics, NIH, Bldg. 36 Room 2A29, Bethesda, MD
20205
LISMAN, JOHN E., Department of Biology, Brandeis University, Waltham, MA 02254
LITTLE, E. P., 216 Highland Street, West Newton, MA 02158
LIUZZI, ANTHONY, Department of Physics, University of Lowell, Lowell, MA 01854
MEMBERS OF THE CORPORATION 1 7
LLINAS, RODOLFO R., Department of Physiology and Biophysics, New York University,
Medical Center, New York, NY 10016
LOEWENSTEIN, WERNER R., Department of Physiology and Biophysics, University of Miami,
P.O. Box 016430, Miami, FL 33101
LOEWUS, FRANK A., Department of Agricultural Chemistry, Washington State University,
Pullman, WA 99164
LOFTFIELD, ROBERT B., Department of Biochemistry, School of Medicine, University of New
Mexico, 900 Stanford, NE, Alburquerque, NM 87105
LONDON, IRVING, M., Massachusetts Institute of Technology, 1 6-5 1 2, Cambridge, MA 02 1 38
LONGO, FRANK J., Department of Anatomy, University of Iowa, Iowa City, IA 65442
LORAND, LASZLO, Department of Biochemistry and Molecular Biology, Northwestern Uni-
versity, Evanston, IL 60201
LURIA, SALVADOR E., Department of Biology, Massachusetts Institute of Technology, Cam-
bridge, MA 02 1 39
LYNCH, CLARA J., 4800 Fillmore Avenue, Alexandria, VA 2231 1
MACAGNO, EDUARDO R., 1003B Fairchild, Columbia University, New York, NY 10022
MACNICHOL, E. F., JR., Laboratory of Sensory Physiology, Marine Biological Laboratory,
Woods Hole, MA 02543
MAHLER, ROBERT, Department of Biochemistry, Indiana University, Bloomington, IN 47401
MAINER, ROBERT, Senior Vice President, The Boston Company, One Boston Place, Boston,
MA 02108
MALKIEL, SAUL, Sidney Farber Cancer Center, 35 Binney Street, Boston, MA 021 16
MANALIS, RICHARD S., RR #4, Columbia City, IN 46725
MANGUM, CHARLOTTE P., Department of Biology, College of William and Mary, Williams-
burg, VA 23185
MARGULIS, LYNN, Department of Biology, Boston University, Boston MA 02215
MARINUCCI, ANDREW C, The Ecosystems Center, Marine Biological Laboratory, Woods
Hole, MA 02543
MARSH, JULIAN B., Department of Biochemistry and Physiology, Medical College of Penn-
sylvania, 3300 Henry Avenue, Philadelphia, PA 19129
MARTIN, LOWELL V., Marine Biological Laboratory, Woods Hole, MA 02543
MARUO, TAKESHI, Department of Obstetrics and Gynecology, Kobe University, Ikuta-ku,
Kobe 650, Japan
MASER, MORTON, 100 Hackmatak Way, Falmouth, MA 02540
MASTROIANNI, LUIGI, JR., Department of Obstetrics and Gynecology, University of Penn-
sylvania, Philadelphia, PA 19174
MATHEWS, RITA W., c/o A. J. Johnson, New York University, Medical Center, New York,
NY 10016
MAUTNER, HENRY G., Department of Biochemistry and Pharmacology, Tufts University,
136 Harrison Avenue, Boston, MA 021 1 1
MAUZERALL, DAVID, The Rockefeller University, 1230 York Avenue, New York, NY 10021
MAZIA, DANIEL, Hopkins Marine Station, Pacific Grove, CA 93950
McCANN, FRANCES, Department of Physiology, Dartmouth Medical School, Hanover, NH
03755
MCCLOSKEY, LAWRENCE R., Department of Biology, Walla Walla College, College Place,
WA 99324
MCLAUGHLIN, JANE A., P.O. Box 187, Woods Hole, MA 02543
McMAHON, ROVERT F., Department of Biology, Box 19498, University of Texas, Arlington,
TX 76019
MEEDEL, THOMAS, Boston University Marine Program, Marine Biological Laboratory,
Woods Hole, MA 02543
MEINERTZHAGEN, IAN A., Department of Psychology, Life Sciences Center, Dalhousie Uni-
versity, Halifax, Nova Scotia, Canada B3H 451
MEINKOTH, NORMAN A., Department of Biology, Swarthmore College, Swarthmore, PA
19081
MEISS, DENNIS E., Department of Biology, Clark University, Worcester, MA 01610
18 MARINE BIOLOGICAL LABORATORY
MELILLO, JERRY M., Ecosystems Center, Marine Biological Laboratory, Woods Hole, MA
02543
MELLON, DEFOREST, JR., Department of Biology, University of Virginia, Charlottesville, VA
22903
MELLON, RICHARD P., P.O. Box 187, Laughlintown, PA 15655
METUZALS, JANIS, Department of Anatomy, Faculty of Medicine, University of Ottawa,
Ottawa, Ontario, Canada KIN 9A9
METZ, CHARLES B., Institute of Molecular and Cellular Evolution, University of Miami, 521
Anastasia Avenue, Coral Gables, FL 33134
MIDDLEBROOK, ROBERT, 86 Station Road, Burley-in-Warfedale, West Yorks, England,
U. K.
MILKMAN, ROGER, Department of Zoology, University of Iowa, Iowa City, IA 52242
MILLS, ERIC L., Institute of Oceanography, Dalhousie University, Halifax, Novia Scotia
MILLS, ROBERT, 56 Worcester Court, Falmouth, MA 02540
MITCHELL, RALPH, Pierce Hall, Harvard University, Cambridge, MA 02138
MIZELL, MERLE, Department of Biology, Tulane University, New Orleans, LA 701 18
MONROY, ALBERTO, Stazione Zoologica, Villa Communale, Napoli, Italy
MONTROLL, ELIOTT W., Institute for Fundamental Studies, Department of Physics, Roch-
ester, NY 14627
MOORE, JOHN W., Department of Physiology, Duke University, Medical Center, Durham,
NC 27710
MOORE, LEE E., Department of Physiology and Biophysics, University of Texas, Medical
Branch, Galveston, TX 77550
MORAN, JOSEPH F., JR., 23 Foxwood Drive, RR #1, Eastham, MA 02642
MORIN, JAMES G., Department of Biology, University of California, Los Angeles, CA 90024
MORRELL, FRANK, Department of Neurological Sciences, Rush Medical Center, 1753 W.
Congress Parkway, Chicago, IL 60612
MORRILL, JOHN B., JR., Division of National Sciences, New College, Sarasota, FL 33580
MORSE, RICHARD S., 193 Winding River Road, Wellesley, MA 02181
MORSE, ROBERT W., Associate Director, Woods Hole Oceanographic Institution, Woods
Hole, MA 02543
MORSE, STEPHEN SCOTT, Department of Biological Sciences, Rutgers University, Nelson
Biological Laboratories, New Brunswick, NJ 08903
MOSCONA, A. A., Department of Biology, University of Chicago, 920 East 58th Street, Chi-
cago, IL 60637
MOTE, MICHAEL I., Department of Biology, Temple University, Philadelphia, PA 19122
MOUNTAIN, ISABEL, Vinson Hall #112, 6251 Old Dominion Drive, McLean, VA 22101
MULLEN, GEORGE, President, Mohawk Carpets, Amsterdam, NY 12010
MUSACCHIA, XAVIER J., Graduate School, University of Louisville, Louisville, KY 40295
NABRIT, S. M., 686 Beckwith Street, SW, Atlanta, GA 30314
NACE, PAUL F., 5 Bowditch Road, Woods Hole, MA 02543
NAKA, KEN-ICHI, National Institute for Basic Biology, Okazaki, Japan 444
NAKAJIMA, SHIGEHIRO, Department of Biological Sciences, Purdue University, West Lafay-
ette, IN 47907
NAKAJIMA, YASUKO, Department of Biological Sciences, Purdue University, West Lafayette,
IN 47907
NARAHASHI, TOSHIO, Department of Pharmacology, Medical Center, Northwestern Univer-
sity, 303 East Chicago Avenue, Chicago, IL 606 1 1
NASATIR, MAIMON, Department of Biology, University of Toledo, Toledo, OH 43606
NELSON, LEONARD, Medical College of Ohio, Department of Physiology, Toledo, OH 43699
NELSON, MARGARET C., Section on Neurobiology and Behavior, Cornell University, Ithaca,
NY 14850
NICHOLLS, JOHN G., Department of Neurobiology, Stanford University, Stanford, CA 94305
NICOSIA, SANTO V., Department of OB-GYN, Division of Reproductive Biology, University
of Pennsylvania, Philadelphia, PA 19174
NIELSEN, JENNIFER B. K., Waksman Institute for Microbiology, Piscataway, NJ 08854
MEMBERS OF THE CORPORATION 1 9
NOE, BRYAN D., Department of Anatomy, Emory University, Atlanta, GA 30345
OBAID, ANA LIA, Department of Physiology and Pharmacy, University of Pennsylvania,
School of Dental Medicine, 4001 Spruce Street, Philadelphia, PA 19104
OCHOA, SEVERO, 530 East 72nd Street, New York, NY 10021
ODUM, EUGENE, Department of Zoology, University of Georgia, Athens, GA 30701
OERTEL, DONATA, Department of Neurophysiology, University of Wisconsin, 283 Medical
Science Bldg., Madison, WI 53706
O'HERRON, JONATHAN, Lazard Freres and Company, 1 Rockefeller Plaza, New York, NY
10020
O'MELIA, ANNE F., George Mason University, 4400 University Drive, Fairfax, VA 22030
OLSON, JOHN M., Institute of Biochemistry, Odense University, Campusvej 55, DK 5230
Odense M, Denmark
OSCHMAN, JAMES L., Marine Biological Laboratory, Woods Hole, MA 02543
PALMER, DOUGLAS W., 21 Stanford Road, Wellesley, MA 02181
PALMER, JOHN D., Department of Zoology, University of Massachusetts, Amherst, MA 01002
PALTI, YORAM, Department of Physiology and Biophysics, Israel Institute of Technology. 12
Haaliya Street, BAT-GALIM,*POB 9649, Haifa/Israel
PANT, HARISH C., Laboratory for Preclinical Studies, National Institute on Alcohol Abuse
and Alcoholism, 12501 Washington Avenue, Rockville, MD 20852
PAPPAS, GEORGE D., Department of Anatomy, College of Medicine, University of Illinois,
808 South Wood Street, Chicago, IL 60612
PARDEE, ARTHUR B., Department of Pharmacology, Harvard Medical School, Boston, MA
02115
PARDY, ROSEVELT L., School of Life Sciences, University of Nebraska, Lincoln, NE 27710
PARMENTIER, JAMES L., Department of Anesthesiology, Duke University Medical Center,
Durham, NC27710
PASSANO, LEONARD M., Department of Zoology, Birge Hall, University of Wisconsin, Mad-
ison, WI 53706
PEARLMAN, ALAN L, Department of Physiology, School of Medicine, Washington University,
St. Louis, MO 63110
PEDERSON, THORU, Worcester Foundation for Experimental Biology, Shrewsbury, MA 01545
PERKINS, C. D., National Academy of Engineering, 2101 Constitution Avenue, NW, Wash-
ington, DC 20418
PERSON, PHILIP, Special Dental Research Program, Veterans Administration Hospital. Brook-
lyn, NY 11219
PETERSON, BRUCE J., Ecosystems Center, Marine Biological Laboratory, Woods Hole, MA
02543
PETHIG, RONALD, School of Electronic Engineering Science, University College of North
Wales, Bangor, Gwynedd, LL57 1UT
PETTIBONE, MARIAN H., Division of Worms, W-213, Smithsonian Institution, Washington,
DC 20560
PFOHL, RONALD J., Department of Zoology, Miami University, Oxford, OH 45056
PIERCE, SIDNEY K., JR., Department of Zoology, University of Maryland, College Park, MD
20740
POLLARD, HARVEY B., NIH, F Building 10 Room 10B17, Bethesda. MD 20205
POLLARD, THOMAS, D., Director, Department of Cell Biology and Anatomy, Johns Hopkins
University, 725 North Wolfe Street, Baltimore, MD 21205
POLLOCK, LELAND W., Department of Zoology, Drew University, Madison, NJ 07940
PORTER, BEVERLY H., 14433 Taos Court, Wheaton, MD 20906
PORTER, KEITH R., 748 Eleventh Street, Boulder, CO 80302
POTTER, DAVID, Department of Neurobiology, Harvard Medical School, Boston, MA 02 1 1 5
POTTER, H. DAVID, Neural Sciences Program, Chemistry Building, Indiana University,
Bloomington, IN 47404
POTTS, WILLIAM T., Department of Biology, University of Lancaster, Lancaster, England,
U. K.
POUSSART, DENIS, Department of Electrical Engineering, Universite Laval, Quebec, Canada
20 MARINE BIOLOGICAL LABORATORY
PRENDERGAST, ROBERT A., Department of Pathology and Ophthalmology, Johns Hopkins
University, Baltimore, MD 21205
PRICE, CARL A., Waksman Institute of Microbiology, Rutgers University, P.O. Box 759,
Piscataway, NJ 08854
PRICE, CHRISTOPHER H., Biological Science Center, 2 Cummington Street, Boston, MA
02215
PRIOR, DAVID J., Department of Biological Sciences, University of Kentucky, Lexington,
KY 40506
PROVASOLI, LUIGI, Haskins Laboratories, 165 Prospect Street, New Haven, CT 06510
PRUSCH, ROBERT D., Department of Life Sciences, Gonzaga University, Spokane, WA 99258
PRZYBYLSKJ, RONALD J., Department of Anatomy, Case Western Reserve University, Cleve-
land, OH 44104
QUIGLEY, JAMES, Department of Microbiology and Immunology, SUNY, Downstate Medical
Center, Brooklyn, NY 11203
RABIN, HARVEY, P.O. Box 239, Braddock Heights, MD 21714
RAFF, RUDOLF A., Department of Biology, Indiana University, Bloomington, IN 47405
RAKOWSKI, ROBERT F., Department of Physiology and Biophysics, Washington University
School of Medicine, St. Louis, MO 631 10
RAMON, FIDEL, Departamento de Fisilogia y Biofisca, Centre de Investigacion y de Estudius
Avanzados del 1PN, Apurtado Postal 14-740, Mexico D. F., Mexico 07000
RANZI, SILVIO, Department of Zoology, University of Milan, Milan, Italy
RATNER, SARAH, Department of Biochemistry, Public Health Research Institute, 455 First
Avenue, New York, NY 10016
REBHUN, LIONEL I., Department of Biology, Gilmer Hall, University of Virginia, Charlottes-
ville, VA 22901
REDDAN, JOHN R., Department of Biological Sciences, Oakland University, Rochester, MI
48063
REDFIELD, ALFRED C, 10 Maury Lane, Woods Hole, MA 02543 (Deceased March 1983)
REESE, THOMAS S., Section on Functional Neuroanatomy, NIH, Bethesda, MD 20205
REINER, JOHN M., Department of Biochemistry, Albany Medical College of Union Univer-
sity, Albany, NY 12208
REINISCH, CAROL L., Tufts University School of Veterinary Medicine, 203 Harrison Avenue,
Boston, MA 02115
REUBEN, JOHN P., Department of Neurology, College of Physicians and Surgeons, Columbia
University, New York, NY 10032
REYNOLDS, GEORGE T., Department of Physics, Jadwin Hall, Princeton University, Prince-
ton, NJ 08540
RICE, ROBERT V., Carnegie Mellon Institute, 4400 Fifth Avenue, Pittsburgh, PA 15213
RICKLES, FREDERICK R., University of Connecticut, School of Medicine, Veterans Admin-
istration Hospital, Newington, CT 06 1 1 1
RIPPS, HARRIS, Department of Ophthalmology, School of Medicine, New York University,
New York, NY 10016
ROBERTS, JOHN L., Department of Zoology, University of Massachusetts, Amherst, MA
01002
ROBINSON, DENIS M., High Voltage Engineering Corporation, Burlington, MA 01803
ROCKSTEIN, MORRIS, 335 Fluzia Avenue, Miami, FL 33134
RONKIN, RAPHAEL R., 3212 McKinley Street, NW, Washington, DC 20015
ROSBASH, MICHAEL, Rosenstiel Basic Medical Research Center, Department of Biology,
Brandeis University, Waltham, MA 02154
ROSE, BIRGIT, Department of Physiology R-430, School of Medicine, University of Miami,
P.O. Box 016430, Miami, FL 33152
ROSE, S. MERYL, Box 309W, Waquoit, MA 02536
ROSENBAUM, JOEL L., Department of Biology, Kline Biology Tower, Yale University, New
Haven, CT06510
ROSENBERG, PHILIP, School of Pharmacy, Division of Pharmacology, University of Con-
necticut, Storrs, CT 06268
MEMBERS OF THE CORPORATION 2 1
ROSENBLUTH, JACK, Department of Physiology, School of Medicine, New York University,
550 First Avenue, New York, NY 10016
ROSENBLUTH, RAJA, 3380 West 5th Avenue, Vancouver 8 BC, Canada V6R 1R7
ROSENK.RANZ, HERBERT S., Department of Microbiology, New York Medical College, Val-
halla, NY 10595
ROSLANSKY, JOHN, Box 208, Woods Hole, MA 02543
ROSLANSKY, PRISCILLA F., Box 208, Woods Hole, MA 02543
Ross, WILLIAM N., Department of Physiology, New York Medical College, Valhalla, NY
10595
ROTH, JAY S., Division of Biological Sciences, Section of Biochemistry and Biophysics, Uni-
versity of Connecticut, Storrs, CT 06268
ROWE, DOROTHY, 88 Chestnut Hill, Boston, MA 02165
ROWLAND, LEWIS P., Neurological Institute, 710 West 168th Street, New York, NY 10032
RUDERMAN, JOAN V., Department of Anatomy, Harvard Medical School, Boston, MA 02 1 1 5
RUSHFORTH, NORMAN B., Department of Biology, Case Western Reserve University, Cleve-
land, OH 44106
RUSSELL-HUNTER, W. D., Department of Biology, 1 10 Lyman Hall, Syracuse University,
Syracuse, NY 13210
RUSTAD, RONALD C., Radiology Department, Case Western Reserve University, Cleveland,
OH 44106
SAGER, RUTH, Sidney Farber Cancer Institute, 44 Binney Street, Boston, MA 021 15
SALAMA, GUY, Department of Physiology, University of Pittsburgh, Pittsburgh, PA 15261
SALMON, EDWARD D., Department of Zoology, University of North Carolina, Chapel Hill,
NC 27514
SALZBERG, BRIAN H., Department of Physiology, University of Pennsylvania, 4010 Locust
Street, Philadelphia, PA 19174
SANDERS, HOWARD, Woods Hole Oceanographic Institution, Woods Hole, MA 02543
SANGER, JEAN M., Department of Anatomy, School of Medicine, University of Pennsylvania,
36th and Hamilton Walk, Philadelphia, PA 19174
SANGER, JOSEPH, Department of Anatomy, School of Medicine, University of Pennsylvania,
36th and Hamilton Walk, Philadelphia, PA 19174
SATO, HIDEMI, Sugashima Marine Biological Laboratory, Nagoya University, Sugashima-
cho, Toba-shi, Mie-Ken 517, Japan
SAUNDERS, JOHN, JR., Department of Biological Sciences, SUNY, Albany, NY 12222
SAZ, ARTHUR K., Medical and Dental Schools, Georgetown University, 3900 Reservoir Road,
NW, Washington, DC 2005 1
SCHACHMAN, HOWARD K., Department of Molecular Biology, University of California,
Berkeley, CA 94720
SCHIFF, JEROME A., Institute for Photobiology of Cells and Organelles, Brandeis University,
Waltham, MA 02154
SCHLESINGER, R. WALTER, Department of Microbiology, College of Medicine and Dentistry,
Rutgers University, P.O. Box 101, Piscataway, NJ 08854
SCHMEER, SISTER ARLINE C., Mercenene Cancer Research Hospital of Saint Raphael, New
Haven, CT 06511
SCHNEIDERMAN, HOWARD K., Monsanto Company, 800 North Lindberg Blvd., D1W, St.
Louis, MO 63 166
SCHOPF, THOMAS, J. M., Department of Geophysical Sciences, University of Chicago, 5734
South Ellis Avenue, Chicago, IL 60637
SCHOTTE, OSCAR E., Department of Biology, Amherst College, Amherst, MA 01002
SCHUEL, HERBERT, Department of Anatomical Sciences, SUNY, Buffalo, NY 14214
SCHUETZ, ALLEN W., School of Hygiene and Public Health, Johns Hopkins University,
Baltimore, MD 21205
SCHWAB, WALTER E., Department of Biology, Virginia Polytechnical Institute and State
University, Blacksburg, VA 24601
SCHWARTZ, JAMES H., College of Physicians and Surgeons, Columbia University, New York,
NY 10032
22 MARINE BIOLOGICAL LABORATORY
SCHWARTZ, MARTIN, Department of Biological Sciences, University of Maryland, Baltimore
County, Catonsville, MD 21228
SCHWARTZ, TOBIAS L., Biological Sciences Group, University of Connecticut, Storrs, CT
06268
SCOTT, ALLAN C, 1 Nudd Street, Waterville, ME 04901
SCOTT, GEORGE T., 10 Orchard Street, Woods Hole, MA 02543
SEARS, MARY, P.O. Box 152, Woods Hole, MA 02543
SEGAL, SHELDON J., Director, Population Division, The Rockefeller Foundation, 1 133 Av-
enue of the Americas, New York, NY 10036
SELIGER, HOWARD H., Johns Hopkins University, McCollum-Pratt Institute, Baltimore, MD
21218
SELMAN, KELLY, Department of Anatomy, College of Medicine, University of Florida,
Gainesville, FL 32601
SENFT, JOSEPH, 378 Fairview Street, Emmaus, PA 18049
SHANKLIN, DOUGLAS R., P.O. Box 1267, Gainesville, FL 32602
SHAPIRO, HERBERT, 6025 North 13th Street, Philadelphia, PA 19141
SHAVER, GAIUS R., Ecosystems Center, Marine Biological Laboratory, Woods Hole, MA
02543
SHAVER, JOHN R., Department of Zoology, Michigan State University, E. Lansing, MI 48823
SHEPARD, DAVID C., P.O. Box 44, Woods Hole, MA 02543
SHEPRO, DAVID, Department of Biology, Boston University, Boston, MA 02215
SHERMAN, I. W., Division of Life Sciences, University of California, Riverside, CA 92502
SHILO, MOSHE, Head, Department of Microbiological Chemistry, Hebrew University, Je-
rusalem, Israel
SHOUKIMAS, JONATHAN J., Laboratory of Biophysics, NINCDS, Marine Biological Labo-
ratory, Woods Hole, MA 02543
SHRIVASTAV, BRIJ S., Department of Pharmacology, Duke University Medical Center, Dur-
ham, NC27710
SIEGEL, IRWIN M., Department of Ophthalmology, New York University, Medical Center,
New York, NY 10016
SIEGELMAN, HAROLD W., Department of Biology, Brookhaven National Laboratory, Upton,
NY 11973
SIMON, ERIC J., New York University, Medical School, New York, NY 10016
SJODIN, RAYMOND A., Department of Biophysics, University of Maryland, Baltimore, MD
21201
SKINNER, DOROTHY M., Biology Division, Oak Ridge National Laboratory, Oak Ridge, TN
37830
SLOBODA, ROGER D., Department of Biological Sciences, Dartmouth College, Hanover, NH
03755
SLOBODKIN, LAWRENCE B., Department of Biology, SUNY, Stony Brook, NY 1 1790
SMITH, HOMER P., General Manager, Marine Biological Laboratory, Woods Hole, MA 02543
SMITH, MICHAEL A., Foreign and Commonwealth Office, King Charles Street, London SW1 A
2AH, England, U. K.
SMITH, PAUL F., P.O. Box 264, Woods Hole, MA 02543
SMITH, RALPH I., Department of Zoology, University of California, Berkeley, CA 94720
SORENSON, ALBERT L., Department of Physiology, Albert Einstein College of Medicine, 1300
Morris Park Avenue, Bronx, NY 10461
SPECK, WILLIAM T., Department of Pediatrics, Case Western Reserve University, Cleveland,
OH 44106
SPECTOR, A., College of Physicians and Surgeons, Black Bldg. Room 1516, Columbia Uni-
versity, New York, NY 10032
SPIEGEL, EVELYN, Department of Biological Sciences, Dartmouth College, Hanover, NH
02755
SPIEGEL, MELVIN, Department of Biological Sciences, Dartmouth College, Hanover, NH
02755
MEMBERS OF THE CORPORATION 23
SPRAY, DAVID C., Department of Neurosciences, Albert Einstein College of Medicine, 1300
Morris Park Avenue, Bronx, NY 10461
STARZAK, MICHAEL E., Department of Chemistry, SUNY, Binghamton, NY 13901
STEELE, JOHN HYSLOP, Director, Woods Hole Oceanographic Institution, Woods Hole, MA
02543
STEINACHER, ANTOINETTE, Department of Biophysics, The Rockefeller University, New
York, NY 10021
STEINBERG, MALCOLM, Department of Biology, Princeton University, Princeton, NJ 08540
STEPHENS, GROVER C., Department of Developmental and Cell Biology, University of Cal-
ifornia, Irvine, CA 92717
STEPHENS, RAYMOND E., Marine Biological Laboratory, Woods Hole, MA 02543
STETTEN, MARJORIE R., NIH, Bldg. 10 9B-02, Bethesda, MD 20205 (Deceased May 1983)
STETTEN, DEWITT, JR., Senior Scientific Advisor, NIH, Bldg. 16 Room 1 18, Bethesda, MD
20205
STOKES, DARRELL R., Department of Biology, Emory University, Atlanta, GA 30322
STRACHER, ALFRED, Downstate Medical Center, SUNY, 450 Clarkson Avenue, Brooklyn,
NY 11203
STREHLER, BERNARD L., 2235 25th Street, #217, San Pedro, CA 90732
STUART, ANN E., Medical Sciences Research Wing 206H, Department of Physiology, Uni-
versity of North Carolina, Chapel Hill, NC 27514
SUMMERS, WILLIAM C., Huxley College, Western Washington State College, Bellingham,
WA 98225
SUSSMAN, MAURICE, Department of Life Sciences, University of Pittsburgh, Pittsburgh, PA
15260
SWENSON, RANDOLPHS P., JR., Department of Physiology G-4, University of Pennsylvania,
Philadelphia, PA 19174
SZABO, GEORGE, Harvard School of Dental Medicine, 188 Longwood Avenue, Boston, MA
02 1 1 5
SZAMIER, R. BRUCE, Harvard Medical School, Berman-Gund Laboratory, Massachusetts Eye
and Ear Infirmary, 243 Charles Street, Boston, MA 021 14
SZENT-GYORGYI, ALBERT, Marine Biological Laboratory, Woods Hole, MA 02543
SZENT-GYORGYI, ANDREW, Department of Biology, Brandeis University, Waltham, MA
02154
TAKASHIMA, SHIRO, Department of Bioengineering, University of Pennsylvania, Philadel-
phia, PA 19174
TAMM, SIDNEY L., Boston University Marine Program, Marine Biological Laboratory, Woods
Hole, MA 02543
TANZER, MARVIN L., Department of Biochemistry, Box G, Medical School, University of
Connecticut, Farmington, CT 06032
TASAKI, ICHIJI, Laboratory of Neurobiology, NIMH, NIH, Bethesda, MD 20205
TAYLOR, DOUGLASS L., Biological Sciences, Mellon Institute, 4400 Fifth Avenue, Pittsburgh,
PA 15213
TAYLOR, ROBERT E., Laboratory of Biophysics. NINCDS, NIH, Bethesda, MD 20205
TAYLOR, W. ROWLAND, 4800 Atwell Road, Shady Side. MD 20764
TELFER, WILLIAM H., Department of Biology, University of Pennsylvania, Philadelphia, PA
19174
THORNDIKE, W. NICHOLAS, Wellington Management Company, 28 State Street, Boston,
MA 02 109
TIFFNEY, WESLEY N., 226 Edge Hill Road, Sharon, MA 02067 (Deceased January 1983)
TRACER, WILLIAM, The Rockefeller University, 1230 York Avenue, New York, NY 10021
TRAVIS, D. M., Veterans Administration Medical Center, Fargo, ND 58102
TREISTMAN, STEVEN N., Worcester Foundation for Experimental Biology, Shrewsbury, MA
01545
TRIGG, D. THOMAS, 1215 Grove Street, Wellesley, MA 02181
TRINKAUS, J. PHILIP, Osborn Zoological Labs, Department of Zoology, Yale University, New
Haven, CT 06510
24 MARINE BIOLOGICAL LABORATORY
TROLL, WALTER, Department of Environmental Medicine, College of Medicine, New York
University, New York, NY 10016
TROXLER, ROBERT F., Department of Biochemistry, School of Medicine, Boston Universtiy,
80 East Concord Street, Boston, MA 021 18
TURNER, RUTH D., Mollusk Department, Museum of Comparative Zoology, Harvard Uni-
versity, Cambridge, MA 02138
TWEEDELL, KENYON S., Department of Biology, University of Notre Dame, Notre Dame,
IN 46656
TYTELL, MICHAEL, Department of Anatomy, Bowman Gray School of Medicine, Winston-
Salem, NC 27103
URETZ, ROBERT B., Division of Biological Sciences, University of Chicago, 950 East 59th
Street, Chicago, IL 60637
VALIELA, IVAN, Boston University Marine Program, Marine Biological Laboratory, Woods
Hole, MA 02543
VALOIS, JOHN, Marine Biological Laboratory, Woods Hole, MA 02543
VAN HOLDE, KENSAL, Department of Biochemistry and Biophysics, Oregon State University,
Corvallis, OR 97331
VILLEE, CLAUDE A., Department of Biological Chemistry, Harvard Medical School, Boston,
MA 02115
VINCENT, WALTER S., School of Life and Health Sciences, University of Delaware, Newark,
DE 19711
WAINIO, WALTER W., Box 1059 Nelson Labs, Rutgers Biochemistry, Piscataway, NJ 08854
WAKSMAN, BYRON, National Multiple Sclerosis Society, 205 East 42nd Street, New York,
NY 10017
WALKER, CHARLES A., 3113 Shamrock South, Tallahassee, FL 32303
WALL, BETTY, Marine Biological Laboratory, Woods Hole, MA 02543
WALLACE, ROBIN A., Department of Anatomy, College of Medicine, University of Florida,
Gainesville, FL 32610
WANG, AN, Bedford Road, Lincoln, MA 01773
WARNER, ROBERT C., Department of Molecular Biology and Biochemistry, University of
California, Irvine, CA 927 1 7
WARREN, KENNETH S., The Rockefeller Foundation, 1133 Avenue of the Americas, New
York, NY 10036
WARREN, LEONARD, Department of Therapeutic Research, School of Medicine, Anatomy-
Chemistry Building Room 337, University of Pennsylvania, Philadelphia, PA 19174
WATERMAN, T. H., Yale University, 610 Kline Biology Tower, New Haven, CT 06510
WATSON, STANLEY, Woods Hole Oceanographic Institution, Woods Hole, MA 02543
WEBB, H. MARGUERITE, Marine Biological Laboratory, Woods Hole, MA 02543
WEBER, ANNEMARIE, Department of Biochemistry, School of Medicine, University of Penn-
sylvania, Philadelphia, PA 19174
WEBSTER, FERRIS, 800 25th Street, NW, Washington, DC 20037
WEIDNER, EARL, Department of Zoology and Physiology, Louisiana State University, Baton
Rouge, LA 70803
WEISS, LEON P., Department of Animal Biology, School of Veterinary Medicine, University
of Pennsylvania, Philadelphia, PA 19174
WEISSMANN, GERALD, Division of Rheumatology, New York University, School of Medicine,
New York, NY 10016
WERMAN, ROBERT, Neurobiology Unit, The Hebrew University, Jerusalem, Israel
WESTERFIELD, R. MONTE, The Institute of Neuroscience, University of Oregon, Eugene, OR
37403
WHITTAKER, J. RICHARD, Director, Boston University Marine Program, Marine Biological
Laboratory, Woods Hole, MA 02543
WIERCINSKI, FLOYD J., Department of Biology, Northeastern Illinois University, 5500 North
St. Louis Avenue, Chicago, IL 60625
WIGLEY, ROLAND L., 35 Wilson Road, Woods Hole, MA 02543
MEMBERS OF THE CORPORATION
25
WILBER, CHARLES G., Department of Zoology, Colorado State University, Fort Collins, CO
80523
WILSON, DARCY B., Department of Pathology, School of Medicine, University of Pennsyl-
vania, Philadelphia, PA 19174
WILSON, EDWARD O., Department of Zoology, Harvard University, Cambridge, MA 02138
WILSON, T. HASTINGS, Department of Physiology, Harvard Medical School, Boston, MA
02115
WILSON, WALTER L., Department of Biology, Oakland University, Rochester, MI 48063
WITKOVSKY, PAUL, Department of Ophthalmology, New York University, Medical Center,
New York, NY 10016
WITTENBERG, JONATHAN B., Department of Physiology and Biochemistry, Albert Einstein
College of Medicine, 1300 Morris Park Avenue, Bronx, NY 10461
WOELKERLING, WILLIAM J., Department of Botany, Latrobe University, Bundoora, Victoria,
Australia 3083
WOLF, DON P., Department of OB-GYN, University of Texas Health Sciences Center, 6431
Fannin, Houston, TX 77030
WOODWELL, GEORGE M., Director, Ecosystems Center, Marine Biological Laboratory,
Woods Hole, MA 02543
Wu, CHAU HSIUNG, Department of Pharmacology, Northwestern University Medical School,
Chicago, IL 60611
WYTTENBACH, CHARLES R., Department of Physiology and Cell Biology, University of Kan-
sas, Lawrence, KS 06045
YAMIN, MICHAEL A., The Rockefeller University, 1230 York Avenue, New York, NY 10021
YEH, JAY Z., Department of Pharmacology, Northwestern University Medical School, 303
E. Chicago Avenue, Chicago, IL 6061 1
YOUNG, RICHARD, 100 Royalston Road, Wellesley Hills, MA 02181
YPHANTIS, DAVID A., Department of Biochemistry and Biophysics, University of Connec-
ticut, Storrs, CT 06268
ZIGMAN, SEYMOUR, School of Medicine and Dentistry, University of Rochester, 260 Crit-
tenden Blvd., Rochester, NY 14620
ZIMMERMAN, A. M., Department of Zoology, University of Toronto, Toronto 5, Ontario,
Canada
ZUCKER, ROBERT S., Department of Physiology, University of California, Berkeley, CA 94720
ASSOCIATE MEMBERS
ACKROYD, DR. AND MRS. FREDERICK W.
ADELBERG, DR. AND MRS. EDWARD A.
ADELMAN, DR. AND MRS. WILLIAM J.
AHEARN, MR. AND MRS. DAVID C.
ALLEN, Miss CAMILLA K.
ALLEN, DRS. ROBERT D. AND NINA S.
AMBERSON, MRS. WILLIAM R.
ANDERSON, DRS. JAMES L. AND HELENE
M.
ARMSTRONG, DR. AND MRS. SAMUEL C.
ARNOLD, DR. AND MRS. JOHN M.
ATWOOD, DR. AND MRS. K.IMBALL C.
BALL, MRS. ERIC G.
BALLANTINE, DR. AND MRS. H. T., JR.
BANG, MRS. FREDERIK B.
BANKS, MR. AND MRS. WILLIAM L.
BARROWS, MRS. ALBERT W.
BEERS, DR. AND MRS. YARDLEY
BENNETT, DR. AND MRS. MICHAEL V. L.
BERNHEIMER, DR. ALAN W.
BERNSTEIN, MR. AND MRS. NORMAN
BIGELOW, MRS. ROBERT O.
BLACKBURN, DR. AND MRS. GEORGE L.
BODEEN, MR. AND MRS. GEORGE H.
BOETTIGER, DR. AND MRS. EDWARD G.
BOLTON, MR. AND MRS. THOMAS C.
BORGESE, DR. AND MRS. THOMAS A.
BOTKIN, DR. and MRS. DANIEL B.
BOWLES, DR. AND MRS. FRANCIS P.
BRADLEY, DR. AND MRS. CHARLES C.
BRONSON, MRS. SAMUEL C.
BROWN, MRS. DUGALD E. S.
BROWN, DR. AND MRS. FRANK A., JR.
BROWN, DR. AND MRS. THORNTON
BUCK, MRS. JOHN B.
BUFFINGTON, MRS. ALICE H.
BUFFINGTON, MRS. GEORGE
BURGER, DR. AND MRS. MAX M.
BURROUGH, MRS. ARNOLD H.
BURT, MR. AND MRS. CHARLES E.
26
MARINE BIOLOGICAL LABORATORY
BUTLER, MR. AND MRS. RHETT W.
CALKINS, MR. AND MRS. G. N., JR.
CAMPBELL, DR. AND MRS. DAVID G.
CAMPBELL, MR. AND MRS.
WORTHINGTON, JR.
CAPOBIANCO, MR. AND MRS. PAT J.
CARLSON, DR. AND MRS. FRANCIS
CARLTON, MR. AND MRS. WINSLOW G.
CASHMAN, MR. AND MRS. EUGENE R.
CHAMBERS, DR. AND MRS. EDWARD L.
CHENEY, DR. AND MRS. RALPH H.
CLAFF, MR. AND MRS. MARK
CLARK, MR. AND MRS. HAYS
CLARK, MR. AND MRS. JAMES McC.
CLARK, DR. AND MRS. LEONARD B.
CLARK, MR. AND MRS. LEROY, JR.
CLARK, MRS. W. VAN ALAN
CLEMENT, DR. AND MRS. A. C.
CLOWES FUND, INC.
CLOWES, DR. AND MRS. ALEXANDER W.
CLOWES, MR. ALLEN W.
CLOWES, DR. AND MRS. G. H. A., JR.
COHEN, DR. AND MRS. SEYMOUR
COLEMAN, DR. AND MRS. JOHN
CONNELL, MR. AND MRS. W. J.
COOPER, MR. AND MRS. JOHN H., JR.
COPELAND, MRS. D. EUGENE
COPELAND, MR. AND MRS. PRESTON S.
COSTELLO, MRS. DONALD P.
CRAIN, MR. AND MRS. MELVIN C.
CRAMER, MR. AND MRS. IAN D. W.
CRANE, MRS. JOHN
CRANE, JOSEPHINE B., FOUNDATION
CRANE, MRS. W. CAREY
CROSS, MR. AND MRS. NORMAN C.
CROSSLEY, MR. AND MRS. ARCHIBALD M.
CROWELL, DR. AND MRS. SEARS
DAIGNAULT, MR. AND MRS.
ALEXANDER T.
DANIELS, MR. AND MRS. BRUCE G.
DAVIS, MR. AND MRS. JOEL F.
DAY, MR. AND MRS. POMEROY
DICKSON, DR. WILLIAM A.
DRUMMOND, MR. AND MRS. A. H., JR.
DuBois, DR. AND MRS. ARTHUR B.
DUNKERLEY, MR. AND MRS. H. GORDON
DUPONT, MR. A. FELIX, JR.
DYER, MR. AND MRS. ARNOLD W.
EBERT, DR. AND MRS. JAMES D.
EDWARDS, DR. AND MRS. ROBERT L.
EGLOFF, DR. AND MRS. F. R. L.
ELLIOTT, MRS. ALFRED M.
ELSMITH, MRS. DOROTHY O.
EPPEL, MR. AND MRS. DUDLEY
EVANS, MR. AND MRS. DUDLEY
EWING, DR. AND MRS. GIFFORD C.
FENNO, MRS. EDWARD N.
FERGUSON, DR. AND MRS. JAMES J., JR.
FINE, DR. AND MR. JACOB
FISHER, MRS. B. C.
FISHER, MR. FREDERICK S., Ill
FISHER, DR. AND MRS. SAUL H.
FRANCIS, MR. AND MRS. LEWIS W., JR.
FRIENDSHIP FUND
FRIES, DR. AND MRS. E. F. B.
FYE, DR. AND MRS. PAUL M.
GABRIEL, DR. AND MRS. MORDECAI L.
GAISER, DR. AND MRS. DAVID W.
GARFIELD, Miss ELEANOR
CARREY, DR. AND MRS. WALTER E.
GELLIS, DR. AND MRS. SYDNEY
GERMAN, DR. AND MRS. JAMES L., Ill
GIFFORD, MR. AND MRS. JOHN A.
GIFFORD, DR. AND MRS. PROSSER
GILBERT, DR. AND MRS. DANIEL L.
GILBERT, MRS. CARL J.
GILDEA, DR. MARGARET C. L.
GILLETTE, MR. AND MRS. ROBERT S.
GLASS, DR. AND MRS. H. BENTLEY
GLAZEBROOK, MRS. JAMES R.
GLUSMAN, DR. AND MRS. MURRAY
GOLDMAN, DR. AND MRS. ALLEN S.
GOLDSTEIN, MR. AND MRS. MOISE H., JR.
GRANT, DR. AND MRS. PHILIP
GRASSLE, MR. AND MRS. J. F.
GREEN, Miss GLADYS M.
GREENE, MR. AND MRS. WILLIAM C.
GREER, MR. AND MRS. W. H., JR.
GROSCH, DR. AND MRS. DANIEL S.
GROSS, MRS. PAUL C.
GRUSON, MRS. MARTHA R.
GUNNING, MR. AND MRS. ROBERT
HAAKONSEN, DR. HARRY O.
HALVORSON, DR. AND MRS. HARLYN O.
HANDLER, MRS. PHILIP
HARVEY, DR. AND MRS. RICHARD B.
HASSETT, MR. AND MRS. CHARLES
HASTINGS, DR. AND MRS. J. WOODLAND
HEFFRON, DR. AND MRS. RODERICK
HENLEY, DR. CATHERINE
HIATT, DR. AND MRS. HOWARD
HILL, MRS. SAMUEL E.
HlLSINGER, MR. AND MRS. ARTHUR
HlRSCHFELD, MRS. NATHAN B.
HOBBIE, DR. AND MRS. JOHN
HOCKER, MR. AND MRS. LON
HOFFMAN, REV. AND MRS. CHARLES
HORWITZ, DR. AND MRS. NORMAN H.
HOUSTON, MR. AND MRS. HOWARD E.
HUETTNER, DR. AND MRS. ROBERT J.
HUNZIKER, MR. AND MRS. HERBERT E.
HYNES, MR. AND MRS. THOMAS J. JR.
MEMBERS OF THE CORPORATION
27
INOUE, DR. AND MRS. SHINYA
IRELAND, MRS. HERBERT A.
ISSOKSON, MR. AND MRS. ISRAEL
IVENS, DR. SUE
JACKSON, Miss ELIZABETH B.
JANNEY, MRS. F. WISTAR
JEWETT, G. F., FOUNDATION
JEWETT, MR. AND MRS. G. F., JR.
JONES, MR. AND MRS. DEWITT C., Ill
JONES, MR. AND MRS. FREDERICK, III
JORDAN, DR. AND MRS. EDWIN P.
KAAN, DR. HELEN W.
KAHLER, MR. AND MRS. GEORGE A.
KAHLER, MR. AND MRS. ROBERT W.
KAMINER, DR. AND MRS. BENJAMIN
KARUSH, DR. AND MRS. FRED
KEITH, MRS. JEAN R.
KELLEHER, MR. AND MRS. PAUL R.
KENDALL, MR. RICHARD E.
KEOSIAN, MRS. JESSIE
KlEN, MR. AND MRS. PlETER
KINNARD, MRS. L. RICHARD
KIVY, DR. AND MRS. PETER
KOHN, DR. AND MRS. HENRY I.
KOLLER, DR. AND MRS. LEWIS R.
KUFFLER, MRS. STEPHEN W.
LADERMAN, MR. AND DR. AIMLEE EZRA
LASH, DR. AND MRS. JAMES
LASTER, DR. AND MRS. LEONARD
LAUFER, DR. AND MRS. HANS
LAVIGNE, MRS. RICHARD J.
LAWRENCE, MR. FREDERICK V.
LAWRENCE, MR. AND MRS. WILLIAM
LAWRENCE SAUNDERS FUND
LAZAROW, MRS. ARNOLD
LEATHERBEE, MRS. JOHN H.
LEMANN, MRS. LUCY B.
LENHER, DR. AND MRS. SAMUEL
LEVINE, DR. AND MRS. RACHMIEL
LEWIS, MR. JOHN T.
LITTLE, MRS. ELBERT
LOEB, MRS. ROBERT F.
LOVELL, MR. AND MRS. HOLLIS R.
LOWE, DR. AND MRS. CHARLES W.
LOWENGARD, MRS. JOSEPH
MACKEY, MR. AND MRS. WILLIAM K.
MACLEISH, MRS. MARGARET
MACNARY, MR. AND MRS. B. GLENN
MACNlCHOL, DR. AND MRS. EDWARD F.,
JR.
MAHER, Miss ANNE CAMILLE
MARKS, DR. AND MRS. PAUL A.
MARSLAND, DR. DOUGLAS
MARTYNA, MR. AND MRS. JOSEPH C.
MARVIN, DR. DOROTHY H.
MASER, DR. AND MRS. MORTON
MASTROIANNI, DR. AND MRS. LUIGI, JR.
MATHER, MR. AND MRS. FRANK J., Ill
MATTHIESSEN, MR. AND MRS. G. C.
MCCUSKER, MR. AND MRS. PAUL T.
MCELROY, MRS. NELLA W.
MCLANE, MRS. T. THORNE
MEIGS, MR. AND MRS. ARTHUR
MEIGS, DR. AND MRS. J. WISTER
MELILLO, DR. AND MRS. JERRY M.
MELLON, RICHARD KING, TRUST
MELLON, MR. AND MRS. RICHARD P.
MENKE, DR. W. J.
METZ, DR. AND MRS. CHARLES B.
MEYERS, MR. AND MRS. RICHARD
MILLER, DR. DANIEL A.
MIXTER, MR. AND MRS. WILLIAM J., JR.
MONTGOMERY, DR. AND MRS. CHARLES
H.
MONTGOMERY, DR. AND MRS. RAYMOND
P.
MOORE, MR. AND MRS. BEERIEN, III
MORSE, MR. AND MRS. CHARLES L., JR.
MORSE, MR. AND MRS. RICHARD S.
MOUL, DR. AND MRS. EDWIN T.
NEWTON, MR. AND MRS. WILLIAM F.
NlCKERSON, MR. AND MRS. FRANK L.
NORMAN, MR. AND MRS. ANDREW E.
NORMAN FOUNDATION
O'HERRON, MR. AND MRS. JONATHAN
ORTINS, MR. AND MRS. ARMAND
O'SULLIVAN, DR. RENEE BENNETT
PALMER, MRS. DOUGLAS W.
PAPPAS, DR. AND MRS. GEORGE D.
PARK, MRS. FRANKLIN A.
PARK, MR. AND MRS. MALCOLM S.
PARMENTER, Miss CAROLYN L.
PARMENTIER, MR. AND MRS. GEORGE L.
PENDERGAST, MRS. CLAUDIA
PENDLETON, DR. AND MRS. MURRAY E.
PENNINGTON, Miss ANNE H.
PERKINS, MR. AND MRS. COURTLAND D.
PERSON, DR. AND MRS. PHILIP
PETERSON, MR. AND MRS. E. GUNNAR
PETERSON, MR. AND MRS. E. JOEL
PETERSON, MR. RAYMOND W.
PHILIPPE, MR. AND MRS. PIERRE
PORTER, DR. AND MRS. KEITH R.
PROSSER, DR. AND MRS. C. LADD
PUTNAM, MR. ALLAN RAY
PUTNAM, MR. AND MRS. WILLIAM A., Ill
PYNE, Miss RUTH
RAYMOND, DR. AND MRS. SAMUEL
READ, Ms. LEE
REDFIELD, DR. AND MRS. ALFRED C.
RENEK, MR. AND MRS. MORRIS
REYNOLDS, DR. AND MRS. GEORGE
28
MARINE BIOLOGICAL LABORATORY
REYNOLDS, MRS. JAMES T.
REZNIKOFF, DR. AND MRS. PAUL
RICCA, DR. AND MRS. RENATO A.
RIGGS, MR. AND MRS. LAWRASSON, III
RIINA, MR. AND MRS. JOHN R.
ROBB, MRS. ALISON A.
ROBERTSON, MRS. C. STUART
ROBERTSON, DR. AND MRS. C. W.
ROBINSON, DR. AND MRS. DENIS M.
ROGERS, MRS. JULIAN
ROOT, MRS. WALTER S.
Ross, DR. VIRGINIA
ROWE, MR. AND MRS. WILLIAM S.
RUBIN, DR. JOSEPH
RUGH, MRS. ROBERTS
RUSSELL, MR. AND MRS. HENRY D.
RYDER, MR. AND MRS. FRANCIS C.
SAUNDERS, DR. AND MRS. JOHN W.
SAUNDERS, MRS. LAWRENCE
SAUNDERS, LAWRENCE FUND
SAWYER, MR. AND MRS. JOHN E.
SCHLESINGER, DR. AND MRS. R. WALTER
SCOTT, DR. AND MRS. GEORGE T.
SCOTT, MR. AND MRS. NORMAN E.
SEARS, MR. AND MRS. HAROLD B.
SEGAL, DR. AND MRS. SHELDON J.
SHAPIRO, MRS. HARRIET S.
SHEMIN, DR. AND MRS. DAVID
SHEPRO, DR. AND MRS. DAVID
SMITH, MRS. HOMER P.
SMITH, MR. VAN DORN C.
SNIDER, MR. ELIOT
SOLOMON, DR. AND MRS. A. K.
SPECHT, MRS. HEINZ
SPIEGEL, DR. AND MRS. MELVIN
STEELE, MRS. M. EVELYN
STEINBACH, MRS. H. BURR
STETSON, MRS. THOMAS J.
STETTEN, DR. AND MRS. DEWITT, JR.
STRACHER, DR. AND MRS. ALFRED
STUNKARD, DR. HORACE
STURTEVANT, MRS. A. H.
SWANSON, DR. AND MRS. CARL P.
SWOPE, MR. AND MRS. GERARD L.
SWOPE, MRS. GERARD, JR.
TAYLOR, DR. AND MRS. W. RANDOLPH
TYLOR, MRS. MARJORIE G.
TIETJE, MR. AND MRS. EMIL D., JR.
TODD, MR. AND MRS. GORDON F.
TOLKAN, MR. AND MRS. NORMAN N.
TOMPKINS, MRS. B. A.
TRACER, MRS. WILLIAM
TROLL, DR. AND MRS. WALTER
TULLY, MR. AND MRS. GORDON F.
VALOIS, MR. AND MRS. JOHN
VAN BRUNT, MR. AND MRS. A. H., JR.
VEEDER, MRS. RONALD A.
WAINIO, MRS. WALTER
WAITE, MR. AND MRS. CHARLES E.
WAKSMAN, DR. AND MRS. BYRON H.
WARE, MR. AND MRS. J. LINDSAY
WATT, MR. AND MRS. JOHN B.
WEISBERG, MR. AND MRS. ALFRED M.
WHEATLEY, DR. MARJORIE A.
WHEELER, DR. AND MRS. PAUL S.
WHEELER, DR. AND MRS. RALPH E.
WHITNEY, MR. AND MRS. GEOFFREY G.,
JR.
WlCHTERMAN, DR. AND MRS. RALPH
WICK.ERSHAM, MR. AND MRS. A. A.
TILNEY
WlCKERSHAM, MR. AND MRS. JAMES H.,
JR.
WILHELM, DR. HAZEL S.
WlTMER, DR. AND MRS. ENOS E.
WOLFINSOHN, MR. AND MRS. WOLFE
WOODWELL, DR. AND MRS. GEORGE M.
YNTEMA, MRS. CHESTER L.
ZINN, DR. AND MRS. DONALD J.
ZIPF, DR. ELIZABETH
ZWILLING, MRS. EDGAR
III. CERTIFICATE OF ORGANIZATION
(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
BYLAWS 29
fifteenth chapter of the Public Statutes of the Commonwealth of Massachusetts, and the Acts
in amendment thereof and in addition thereto.
The name by which the Corporation shall be known is THE MARINE BIOLOGICAL LAB-
ORATORY.
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 instruction 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.
//; 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 Stan-
ford Stevens, Anna D. Phillips. Susan Mims, 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 Wimess Whereof, we have hereunto signed our names, this thirteenth day of 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:
/ 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 requirements 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 compiled with
and I hereby approve said certificate this twentieth day of March A.D. eighteen hundred and
eighty-eight.
CHARLES ENDICOTT
Commissioner of Corporations)
IV. ARTICLES OF AMENDMENT
(On File in the Office of the Secretary of the Commonwealth)
We, James D. Ebert, President, and David Shepro, Clerk of the Marine Biological Laboratory,
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 meetings 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
30 MARINE BIOLOGICAL LABORATORY
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,
meetings 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 1 80, 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,
1975.
PAUL GUZZI
Secretary of the Commonwealth)
V. BYLAWS OF THE CORPORATION OF THE MARINE
BIOLOGICAL LABORATORY
(Revised August 1 1, 1978)
I. (A) The name of the Corporation shall be The Marine Biological Laboratory. The
Corporation's purpose shall be to establish and maintain a laboratory or station for scientific
study and investigation, and a school for instruction in biology and natural history.
(B) Marine Biological Laboratory admits students without regard to race, color, sex,
national and ethnic origin to all the rights, privileges, programs and activities generally ac-
corded or made available to students in its courses. It does not discriminate on the basis of
race, color, sex, national and ethnic origin in employment, administration of its educational
policies, admissions policies, scholarship and other programs.
II. (A) The members of the Corporation ("Members") shall consist of persons elected
by the Board of Trustees, upon such terms and conditions and in accordance with such
procedures, not inconsistent with law or these Bylaws, as may be determined by said Board
of Trustees. Except as provided below, any Member may vote at any meeting, either in person
or by proxy executed no more than six months prior to the date of such meeting. 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 attained
the age of seventy years or has retired from his home institution shall automatically be
designated a Life Member provided he signifies his wish to retain his membership. Life
Members shall not have the right to vote and shall not be assessed for dues.
BYLAWS 3 1
(B) 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.
III. The officers of the Corporation shall consist of a Chairman of the Board of Trustees,
President, Director, Treasurer and Clerk, elected or appointed by the Trustees as set forth
in Article IX.
IV. The Annual Meeting of the Members shall be held on the Friday following the Second
Tuesday in August in each year at the Laboratory in Woods Hole, Massachusetts, at 9:30
a.m. Subject to the provisions of Article VIII(2), at such meeting the Members shall choose
by ballot six Trustees to serve four years, and shall transact such other business as may
properly come before the meeting. Special meetings of the Members may be called by the
Chairman or Trustees to be held at such time and place as may be designated.
V. Twenty five 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 or by proxy at a meeting attended by a quorum (present in person or by proxy)
shall constitute action on behalf of the Members.
VI. (A) Inasmuch as the time and place of the Annual Meeting of Members are fixed by
these Bylaws, no notice of the Annual Meeting need be given. Notice of any special meeting
of Members, however, shall be given by the Clerk by mailing notice of the time and place
and purpose of such meeting, at least 1 5 days before such meeting, to each Member at his
or her address as shown on the records of the Corporation.
(B) Any meeting of the Members may be adjourned to any other time and place by the
vote of a majority of those Members present or represented at the meeting, whether or not
such Members constitute a quorum. It shall not be necessary to notify any Member of any
adjournment.
VII. The Annual Meeting of the Trustees shall be held promptly after the Annual Meeting
of the Corporation at the Laboratory in Woods Hole, Massachusetts. Special meetings of the
Trustees shall be called by the Chairman, the President, or by any seven Trustees, to be held
at such time and place as may be designated. Notice of Trustees' meetings may be given
orally, by telephone, telegraph or in writing; and notice given in time to enable the Trustees
to attend, or in any case notice sent by mail or telegraph to a Trustee's usual or last known
place or residence, at least one week before the meeting shall be sufficient. Notice of a meeting
need not be given to any Trustee if a written waiver of notice, executed by him before or
after the meeting is filed with the records of the meeting, or if he shall attend the meeting
without protesting prior thereto or at its commencement the lack of notice to him.
VIII. (A) There shall be four groups of Trustees:
(1) Trustees (the "Corporate Trustees") elected by the Members according to such pro-
cedures, not inconsistent with these Bylaws, as the Trustees shall have determined. Except
as provided below, such Trustees shall be divided into four classes of six, one class to be
elected each year to serve for a term of four years. Such classes shall be designated by the
year of expiration of their respective terms.
(2) Trustees ("Board Trustees") elected by the Trustees then in office according to such
procedures, not inconsistent with these Bylaws, as the Trustees shall have determined. Except
as provided below, such Board Trustees shall be divided into four classes of three, one class
to be elected each year to serve for a term of four years. Such classes shall be designated by
the year of expiration of their respective terms. It is contemplated that, unless otherwise
32 MARINE BIOLOGICAL LABORATORY
determined by the Trustees for good reason. Board Trustees shall be individuals who have
not been considered for election as Corporate Trustees.
(3) Trustees ex officio, who shall be the Chairman, the President, the Director, the Trea-
surer, and the Clerk.
(4) Trustees emeriti who shall include any Member who has attained the age of seventy
years (or the age of sixty five and has retired from his home institution) and who has served
a full elected term as a regular Trustee, provided he signifies his wish to serve the Laboratory
in that capacity. Any Trustee who qualifies for emeritus status shall continue to serve as a
regular Trustee until the next Annual Meeting whereupon his office as regular Trustee shall
become vacant and be filled by election by the Members or by the Board, as the case may
be. The Trustees ex officio and emeriti shall have all the rights of the Trustees, except that
Trustees emeriti shall not have the right to vote.
(B) The aggregate number of Corporate Trustees and Board Trustees elected in any year
(excluding Trustees elected to fill vacancies which do not result from expiration of a term)
shall not exceed nine. The number of Board Trustees so elected shall not exceed three and
unless otherwise determined by vote of the Trustees, the number of Corporate Trustees so
elected shall not exceed six.
(C) The Trustees and Officers shall hold their respective offices until their successors are
chosen in their stead.
(D) 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 so to be removed shall have been given reasonable notice and opportunity to be
heard before the body proposing to remove him.
(E) Any vacancy in the number of Corporate Trustees, however arising, may be filled by
the Trustees then in office unless and until filled by the Members at the next Annual Meeting.
Any vacancy in the number of Board Trustees may be filled by the Trustees.
(F) A Corporate Trustee or a Board Trustee who has served an initial term of at least 2
years duration shall be eligible for re-election to a second term, but shall be ineligible for re-
election to any subsequent term until two years have elapsed after he last served as Trustee.
IX. (A) The Trustees shall have the control and management of the affairs of the Cor-
poration. They shall elect a Chairman of the Board of Trustees who shall be elected annually
and shall serve until his successor is selected and qualified and who shall also preside at
meetings of the Corporation. They shall elect a President of the Corporation who shall also
be the Vice Chairman of the Board of Trustees and Vice Chairman of meetings of the
Corporation, and who shall be elected annually and shall serve until his successor is selected
and qualified. They shall annually elect a Treasurer who shall serve until his successor is
selected and qualified. They shall elect a Clerk (a resident of Massachusetts) who shall serve
for a term of 4 years. Eligibility for re-election shall be in accordance with the content of
Article VIII (F) as applied to Corporate or Board Trustees. They shall elect Board Trustees
as described in Article VIII (B). They shall appoint a Director of the Laboratory for a term
not to exceed five 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 may choose such
other officers and agents as they may think best. They may fix the compensation and define
the duties of all the officers and agents of the Corporation and may remove them at any time.
They may fill vacancies occurring in any of the offices. The Board of Trustees shall have the
power to choose an Executive Committee from their own number as provided in Article X,
and to delegate to such Committee such of their own powers as they may deem expedient
in addition to those powers conferred by Article X. 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) The Board of Trustees shall also have the power, by vote of a majority of the Trustees
then in Office, to elect an Investment Committee and any other committee and, by like vote,
BYLAWS 33
to delegate thereto some or all of their powers except those which by law, the Articles of
Organization or these Bylaws they are prohibited from delegating. The members of any such
committee shall have such tenure and duties as the Trustees shall determine; provided that
the Investment Committee, which shall oversee the management of the Corporation's en-
dowment funds and marketable securities, shall include the Chairman of the Board of Trust-
ees, the Treasurer of the Corporation, and the Chairman of the Corporation's Budget Com-
mittee, as ex officio members, 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.
X. (A) The Executive Committee is hereby designated to consist of not more than ten
members, including the ex officio Members (Chairman of the Board of Trustees, President,
Director and Treasurer); and six additional Trustees, two of whom shall be elected by the
Board of Trustees each year, to serve for a three-year term.
(B) The Chairman of the Board of Trustees shall act as Chairman of the Executive
Committee, and the President as Vice Chairman. A majority of the members of the Executive
Committee shall constitute a quorum and the affirmative vote of a majority of those voting
at any meeting at which a quorum is present shall constitute action on behalf of the Executive
Committee. The Executive Committee shall meet at such times and places and upon such
notice and appoint such sub-committees as the Committee shall determine.
(C) The Executive Committee shall have and may exercise all the powers of the Board
during the intervals between meetings of the Board of Trustees 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 Trustees, 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 of Trustees.
(D) The Executive Committee shall keep appropriate minutes of its meetings and its
action shall be reported to the Board of Trustees.
(E) The elected Members of the Executive Committee shall constitute as a standing
"Committee for the Nomination of Officers," responsible for making nominations, at each
Annual Meeting of the Corporation, and of the Board of Trustees, for candidates to fill each
office as the respective terms of office expire (Chairman of the Board, President, Director,
Treasurer, and Clerk).
XI. A majority of the Trustees, the Executive Committee, or any other committee elected
by the Trustees shall constitute a quorum; and a lesser number than a quorum may adjourn
any meeting from time to time without further notice. At any meeting of the Trustees, the
Executive Committee, or any other committee elected by the Trustees, the vote of a majority
of those present, or such different vote as may be specified by law, the Articles of Organization
or these Bylaws, shall be sufficient to take any action.
XII. Any action required or permitted to be taken at any meeting of the Trustees, the
Executive Committee or any other committee elected by the Trustees as referred to under
Article IX may be taken without a meeting if all of the Trustees or members of such committee,
as the case may be, consent to the action in writing and such written consents are filed with
the records of meetings. The Trustees or members of the Executive Committee or any other
committee appointed by the Trustees may also participate in meeting by means of conference
telephone, or otherwise take action in such a manner as may from time to time be permitted
by law.
XIII. The consent of every Trustee shall be necessary to dissolution of the Marine Bio-
logical Laboratory. In case of dissolution, the property shall be disposed of in such manner
34 MARINE BIOLOGICAL LABORATORY
and upon such terms as shall be determined by the affirmative vote of two-thirds of the Board
of Trustees then in office.
XIV. 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) the provisions of these Bylaws governing (i) the
removal of Trustees and (ii) 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 the 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 Corporation 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.
XV. The account of the Treasurer shall be audited annually by a certified public ac-
countant.
XVI. The Corporation will indemnify every person who is or was a trustee, officer or
employee of the Corporation or a person who provides services without compensation to an
Employee Benefit Plan maintained by the Corporation, for any liability (including reasonable
costs of defense and settlement) arising by reason of any act or omission affecting an Employee
Benefit Plan maintained by the Corporation or affecting the participants or beneficiaries of
such Plan, including without limitation any damages, civil penalty or excise tax imposed
pursuant to the Employee Retirement Income Security Act of 1974; provided, (1) that the
Act or omission shall have occurred in the course of the person's service as trustee or officer
of the Corporation or within the scope of the employment of an employee of the Corporation
or in connection with a service provided without compensation to an Employee Benefit Plan
maintained by the Corporation, (2) that the Act or omission be in good faith as determined
by the Corporation (whose determination made in good faith and not arbitrarily or capri-
ciously shall be conclusive), and (3) that the Corporation's obligation hereunder shall be offset
to the extent of any otherwise applicable insurance coverage, under a policy maintained by
the Corporation or any other person, or other source of indemnification.
VI. REPORT OF THE DIRECTOR
Introduction: Reaction Kinetics
Elementary processes may be discontinuous, all-or-none; but the behavior of
macroscopic things is not. Time passes between the beginning and the end of change,
and the course is, at least, a smooth curve; almost never a square-wave. For cells,
organisms, societies, smoothness of the curve is an illusion: Occam's Razor aside,
change in the people and institutions we know is never simple, nor smooth. There
is always fine-structure: large transients are composed of small ones.
We have seen, recently, one of the transients of a broader process: the recon-
struction of the MBL's physical plant. The initial state occupied much attention of
standing committees of the Corporation, and was reported with many dire predic-
tions, at Corporation meetings held in the 1970's. The final state is that condition
of buildings, housing, and equipment which we have made our goal for the Lab-
oratory's hundredth birthday in 1988. The recent transient was the start and sue-
REPORT OF THE DIRECTOR 35
cessful finish of Phase One of the Second Century Fund campaign, between 1979
and 1982.
Fine new Library space and the $1.5 million rehabilitation of Lillie are facts,
as are the new Environmental Sciences Center, the handsome Candle House (which
not only removed administrative offices from the laboratory and Library space they
had preempted, and gave us some fine new lecture rooms, but even won the MBL
an award this year), and a host of smaller, but not unimportant projects. Most of
these changes were visible enough, in 1981-82, and they were discussed in the 1981
Director's Report and elsewhere.
If, however, it was physical change commanding attention that year, it is change
in operations and management that accounts for most of the fine-structure in the
year to which this Report refers. The change is not so immediately striking to the
eye as a new or restored building; but in the end, it will be more important in
determining the new equilibrium state of the MBL. By operations, I mean not only
the work and leadership of the support Departments, where, after a longish period
of quiescence, there have recently been a good many bumps on the curve. I refer
also to the big issue of running year-round research and educational programs at
the MBL, and the relationship thereof to the traditional summer activities.
We are not yet at the new plateau, as that was planned and set forth in my
report for the Trustees in 1979, but the changes have been occurring faster and on
a larger scale than at any time since they began in 1975. We are much closer now
to the plateau than we are to the initial state. Although the changes have taken place
quietly, they have made plenty of fine-structure during the past year.
The same holds for management of the Laboratory. There are new faces among
the Department Heads and new people working in the Central Administration. The
changes thus represented are, again, but component parts of a much larger change,
which must occur over time in consonance with the change of operating scale and
style. They, too, have made for not a little jiggling of the curve, which turns out,
however, to be not noise, but signal.
This jiggling is a part of the fine-structure of change that I hope and expect will
diminish somewhat for the next two or three years, for we must now return to
emphasis upon the other aspects of the MBL's plan for itself: further improvements
in the physical plant, including, notably, housing facilities; and even more urgently,
the achievement of a higher level of financial independence than the Laboratory
has ever enjoyed, through new, long-term funding of key programs, not solely by
government, and through new and significant endowment funds.
I try here to provide an accurate impression, but as usual not a detailed account,
of the changes in operations and management that were visible to those of us who
were here the full year or last summer. They may not have been visible to those
Corporation members who did not come to Woods Hole in 1982, or whose visits
were brief; but all members should be aware of them and of what they signify.
"Omnia mutant ur, nos et mulamur in illis.1'
Department Leadership: Changes
Robert Gunning, the popular Superintendent of Buildings and Grounds, re-
mained at his post for a full year beyond what would have been his normal date
of retirement. This was a generous decision on his part, and the Laboratory had the
benefit, thereby, of his long years of experience while two important processes were
1 Freely: "All things must change; we had better change with them."
36 MARINE BIOLOGICAL LABORATORY
brought to a conclusion: the Lillie rehabilitation and the selection of his successor.
The work on Lillie was brought to a successful conclusion; and the succession has
been seen to.
In this case the new Department Head is one of our own: Donald B. Lehy,
himself a veteran of no short period of service to the MBL. Since Lehy had earlier
shown his form on many occasions — serious work, no-nonsense attention to detail,
high technical competence, loyalty to the institution — it is no surprise that the
transition has been smooth. Bob Gunning was honored at a splendid party, with
a sufficiency of chicken wings and wine, in the Meigs Room, organized by the
support staff at the time of his retirement. He was presented with the key to the
MBL. Some little trouble may be encountered in its use, since it is five feet long
and a bit thick for most of our locksets, but that seems not to have cast any sort
of pall over the occasion.
The departure of Dr. Morton Maser, whose position as Assistant Director for
Research and Educational Services included leadership of the Research Services
Department, left the Laboratory with several important management responsibilities
unassigned. After much discussion among Corporation members and Trustees, the
decision was made to divide those responsibilities, and appropriate searches for new
people were instituted.
The Research Services Department was in any case due for some reorganization
of its subdivisions, which include the stockrooms ("Chem. Room"), the graphics
facility ("Photo Lab"), purchasing, machine shop, and the provision of specialized
equipment and techniques through the apparatus department, the radiation labo-
ratories, the general-use equipment rooms, and the electron microscopy facility. The
management of such a spectrum of services is no small assignment, and we are very
fortunate to have been able to attract an ideally-qualified person as the new De-
partment Head.
He is Mr. Barry T. O'Neil, recently Steward of the Lemuel Shattuck Hospital
in Boston, and a highly-experienced manager of people, machines, and services. As
it happens, he is also a talented medical artist and scientific illustrator, and a sailor
with a large family of sailors. Whence, as for many of us who reside here the year
round, there come certain pleasures in the job that cannot be found elsewhere. There
is every reason to expect that the new esprit of the Department, which is already
evident, will be matched by a new efficiency in the Department's work, and by a
new atmosphere of good cheer and cooperativeness.
The electron microscopy facilities are under the immediate supervision of Dr.
Eugene Copeland, a colleague whose scientific achievements and expertise in aca-
demic management have on several occasions been made available, when they were
most needed, to the Laboratory. Louis Kerr continues to be in charge of the radiation
laboratories as well as to operate the EM services: a current issue of the MBL
NEWSLETTER provides a properly laudatory review of his excellent work. A new
committee of users has been formed to advise Gene Copeland, and the standing
committee on radiation is in the process of expansion and reorganization. Dr. An-
thony Liuzzi, who has been our Health Physicist and Radiation Officer for a good
many years, continues in the post with an enlarged responsibility and an expanded
schedule of visits to the MBL, during some of which he will carry on his own
research.
Dr. Maser's responsibilities in management of the instructional program were
very broad, and occupied much of his always-busy time. They included responsibility
for Admissions, which office was operated by an Admissions Officer reporting to
him, the initiation and management of short courses, oversight of the January se-
REPORT OF THE DIRECTOR 37
mester, and general coordination of facilities for the summer courses. This range
of duties has been redefined so as to make it more tractable to the administrator
and more readily understood by the faculties. Following a careful search and much
discussion of the program by the Executive Committee and those concerned directly
with the courses, including the Committee on Instruction, the job was assigned to
Joan E. Howard, who was at the time the Laboratory's Grants and Contracts Ad-
ministrator.
She had accumulated, in that office, much experience in handling the finances
of the summer courses, most of which are supported in part by training grants,
government and private. This was expected to prove helpful, as it has, in a period
of transition for management of the instructional program. Ms. Howard continues
to be in charge of grants and contracts, but she now oversees Admissions as well
(where Joanne Foley is the Admissions Officer), and she coordinates the financing,
budgeting, and equipping of all courses. In these activities she works effectively and
closely with other staff, e.g., with the Controller in finance and budget-making; with
Mr. O"Neil in meeting the equipment needs of the courses; with Dr. Copeland; with
Mr. Smith in the assignment of laboratory spaces; and with the Director in the
design of a new and expanded program of non-summer courses. Her official title
is "Coordinator for Grants and Educational Services.1' Here, too, a transition that
might well have been troublesome has occurred with no more than minor dislo-
cations, and in many respects the component responsibilities are being met better
than ever before.
Dr. Wesley N. Tiffney, Director of the George M. Gray museum, died in January,
1983 after a long illness. He was an active and devoted curator as well as a valued
colleague. Although the event occurred beyond the interval that is covered, formally,
by this report, it is mentioned here because the museum is important to many MBL
scientists, and in the minds of many of those, the museum was, at least recently,
Wes Tiffney. Once again, we are fortunate in having had some depth in an important
facility. Dr. Louise F. Bush, who works in the museum and shared the running of
it with Tiffney, has kindly and articulately consented to take charge. Dr. Edwin T.
Moul continues to assist with the identification of plant material, as do Dr. Arthur
Humes for Crustacea and Dr. Ruth Turner for Mollusca.
There is evidence, from plans and suggestions offered by Dr. Bush, that far from
a holding action, current work of the museum staff will make the facility more
useful, and far busier in the immediate future than it was in the past. We look
forward to participation of the museum staff in forthcoming discussions with ar-
chitects who will be conducting a new feasibility study for the Marine Resources
Center. The best possibility for a proper and permanent home for the museum is
in that building.
Department Leadership: No Change
It would violate the intention of devoting this report to change were I to set
down, systematically, the news from each of the operating departments, distributing
kudos and concerns as judgment required. I can illustrate the pervasiveness of
change, however, by reference to any one of the Departments in which there has
not been a major change of leadership during the past year, and yet pointing up the
implications of this or that new activity. Here, then, is my example, employing the
words of John Valois, Manager of the Department of Marine Resources. He wrote
them for me in his regular report of the year's work. They have to do with squid
in 1982.
38 MARINE BIOLOGICAL LABORATORY
"The spring squid came late (May 1 1th), but the numbers and size were excellent
. . . catches remained good through most of July, until the last week. The usual
slow catches lasted into mid-August, and a slight revival was followed by three days
of very poor catches. By September, there was some strengthening of size and num-
bers, until a large wave of adult squid came into the collecting areas in late Sep-
tember, and stayed until the last part of November. We are going to watch this
change in the migration . . . hoping that if it remains ... we might be able to
encourage some of our neurobiologists to extend their research season . . . finally,
it is doubtful that the majority of our squid users understand the financial effort
that this Laboratory makes, and the skills that are used to supply squid to about
forty laboratories.
"... Dr. Louis Leibovitz and I have been interested in furthering observations
... on holding squid in cold water. A remarkable improvement was made in 1982
in squid survival through redesigning squid tanks and improvement of water flow.
With better-trained technicians, more time available for good record-keeping, and
facilities purchased recently through the help of Dr. Leibovitz, the Department
. . . has gained valuable information on the causes of mortality. We have designed
a pilot system using cold water in the Marine Resources building for Dr. Leibovitz.
There will be suitable controls ... a copy of the design is enclosed . . . MBL sea
water is monitored daily by a technician from the Water Quality laboratory . . .
elements of the nitrogen cycle are sampled bi-weekly, as are . . . O2, phosphates,
etc. . . . We would like to begin a program of ... exploration of plankton and its
effect on sea water . . . seasonal variations must play an important role . . . the
commonly-held belief that anoxia and mortality are directly related to overcrowding
is an oversimplification.
". . . Most of our employees have a general understanding of the research goals
at the MBL, but more importantly, they treat our scientists with respect and po-
liteness."
This last reminds me to report the very sad news of the deaths of Lew Lawday
and Bruno Trapasso, two members of the Department of whom it is just, indeed,
to say that they had an understanding of research goals, and that they treated ev-
eryone with respect and politeness. They also got them in return.
The implication for change, in this extract from John Valois's report, should be
clear. The MBL now has, in Dr. Leibovitz, a resident marine animal pathologist of
the highest scientific stature; a functioning Laboratory of Marine Animal Health;
and a strong and growing interaction between research in that field and the practical
business of collecting and holding animals. Collaboration goes, in fact, much deeper
than is evidenced in a casual glance. Ongoing and planned collaborative studies are
concerned not only with holding, but with the culture of marine animals.
The transients implied by the quoted material are just that: local peaks of a
broader and much larger variation, in which the final state will be the presence on
this campus of a splendid new facility for Marine Resources. There, research and
supply will operate side-by-side, with two goals: ( 1 ) the advance of knowledge of
marine animals, in health and disease; their ecology, genetics, development, behav-
ior, population biology, and pathologies; and (2) the development of entirely new,
research-based techniques for collecting, holding, and culturing those animals, so
as to free the Laboratory of day-to-day dependence upon the chances of fishing. In
the end, the process could give scientists living marine material for research that is
no less uniform, predictable, and available than the inbred rodents that are now so
universally employed.
REPORT OF THE DIRECTOR 39
Financial Management; Development; Public Information
The arrival of Controller John Speer was reported last year, and is hence not,
in itself, news. The consequent change is news, and I can do no more here than
touch upon its components. Most important of all is the evidence that there is finally
someone in charge, who has the knowledge, the technical background, the authority,
and the will to be in charge, of so specialized and complex task as keeping the MBL
in sound financial health. We have needed such a person — a Controller in fact —
for a very long time. He has managed, with no significant enlargement of the regular
staff, to routinize what was heretofore ad hoc in the system, to speed reporting and
response, and in general to give outsiders dealing with the Laboratory confidence
in its fiscal procedures.
The conversion of the MBL's accounting systems from manual to computer-
based is nearly complete. This has already yielded some returns, and there will be
many more in the year ahead. Tracking of income and expense is, for example,
faster and more accurate than before. It is continuous: MBL people no longer have
to wait days, or even weeks, for outside organizations to process data, issue checks,
or prepare reports. The availability of administrative data processing and computing
in-house has had another benefit, just now making itself felt throughout the Lab-
oratory: word processing capability2 alongside direct access to data in memory.
Purchase, installation, and de-bugging of the system have not been without typical
troubles: but the shakedown period is nearly over at the time of writing, and the
consequences of a modern data-processing capability will be felt — as a great advance
in convenience — by all who come here for the summer of 1983.
The long-delayed review and re-structuring of the MBL's overhead cost recovery
system has begun, now that there are people here to be in charge of it and to
communicate as peers with the government's cost-control people and auditors. The
old system is, as I have so often tried to explain, not unfair to users of the MBL:
it is unfair to the MBL as an institution. It was designed purposely to make income
from overhead payments less than the actual operating expenses of the Laboratory.
Some parts of the MBL's scientific mission, such as the instructional program, yield
no overhead at all, by law, or allow a merely token amount.
The loss is not passed on to programs that do provide for overhead, e.g., research.
Please note that: research does not pay for education. It is simply a loss, made up
from funds raised in the private sector. The result has been, for many years, that
among places in which to conduct research, the MBL is one of the best bargains
in the world, cheaper by several-fold than its neighbors. But the bargain has to be
paid for. In this case it is paid for by private funds that could, and should, serve
more important purposes than the mere payment of current bills. It is also paid for
by the time of the Director, the Chairman of the Board, and all those engaged in
fund raising for the MBL.
Our goals in the design and eventual establishment of a new system will be to
bring overhead payments much closer to actual operating expense (which is lower
than at universities operating laboratories of similar sophistication), and, at the same
2 This had had no positive effects, so far as I can see, upon spelling, punctuation, or the use of the
apostrophe. Neither, however, did my broadcast distribution of the paperback edition of Strunk & White,
a few years ago. There is a strong movement toward the purchase of a software package that corrects
spelling. I have resisted it mercilessly, out of pure anachronistic impulse. This I communicate here to
balance the emphasis, in the text, upon "change."
40 MARINE BIOLOGICAL LABORATORY
time, to reduce or eliminate entirely the impact of MBL cost recovery upon the
research grants and other resources of MBL investigators, summer and year-round.
The reader may be reassured, if that sounds impossible, that it is not. It is merely
very, very difficult.
About the state of the Laboratory's finances I need say little: appended reports
from the Controller and from our excellent Treasurer cover the territory very well.
It is important to note here, however, that with major efforts of the Controller's
Department, and — since his arrival — close cooperation with other Departments, the
MBL has achieved the remarkable result of completing a $4.5 million campus re-
habilitation, in a time of rampaging cost inflation, with a net overrun of less than
ten percent on the original estimates. Most of that overrun was caused by changes
in a single project, the Environmental Sciences Center; but those changes added far
more than their instantaneous dollar value to the asset value of the facility. During
the year, as a result, there was a temporary cash-flow problem — nothing in the
slightest unusual for an academic organization engaged in major construction —
which was dealt with firmly and properly by the Controller, the Treasurer, and the
Executive Committee.
It is commonplace for financial officers and administrators to grumble, in re-
search-centered organizations such as the MBL, ". . . we must get our costs under
control." A little study of the financial reports in this volume, including those of
the Controller and Treasurer, will make it clear that the MBL has its costs under
control, and quite tight control at that. It is income that the MBL must get under
control, and I have the conviction, now as not in the prior four years, that we shall
be able to do that before long.
To the extent that private-sector funds are now, and will remain for at least a
few years, an important part of working income, independently of their use in the
acquisition of new facilities and programs, the Development Office has its work cut
out. Here, too, is change, and again, the change was reported last year, with notice
that Ms. Carol Salguero had joined the MBL as Director of Development. Since
then the conversion to a self-contained, in-house fund raising program has been
accomplished. After the inevitable months of learning and form-fitting, it appears
to have settled into decently routine operation. The flow of proposals outward and
gifts inward has resumed, after a year and a half of slow-down; several events, in
which the MBL's case has been put to appropriate listeners among Board Trustees,
business executives, and philanthropists, have been organized and carried off with
success. There will have to be an increasing number of those in the future.
In that connection, the MBL's regular publications, such as the NEWSLETTER,
releases to the press, and occasional sponsored articles in magazines, are public
"events" of a special importance. The new Public Information Officer, James
Shreeve, gives evidence of being, not only a highly skilled writer, but — and this is
critically important — a Quick Study, able to learn from reading and conversation
what MBL scientists are doing and thinking about, and to turn what he has learned
into accurate, comprehensible, and stimulating prose.
Public information is a domain in which, as many Trustees are aware, the MBL
has been backward in relation to its peer-institutions. It is an activity in which the
expository styles and approaches of proposal-writing, be it for public or private
agencies, are not merely inappropriate, but actually counter-productive. Good public
relations work requires its own kind of expository skill and imagination, and people
who have or can learn those are very rare. The early indications are that James
Shreeve and his assistant. Arch Maclnnes, have picked up from where Barbara
REPORT OF THE DIRECTOR 4 1
Haskell left off, a year ago, and are moving to an altogether higher level of achieve-
ment in presenting the story of the MBL to the educated layman.
Research: The Year Round Programs
I have understood, from conversation with many Corporation members and by
the precedent of former Directors, that these Director's Reports do not have, among
their purposes, the citation or description of research achievements, except in very
unusual cases. The reason for such a variance from the typical content of Presidents'
and Directors' Reports is that the October Biological Bulletin prints abstracts from
the General Scientific Meetings of the prior August, and those, in turn, represent
accurately the activity of the scientific community of the MBL. Readers of these
Director's Reports being almost all biologists themselves, my re-summarizing the
summaries for them would be gilding the lily. So be it: I have, for that reason, not
cited anything like all the important accomplishments of MBL investigators, nor
even a representative set of them, in any year. To do so, in fact, would be (as I
suggested in a musical digression last year) to court trouble.
This year's report is concerned with change, however, and to complete the rep-
resentation of change in progress at the MBL, I must devote some little space to the
year round research programs. I do so, not to illuminate specific advances or to
comment upon progress, but to give the reader a sense of the magnitude and breadth
of the program as it is today, roughly eight years after the decision was made to
allow a growth of serious year-round science at the MBL. It is also four years after
the contentious issue of year-round versus summer use of the facilities was addressed
in my 1979 report to the Trustees.
The most important point of that report was as much an undertaking, or a
promise, as it was a recommendation. It would surely never have received the
unanimous approval voted had there not been a promise. My report called, in brief,
for a considerable growth of the year round research program, in all — not just one —
of the main disciplines of MBL biology, i.e.. Cell and Developmental Biology; Neu-
robiology and Biophysics; Ecology; and Marine Biomedicine. The promise was that
such growth would be scrupulously controlled as to quality of the science and the
scientists; and that excluding transient fluctuations over the course of one or two
years, the absolute amount of space and the research facilities reserved for summer
investigators and for teaching would not decline.
Among the Trustees there must have been some skeptics, even though I explained
that the trick would be accomplished by the construction of new space. We have
indeed acquired new space: the Candle House, in accommodating the Central Ad-
ministration, made almost 6,000 square feet available in Lillie, most of which be-
came the expanded Library, but some of which has become fine laboratories in use
today. The Environmental Sciences Center provided a home for most (but not all)
MBL ecologists, and released a large amount of valuable teaching space in Loeb.
Some day the Marine Resources Center will provide us with thousands of square
feet of still newer laboratories, and they will, by every test we can make, all be
occupied.
In any case, the promise has been kept thus far: recommended growth of the
year round scientific program has taken place, and we still have the full summer
program. The resulting change has come about with minimum fanfare, but it is a
change, and a big one. To indicate its dimensions, I will simply list programs— not
all of them, but a good sampling — that today have a year round home at the MBL,
42 MARINE BIOLOGICAL LABORATORY
and that contribute to its international stature, without any one of them having
infringed in an important way upon the MBL's indispensable teaching programs
nor its "observatory" function.
The Ecosystems Center, one of the oldest of such programs, and perhaps the
best known to readers of this Report from its own Annual Report.
The Laboratory ofShinya Inoue, a world resource for polarization microscopy,
quantitative light microscopy, and for study of the molecular organization of motility
processes in living cells.
The Laboratory of Sensory Physiology, headed by E. F. MacNichol, Jr., an
unique center for the biophysical investigation of vision.
The Boston University Marine Program, whose Director is MBL Trustee J. R.
Whittaker, and whose faculty are leaders in the following fields: developmental
biology; animal behavior and invertebrate physiology; systematics; primitive motility
processes; ecology; and marine biology. This program is responsible for the education
of some thirty graduate students who are in residence at the MBL, and who have,
since its founding, produced an outstanding crop of dissertations and research
papers.
The Laboratory of Noel de Terra, who studies the mechanisms and control of
cell division in the ciliate Stentor.
The Laboratory of Carl J. Berg, Jr., whose group investigates mariculture of
marine animals and their reproductive biology, especially in relation to managed
systems.
The Laboratory of Biophysics, headed by W. J. Adelman, in which two very
large programs of the NIH (headed by Adelman — membrane biophysics — and Dan
Alkon — cellular basis of learning in Hermissenda), totalling twenty investigators on
the average, are on permanent location at the MBL.
The Laboratory of Raymond E. Stephens, which is recognized internationally
for its research on tubulins and microtubules.
The Laboratory ofOsamu Shimomura, whose head, the discoverer of Aequorin,
pursues a lifelong quest for fundamental mechanisms, at the molecular and atomic
levels, of bioluminescent processes.
The Laboratory of Judith P. Grassle, in which there is an expanding program
of research on the population genetics; adaptation to special environments; and
responses to pollutants, of marine invertebrate animals.
The National Vibrating Probe Facility, Lionel Jaffe, Director, in which, for the
first time, a research and service facility is to be generally available for investigators
wishing to analyze microscopic, bioelectric fields with the non-invasive vibrating
probe system. The facility is expected to have several two-dimensional vibrating
probes ("nutating") ready for visiting-scientist use in the summer 1983.
The Laboratory of D. E. Copeland, where Dr. Copeland continues his fine-
structure and physiological studies on the eye and the swimbladder of fishes.
I stop the listing at this point, but not without noting again that it is incomplete;
and that there will almost certainly be at least one distinguished addition next year,
in a subdiscipline of neuroscience not already represented in the year round com-
munity. The list is not really different from one that appears, each year, in the
Laboratory's Annual Bulletin. It is my hope, however, that by seeing it here, in the
form and context given, Corporation members may get a more accurate idea than
they would get from the Bulletin of the magnitude of the entire effort. It is, I am
sure, something of which any great university would be proud, let alone a small,
private laboratory, starting with no applicable endowment and no significant sources
REPORT OF THE DIRECTOR 43
of tuition or other direct income. It is, certainly, a change from just a few years ago.
I see in it little to fear, much to take pride in and to consider with care.
It is this "consideration with care" that must be the final point. We have come
about halfway in the plan that the Trustees received, debated, and approved without
dissent in 1979. This is a good time to look at the results; at the MBL as it appears,
as it works, as it feels to those several hundred biologists for whom it is, as it was
for their predecessors, one of this country's scientific treasures.
Does it feel all right? Has the change so far — and I hope it is now clear that
there has been plenty of change — been positive or negative? And when you have
an answer to that, ask then: If positive, was it preordained, or automatic, or was it
engineered? If negative, was it bound to happen anyway, as a consequence of the
way the world works, or is it a result of policy? Even a minimal searching of souls
in such terms will produce directed, apposite discussion and argument at the next
Corporation meeting, and among the Trustees. That, after all, is what identifies the
MBL. Unlike any other place known to you or me, this research and teaching
organization is owned by the people who investigate and teach in it, and their
questions and votes do, in fact, determine what happens next.
VII. REPORT OF THE TREASURER
In a separate report, the Controller will review the Laboratory's income and
expenses for 1982. Here, I hope to contribute a perspective on MBL's financial
affairs.
The MBL defines its purpose in terms of science and education. The Laboratory
acquires, organizes, and applies talent and money to the pursuit of its purposes.
Unlike a business enterprise, it does not seek to earn a surplus for the benefit of its
owners. Thus, the MBL's financial objective each year is to match planned expenses
against expected revenues; i.e. to maintain a balanced budget.
The MBL is not richly endowed. Therefore, its approach to financial manage-
ment must be aggressive. The alternative to compromise in its scientific and edu-
cational purposes is to intensify its efforts to acquire resources. Although the MBL
uses an annual iterative budgeting cycle to match plans to resources, the require-
ments for good science and good education tend to drive the revenue targets.
Looking back over the past ten years (see the accompanying Exhibit), one ob-
serves that the MBL has been ambitious in setting its revenue targets. In seven of
the past ten years, the expenses of the Laboratory's endeavors exceeded its income.
This was the case in 1982.
At the MBL, income generation is spread throughout the fiscal year while ex-
penses are heavier during the peak activity of the summer months. Unavoidably,
some expense commitments must be made before the year's income is precisely
known. Nevertheless, both surpluses and deficits have been modest in relation to
the MBL's total "throughput" (unrestricted funds plus restricted grants for research
and programs). The total throughput is a better gauge of the management task than
is the magnitude of expenses met with unrestricted funds. If the surplus or deficit
in each of the past ten years is computed as a per cent of the total throughput, the
average is less than 4 per cent and the range is narrow. Thus, in its efforts to strike
a balanced budget each year, the MBL has missed its target by relatively small
amounts and its performance has been stable. However, the challenge has been
increasing; the throughput of unrestricted and restricted grant monies has more than
trebled since 1973.
44
MARINE BIOLOGICAL LABORATORY
<N
00
Os
so r— —
•* o so
sq OO —
rn rn
oo
fN
fN
<N
vo r4 ^ oo
— oo r- oo
o
rn
vq
o
Ov OO ^t r«1
OO vO vo —
vO vO O OO
so' O — '
00
Ov
O
t~-
O
° S
•n' ^
OO v~i f) O
O O O vO
oo rn Tt r-
so' oo' — '
Ov
sOr sO
fN fN
— r~
(N Tj-'
r-
d
— so r- <N
t-~ oo o oo
•N
o\
rj oo so
O r"l r^i
(N (N
sO
i/1
r-
oo
o
OO vO <^
f) ro — .
>/i oo' —
I
rv-^
Is.
Ov
so
t\
Ov
— r^ so
00 SO 00
o — ^
fN fN
O TT sO
r) — O
Ov OO —
c
£
OO
o
SO
o
— fN
sO
I—
[^ m so 1^-
n r<-) os Ov
sO Ov v~t Ov
•^•' oo'
r*1 <^i OO O
^ >f^ oo' — '
Ov
so oo rx|
sO Os r»1
fN r-
ON Ov
— <N Ov ON
OO O) fN —
^ O — O,
>f) Ov — *
Ov
r-~ i~- o
— Ov OO
sO
<rt
sO
VI
fN
<N
so
so O r- 1-
1— sO 'T —
— •_ <N TT so
^^ Tt O\
Ov
ON SO r~-
O OV OO
O
OO
so
ON
p~ m oo •^•
CO fN fN fN
Ov
O
1
Q
t/1 -^^
lit
8 *•*
Ji X 3
I
* E IN-B!
J= g « 51 «« « s
S 8> § | ! Q ^
w ^ ° i ^ - ^
>5
— UJ V5
.JJ
(£ H "S.
0<
^
[j
k.
E-
— fN ff-,
*5
^ **~^ &
O
O
03
c
J2 c a, ;=
ea % — ^
03 E % w
"3 w '5. 5
ti -^ ^» "^
3 f. s: 'C,
U_ — Cu CU
REPORT OF THE TREASURER 45
The accompanying Exhibit also shows that the MBL has demonstrated increasing
effectiveness in attracting resources over the past ten years. Private gifts in the early
1970s averaged a few hundred thousand dollars per year; more recently, the level
of private support has been substantially greater. In 1982, a shift from dependence
on outside consulting services to an internal development office resulted in a tem-
porary discontinuity in the MBL's fundraising efforts. The Laboratory now has in
place the capacity with which to raise in excess of $ 1 million annually, and at that
level of success the 1982 deficit would have been avoided.
I call attention to several other facts in the accompanying Exhibit. The MBL's
Fund Balance account, which can be thought of as a surplus account, has remained
positive since 1976. Investment and endowment funds have increased steadily over
the past ten years; the figures shown are on the basis of book value, and market
value would show even greater growth. After seven years of declining balances in
the Physical Plant account due to the bookkeeping effect of annual depreciation,
the recent infusions of capital into campus rehabilitation and new construction is
apparent. Most significantly, the MBL remains unburdened by either long or short
term debt.
Although I hope these observations have helped to place the MBL's 1982 results
into a useful framework for understanding, I do not wish to suggest that the MBL
can afford complacency. The Laboratory needs the direct and indirect assistance of
every person friendly to its purposes. The challenge lies not so much in expense
reduction, for the economical character of MBL's operations is well known, but
rather in revenue development.
Before closing my report, I wish to highlight two particularly important accom-
plishments in 1982. The Investment Committee has given excellent policy guidance
to the management of the MBL's investments. Invested funds increased 10 per cent
in 1982 despite withdrawals totaling $565,000 for construction and operating pur-
poses. In another category of accomplishment, we are grateful to cooperative in-
vestigators and to diligent administrators who have helped in the reduction of MBL's
receivables from $623,658 at the end of 1981 to $237,859 by the close of 1982.
Robert Mainer
Treasurer
VIII. REPORT OF THE CONTROLLER
The most obvious outcome of our 1982 audit is that we experienced an operating
deficit of $160,591. This, together with the drop in our fund balance, represents a
problem that must be addressed in 1983 and beyond. While the operating deficit
and the decrease in the fund balance are serious, they should not cause undue
concern. We have achieved several important objectives which should enable us to
improve the Laboratory's finances in the future. Among them are:
(1) a firm plan to put the educational program on a financially sound basis,
with the ultimate goal of a fair recovery of costs;
(2) a newly developed financial management system that has already made a
positive contribution to budgetary control;
(3) several changes in policy to provide us with more sensitive and immediate
control over our financial resources;
(4) fundraising, that in the second half of 1982 increased significantly over the
first half, reversing an earlier decline in private sector gifts.
46 MARINE BIOLOGICAL LABORATORY
The point of all this is that we must view 1982, and to some extent 1983, as
years of transition that will lead to much better financial management in 1984 and
beyond.
Having said that, let me now address the key financial indicators for this past
year in more detail.
Revenues
Overall, our revenues increased by less than 3%, which for the second straight
year amounts to "level funding." While we showed an increase of 13.5% in lab fees
and 9.6% in Research Services, most other forms of income decreased or had very
modest increases. Revenue generated by the Library, The Biological Bulletin, and
Marine Resources was down, as were unrestricted gifts and investments. The need
to increase revenues for 1983 is thus our most serious challenge. The keys to this
are increased unrestricted gifts and more appropriate overhead recovery, principally
from the year-round research and the educational programs.
Expenditures
"Unrestricted" expenditures increased by a very modest 5.8%, reflecting, in part
a diminution in the inflation rate, but also a concerted effort on the part of all
support staff to hold the line on expenditures. Direct costs of instruction were reduced
by almost 45% from the previous year, the result of some selective actions that we
hope will lead to a balanced education budget in 1983 and possibly even some small
amount of overhead recovery. This is a major accomplishment given the reductions
and changing emphasis in federal funding in training. We have also reduced the
number of people on the regular payroll through a selective hiring freeze.
It is important to remember, however, that reductions in expenditures can have
negative results, if they are not carefully applied. Reductions in capital expenditures
for laboratory equipment, library periodicals, and deferred maintenance of the phys-
ical plant may appear attractive budget-balancers in the short run, but over time
they can be devastating. Evidence of this can be shown in the summer cottages and
in the capital equipment budget for Apparatus, because for the past several years
both have been severly underfunded. We must continue to redress the neglect in
the physical plant and in the replacement of capital scientific equipment, even while
we hold overall expenditures in line with our (modest) revenue projections.
Looking ahead to 1983, we see challenges and opportunities in several areas.
First, we must continue our efforts in development to support all facets of the MBL.
Second, we must increase our overhead recovery through a better, more equitable
system. Third, we must press on toward the goal of recovering the indirect costs of
the educational programs. Finally, we must increase regular expenditures for capital
equipment and in deferred maintenance, within the general context of a balanced
operating budget.
John W. Speer
Controller
REPORT OF THE CONTROLLER 47
Coopers
&Ly brand
certified public accountants
To the Trustees of
Marine Biological Laboratory
Woods Hole, Massachusetts
We have examined the balance sheets of Marine Biological
Laboratory as of December 31, 1982 and 1981, and the related state-
ments of current funds revenue and expenses and changes in fund
balances for the years then ended. Our examinations were made in
accordance with generally accepted auditing standards and, accord-
ingly, included such tests of the accounting records and such other
auditing procedures as we considered necessary in the circumstances.
In our opinion, the financial statements referred to above
present fairly the financial position of Marine Biological Laboratory
at December 31, 1982 and 1981, and its current funds revenue and
expenses and the changes in fund balances for the years then ended,
in conformity with generally accepted accounting principles applied
on a consistent basis.
u
Boston, Massachusetts
April 22, 1983
48 MARINE BIOLOGICAL LABORATORY
MARINE BIOLOGICAL LABORATORY
BALANCE SHEETS
December 31, 1982 and 1981
Assets 1982 1981
Current funds:
Unrestricted:
Cash and savings deposits $ 198,102 $ 212,262
Money market securities
(Note F) 665,000 1,850,000
Accounts receivable, net of
allowance for
uncollectible accounts 237,859 623,658
Other assets 5,490 19,531
Due to restricted current
funds (192,134) (597,747)
Due to invested funds (93,335) (90,133)
Due to restricted plant fund (163.676) (720.535)
Total unrestricted 657.306 1.297.036
Restricted:
Accounts receivable 368,958 346,828
Investments, at cost (Notes B
andF) 2,194,297 2,179,531
Due from unrestricted
current fund 192,134 597,747
Due from invested funds 66,203 350,967
Total restricted 2,821,592 3,475,073
Total current funds $ 3,478,898 $ 4.772,109
Invested funds:
Investments, at cost (Notes B
and F) 4,630,893 4,488,885
Due from unrestricted current
fund 93,335 90,133
Due to restricted current funds (66,203) (350,967)
Total invested funds $ 4,658,025 $ 4,228,051
Plant funds:
Unrestricted:
Land, buildings and
equipment (Note C) 16,945,601 14,907,184
Less accumulated
depredation 5,203.404 4.843,425
Total unrestricted 11,742,197 10.063,759
Restricted:
Due from unrestricted
current fund 163,676 720.535
Total restricted 163,676 720,535
Total plant funds $11,905,873 $10,784,294
The accompanying notes are an integral part of the financial statements.
REPORT OF THE CONTROLLER 49
MARINE BIOLOGICAL LABORATORY
BALANCE SHEETS
December 31, 1982 and 1981
Liabilities and Fund Balances 1982 1981
Current funds:
Unrestricted:
Accounts payable and
accrued expenses $ 413,459 $ 530,917
Deferred income 80,089 77,138
Fund balance 163.758 688,981
Total unrestricted 657,306 1.297.036
Restricted funds:
Unexpended gifts and grants 2.792.419 3,373,696
Unexpended income of
endowment funds 29-173 101,377
Total restricted 2.821,592 3.475.073
Total current funds $3,478,898 $4,772,109
Invested funds:
Endowment funds 2,184,297 2,218,669
Quasi-endowment funds 1,209,204 934,143
Retirement fund (Note D) 1.264.524 1.075,239
Total invested funds $ 4,658,025 $ 4,228,05 1
Plant funds:
Unrestricted 11,742,197 10,063,759
Restricted 163,676 720.535
Total plant funds $11.905,873 $10,784,294
The accompanying notes are an integral part of the financial statements.
50
MARINE BIOLOGICAL LABORATORY
00 ON
00 CN
ON — .'
in CN
m CN
ri
<N
00
o\
NO o
o m
oo oo
(N
• — ' ^ TJ- in
O oo m NO
ON r~ m — •
O NO in NO
Tt ON <N
•^- CN Tt r- ON os in
m rf m — TJ- CN so
ON r-~ — oo r-- NO oo
— r CN ^-' fN r~' O\ CX3
OO ON Tf O O ON —
»/"> O
o —
t— o
O
ON
r
<"*") Tj" CN
— CN OO
OO f*i r*")
O*\ O\ OO
Os (N
ON O
OO O
r<1 (N —
— OO rn OO
ON ON t~-' </1 O
ON rn OO OO
•* O
oo <N
O O
r
ri
CO
UJ
Z
UJ
§
cu
i-H
K*
X
UJ
00
ON
_ SJ
C
(s*
oc,
Q
•o
OS
o
Z
C
C3
^
UJ
fN
fV
D
OO
o
Z
2
03
>
'
UJ
•t r-
00 ON
oq r|
ON' — '
m rj
rn CNJ
(N
NO O
r- —
oo' in'
oo oo
H- 1
MH
on
u
"•"j
Q
_c
U
Z
D
(U
BIOLOG1
URRENT F
(U
Q
-o
•o
"S
V
&5
SJ
_s:
o
w
U
t/5
Z
u.
o3
5
0
<u
en
^<
H
u
5
Z
>->
UJ
I—
UJ
<-2
H
H
on
ON
NO'
f"l ON
oo «n
CN — •
ON TJ-
— • 00'
\O rn TJ-
r<1 — CN
NO OO fi
m ON ON
CN
OO
CN
q
r<->
NO
CN
NO
_
Tf ^t
m
,-j.
CN
•<t r~
0^
ON in
O
__
ro
00
o
oo in
*O
in
*~T
*n —
CN NO
00
n
— -
ON
CN
ON
t~~
~~, °°,
r—
^C 00
ON
q
q
in
— '
NO' rK
r-4
, — ',
r'l
•*' CN
[^T
ON OO'
r-'
r-'
ri
. — .
o
sO in
OO
ON
t O
0
ON —
^~
oo
•TJ-
in
ON
CN
f*1
f^
CN
r4
in
v»
~^
rn
ON
r i
r-f
<^>
NO f>
ON
NO
O
O
, — ,
ON
0
r-
ON
»}•
f —
oo
CN
T}-
o-.
NO
^^
OO r^)
oo
CN
ON
r--
—
m'
in
ON
ON
r-' in
o'
T}-'
oo'
__'
NO'
oo
0
rn
ON
r^> oo
OO
r^i
—
<^
^
r*^
ri
—
CO
rn
CN
ri
NO
^^
ro
1
ts
h
'•3
s related to researc!
3
C
«t ,
(U
t reimbursement o
Instruction
Research
very of indirect co:
and instruction
Summer program
Year-round progra
Other
on
ort activities:
Dormitory
Dining hall
Library
Biological Bulletin
Research services
Marine resources
Other
1 support activities
itment income
a
1
c
o
•s n
w
£ £
<a
o
•- Q.
6
V
p 3
H on
o
H
IS
REPORT OF THE CONTROLLER
51
s
so
in
IT,
SO
tO
in
so
fN
Cs
Tj" rj ON
m so
0 O
-t
O
^_
_^
rn oo m
rn rj-
oo O
sD
0
sO
r-
O rj >/->
Wi ON
sO ON
rn
rn
SO
oo
•^t fN r~-
in rn
ON OO
oo
O
so'
rn
oo ON r~-
O oo
so r-
oo
oo
rn
•^J-
— — m
— ^
rj oo
oc
rn
so
• — '
rn
so'
M
00
r- oo —
ON ^J- V~i
ITl OO SO
r--' rsf ON
— m —
>A-I oo m
rsf
fN m
ON OO
r- r-
00 sD
fN
oo
|
_ r-.
^" m
o —
O rn
T^-
rn
ON
ON r-
r-~ — •
^° ON
in ^
OO
ON
m
ON OO
r-- so
rn r~
— • r--
ri
ri
O
— O
oo —
OO ON
oo —
— ^
oo
sO
rj ri
rn —
•* rj
ON ON
sO
rs
rn
•^
^
§
^- r- in
sO so SO
m m w-i
§O v"^
ri so
rt ^f r%i
ri
ON
ri
O
so"
-r
OO so
ON "/">
sO
r-
rj
r-
r i
<N
r- rn
o o
r4 oo
rs| — i
ri iy,
fN sO
ON
— rj
O so
ON ri
ON
O
sO
r i
ri
so'
r-
r-
oo
oo
so rn
sO
m
~t
fN
O> "">
sC
-
r~
O
Tf
ON
f^~,
OO
rn rr,
-^~
00
— >
o
fN
OO
~
fN
in iy~>
a-
so
•y-
oo
r-
-t
r i
— Vl
rn
ON
r- 1
>/">
m'
</~i
oo
ON
rn O
oo
SO
r-
OO
rj
ON
n —
(N
oo
oo
rj
<N
rn sO
m r^
00 sO
ON ON
O rj
fN
>n
rn
OO
^
r-~
r~
,j.
m
rj
- — •
—
r~
0
rn
0
—
o
r~
m
ON
3>
r-
("*",
— '
t
ON
xO
— .
fN
[ —
m
^
oo
3^
SO
rn
[^-
r^
^j-
r-*
so'
O
^
~
^~
— M
oo
ON
r~
KM
sO
O
sO
ri
fN
fN
—•
T
fN
ON
•y-
oo
rn
rn
W
C
<u
2
'o
c
c
It
C
03
O.
II
3
C
T3
C
and
p
cti
s
enses:
.a j=
%%«
3 O3 —
S S-S
= ^^
Support ac
)ormitory
73
ofl
C
E
/~\
c
a
xi
c
~-
H
~3
CQ
"?5
o
'5j
o
0
Research services
Marine resources
idministration
'lant operation
Total support ac
D.
X
<L>
o
pe
Exce
03
u*
00
c
c
accompanying no
52
MARINE BIOLOGICAL LABORATORY
o r-
\O O*\ ^^
i/^ O O
O^ ^^ O*1
2 •$
t3 ^
fN
(N
O —
SO sO sO O4
sD OO sO "/I
so r-- ir> o
rn so 04 — '
sO so ' — ' rn
so' ^ ^~
Os
ON
ON
rn
so'
ON
ft,
s;
> ^
•~ O4
o
o
O
>x
sO
rn
oo
so
so'
O
sO
O
oo'
ON
oi
so
I/I
m
tr>
o
ri
r-
so*
sO
ON
r-
r~.
i/1
i/n
r^
r--
r-'
so'
rn
O
SO
04
O
O
^^
oo
\S) \O O^
oo
UJ
00
ON
1
0
^ o^ ^^
r-; o o
\o *^"* r^*"
^*
U
£;
00
Tt rn rn
*/
Z
T3
00
"^
0
J
C
c3 ^
^
M
^
CQ
rN 2
^_
^J
oo 5
^
r*"i
o
oa
Z
— ^,
'i 1
3 =
I
•^
U.
m Z
_s:
^
Z
W- C!
^
6(9
U
5
o
HANGES
<u
o
OJ
Q
ndowmeni
o
o
1/1
r-'
r-
o
NO O
r- Tj-
04 ~
j
U
•o
ty
</^
o
5
UH
0
OO
•o
c
(U
^
04
O O OO </1
•^- oo oo so^
z
(-
Z
UJ
(/5
^
S!
o
ON O rn rn
O oo O •t
f*^~ ^^ f^
5
S
>, r§
£^
rn
rn
<^
R
(U "^
%
<
•*-• s
--3
ON
rr m —
H
i_ b
Xt
m
— 04 oo
C/D
^^ ^.
«2 c
• ii
ON
^r o or
55
O
rs TT oo
gj
OO
O OJ O4
^:
o
*
rn
ON
SO
so
04 ON
r- \O
O4 OO
O s£
rn
00
rn
sO
rr
O
O
o
r-
oo
oo
O
ON
rn
r-
o
m
ON
ON
sO
sO
oo'
(N
rn
o
r~
rn
oo
ON
oo'
oo
so
0
oo
ON
rn
nt funds revenue
"o
15
t/5
C
u °
'o
en
O
Q.
•5
c
o
c
CM
c
u
E
2
m
00
«
0
loners
lances before
00
ON
m
u
(L)
E
E S
r\ • —
1/5
If, . —
K
l«
03
(L)
i/i
S
C
^*~i.
i_
mces at Decemb
eases:
Unrestricted cu
<A
+*
C <s>
ss
oo
Investment ina
Realized net ga
investment
Realized net ga
fixed assets
Addition to reti
Tuition
reuses:
Instruction, res<
expenses
Indirect costs
Payments to pe
Depreciation
-a
c
.5 £
<U ,U
£>£
03 03
J^ l-
O ^
1?
,1
c
*>Si
^~
1
o
Q
on
4J
O
N
03
m
fe.
"o
Q
1J
&
m
REPORT OF THE CONTROLLER
53
— oo r-- o\
r*~> n os r-
^T Tf — r«-i
in — •" <•«•>' r-'
Os OO OO OO
— m OO in
O
-I-
Cs
o
in
O
o
v~it O
O rj
Os
oo
(N
SO
n
Os
r
r—
oo
t
<N
Os
&
Os O
Os' Os'
SO
<N
ON
00
sO
r-
sq
rr-i
SO
SO
oo
Os
OO
o
c
^
O
o
O
Tt
Os
r-
Os
i--
oo
V)
m
O
SO
rj
oo
ri
Os'
OO
sO
r-
•8
sO
rn
o
Os'
o
<N
rn
r-
Os_
m'
fN
s:
^
I
G
oo
Os m
m
^o o^
o
— O
fN
C
rj
sO Os
O >O
ON
ri TT
rt m
Os
O
— oo
Os ^~
Os
E
— .'
r~ ri
ri
ri
so' rn
^-'
r*( in
_'
<L>
^o
i • —
(N
i~- r-
sO
r- — '
rj
C3
"1
r\i m
Tf t-~
<n
oo
to
<->
r*{ ^"
ri
Is
'o
C
—
O O
rT
' ^
r7
oo
C8
oo — i
Os
1^
r- os
r^
SO
[--.
"
«n
— ' oo'
so'
O
,-j."
rn
r;
OS
f*~l ~—
O
SO
so
SO
~~*
r-j <N
oo
—
rn
CM
CO
rn
^•^
^^
O
' '
ft
C
C3
D.
1>
4
—
(M
o
_
S
1
to
u
u
U
4-*
OJ
c
cd
y
u«
C
1_
GO
o
•eases:
Unrestricted current funds i
Grants
Gifts
Investment income
Realized net gains (losses) c
of investments
Realized net gains on dispo
fixed assets
Addition to retirement func
Tuition
rawes:
Instruction, research and ge
expenses
Indirect costs
Payments to pensioners
Depreciation
change in fund balances bef
transfers
•75/eri- /o (/rowy funds:
Acquisition of fixed assets
Reclassification (Note A)
Other
nces at December 31, 1982
accompanying notes are an
C
S
M\
5 J2 <u
£
Q
z
r C^ J—
K OQ H
54 MARINE BIOLOGICAL LABORATORY
NOTES TO FINANCIAL STATEMENTS
A. Purpose of the Laboratory:
The purpose of Marine Biological Laboratory (the "Laboratory") is to establish and maintain a
laboratory or station for scientific study and investigations, and a school for instruction in biology
and natural history.
B. Significant Accounting Policies:
Basis of Presentation — Fund Accounting
In order to ensure observance of limitations and restrictions placed on the use of resources available
to the Laboratory, the accounts of the Laboratory are maintained in accordance with the principles
of "fund accounting." This is the procedure by which resources are classified into separate funds
in accordance with specified activities or objectives. In the accompanying financial statements,
funds that have similar characteristics have been combined.
Externally restricted funds may only be utilized in accordance with the purposes established by the
source of such funds. However, the Laboratory retains full control over the utilization of unrestricted
funds. Restricted gifts, grants, and other restricted resources are accounted for in the appropriate
restricted funds. Restricted current funds are reported as revenue when expended for current
operating or other purposes. Unrestricted revenue is reported as revenue in the unrestricted current
fund when earned.
Endowment funds are subject to restrictions requiring that the principal be invested with income
available for use by the Laboratory. Quasi-endowment funds have been established by the Labo-
ratory for the same purposes as endowment funds; however, any portion of these funds may be
expended.
Reclassifications
The financial statements for 1982 reflect certain changes in classification of revenue. Similar re-
classifications have been made to amounts previously reported in order to provide consistency of
the financial statements. In addition, certain invested funds' balances have been reclassified to
appropriately reflect the donors' intentions.
Investments
Investments purchased by the Laboratory are carried at cost. Investments donated to the Laboratory
are carried at fair market value at date received. For determination of gain or loss upon disposal
of investments, cost is determined based on the average cost method.
Investment Income and Distribution
The Laboratory follows the accrual basis of accounting except that investment income is recorded
on a cash basis. The difference between such basis and the accrual basis does not have a material
effect on the determination of investment income earned on a year-to-year basis.
Investment income includes income from the investments of specific funds and from the pooled
investment account. Income from the pooled investment account is distributed to the participating
funds on the basis of the market value at the beginning of the quarter, adjusted for the cost of any
additions or disposals during the quarter.
C. Land, Buildings and Equipment:
Following is a summary of the unrestricted plant fund assets:
Classification 1982 1981
Land $ 720,125 $ 719,798
Buildings 14,360,395 12,535,197
Equipment 1,865,081 1,652,189
16,945,601 14,907,184
REPORT OF THE CONTROLLER 55
Less accumulated depreciation 5,203.404 4.843.425
$11,742,197 $10,063,759
Depreciation is computed using the straight-line method over estimated useful lives.
D. Retirement Fund:
The Laboratory has a noncontributory pension plan for substantially all full-time employees which
complies with the requirements of the Employee Retirement Income Security Act of 1974. The
actuarially determined pension expenses charged to operations in 1982 and 1981 were $160,554
and $ 1 37,009, respectively. The Laboratory's policy is to fund pension costs accrued, as determined
under the aggregate level cost method. As of the latest valuation date, based on benefit information
obtained January 1, 1983, the actuarial present values of vested and nonvested benefits, assuming
an investment rate of return of 6%, were approximately $1,076,652 and $39.562, respectively. At
January 1, 1983 net assets of the plan available for benefits, were approximately $1,364,107.
In addition, the Laboratory has a pension plan funded by contributions to the Teachers Insurance
and Annuity Association.
E. Pledges and Grants:
As of December 31, 1982 and 1981, the following amounts remain to be received on gifts and
grants for specific research and instruction programs, and are expected to be received as follows:
December 31. 1982 December 31, 1981
Unrestricted Restricted Unrestricted Restricted
1982 $20,000 $ 96,800
1983 $63,000 $ 83,400 95,000
1984 52,000 40.000
$63,000 $135,400 $20.000 $231.800
In February 1979, the Laboratory initiated the MBL Second Century Fund, a phased effort, to
secure $23,000,000 in support of capital rehabilitation, new construction, and endowment. As of
December 31, 1982, the Laboratory has received pledges related to this effort of approximately
$4,553,000 of which a substantial portion has been collected.
56
MARINE BIOLOGICAL LABORATORY
TD
ca
O fN u~l —
~- fN o •* r~
Tt OO
ON >/•>
NO
m
income
Income
§> fN OO' Tfr O
i^ O fN 10 fN
t/~i r —
O fN
r-'
I--
<N
itment
iment
>
3
fN NO ^J" OO
O ON"
^^
c
£
r oo oo r- fi
rJJ CO l/^ <"*"} V~l
QO ^ ^
oo <•<•>
5
1
Os t~~ O ON Tf
"-• ON fN OO fN
m fN
oo'
o
s
•o
c
O O r . ON
r-
ON f"> r) O ^t
ON
'
( — ON CO fN */~t
T
00
OQ — r- ^t oo r-
a^
O\ f^l NO CO ^ ^™
t*}
—
~^ oo — m fN
t~~t
T3
m
TT
C
</5
*^
00
*
ri
^^
<f}
c
ON
m
m
^
o
lo
.7,
oo
u
QO ^o
r-
m
rr
r
ON
f~
O^ **O
1^^
r-
fN
—
m
E
~- — .
f>
"~'.
fN
fi
o
^
<>e
0
Q
fa
oo
cS
TD
C
ca
r4
_^ t^
O
ON
5
c:
ON
oo
00
C
O
fN
oo
QO oo
Ov —
r^
.c
DC
t~-
oo'
oo
E
—
-^ ON
eN
fx
(-)
5
c
u
fc S
~c
1 ^
g\
o
3
CO
o
Q
OO
ON
r-
0
ON
1*1
ON
OO
•p
Oo fN
_;
•/•>
Tj-
r-
o
O
u
Ov ON
NO
ro
ri
—
x
T3
-^, —
f*^
00
fN
NO
—
CO
C
—
<N
Tfr
E
C
(A
fa"
TD
ca
C
ca
ca
u
o
4^
o
U.
u
(O
_f^
(/)
*— •
U
c
"o
o
^^
o
«
^
c
oi
ca
a surnm;
estment
•c
3
CJ
^
G
E
0
unties
to
O
O
fa
cyi
c
C
c
o
u
C/5
OJ
OO
r ,
U
-o
•^
ca
_^
C
_o
o
c
,o
E
c
u
"^3 O
<n
X
'-C
2
0
C
C
0
1
_§
5
O
H
c
ca
H
o
o.
(/I
'•o
1 °
"S5 ^
•Si ^
1
0
o
E
E
o
U
>\
c
o
cj
"ca
(2
'•5
o
3
O
REPORT OF THE CONTROLLER
57
r*"} ^1" i — f""*)
•* VO
OO
o
^j.
\O X: — i </~>
— IN
OO
oo
0
~-, 1 1 Tf Tf
<N —
O
ON
r—
<x>
o= ON o rn rtj
r — ON
oo'
r-'
£
ON ' —
00
r-
2
o
*~:
^j
3
%
ON
OO 1 — NO
o
ON
_
K
<N
00
OO
O
oo O ON
r-
oo
ON
ON
r-
00
ex
^*
oo' o' rn
rn
O
IN
*^
NO </"> —
o
ON
,3.
_r
oo
u-i
, ,
g
r-
o
r-
It
oo
p^_
""l
rn
oo
r-
O
•
m'
O
rn
vO*
r-'
rn
OO
IN
ri
O
rn
oo
IN
NO
ON
r-
<**!
r-
ro
ON
ON
u^
IN
r}- O
r- —
OO ON
<X oo'
i/~i —
<N
ON
r-
rn
r-'
oo
§
g
f \ •/") r*~j ^^
c~ i r* ) m ON
r^j rn — • »/%
•"t O — ' rn
ON — ON —
rl rn OO OO
TJ- rn OO "/I
— ^ ^^ l/°"l ON
I — ^" OO rn
>/~t m (^j <N
ON
>r>
O
1^1
oo
r i
IN
NO
NO
rn
!/^J
r- NO
^^
o
JO
^f ^^
rn
_^
ON
ON
O "~>
</^
00
!/">
ON
O m'
ON
o\
•/^
1-
O t^
p-
~^
ON
— '
ON —
— ^
ri
q
o
OO *S~i NO OO
P-^
O OO ^t "/"l
ON
fN
f"^l ON r*~l ON
00
r--' ON oo' oo'
Tj-
ON
I — ^T IN rn
ON
"-<
»/~i m o i^J
—
IN
NO
en
(U
c
_o
Si
z
o
'C
(U
C/1
il
3
O
E
o
•c
C/5
1
C
c
<Sl
0
^
<L>
•o
l .
-Q'
e
c
x
0
0
u
^
k^
1)
<u
c
n
G
^
0
^3
o
1
O
06
1
0
omm
c
o
S
b
O
U
~^i
T3
C
3
to
HJ
C
05
3
O
- ^
03
g
U
C
o
irrent use:
current
3
^
O ^
03
(U
"C
i« CX
o E
3
•o
O
'C
^ -a
3
O
C
T3
U
N
p^
0 C
u-
"™ —
O3 f^" C
2 §
3
C/3
c
.0 -o
"*^ Qj
o .2
°.a-s
o c-
H H
IU
0
0
_o
2
l'5
>
<
^— 1
D
E
•u
0
a.
G
58 MARINE BIOLOGICAL LABORATORY
IX. REPORT OF THE LIBRARIAN
1982 was the year that the walls literally came tumbling down on three floors
of the Lillie Building that housed the Library and Administration offices. It was a
strange time for the staff. We placed five staff members at reserved desks in the stack
wing and the rest of the staff occupied four summer labs in the Crane wing. We
communicated by intercom for five months.
The entire office area (copy center and catalog room) was renovated with new
walls and lowered ceilings. The Reading Rooms remained intact and so did the five
floors of stacks. Therefore we were able to operate with access to the collection on
the usual 24 hour basis, provided that the users were able to concentrate with the
jackhammers, drills, and DUST. It was surprising how many did and seemed obliv-
ious to the din.
The contractors left in May and returned in September to complete the work.
We moved the entire book collection to a new section of the Library on the third
floor at the beginning of 1983, and the Rare Books collection was returned to the
Library and placed on the first floor in what was formerly the Administration area.
The summer exhibits were moved from the Lillie Lobby to the first floor of
Swope because of a confusing pattern of traffic due to the remodeling. We were able
to accomodate more exhibitors due to the increased amount of space but the general
opinion was that the lobby in Lillie is a more central area for viewing exhibits. They
will return there in 1983.
The Library Users Committee spent a number of meetings discussing a grant
from the Rockefeller Foundation which will enable us to conduct a major study of
utilization patterns of our periodical collection. This User Study (carried out over
a twelve month period) will enable us to develop long-range plans and policy, as
well as establish cost effective procedures in acquisitions. The Study will start at the
beginning of 1983.
X. EDUCATIONAL PROGRAMS
SUMMER
BIOLOGY OF PARASITISM
Instructor-in-chief
DAVID, JOHN, Harvard Medical School/Harvard School of Public Health
Other faculty, staff, and lecturers
ASKENASE, PHILIP, Yale University School of Medicine
AVERY, ROBIN, Harvard University
CANTOR, HARVEY, Sidney Farber Cancer Center
CAULFIELD, JOHN, Harvard Medical School
CROSS, GEORGE, Rockefeller University
DAVID, ROBERTA, Brigham and Women's Hospital
DESSEIN, ALAIN, Harvard Medical School
DWYER, DENNIS, National Institutes of Health
ENGLUND, PAUL, Johns Hopkins University School of Medicine
FEARON, DOUGLAS, Harvard Medical School
GIGLI, IRMA, New York University School of Medicine
GITLER, CARLOS, Weizmann Institute of Science, Israel
HALS, GARY, Capitol University
HARN. DONALD, Harvard Medical School
EDUCATIONAL PROGRAMS 59
LANDFEAR, SCOTT, Harvard School of Public Health
LODISH, HARVEY, Massachusetts Institute of Technology
MARSDEN, PHILIP, Federal University of Brasilia, Brazil
METZGER, HENRY, National Institutes of Health
NATHAN, CARL, Rockefeller University
NELSON, GEORGE, Liverpool School of Tropical Medicine, England
NUSSENSWEIG, RUTH, New York University School of Medicine
OTTESON, ERIC, National Institutes of Health
PEREIRA, MIERCIO, Tufts University School of Medicine
PERKINS, MARGARET, Rockefeller University
PFEFFERKORN, ELMER, Dartmouth Medical School
PRATT, DIANNE, Harvard Medical School
RIFKIN, MARY, Rockefeller University
ROBERTS, BRYAN, Harvard Medical School
SAMUELSON, JOHN, Harvard Medical School
SHER, ALAN, National Institutes of Health
SHERMAN, IRWIN, University of California
SPIELMAN, ANDREW, Harvard School of Public Health
TRACER, WILLIAM, Rockefeller University
WALSH, CHRIS, Massachusetts Institute of Technology
WIRTH, DYANN, Harvard School of Public Health
WYLER, DAVID, Tufts University School of Medicine
Students*
*AVRON, BOAZ, Weizmann Institute of Science, Israel
*BANGS, JAMES, Johns Hopkins University School of Medicine
*BARKER, ROBERT, Brown University
*BOSWELL, CARL, Oregon State University
BUCK, GREGORY, Institut Pasteur, France
*DELAUW, MARIE-FRANCE, Beaumont, Belgium
HALDAR, KASTURI, Massachusetts Institute of Technology
"JUNGERY, MICHELE, University of Oxford, England
*LANGER, PAMELA, Wellcome Trust Research Laboratories, Kenya
*MORIEARTY, PAMELA, Fundacao Oswaldo Cruz, Brazil
*PAMMENTER, MARTIN, Research Institute for Diseases in a Tropical Environment, South
Africa
*ROWSE-£AGLE, DEBRA, Yale University
TSENG, PETER, Johns Hopkins University School of Medicine
*ULISSES DE CARVALHO, TECIA MARIA, Instituto de Biofisica, Brazil
*ZEICHNER, STEVEN, University of Chicago
*ZILBERSTEIN, DAN, Hebrew University, Israel
EMBRYOLOGY
Instructor-in-chief
RAFF, RUDOLF, Indiana University
ANGERER, LYNNE, University of Rochester
ANGERER, ROBERT, University of Rochester
BEDARD, ANDRE, McGill University, Canada
BEGG, DAVID, Harvard Medical School
1 All summer students listed completed the formal course programs. Asterisk indicates those completing
post-course research sessions.
60 MARINE BIOLOGICAL LABORATORY
BENNETT, JEAN, University of California at Berkeley
BLUMENTHAL, THOMAS, Indiana University
BRANDHORST, BRUCE, McGill University, Canada
CROUCH, MARTY, Indiana University
DOHMEN, RENE, University of Utrecht, Netherlands
EPEL, DAVID, Stanford University
FREEMAN, GARY, University of Texas at Austin
GERHARDT, JOHN, University of California at Berkeley
GRAINGER, ROBERT, University of Virginia
GROSS, PAUL, Marine Biological Laboratory
HARKEY, MICHAEL, University of Washington
HENDERSON, JUDY, State University of New York, Buffalo
HENRY, JONATHAN, University of Texas at Austin
HEREFORD, LYNNA, Sidney Farber Cancer Institute
HILL, DAVID, Harvard Medical School
HILLE, MERRILL, University of Washington
HORVITZ, ROBERT, Massachusetts Institute of Technology
HUMPHREYS, THOMAS, University of Hawaii
JEFFERY, WILLIAM, University of Texas at Austin
JOHNSON, MARTIN, Cambridge University, England, U. K.
KALTOFF, KLAUS, University of Texas at Austin
KAUFMAN, THOMAS, Indiana University
KLEIN, WILLIAM, Indiana University
KOMAROFF, LYDIA, University of Massachusetts Medical School
LEAF, DAVID, Indiana University
MAHOWALD, ANTHONY, Case Western Reserve University
McCLAY, DAVID, Duke University
MELTON, DOUGLAS, Harvard University
NEWROCK, KENNETH, McGill University, Canada
PENMAN, SHELDON, Massachusetts Institute of Technology
POLISKY, BARRY, Indiana University
POSAKONY, JAMES, Harvard University
RAFF, BETH, Indiana University
RENDER, JoANN, University of Texas at Austin
RICH, JESSICA, Brown University
RUDERMAN, JOAN, Harvard Medical School
SADOWNICK, BRUCE, Harvard University
SANDER, KLAUS, University of Freiburg, West Germany
SOWERS, Louis, Indiana University
STERNBERG, PAUL, Massachusetts Institute of Technology
STUHL, KEVIN, Harvard Medical School
TRINKAUS, JOHN, Yale University
VACQUIER, VICTOR, Scripps Oceanographic, University of California at San Diego
WHITTAKER, J. RICHARD, Boston University Marine Program
WOOD, WILLIAM, University of Colorado
Students1
*BROWN, NICHOLAS, Harvard University
*£MERSON, JULIA, University of California at San Francisco
*GOULD, MITCHELL, Emory University
*HouGAN, LINDA, McGill University, Canada
*LEBLANC, JANINE, Wesleyan University
*LESK, MARK, McGill University, Canada
*LINGAPPA, JAIRAM,- Harvard University
*MAPLES, PHILLIP, Oklahoma University
*MARTINDALE, MARK, University of Texas at Austin
EDUCATIONAL PROGRAMS 6 1
*MARTONE, ROBERT, University of Vermont
*MERLINO, GLENN, National Institutes of Health
*PEARMAN, BRADLEY, University of Tennessee
*PERRY, HEATHER, University of Chicago
*POZNANSKI, ANN, University of California at San Francisco
*PULTZ, MARY ANNE, Indiana University
*SUTHERLAND, ANN, University of California at San Francisco
*ST. JOHNSTON, DANIEL, Harvard University
*THOMSEN, GERALD, Washington University
*VAFOPOULOU-MANDALOS, XANTHE, University of Connecticut
"VERAKALASA, PACHARA, University of Hawaii
*WHARTON, KRISTI, Yale University
*WILSON, LINDA, University of Texas at Austin
*ZWIEBEL, LAURENCE, University of Michigan
MARINE ECOLOGY
Instructors-in-chief
TEAL, JOHN, Woods Hole Oceanographic Institution
VALIELA, IVAN, Boston University Marine Program/Marine Biological Laboratory
Other faculty, staff, and lecturers
ALBERTE, RANDALL, University of Chicago
ANDERSON, DONALD, Woods Hole Oceanographic Institution
CONNELL, JOSEPH, University of California at Santa Barbara
DACEY, JOHN, Woods Hole Oceanographic Institution
DAVIS, CABELL, Boston University Marine Program/Marine Biological Laboratory
DENNISON, WILLIAM, University of Chicago
GIBLIN, ANNE, Woods Hole Oceanographic Institution
GLIBERT, PATRICIA, Woods Hole Oceanographic Institution
GRASSLE, FREDERICK, Woods Hole Oceanographic Institution
GRASSLE, JUDITH, Marine Biological Laboratory
GROSSBERG, RICHARD, Yale University
HOBBIE, JOHN, Marine Biological Laboratory
HUMES, ARTHUR, Marine Biological Laboratory
JANNASCH, HOLGER, Woods Hole Oceanographic Institution
JEFFRIES, ROBERT, University of Toronto, Canada
KOEHL, MIMI, University of California at Berkeley
LEVINTON, JEFFREY, SUNY at Stony Brook
MADIN, LARRY, Woods Hole Oceanographic Institution
MANN, ROGER, Woods Hole Oceanographic Institution
NIXON, SCOTT, University of Rhode Island
ODUM, WILLIAM, University of Virginia
PETERSON, BRUCE, Marine Biological Laboratory
PETERSON, SUSAN, Woods Hole Oceanographic Institution
STOECKER, DIANNE, Woods Hole Oceanographic Institution
TAGHON, GARY, Woods Hole Oceanographic Institution
WATKINS, WILLIAM, Woods Hole Oceanographic Institution
WIEBE, PETER, Woods Hole Oceanographic Institution
WOODWELL, GEORGE, Marine Biological Laboratory
Students*
ABAD, MARK, University of Chicago
*ANUTH, CRAIG, Oberlin College
BOUTROS, OSIRIS, University of Pittsburgh
62 MARINE BIOLOGICAL LABORATORY
BROWN, ALEXIS, California State University at Dominguez Hills
DEMUTH, ROBIN, Childrens Hospital, Boston
DOETKOTT, CURT, North Dakota State University
ENGLER, MARLIES, Weiterbildungsschule/Diplommittelschule des Kantons Zug,
Switzerland
GROSS, CHARLES, Southeastern Massachusetts University
KOWALLIS, GEORGE, New York Medical College
LUBE, FATIMA, Rio de Janiero, Brasil
*MARZOLF, ERICH, Colorado College
MEROW, ALISON, Stanford University
MORROW, LAURA, University of Texas at Austin
OLSEN, SCOTT, Lehigh University
REVELAS, EUGENE, State University of New York at Stony Brook
*SENIE, ALLYSON, Ithaca College
*SMITH, ROBERT, University of Chicago
*STODDARD, JEFFREY, University of Wisconsin at Madison
TREGGOR, JOSEF, Central Connecticut State College
*WEISSBURG, MARC, University of California at Berkeley
MICROBIAL ECOLOGY
Instructor-in-chief
HALVORSON, HARLYN, Brandeis University
Other faculty, staff, and lecturers
ALEXANDER, MARTIN, Cornell University
ATWOOD, KIMBALL, Columbia University
BREZNAK, JOHN, Michigan State University
CASTENHOLZ, RICHARD, University of Oregon
DAVIS, BERNARD, Harvard Medical School
DWORKIN, MARTIN, University of Minnesota at Minneapolis
GARDNER, JEFFREY, University of Illinois at Urbana
GREENBERG, PETER Cornell University
HANSON, RICHARD, University of Minnesota at Minneapolis
JANNASCH, HOLGER, Woods Hole Oceanographic Institution
KEYNAN, ALEX, Hebrew University of Jerusalem, Israel
KORNBERG, HANS, Cambridge University, England, U. K.
MARRS, BARRY, University of Connecticut at Storrs
NICKERSON, KENNETH, University of Nebraska at Lincoln
POINDEXTER, JEANNE, Public Health Research Institute, New York
POTRIKUS, CATHERINE, Harvard University
REZNIKOFF, WILLIAM, University of Wisconsin at Madison
ROMESSER, JAMES, Dupont Corporation
RUBY, EDWARD, University of California at Los Angeles
SCHWEIGER, HANS, Max-Planck Institute, West Germany
SINNIS, FRANNIE, Woods Hole Oceanographic Institution
SLATER, HOWARD, Cambridge University, England, U. K.
TAYLOR, CRAIG, Woods Hole Oceanographic Institution
UHLINGER, DAVID, Florida State University
VINCENT, WALTER, University of Delaware
WEISBLUM, BERNARD, University of Wisconsin at Madison
WHITE, DAVID, Florida State University
Students*
*BOUTROUS, SUSAN, University of Pittsburgh
*BRATBAK, GUNNAR, University of Bergen, Norway
EDUCATIONAL PROGRAMS 63
*FATTUM, ALI, Hebrew University of Jerusalem, Israel
*FOSNAUGH, KATHY, Cornell University
*HALL, ROBERT, Nantucket High School
*HAPPEL, ANNE, Purdue University
"HEIMBROOK, MARGARET, University of Northern Colorado
*HULLAR, MEREDITH, Tallahassee, Florida
*KERKHOF, LEE, Harvard University
*MAY, HAROLD, Virginia Polytechnic Institute and State University
*PADEN, CYNTHIA, Scripps Institution of Oceanography
*PERNACK, TINA, Arizona State University
*SCHMIDT, THOMAS, Ohio State University
*SCHNELL, DANNY, University of Nebraska at Lincoln
*STAHL, DAVID, National Jewish Hospital and Research Center
*VENTOSA, ANTONIO, University of Sevilla, Spain
*WIER, PATRICIA, University of Colorado at Boulder
*WOGRIN, NANCY, University of Massachusetts at Amherst
NEURAL SYSTEMS AND BEHAVIOR
Instructors- in-ch ief
HOY, RONALD, Cornell University
MACAGNO, EDUARDO, Columbia University
Other faculty, staff, and lecturers
CALABRESE, RONALD, Harvard University
CAREW, THOMAS, Columbia University
ERBER, JOCHEM, Free University of Berlin, Germany
FARLEY, JOSEPH, Princeton University/Marine Biological Laboratory
GELPERIN, ALAN, Princeton University
HARRIS-WARRICK, RONALD, Cornell University
KELLEY, DARCY, Princeton University
KROODSMA, DONALD, University of Massachusetts
LEVINTHAL, CYRUS, Columbia University
LLINAS, RODOLFO, New York University/Marine Biological Laboratory
NELSON, MARGARET, Cornell University
NOTTEBOHM, FERNANDO, Rockefeller University
O'NEILL, WILLIAM, University of Rochester
PALKA, JOHN, University of Washington
WURTZ, ROBERT, National Eye Institute
ZIPSER, BIRGIT, Cold Spring Harbor Laboratory
Students1
BERARDUCCI, ALBERT, University of Massachusetts Medical School
CAHILL, GREGORY, University of Oregon Institute of Neurosciences
CRAWFORD, JOHN, Cornell University
EDGECOMB, ROBERT, Purdue University
FERME, PAOLA, Boston University
HOCH, DAVID, Albert Einstein College of Medicine
HOOPES, CHARLES, Wake Forest University
*KRAFT, TIMOTHY, University of Minnesota
LEWENSTEIN, LISA, New York Medical College
*MARLER, JENNIFER, McGill University, Canada
*NICOL, DIANNE, Dalhousie University, Canada
*NORRIS, BRIAN, Texas Tech University
PIRES, ANTHONY, Harvard College
RANKIN, CATHERINE, City University of New York
64 MARINE BIOLOGICAL LABORATORY
REDMOND, TIM, Case Western Reserve University
REHDER, VINCENT, Free University of Berlin, Germany
RUSAK, BENJAMIN, Dalhousie University, Canada
SAUNDERS, JAMES, University of Oklahoma
SCHUTRUMPF, ANDREW, Northeastern University
SMITH, KENNETH, Columbia University
NEUROBIOLOGY
Instnictors-in-chief
HILDEBRAND, JOHN, Columbia University
REESE, THOMAS, NINCDS/National Institutes of Health
Other faculty, staff, and lecturers
AUERSWALD, COLLETTE, Radcliffe College
ARMSTRONG, CLAY, University of Pennsylvania
BATTELLE, BARBARA, NEI/National Institutes of Health
BURD, GAIL, Massachusetts General Hospital
CHRISTAKIS, NICHOLAS, Yale College
DUDAI, YADIN, Weizmann Institute, Israel
DUNLAP, KATHLEEN, Tufts University Medical School
FISCHBACH, GERALD, Washington University School of Medicine
FURSHPAN, EDWIN, Harvard Medical School
GOODMAN, COREY, Stanford University
GOULD, ROBERT, New York Institute for Basic Research in Mental Retardation
GOY, MICHAEL, Harvard Medical School
GRAHAM, WILLIAM, NINCDS/National Institutes of Health
GRANT, PHILIP, University of Oregon
GRAYBIEL, ANN, Massachusetts Institute of Technology
HALL, LINDA, Albert Einstein College of Medicine
HERBERT, EDWARD, University of Oregon
HORVITZ, ROBERT, Massachusetts Institute of Technology
HUTTNER, SUSANNE, University of California at Los Angeles
KACHAR, BECHARA, NINCDS/National Institutes of Health
KENT, KARLA, Columbia University
KRAVITZ, EDWARD, Harvard Medical School
LAFRATTA, JAMES, Harvard Medical School
LANDIS, DENNIS, Massachusetts General Hospital
LANDIS, STORY, Harvard Medical School
LANE, NANCY, University of Cambridge, England, U. K.
LATORRE, RAMON, Harvard Medical School
MANSOUR, RANDA, University of Rhode Island
MATSUMOTO, STEVEN, Harvard Medical School
NICHOLLS, JOHN, Stanford University
NISHI, RAE, Harvard Medical School
O'CoNNELL, MAUREEN, NINCDS/National Institutes of Health
O'LAGUE, PAUL, University of California at Los Angeles
PAGANO, RICHARD, Carnegie Institution
POTTER, DAVID, Harvard Medical School
RAHAMIMOFF, RAMI, Hebrew University Medical School, Israel
RAND, PETER, Brock University
RAVIOLA, ELIO, Harvard Medical School
REESE, BARBARA, NINCDS/National Institutes of Health
SEJNOWKI, TERRANCE, Harvard Medical School
SHEPHERD, GORDON, Yale University
EDUCATIONAL PROGRAMS 65
WEINSTEIN, JOHN, NCI/National Institutes of Health
WALROND, JOHN, NINCDS/National Institutes of Health
WHITE, EDWARD, Boston University School of Medicine
WIESEL, TORSTEN, Harvard Medical School
WOLF, DAVID, Worcester Foundation for Experimental Biology
ZIGMOND, RICHARD, Harvard Medical School
Students1
*CHANG, DONALD, Baylor College of Medicine
*COOK-DEEGAN, ROBERT, University of Colorado
*DAVID, SAMUEL, Montreal General Hospital, Canada
*Fujn, JOANNE, University of California at San Diego
*HISHINUMA, AKIRA, Columbia University
*HUETTNER, JAMES, Harvard Medical School
*KELL, MICHAEL, Emory University School of Medicine
*LERNER, MICHAEL, Washington University School of Medicine
*LUMMIS, SARAH, University of Cambridge, England
*MiLLS, LINDA, McMaster University, Canada
*NAWROCKJ, LEON, University of Oregon
"O'CONNOR, PATRICIA, University of California at Berkeley
PHYSIOLOGY
Instmctor-in-chief
ROSENBAUM, JOEL, Yale University
Other faculty, staff, and lecturers
ACKERS, GARY, Johns Hopkins University
ALBRECHT, GUENTER, Cold Spring Harbor Laboratory
ALLEWELL, NORMA, Wesleyan University
BARNARD, STEVE, Boston College
BECKWITH, JON, Harvard Medical School
BEYER, ANN, Worcester Foundation for Experimental Biology
BORISY, GARY, University of Wisconsin
BRADY, SCOTT, Case Western Reserve University
BRANTON, DAN, Bio Labs
BRAY, DENNIS, Medical Research Council, England, U. K.
BROGLIE, RICHARD, Rockefeller University
CHILD, ALICE, Tufts University
CHISHOLM, REX, Massachusetts Institute of Technology
CONDEELIS, JOHN, Albert Einstein College of Medicine
CROUCH, MARTHA, Indiana University
DEMAY, JAN, Janssen Pharmaceutical Laboratory of Oncology, Belgium
DILL, KENNETH, University of Florida
GOLDMAN, ROBERT, Northwestern University School of Medicine
GRINNEL, FREDERICK, University of Texas Southwest Medical School
HARTWELL, LEE, University of Washington
HEREFORD, LYNNA, Brandeis University
HOBBIE, LAWRENCE, Yale University
HUNT, TIM, Cambridge University, England, U. K.
INOUE, SHINYA, Marine Biological Laboratory
JAFFE, LIONEL, Marine Biological Laboratory/Purdue University
JOHNSON, KENNETH, Pennsylvania State University
KARN, JOHN, Medical Research Council, England, U. K.
KILMARTIN, JOHN, Medical Research Council, England, U. K.
66 MARINE BIOLOGICAL LABORATORY
KORNBERG, ROGER, Stanford University Medical School
KUMAR, AJIT, George Washington University Medical Center
DELAYRE, JEAN, Harvard Medical School
MATSUMURA, FUMIO, Cold Spring Harbor Laboratory
MOOSEKER, MARK, Yale University
MURRAY, ANDREW, Sidney Farber Cancer Research Institute
OLMSTED, JOANNA, University of Rochester
PEDERSON, THORU, Worcester Foundation for Experimental Biology
POLISKY, BARRY, Indiana University
RAFTERTY, MICHAEL, California Institute of Technology
REID, MARTHA, Earlham College
RICH, ALEXANDER, Massachusetts Institute of Technology
ROSENTHAL, ERIC, Harvard Medical School
SCHACHMAN, HOWARD, University of California at Berkeley
SILFLOW, CAROLYN, University of Minnesota
SLOBODA, ROGER, Dartmouth College
SOLL, DAVID, University of Iowa
SPUDICH, JOHN, Albert Einstein College of Medicine
STEINBERG, JULIE, McAlester College
SZOSTAK, JACK, Sidney Farber Cancer Institute
TILNEY, LEWIS, University of Pennsylvania
TRINKAUS, J. P., Yale University
VILLA, LYDIA, University of Massachusetts Medical Center
WANG, JAMES, Harvard University
WEISENBERG, RICHARD, Temple University
WHITMAN, GEORGE, Worcester Foundation for Experimental Biology
WIEBEN, ERIC, Worcester Foundation for Experimental Biology
Students*
*BRONSON, REBECCA, Boston University
*CENTONZE, VICTORIA, Dartmouth College
*COLUCCIO, LYNNE, Rensselaer Polytechnic Institute
*CONZELMAN, KAREN, Yale University
*DALEY, GEORGE, Harvard University
*DISTEL, DANIEL, Scripps Institution of Oceanography
*FATH, KARL, Case Western Reserve University
*FiNi, ELIZABETH, Dartmouth College
*FRANCIS, RALPH, Oregon State University
*GALLATI, MICHELE, George Washington Medical Center
*GORBSKY, GARY, Princeton University
GUYER, DAVID, Yale College
*HANNEKEN, ANNE, Medical College of Wisconsin
*JosEPH-SiLVERSTEiN, JACQUELYN, Hunter College of City University of New York
*KAMIYA, RITSU, Nagoya University, Japan
*KELLY, WILLIAM, University of Maryland
"KULAKOSKY, PETER, University of Pennsylvania
*LEE, HEIDE, Brown University
LOUIE, DIANE, Yale College
*LUFKIN, THOMAS, Cornell University Medical College
*MORGANELLI, CHRISTINE, Dartmouth College
*NELSON, JAMES, Purdue University
*POOLE, THOMAS, Harvard Medical School
*PORTER, DONALD, Scripps Institution of Oceanography
*QuiGLEY, MICHAEL, University of Virginia
*RAUSCH, DIANNE, Northwestern University
*ROGELJ, SNEZNA, Boston University
EDUCATIONAL PROGRAMS 67
*ROZDZIAL, MOSHE, University of California at Riverside
*STEPHENS, LAURIE, University of Virginia
TAYLOR, LAVENTRICE, University of North Carolina
*WILLIAMS, BENJAMIN, Yale University
*YOUNGBLOM, JAMES, University of Minnesota
JANUARY
BEHAVIOR
Instructor-in-chief
ATEMA, JELLE, Boston University Marine Program/Marine Biological Laboratory
Other faculty, staff, and lecturers
BARLOW, ROBERT, Syracuse University
BERG, CARL, Marine Biological Laboratory
BRIDGES, ROBERT, Harvard Medical School
BRISBIN, I. LEHR, Savannah River Ecology Program
CALLARD, GLORIA, Boston University
CAREY, FRANCIS, Woods Hole Oceanographic Institution
DETHIER, VINCENT, University of Massachusetts
DOLPHIN, WILLIAM, Boston University
DORSEY, ELLIE, Payne Laboratories
ELGIN, RANDALL, Boston University Marine Program/Marine Biological Laboratory
FRANCIS, ELIZABETH, Bates College
FRAZIER, JEAN, Boston University
HAUSFATER, GLEN, Cornell University
KALMIJN ADRIANUS, Scripps Institute of Oceanography
KAMIL, AL, University of Massachusetts
KREITHEN, MEL, University of Pittsburgh
LANGBAUER, WILLIAM, Boston University Marine Program/Marine Biological Laboratory
LEVINE, JOSEPH, Boston College
MARLER, PETER, Rockfeller University
MOLLER, PETER, American Museum of Natural History
PAYNE, KATY, Lincoln, Massachusetts
RISTAU, CAROLYN, Rockefeller University
STUART, ALASTAIR, University of Massachusetts
SWAIN, TONY, Boston University
TERMAN, MICHAEL, Northeastern University
TRANIELLO, JAMES, Boston University
WILCOX, STIMSON, State University of New York, Binghamton
WILLIAMS, JANET, Swarthmore College
WILLIAMS, TIMOTHY, Swarthmore College
Students
FORBES, MARK, Acadia University, Canada
GIORDANO, AMELIA, Station Marine D'Endoume, France
GINSBURG, PAULETTE, State University of New York at Binghamton
GULBRANSEN, THOMAS, University of Rhode Island
HALL, VALERIE, Boston University
JOHNSON, BRUCE, University of North Carolina Medical Center
KOLLMER, MARY, Russel Sage College
LEVIN, DEBORAH, Bowdoin College
MclSAAC, HUGH, University of Pittsburgh
MERRILL, CARL, University of Delaware
68 MARINE BIOLOGICAL LABORATORY
MICHAUD, JAYNE, University of Maine/Boston University
MOONEY, SUSAN, Stonehill College
MURPHY, SHAWN, New College of University of South Florida
SPRINT, MICHELLE, Hollins College
WOOD, SUSAN, St. Jude Childrens Research Hospital
WEILGART, LINDA, Luther College
YUEH-JUNG, CHANG, Mount Holyoke College
COMPARATIVE PATHOLOGY OF MARINE INVERTEBRATES
Instructors-in-chief
BANG, BETSY, Johns Hopkins University/Marine Biological Laboratory
REINISCH, CAROL, Harvard University
Other faculty, staff, and lecturers
CHANG, PEI WEN, University of Rhode Island
DA WE, CLYDE, National Institutes of Health
DUCKLOW, HUGH, Columbia University
HDDS, KENNETH, State University of New York, Buffalo
ELSTON, RALPH, Battelle Marine Research Laboratory
FARLEY, AUSTIN, National Marine Fisheries Service
HOOVER, KAREN, National Institutes of Health
JANNASCH, HOLGER, Woods Hole Oceanographic Institution
LEIBOVITZ, Louis, Cornell University/Marine Biological Laboratory
LEONARD, LESLIE, Johns Hopkins University
LEVIN, JACK, School of Medicine, University of California at Los Angeles
MICHELSON, EDWARD, Harvard School of Public Health
PEARCE, JOHN, National Marine Fisheries Service
PRENDERGAST, ROBERT, Johns Hopkins University
SILVERSTEIN, ARTHUR, Johns Hopkins University
SINDERMAN, CARL, National Marine Fisheries Service
STRANDBERG, JOHN, Johns Hopkins University
WEBB, MARGARET, Goucher Collge
Students
ANDERSON, LINDA, Seattle Pacific University
BROWN, DENOLA, Texas Southern University
BUCHANAN, JoANN, Northeastern University
DOLL, JOHN, Smith Kline and French
GODDARD, KATHRYN, Boston University
HENDRICK, MATTIE, Tufts University Veterinary School
HOLDERBAUM, ROXANNA, Falmouth, Massachusetts
HORGAN, ERICH, Marine Biological Laboratory
JOHNSON, BETTY, Memorial University of Newfoundland, Canada
MANI, GNANA, State University of New York at Buffalo
MCCAIN, ELIZABETH, New College of University of South Florida
MORRISON, REBECCA, State University of New York at Binghamton
SAKAMOTO, HIDEMI, Tufts University
SCOTT, ALAN, Johns Hopkins University
WILLIAMS, ROBERT, Middleburg, Virginia
DEVELOPMENTAL BIOLOGY
Inst ructor- in-ch ief
EDDS, KENNETH, State University of New York, Buffalo
EDUCATIONAL PROGRAMS 69
Other faculty, staff, and lecturers
BEGG, DAVID, Harvard Medical School
BELL, EUGENE, Massachusetts Institute of Technology
BURGESS, DAVID, Dartmouth College
COLEMAN, ANNETTE, Brown University
COLEMAN, JOHN, Brown University
CROWTHER, ROBERT, Marine Biological Laboratory
FUJIWARA, KUIGI, Harvard Medical School
GROSS, PAUL, Marine Biological Laboratory
HARRIGAN, JUNE, Marine Biological Laboratory
HEIPLE, JEANNE, Boston University
INOUE, SHINYA, Marine Biological Laboratory
JAFFE, LAURINDA, University of Connecticut
LUTZ, DOUGLAS, University of Pennsylvania/Marine Biological Laboratory
MARCUS, NANCY, Woods Hole Oceanographic Institution
MASER, MORTON, Marine Biological Laboratory
O'LOUGHLIN, JOHN, Marine Biological Laboratory
POCCIA, DOMINIC, Amherst College
ROBINSON, KENNETH, University of Connecticut
RUDERMAN, JOAN, Harvard Medical School
SCHUEL, HERBERT, State University of New York, Buffalo
TAMM, SIDNEY, Boston University Marine Program/Marine Biological Laboratory
TAYLOR, D. LANSING, Dartmouth College
WHITTAKER, RICHARD, Boston University Marine Program/Marine Biological Laboratory
Students
BERRY, FAITH, University of Maine at Farmington
BRYANT, VIVIAN, Benedict College
DAUM, HENRY, University of Mississippi Medical Center
DOBRINSKY, BARBARA, Sarah Lawrence College
DUGAN, PATRICIA, Kutztown State College
HAMILTON, LAURA, University of New Hampshire
HAZELL, RHONDA, Fairleigh Dickinson University
HENSON, JOHN, Florida State University
JOHNSON, KJMBERLY, Xavier University
KANE, HELEN, Immaculata College
KELLY, HERBERT, Dillard University
MELVIN, WILLIE, Texas Southern University
TEICHMANN, JEFT, Southampton College of Long Island University
TICE, KJMBERLY, Southampton College of Long Island University
VOPICKA, ELLEN, Mercy College
WILL, CINDY, Marquette University
NEUROBIOLOGY
Instructor-in-chief
ALKON, DANIEL, National Institutes of Health/Marine Biological Laboratory
Other faculty, staff, and lecturers
ACOSTA-URQUIDI, JUAN, NINCDS, National Institutes of Health, and Marine Biological
Laboratory
ADELMAN, WILLIAM, JR., National Institutes of Health/Marine Biological Laboratory
ATWOOD, HAROLD, University of Toronto, Canada
70 MARINE BIOLOGICAL LABORATORY
BAILEY, CLAUDIA, University of Arkansas
BARLOW, ROBERT, JR., Syracuse University
BRIGHTMAN, MILTON, National Institutes of Health
BRODWICK, MALCOLM, University of Texas Medical School
CLAY, JOHN, National Institutes of Health/Marine Biological Laboratory
CONNOR, JOHN, Bell Laboratories
CORSON, D. WESLEY, Marine Biological Laboratory
DEFELICE, Louis, Emory University of Medicine
DOWLING, JOHN, Harvard University
FARLEY, JOSEPH, Princeton University
FEIN, ALAN, Boston University Medical School/Marine Biological Laboratory
GART, SERGE, NINCDS, National Institutes of Health, and Marine Biological Laboratory
GILBERT, CHARLES, Harvard Medical School
GOVIND, C. K., University of Toronto, Canada
GRAHAM, WILLIAM, NINCDS, National Institutes of Health
HAROSI, FERENC, Boston University/Marine Biological Laboratory
HILL, RUSSELL, University of Toronto, Canada
JACKLET, JON, State University of New York, Albany
KAPLAN, EHUD, Rockefeller University
KRAVITZ, EDWARD, Harvard Medical School
KUZIRIAN, ALAN, NINCDS, National Institutes of Health/Marine Biological Laboratory
LEDERHENDLER, I., IZJA, NINCDS, National Institutes of Health/Marine Biological
Laboratory
LLINAS, RODOLFO, New York University Medical Center
MOORE, JOHN, University of Massachusetts
NEARY, JOSEPH, NINCDS, National Institutes of Health/Marine Biological Laboratory
ODETTE, Louis, Eye Research Institute
PAPPAS, GEORGE, University of Illinois
POTTER, DAVID, Harvard Medical School
PRICE, CHRISTOPHER, Boston University
RASMUSSEN, HOWARD, Yale University School of Medicine
RAYMOND, STEPHEN, Massachusetts Institute of Technology
REESE, THOMAS, NINCDS, National Institutes of Health
RICHARDS, WILLIAM, Princeton University
SHEPHERD, GORDON, Yale University School of Medicine
SHOUKJMAS, JONATHAN, NINCDS, National Institutes of Health/Marine Biological
Laboratory
SZUTS, ETE, Marine Biological Laboratory
WEISS, THOMAS, Massachusetts Institute of Technology
Students
BERTHIER, NEIL, University of California at Los Angeles
COULTER, DOUGLAS, University of Rhode Island
DESMOND, JOHN, University of Massachusetts
GIBSON, BARBARA, University of Rhode Island
HASELTON, JANE, Tufts University
HAY, BRUCE, Claremont Mckenna College
HELM, JAMES, Medical College of Wisconsin
MILES, CLAUDIA, Wayland, Massachusetts
MILLER, AMELIA, Tufts University
POITRY, SERGE, University of Geneva, Switzerland/Rockefeller University
SCHMIDEK, HENRY, University of Vermont College of Medicine
SWATT, JOHN, Eisenhower College
TAKEDA, MAJCIO, University of Massachusetts
WASHINGTON, SHEILA, Dillard University
EDUCATIONAL PROGRAMS 7 1
SHORT COURSES
ANALYTICAL AND QUANTITATIVE LIGHT MICROSCOPY
Instructor-in-chief
INOUE, SHINYA, Marine Biological Laboratory
Other faculty, staff, and lecturers
ANDERSON, HELENE, Crimson Camera Technical Sales, Inc.
BOGAN, JOHN, DAGE-MTI
BRENNER, MEL, Nikon, Inc.
CHIASSON, RICHARD, Olympus Corporation of America
CLAYPOOL, DAVID, Crimson Camera Technical Sales, Inc.
ELLIS, GORDON, University of Pennsylvania
GRACE, JOHN, Crimson Camera Technical Sales, Inc.
HAYES, THOMAS, University of North Carolina
HEIPLE, JEANNE, Boston University
HiNSCH, JAN, E. Leitz, Inc.
KELLER, ERNST, Carl Zeiss, Inc.
KERR, Louis, Marine Biological Laboratory
LANGENBACH, UWE, Jeonptik Jena G.m.b.H., West Germany
LAWS, BRIAN, Crimson Camera Technical Sales, Inc.
LUTZ, DOUGLAS, Marine Biological Laboratory
MEYER, DAN, Seiler Instrument Company
OLWELL, PATRICIA, E. Leitz, Inc.
PRESLEY, PHILIP, Carl Zeiss, Inc.
RIKUKAWA, KATSUJI, Nikon, Inc.
SALMON, EDWARD, University of North Carolina
SCHEIER, KURT, Nikon, Inc.
SCOTT, ERIC, Venus Scientific
TAYLOR, D. LANSING, Harvard University
TAYLOR, RICHARD, Colorado Video
THOMAS, PAUL, DAGE-MTI
WICK, ROBERT, Carl Zeiss, Inc.
Students
BYERS, STEPHEN, Georgetown University
DYM, MARTIN, Georgetown University
Hui, CHIU SHUEN, Purdue University
HYLTON, NOLA, Stanford University
KACHAR, BECHARA, National Institutes of Health
KILGREN, LESLIE, University of Pennsylvania
MAYLIE, JAMES, Yale University
ONOGI, KENJI, Nikon, Inc.
PHILP, NANCY, National Eye Institute
RIGNEY, DAVID, Institute for Cancer Research
SAMUELSON, JOHN, Boston, Massachusetts
SANGER, JEAN, University of Pennsylvania Medical School
SILVER, FREDERICK, Rutgers Medical School
STROME, SUSAN, University of Colorado
SZOKA, FRANCIS, University of California School of Pharmacy
USHIDA, KAZUO, Nikon, Inc.
72 MARINE BIOLOGICAL LABORATORY
AUTORADIOGRAPHY IN ELECTRON MICROSCOPY
Instructor- in-ch ief
WILLIAMS, MICHAEL, University of Sheffield, England, U. K.
Other faculty, staff, and lecturers
BAFCER, JOHN, Ciba Geigy Pharmaceuticals Division, England, U. K.
GOULD, ROBERT, Institute for Basic Research in Mental Retardation
KERR, Louis, Marine Biological Laboratory
SUVERK.ROPP, CLAUS, Woods Hole, Massachusetts
Students
BERKLEY, KAREN, Florida State University
CARDELL, EMMA Lou, University of Cincinnati Medical Center
CHIEGO, DANIEL, University of Michigan Dental Research Institute
DAUGHERTY, CYNTHIA, Children's Hospital Medical Center
EDWARDS, CHRIS, University of Michigan Dental Research Institute
EISENBERG, BRENDA, Rush Medical Center
ERLICH, STEPHANIE, University of Southern California School of Medicine
GEISER, ALBERT, Philadelphia, Pennsylvania
GOULD, JACALYN, Gillete Company
HUNG, JULIA, University of Cincinnati
HUNT, RICHARD, University of Mississippi Medical Center
KAPLAN, NANCY, Boston City Hospital
LETOURNEAU, RICHARD, Arthur D. Little, Inc.
VAN ARSDALL, LINDA, University of Louisville School of Dentistry
WRATHALL, JEAN, Georgetown University
ELECTRON MICROSCOPY IN THE BIOLOGICAL SCIENCES
Instructors-in-chief
BOWERS, BLAIR, National Institutes of Health
MASER, MORTON, Marine Biological Laboratory
Other faculty, staff, and lecturers
ANTOL, JOE, Carl Zeiss, Inc.
COPELAND, D., EUGENE, Marine Biological Laboratory
GEISER, ALBERT, Hahnemann Medical Center
HOHMAN, THOMAS, National Institutes of Health
HOUGHTON, SUSAN, Marine Biological Laboratory
KERR, Louis, Marine Biological Laboratory
PEACHEY, LEE, University of Pennsylvania
PORTER, KEITH, University of Colorado
TRICHE, TIMOTHY, National Institutes of Health
Students
BIESIOT, PATRICIA, Woods Hole Oceanographic Institution
BLOCK, BARBARA, Duke University
CAREY, FRANCIS, Woods Hole Oceanographic Institution
CRONCE, DONNA JEAN, University of North Carolina
GOVINDARAJAN, SRINIVASA, East Windsor, New Jersey
KLEESE, WILLIAM, University of Arizona
KMETZ, JOHN, Kean College of New Jersey
EDUCATIONAL PROGRAMS 73
ROMOSER, WILLIAM, Ohio University
WILLIAMS, DONALD, Vassar College
ENERGETICS AND TRANSPORT IN AQUATIC PLANTS
Instructor-in-chief
RAVEN, JOHN, The University of Dundee, Scotland
Other faculty, staff, and lecturers
ANDERSON, DONALD, Woods Hole Oceanographic Institution
BISSON, MARY, State University of New York, Buffalo
CHISHOLM, SALLIE, Massachusetts Institute of Technology
GOLDMAN, JOEL, Woods Hole Oceanographic Institution
MOREL, FRANCOIS, Massachusetts Institute of Technology
MORRIS, IAN, University of Maryland
PLATT, TREVOR, Marine Ecology Laboratory, Canada
STOLZENBACH, KEITH, Massachusetts Institute of Technology
SUNDA, WILLIAM, National Marine Fisheries Service
TAFT, JAY, Chesapeake Bay Institute
TAYLOR, CRAIG, Woods Hole Oceanographic Institution
WATERBURY, JOHN, Woods Hole Oceanographic Institution
WHEELER, PATRICIA. Oregon State University
Students
BINDER, BRIAN, Woods Hole Oceanographic Institution
BOUTROS, OSIRIS, University of Pittsburgh at Bradford
GLIBERT, PAT, Woods Hole Oceanographic Institution
HANDLEY, LINDA, University of Hawaii
HARRISON, GAIL, Massachusetts Institute of Technology
HERING, JANET, Massachusetts Institute of Technology
HOFFMAN, ROSANNE, State University of New York, Buffalo
KRAMER, JONATHAN, State University of New York, Marine Sciences Research Center
LAKICH, MELISSA, Woods Hole Oceanographic Institution
LUBE, FATIMA, Rio de Janeiro, Brasil
OLSON, ROBERT, Massachusetts Institute of Technology
RUETER, JOHN, JR., Portland State University
VAULOT, DANIEL, Woods Hole Oceanographic Institution
WEIDEMANN, ALAN, University of Rochester
WILLIAMS, SUSAN, West Indies Laboratory, Virgin Islands
OPTICAL MICROSCOPY AND IMAGING IN THE BIOMEDICAL SCIENCES
Instructor-in-chief
ALLEN, ROBERT, Dartmouth College
Other faculty, staff, and lecturers
ALLEN, NINA, Dartmouth College
ABRAMOWITZ, MORTIMER, Olympus Corporation of America
AMATO, PHILIP, Carnegie Mellon University
BROWN, DOUGLAS, Dartmouth College
BRUDER, LAWRENCE, Olympus Corporation of America
CHIASSON, RICHARD, Olympus Corporation of America
DECKER, MELVIN, JR., Opti-Quip, Inc.
74 MARINE BIOLOGICAL LABORATORY
FALLON, THOMAS, Datacopy Corporation
GILBERT, SUSAN, Dartmouth College
HANSEN, ERIC, Dartmouth College
HARBISON, RICHARD, Carl Zeiss, Inc.
INOUE, SHINYA, Marine Biological Laboratory
KAY, JEFFREY, E. Leitz, Inc.
KELLER, ERNST, Carl Zeiss, Inc.
KLEIFGEN, JEROME, DAGE-MTI
OLWELL, PATRICIA, E. Leitz, Inc.
PRESLEY, PHILIP, Carl Zeiss, Inc.
REDGRAVE, DAVID, Tech Plus
ROGGENBUCK, PAUL, Kenneth Dawson Company
SATO, MASAHIKO, Dartmouth College
SCHEIER, KURT, Nikon, Inc.
TAYLOR, D. LANSING, Carnegie Mellon University
VACCARELLI, VINCENT, Nikon, Inc.
Students
BANK, HARVEY, Medical University of South Carolina
BRIMIJOIN, STEPHEN, Mayo Clinic
BURNS, ANDREW, Hanover, New Hampshire
FLUCK, RICHARD, Franklin and Marshall College
HAIGH, RAYMOND, Northwest Regional Health Authority, England
HAYAKAWA, TSUYOSHI, Hamamatsu Systems, Inc.
JOHNSON, RICHARD, University of Aberdeen, Scotland
KAMIYA, KJYOSHI, Hamamatsu Systems, Inc.
NEAL, RICHARD, Procter and Gamble Company
PALATINI, DENNIS, American Cyanamid Company
SACK, FRED, Boyce Thompson Institute, Cornell University
SIMSON, ELKIN, Technicon Instruments Corporation
SZIKLAS, ROBERT, Wauwinet Shellfish Company
TEYLER, TIMOTHY, Northeastern Ohio University College of Medicine
TRAUB, RICHARD, Havre-de-Grace, Maryland
WISE, BENJAMIN, Keene State College
YEANDLE, STEPHEN, Naval Medical Research Institute
QUANTITATIVE ANALYSIS OF ELECTRON MICROGRAPHS
Instructor-in-chief
PEACHEY, LEE, University of Pennsylvania
Other faculty, staff, and lecturers
BUSCHMANN, ROBERT, Chicago Veterans Administration Medical Center
HASELGROVE, JOHN, University of Pennsylvania
KERR, Louis, Marine Biological Laboratory
MICHAUD, JAYNE, Marine Biological Laboratory
PALMER, LARRY, University of Pennsylvania
Students
ALLEGRA, SALVATORE, St. Joseph's Hospital
BROWNE, JOY, Tuskegee Institute
BUZZELL, GERALD, University of Alberta, Canada
CAPUZZO, JUDITH, Woods Hole Oceanographic Institution
COHEN, SAMUEL, U. S. Army Natick Research and Development Laboratories
DIAMOND, JACK, McMaster University, Canada
EDUCATIONAL PROGRAMS 75
FLORIDA, ROBERT, Magee-Women's Hospital
FORBES, MICHAEL, University of Virginia School of Medicine
FRENCH, JOHN, National Institute of Environmental Health Sciences
GILLOTT, MARCELLE, University of Illinois
LAUBER, ROBIN, University of Vermont
MACPHERSON, TREVOR, Magee-Women's Hospital
NUTTALL, ROBERT, Emory University
PARTHASARATHY, M. V., Cornell University
ROBBOY, STANLEY, Massachusetts General Hospital
SAWADA, GERIANNE, Upjohn Company
SIMMERMAN, LINDA, University of Kentucky Tobacco and Health Research Institute
SUNDELL, CYNTHIA, University of Pennsylvania
SCANNING ELECTRON MICROSCOPY IN THE BIOLOGICAL SCIENCES
Instructor-in-chief
WETZEL, BRUCE, National Institutes of Health
Other faculty, staff, and lecturers
ALBRECHT, RALPH, University of Wisconsin
KENDIG, ESTHER, National Institutes of Health
KERR, Louis, Marine Biological Laboratory
LANE, W. CURTIS, National Institutes of Health
MASER, MORTON, Marine Biological Laboratory
TOUSIMIS, A. J., Tousimis Research Corporation
Students
AIKEN, GEORGE, Emory University
BIELUNAS, JOAN, Medical College of Pennsylvania
CHASE, ROBERT, Lafayette College
ISENBERG, GEORGE, State University of New York College of Arts and Sciences at
Potsdam
MUNOZ, ELIANA, Frederick Cancer Research Facility
VISSCHER, GEORGE, Sandoz, Inc.
THREE AFTERNOONS TOWARD BETTER SCIENTIFIC WRITING
Instructor- in-ch ief
SCHWARTZ, SUSAN, Berkeley, California
Students
BODZNICK, DAVID, Woods Hole, Massachusetts
BOUTROS, OSIRIS, Marine Biological Laboratory
BOUTROS, SUSAN, Marine Biological Laboratory
BROWN, ELLEN, Woods Hole Oceanographic Institution
VON DER HEYDT, KEITH, Woods Hole, Massachusetts
HUND, GRETCHEN, Woods Hole, Massachusetts
LEVY, ELLEN, East Falmouth, Massachusetts
McKEE, TERRY, Woods Hole, Massachusetts
METUZALS, JANIS, Marine Biological Laboratory
REYNOLDS, LESLIE, Falmouth, Massachusetts
RIETSMA, CAROL, Falmouth, Massachusetts
SALGUERO, CAROL, Marine Biological Laboratory
SCHMITT, RAYMOND, Woods Hole Oceanographic Institution
TRUE, MARY, Woods Hole, Massachusetts
76 MARINE BIOLOGICAL LABORATORY
XI. RESEARCH AND TRAINING PROGRAMS
SUMMER
PRINCIPAL INVESTIGATORS
AGUAYO, ALBERTO, J., Montreal General Hospital, McGill University, Canada
ALBERTINI, DAVID F., Harvard University School of Medicine
ALLEN, NINA S., Dartmouth College
ALLEN, ROBERT D., Dartmouth College
ANDERSON, PETER A. V., Whitney Marine Laboratory, University of Florida
AREAS, EDMUND A., Harvard University
ARMSTRONG, CLAY M., University of Pennsylvania
ARMSTRONG, PETER B., University of California at Davis
ARNOLD, JOHN M., Kewalo Marine Laboratory
BARLOW, ROBERT B., JR., Syracuse University
BARTLETT, GRANT R., Laboratory of Comparative Biochemistry
BENNETT, MICHAEL V. L., Albert Einstein College of Medicine
BEZANILLA, FRANCISCO, University of California at Los Angeles
BODZNICK, DAVID, Wesleyan University
BORGESE, THOMAS A., City University of New York, Lehman College
BORON, WALTER F., Yale University School of Medicine
BREUER, ANTHONY, Cleveland Clinic Foundation
BRODWICK, MALCOLM S., University of Texas Medical Branch
BROWN, JOEL E., State University of New York, Stony Brook
BURDICK, CAROLYN J., Brooklyn College
BURGER, MAX M., University of Basel, Switzerland
CARIELLO, Lucio, Stazione Zoologica di Napoli, Italy
CHAPPELL, RICHARD L., Hunter College
CHOW, IDA, University of California at Irvine
COHEN, LAWRENCE B., Yale University School of Medicine
COHEN, WILLIAM D., Hunter College
COOPERSTEIN, SHERWIN J., University of Connecticut Health Center
COSTELLO, WALTER J., Ohio University
DENTLER, WILLIAM L., University of Kansas
DEWEER, PAUL J., Washington University School of Medicine
DOWLING, JOHN E., Harvard University
DUNHAM, PHILIP B., Syracuse University
DYBAS, LINDA K., Knox College
EATON, DOUGLAS, University of Texas Medical Branch
ECKERT, ROGER, University of California at Los Angeles
EHRENSTEIN, GERALD, National Institutes of Health
EHRING, GEORGE, Northwestern University Medical School
FAHIM, M. A., Case Western Reserve University
FARMANFARMAIAN, A., Rutgers University
FINGER, THOMAS F., University of Colorado School of Medicine
FISHMAN, HARVEY, M., University of Texas Medical Branch
FRENCH, ROBERT J., University of Maryland School of Medicine
GILBERT, DANIEL L., National Institutes of Health
GOODE, DENNIS M., University of Maryland
GOVIND, C. K., Scarborough College, Canada
GRINVALD, AMIRAM, Weizman Institute of Science, Israel
HAIMO, LEAH T., University of California at Riverside
HARDING, CLIFFORD V., Kresge Eye Institute
HASCHEMEYER, AUDREY E. V., Hunter College
HEPLER, PETER K., University of Massachusetts at Amherst
HIGHSTEIN, STEPHEN M., Albert Einstein College of Medicine
RESEARCH AND TRAINING PROGRAMS 77
HILL, SUSAN D., Michigan State University
HOSKIN, FRANCIS C. G., Illinois Institute of Technology
INGOGLIA, NICHOLAS, New Jersey Medical School
IRELAND, LEONARD, Bermuda Biological Station, England, U. K.
JACKLET, JON W., State University of New York, Albany
KAMINER, BENJAMIN, Boston University School of Medicine
KAO, C. Y., State University of New York, Downstate Medical Center
KARWOSKI, CHESTER J., University of Georgia
KATZ, MICHAEL, Brown University
KJNG, GREGORY L., University of North Carolina School of Medicine
KIRK, MARK D., Rice University
LANDOWNE, DAVID, University of Miami
LANGFORD, GEORGE M., University of North Carolina School of Medicine
LASEK, RAYMOND J., Case Western Reserve University
LAUFER, HANS, University of Connecticut
LIPICKY, RAYMOND J., National Institutes of Health
LLINAS, RUDOLFO, New York University Medical Center
LOEWENSTEIN, WERNER R., University of Miami School of Medicine
LONGO, FRANK J., University of Iowa
LORAND, LASZLO, Northwestern University
MAGLOTT, DONNA R., Howard University
MERCIER, A. JOFFRE, JR., University of Calgary, Canada
METUZALS, J., University of Ottawa, Canada
MITCHELL, RALPH, Harvard University
MIYAMOTO, DAVID M., Seton Hall University
MOORE, JOHN W., Duke University Medical Center
MULLINS, L. J., University of Maryland School of Medicine
NAGEL, RONALD L., Albert Einstein College of Medicine
NARAHASHI, TOSHIO, Northwestern University Medical School
NELSON, LEONARD, Medical College of Ohio
NOE, BRYAN D., Emory University
OBAID, ANA LIA, University of Pennsylvania
O'MELIA, ANNE F., George Mason University
OXFORD, GERRY S., University of North Carolina
PAPPAS, GEORGE D., University of Illinois Medical Center
PAUL, DAVID L., Harvard Medical School
PIERCE, SIDNEY K., University of Maryland
POLLARD, HARVEY B., National Institutes of Health
POZNANSKY, MARK J., University of Alberta, Canada
PROENZA, Luis M., University of Georgia
QUIGLEY, JAMES P., State University of New York, Downstate Medical Center
RANE, STANLEY, University of Massachusetts
RIPPS, HARRIS, New York University School of Medicine
Ross, WILLIAM N., New York Medical College
RUDERMAN, JOAN, Harvard Medical School
RUSSELL, JOHN M., University of Texas Medical Branch
SALMON, EDWARD D., University of North Carolina
SALZBERG, BRIAN M., University of Pennsylvania
SANGER, JOSEPH W., University of Pennsylvania School of Medicine
SANGUINETTE, MICHAEL, University of California at Davis
SCHNEIDER, E. GAYLE, University of Nebraska Medical Center
SCHUEL, HERBERT, State University of New York
SCHWAB, WALTER E., Virginia Polytechnic Institute and State University
SEGAL, SHELDON, Rockefeller Foundation
SILVER, WAYNE, Monell Chemical Senses Center
SLUDER, GREENFIELD, Worcester Foundation for Experimental Biology
78 MARINE BIOLOGICAL LABORATORY
SPECK, WILLIAM T., Case Western Reserve University
SPEIGEL, EVELYN, Dartmouth College
SPEIGEL, MELVIN, Dartmouth College
STIMERS, JOSEPH, University of California at Los Angeles
STUART, ANN E., University of North Carolina
SZAMIER, R. BRUCE, Massachusetts Eye and Ear Infirmary
SZENT-GYORGYI, Andrew. Brandis University
TANGUY, JOELLE, Ecole Normale Superieure, France
TASAKI, ICHIJI, National Institutes of Health
TAYLOR, ROBERT E., National Institutes of Health
TELZER, BRUCE R., Pomona College
TILNEY, LEWIS G., University of Pennsylvania
TRIESTMAN, STEVEN N., Worcester Foundation for Experimental Biology
TRINKAUS, J. P., Yale University
TROLL, WALTER, New York University Medical Center
TUCKER, EDWARD B., Vassar College
TYTELL, MICHAEL, Bowman Gray School of Medicine
WALLACE, ROBIN A., University of Florida
WANG, GING Kuo, Harvard University
WEIDNER, EARL, Louisiana State University
WEISSMANN, GERALD, New York University Medical Center
WHITE, MICHAEL M., University of California at Los Angeles
WOODY, CHARLES D., University of California Medical Center at Los Angeles
WORTHINGTON, C. R., Carnegie-Mellon University
ZIGMAN, SEYMOUR, University of Rochester School of Medicine
ZIMMERMAN, ROGER P., Rush University
LIBRARY READERS
ADELBERG, EDWARD A., Yale Medical School
ALBRIGHT, JOHN T., Harvard School of Dental Medicine
ALLEN, GARLAND, Washington University
ANDERSON, EVERETT, Harvard Medical School
APOSHIAN, H. VASKEN, University of Arizona
ARMSTRONG, MARGARET, University of California
BANG, BETSY G., Marine Biological Laboratory
BARKLEY, JOHN J., University of Massachusetts
BASHOR, DAVID P., University of North Carolina at Charlotte
BEAN, CHARLES P., General Electric Research and Development Center
BECKER, FREDERICK F., M. D. Anderson Hospital & Tumor Institute
BECKER, JOHN M., University of South Florida
BEIDLER, LLOYD, Florida State University
BELL, EUGENE, Massachusetts Institute of Technology
BOURNE, DONALD W., Woods Hole Oceanographic Institution
BROYLES, ROBERT H., University of Oklahoma Health Sciences Center
BROWN, FRANK, Woods Hole, Massachusetts
BUCK, JOHN, National Institute of Health
CANDELAS, GRASIELA C., Universidad de Puerto Rico
CARLSON, FRANCIS, John Hopkins University
CARRIERE, RITA, State University of New York, Downstate Medical Center
CHILD, FRANK M., Trinity College
CLARK, ARNOLD, University of Delaware
COHEN, SEYMOUR S., State University of New York, Stony Brook
COLLIER, JACK R., Brooklyn College
COLLIER, MARJORIE M., Saint Peter's College
RESEARCH AND TRAINING PROGRAMS 79
COWLING, VINCENT F., State University of New York, Albany
DETTBARN, WoLF-D., Vanderbilt University Medical Center
DUNCAN, THOMAS, Marine Biological Laboratory
EBERT, JAMES, Carnegie Institute of Washington
ECK.ERT, BARRY S., State University of New York, Buffalo
ECK.ERT ROBERT T., University of New Hampshire
EDDS, KENNETH T., State University of New York, Buffalo
EDDS, LOUISE L., Ohio University
EDER, HOWARD A., Albert Einstein College of Medicine
ELLNER, JERROLD, University Hospitals, Cleveland, Ohio
EPEL, DAVID, Stanford University
FARMANFARMAIAN, A., Rutgers — The State University of New Jersey
FISHER, SAUL, New Y'ork University School of Medicine
FLATNES, OLAUG, Norway
FREINKEL, NORBET, Northwestern University Medical School
FRENCH, KATHLEEN, University of North Carolina School of Medicine
FUSSELL, CATHARINE P., Pennsylvania State University
GABRIEL, MORDECAI L., Brooklyn College
GARDNER, ELIOT L., Albert Einstein College of Medicine
GERMAN, JAMES L., The New York Blood Center
GLASSER, JANE E., University of Georgia
GRANT, PHILIP, University of Oregon
GROSCH, DANIEL S., North Carolina State University
GROSSMAN, ALBERT, New York University Medical Center
GOLDSTEIN, MOISE H., John Hopkins University
GUTTENPLAN, JOSEPH B., New York University College of Dentistry
GUTTMAN, RITA, New York University
HELLMAN, HAL, Leonia, New Jersey
HINSCH, GERTRUDE, University of South Florida
HOCK, ALAN K., Littleneck, New York
HUBBARD, RUTH, Harvard University
ILAN, JOSEPH D., Case Western Reserve University School of Medicine
ILAN, JUDITH, Case Western Reserve University School of Medicine
INOUE, SHINYA, Marine Biological Laboratory/University of Pennsylvania
ISSENBERG, IRVIN, Oregon State University
ISSIDORIDES, MARIETTA R., University of Athens, Eginition Hospital, Greece
JACKLET, JON W., State University New York, Albany
JONES, MEGAN, Harvard University
JOSEPHSON, ROBERT K., University of California
KANE, ROBERT E., University of Hawaii
KALTENBACH, JANE C, Mount Holyoke College
KASS-SIMON, GABRIELE, University of Rhode Island
KELLEY, ROBERT E., University of Illinois College of Medicine
KIRSCHENBAUM, DONALD M., State University of New York, College of Medicine
KLIEN, MORTON, Temple University Medical School
KOGUT, MARGOT, King's College, London, England
LADERMAN, AIMLEE D., Smithsonian Institution
LAZAROW, PAUL B., The Rockfeller University
LEE, JOHN J., City College of New York
LEIGHTON, JOSEPH, The Medical College of Pennsylvania
LEVINE, RACHMIEL, City of Hope Medical Center, California
LEVITZ, MORTIMER, New York University Medical Center
LOCKWOOD, ARTHUR H., Temple University Medical School
MAHLER, HENRY R., Indiana University
MARINE RESEARCH, Falmouth, Massachusetts
MASTROIANNI, LUIGI, Hospital of the University of Pennsylvania
80 MARINE BIOLOGICAL LABORATORY
MAUTNER, HENRY G., Tufts University School of Medicine
MAUZERALL, DAVID, Rockefeller University
MAY, SHELDON, Georgia Institute of Technology
MIZELL, MERLE, Tulane University
MONSANTO COMPANY, St. Louis, Missouri
MONROY, ALBERTO, Stazione Zoologica, Napoli, Italy
MORRELL, FRANK, Rush Medical College
MORSE, STEPHEN, Rutgers University
MULLINS, LORIN J., University of Maryland, Baltimore
OLINS, ADA L., University of Tennessee, Oak Ridge
OLINS, DONALD E., University of Tennessee, Oak Ridge
OSCHMAN, JAMES, Woods Hole, Massachusetts
OTTER, TIMOTHY, Albert Einstein College of Medicine
PALMER, JOHN D., University of Massachusetts
PEISACH, JACK, Albert Einstein College of Medicine
PERSON, PHILIP, V. A. Hopsital, Brooklyn, New York
PLOCKE, DONALD J., Boston College
POLLEN, DANIEL A., Barrow Neurological Institute, Phoenix, Arizona
POSTAL, BILL, North Falmouth, Massachusetts
RICE, ROBERT, Carnegie-Mellon University
RiCH-McCOY, Lois, Palisades, New York
ROTH, EUGENE, Mount Sinai School of Medicine, New York
ROWLAND, LEWIS P., Neurological Institute, New York
RUSHFORTH, NORMAN B., Case Western Reserve University
RUSSELL-HUNTER, W. D., Syracuse University
SAGE, MARTIN, University of Missouri
SAGE, LINDA, University of Missouri
SAUNDERS, JOHN, Waquoit, Massachusetts
SCHWARTZ, MARTIN, University of Maryland Baltimore County
SEGAL, HAROLD L., State University of New York at Buffalo
SHEMIN, DAVID, Northwestern University
SHEPHARD, FRANK, Deep Sea Research
SHEPRO, DAVID, Boston University
SHERMAN, IRWIN W., University of California
SONNENBLICK, R. P., Rutgers University
SPECTOR, A., Columbia University
STEPHEN, MICHAEL J., Rutgers University
STETTEN, MARJORIE R., National Institute of Health
TASHIRO, JAY S., Kenyon College
TRACER, WILLIAM, The Rockefeller University
TROXLER, ROBERT F., Boston University School of Medicine
TWEEDELL, KENYON S., University of Notre Dame
VAN HOLDE, KEN E., Oregon State University
WAGNER, ROBERT R., University of Virginia School of Medicine
WAINIO, WALTER, Rutgers University
WARREN, LEONARD, Instar Institute
WEBB, MARGARET, Woods Hole, Massachusetts
WEISS, LEON, Veterinary School of the University of Pennsylvania
WHEELER, GEORGE E., Brooklyn College
WILBER, CHARLES G., Colorado State University
WITTENBERG, BEATRICE A., Albert Einstein College of Medicine
WITTENBERG, JONATHAN B., Albert Einstein College of Medicine
Yow, FRANK W., Kenyon College
ZACKS, SUMNER L, Miriam Hospital
ZELESKI, ILENE, Deep Sea Research
ZIGMAND, RICHARD E., Harvard Medical School
ZIMMERMAN, MORRIS, Merck Sharp & Dohme Research Laboratories
RESEARCH AND TRAINING PROGRAMS 8 1
OTHER RESEARCH PERSONNEL
ADAMS, CHERYL A., University of Wisconsin
ALLIEGRO, MARK C., State University of New York, Buffalo
ANDERSON, CATHLEEN, Syracuse University
ANTONELLIS, BLENDA, University of Rochester
ARANOW, CYNTHIA, New York University School of Medicine
ARBAS, EDMUND, Harvard University
AUGUSTINE, GEORGE, University of California at Los Angeles
BAKER, ROBERT, New York University Medical Center
BENNETT, NICHOLAS, Riverdale, New York
BERES, LINDA S., University of California at Los Angeles
BETCHAKU, TEIICHI, Yale University Medical School
BIBKO, LISA, Syracuse LIniversity
BLUMER, JEFFERY, Case Western Reserve University
BOOKMAN, RICHARD, University of Pennsylvania
BOWER, JAMES M., New York University Medical Center
BRADY, SCOTT, Case Western Reserve University
BREITWIESER, GERDA E., University of Texas Medical Branch
BREUER, ANTHONY C., Cleveland Clinic Foundation
BROOKS, ALLYN, University of North Carolina at Chapel Hill
BROWN, DOUGLAS T., Dartmouth College
BRUNER-LORAND, JOYCE, Northwestern University
BYRD, WILLIAM, Louisiana State University
CAPUTO, CARLO, Centre de Biofisica, Venezuela
CARIELLO, Lucio, Northwestern University
CARVALHO, ANTONIO, Albert Einstein College of Medicine
CLAPIN, DAVID F., University of Ottawa, Canada
COBUZZI, CAROL A., Mount Vernon, New York
COHEN, JEFFREY MARTIN, New York University Medical Center
COHEN, ROCHELLE S., University of Illinois Medical Center
COLLINS, STEHEN, Case Western Reserve University School of Medicine
CORK, DOUGLAS J., Illinois Institute of Technology
COTRAN, PAUL, Harvard Medical School
COWDER, WILLIAM, Kresge Eye Institute
CURTIS, NANCY, Cornell University
CZINN, STEVEN J., Rainbow Babies and Children's Hospital, University Hospital of
Cleveland
DA-YUAN, CHEN, University of Iowa
DEFELICE, Louis, Emory University
DESIMONE, DOUGLAS, Dartmouth College
DICKER, ADAM D., New York University Medical Center
DiPOLO, REINALDO, Institute Venezolano de Investigaciones Cientificas, Venezula
DIXON, ROBERT, Winchester, Massachusetts
EHRLICH, BARBARA, Albert Einstein College of Medicine
EISELE, LESLIE E., Syracuse University
EL-HAJ, ALICIA, University of Aberdeen, Scotland
ELLNER, JERROLD, Case Western Reserve University
FENNELLY, GLEN J., Dartmouth College
FELDMAN, SUSAN C., New Jersey Medical School
FERNANDEZ, JULIO M., University of California at Los Angeles
FONG, PEYING, Yale University School of Medicine
FORSCHER, PAUL, University of North Carolina
FRACE, ALAN M., University of Texas Medical Branch
FRAIZER, DAN, University of North Carolina
FRANK, DOROTHY M., Rainbow Babies and Children's Hospital
FRANZINI-ARMSTRONG, CLARA, University of Pennsylvania
82 MARINE BIOLOGICAL LABORATORY
FREEDMAN, JOSHUA, Yale University
GAINER, HAROLD, National Institute of Mental Health
GALLANT, PAUL E., National Institute of Mental Health
GARG, AJAY, Albert Einstein College of Medicine
GEDULDIG, ULLA, Saint Johns, Newfoundland, Canada
GILBERT, SUSAN P., Dartmouth College
GIUDITTA, ANTONIO, International Institute for Genetics and Biophysics, Italy
GLASS, NICHOLAS, Buxton School
GLUSMAN, STEVEN, Columbia University
GOULD, ROBERT M., New York Institute for Basic Research in Mental Retardation
GUCHARDI, JOHN, University of Toronto, Canada
GUTERMAN, LEE, Clarkson College
HAGELSTEIN, ERIC B., Northwestern University Medical School
HARONIAN, GRACE, University of Connecticut Health Center
HARRIS, A. L., Albert Einstein College of Medicine
HARRIS, EDWARD M., Duke University Medical Center
HAYASHI, JON H., University of North Carolina
HAYS, TOM, University of North Carolina
HILDESHEIM, RINA, Weissman Institute of Sciences
HILL, RUSSELL, Scarborough College, University of Toronto, Canada
HILL, LENA, Scarborough College, University of Toronto, Canada
HINES, MICHAEL, Duke University Medical Center
HOLEKAMP, LINDA C, Washington University School of Medicine
HOULIHAN, DOMINIC, University of Aberdeen, Scotland
IWASA, KINIHIKO, National Institute of Mental Health
JACOBSEN, FREDA, National Institutes of Health
JAMPEL, JAMES, Bloomfield Hills, Michigan
JASLOVE, STEWART W., Albert Einstein College of Medicine
KALIL, KATHERINE, University of Wisconsin
KAO, PETER N., Columbia University College of Physicians and Surgeons
KAPLAN, EHUD, Rockefeller University
KASS, LEONARD, Syracuse University
KELLER, RAYMOND E., University of California at Berkeley
KIMMEL, CHARLES B., University of Oregon
KIRCHMAN, DAVID, Harvard University
KOIDE, S. S., Population Council
KRACKE, GEORGE R., Washington University School of Medicine
KRAUTHAMER, VICTOR, New York Medical College
LARSEN, JAMES, University of Southern Mississippi
LAWRENCE, ADRIAN, University of North Carolina
LAYTON, BARRY S., Worcester Foundation for Experimental Biology
LEUCHTAG, H. RICHARD, University of Texas Medical Branch
Li, QiNG-Yu, Vassar College
LIMAN, EMILY R., Princeton University
LLANO, ISABEL, University of California Medical School at Los Angeles
Lo, Woo-KuEN, Kresge Eye Institute of Wayne State University
LOPEZ-BARNEO, JOSE, University of Pennsylvania
MACHIDA, KOICHI, University of Miami School of Medicine
MACKIN, ROBERT B., Carleton College
MANCINI, VIVIAN, Hunter College
MANCUSO, CAROL A., Morton, Pennsylvania
MATTESON, DONALD R., University of Pennsylvania
MAUGEL, TIMOTHY K., University of Maryland
MISEVIC, GRADIMIR, University of Basel, Switzerland
MORAN, MICHAEL N., Emory University
OHKI, SHINPEI, State University of New York, Buffalo
RESEARCH AND TRAINING PROGRAMS 83
OLAND, LYNNE A., University of North Carolina
ORBACH, HARRY, Yale University School of Medicine
PANT, HARISH C, National Institute on Alcohol Abuse and Alcoholism
PAXHIA, TERESA, University of Rochester School of Medicine and Dentistry
PEARCE, JOANNE, Scarborough College, Canada
PERSELL, ROGER, Mercy College
POCHAPIN, MARK, University of Pennsylvania School of Medicine
POCRNJIC, ZVONIMIR, Hunter College
PURPURA, KEITH, Rockefeller University
QUINN, RICHARD, University of Maryland
RAKOWSKI, ROBERT F., Washington University School of Medicine
RAO, P. D., PRASADA, New York Medical College
REDDY, VINAY N., Kresge Eye Institute
REQUENA, JAIME, University of Maryland School of Medicine
ROSE, BIRGIT, University of Miami School of Medicine
SAIMI, YOSHIRO, University of Wisconsin
SALTZMAN, CHARLES, University of North Carolina School of Medicine
SANGER, JEAN M., University of Pennsylvania School of Medicine
SARMA, J. VIDYA, University of Maryland
SATO, EIMEI, Population Council
SATO, MASAHIKO, Dartmouth College
SCHAFER, THEO, Biozentrum der Universitat, Switzerland
SCHLUP, VERENA, Biozentrum der Universitat, Switzerland
SCRUGGS, VIRGINIA M., Northwestern University Medical School
SCULPTOREUANU, ADRIAN, Duke University
SELMAN, KELLY, University of Florida College of Medicine
SERHAN, CHARLES, New York University Medical Center
SEYAMA, ISSEI, Northwestern University
SHIMIZU, HIDEAKI, University of Pennsylvania
SIMON, SANFORD, New York University Medical Center
SMALL, MARIA, University of North Carolina
So, FREDERICK, University of Iowa
Socci, ROBIN, Rutgers — The State University of New Jersey
SOMMER, HEIDI, Biozentrum der Universitat, Switzerland
SPRAY, DAVID C., Albert Einstein College of Medicine
STEINACHER, ANTONINETTE, Albert Einstein College of Medicine
STOCKBRIDGE, NORMAN, Duke University Medical Center
STOKES, PATRICIA J., Albert Einstein College of Medicine
STRASSMAN, ANDREW, Albert Einstein College of Medicine
SUGIMORI, MATSUYUKI, New York University Medical Center
SUSAN, STANLEY R., Kresge Eye Institute of Wayne State University
SZENT-GYROGYI, EVA S., Brandeis University
TAATGES, DOUGLAS J., Kansas State University
TAKEUCHI, KIYOSHI, Northwestern University
TANSEY, TERESE, Harvard Medical School
TAYLOR, KEVIN, T., University of Miami School of Medicine
TIMPE, LESLIE C., University of North Carolina
VARNER, JUDITH A., Biozentrum der Universitat, Switzerland
VERGARA, JULIO L., University of California at Los Angeles
VOGEL, STEPHEN M., Northwestern University
WALCH, MARIANNE, Harvard University
WANG, LIN-FANG, Population Council
WEISS, JERRY S., Northwestern University Medical School
WHITE, ROY, Albert Einstein College of Medicine
WHITTEMBURY, JOSE, Universidad Peruana Cayetano Heredia, Peru
WOLNIAK, STEPHEN M., University of Massachusetts at Amherst
84 MARINE BIOLOGICAL LABORATORY
WONG, TERENCE Z., Dartmouth College
WORTHINGTON, A. R., Carnegie-Mellon University
WRIGHT, ANSON E., Harvard University
YAJEYA, JAVIER, University of Pennsylvania School of Medicine
YEH, JAY Z., Northwestern University Medical School
YOUNG, RON E., University of the West Indies, W. I.
ZAKEVICIUS, JANE M., New York University School of Medicine
ZAVILOWITZ, JOSEPH, Albert Einstein College of Medicine
ZWEIFACH, ADAM, New York, New York
YEAR-ROUND PROGRAMS
(All of Marine Biological Laboratory unless otherwise indicated)
BOSTON UNIVERSITY MARINE PROGRAM (BUMP)
Director
WHITTAKER, J. RICHARD, Boston University/Marine Biological Laboratory
Staff (of Boston University unless otherwise indicated)
ALLEN, SARAH
ATEMA, JELLE
COGSWELL, CHARLOTTE, University of Connecticut
CROWTHER, ROBERT
GOVIND, C. K., University of Toronto
HAHN, DOROTHY
HANDRICH, LINDA
HARTMAN, JEAN, University of Connecticut
HILL, RUSSELL
HUMES, ARTHUR
LOESCHER, JANE
MEEDEL, THOMAS
MURRAY-BROWN, MARK
PEARCE, JOANNE
PRICE, CHRISTOPHER
RIETSMA, CAROL, State University of New York, New Paltz
TAMM, SIDNEY
TAMM, SIGNHILD
TAYLOR, MARGERY
VALIELA, IVAN
VAN ETTEN, RICHARD
Students (of Boston University unless otherwise indicated)
BARSHAW, DIANA HOWES, BRIAN
BRYANT, BRUCE JOHNSON, BRUCE
BRYANT, DONALD LAVALLI, KARI
BUCHSBAUM, ROBERT MACIOLEK-BLAKE, NANCY
CARACO, NINA MERRILL, CARL
CHU, KEVIN Moss, ANTHONY
CLARKE, JOANN NEIDINGER, RICHARD
COHEN, ROSALIND PASCOE, NATALIE
COSTA, JOSEPH POOLE, ALAN
COULTER, DOUGLAS TAMSE, ARMANDO
DOJIRI, MASAHIRO TROTT, THOMAS
FERME, PAOLA WEBB, JACQUELINE
FOREMAN, KENNETH WHITE, DAVID
FUJITA, RODNEY WILLIAMS, ISABELLE
GODDARD, KATHRYN WILSON, JOHN
HALL, VALERIE
RESEARCH AND TRAINING PROGRAMS
85
DEVELOPMENTAL AND REPRODUCTIVE BIOLOGY LABORATORY
Director
GROSS, PAUL R.
Staff
HALVORSON, LISA, Brandeis University
O'LOUGHLIN, JOHN
SIMPSON, ROBERT T., National Institutes of Health
THE ECOSYSTEMS CENTER
Director
WOODWELL, GEORGE M.
Staff and consultants
BADENHAUSEN, MARGUERITE M.
BEALE, ELEANORE, M.
BEARD, SARAH H.
BORETOS, DIANE
BOWLES, FRANCIS P.
CARLSON, CHRISTOPHER
CLARK, LYNETTE
COLE, JONATHAN
CORLISS, TERESA A. L.
CUCINATTO, JAMES
DUNCAN, THOMAS
DUNCAN, JENNIFER
ELDRED, KATE
GARRITT, ROBERT W.
GREGG, DAVID
GUTJAHR, RUTH E.
HARTMAN, JEAN
HELFRICH, JOHN V. K.
HOBBIE, JOHN E.
HOUGHTON, RICHARD A.
HOWARTH, ROBERT W.
JUERS, DAVID W.
KJJOWSKI, VOYTEK
LARSSEN, CHERYL
LYNCH, CHRISTINE
MACALUSO, MARYANNE
MARINO, ROXANNE
MARINUCCI, ANDREW C.
MARQUIS, SALLY L.
MCNEILL, JOHN
MELILLO, JERRY M.
MERKEL, SUSAN
MILLINGER, MYRA
MONTGOMERY, ELLYN
MONTGOMERY, MARY LOUISE
Moss, ANN H.
PALM, CHERYL A.
PARSONS, KATHERINE C.
PETERSON, BRUCE J.
QUICK, DEBORAH G.
SECHOKA, ELIZABETH M.
SEMINO, SUZANNE
SHAVER, GAIUS R.
SIMMONS, NANCY S.
STONE, THOMAS
STEUDLER, PAUL A.
TURNER, ANDREA R.
UPTON, JOAN M.
Trainees
BOWDEN, WILLIAM B., North Carolina State University, Year-in-Science
CAVANAUGH, COLLEEN, Harvard University, Year-in-Science
GORDON, DORIA, Intern
SAMPOU, PETER, University of Rhode Island
WILLEY, JOANNE, Intern
LABORATORY OF BIOPHYSICS
Director
ADELMAN, WILLIAM, J., JR. NINCDS-NIH
86 MARINE BIOLOGICAL LABORATORY
Staff (ofNINCDS-NIH unless otherwise indicated)
Section on Neural Membranes
ADELMAN, WILLIAM J., Chief
CLAY, JOHN R.
DEFELICE, Louis J., Emory University
DYRO, FRANCES M., Veterans Administration Medical Center
GOLDMAN, DAVID E., State University of New York, Binghamton
HODGE, ALAN J.
LEONARD, DOROTHY A.
MUELLER, RUTHANNE
RICE, ROBERT V., Carnegie-Mellon University
ROSLANSKY, PRISCILLA F., Bunting Institute of Radcliffe College
TYNDALE, CLYDE L.
VOLKMAN, MARY
WALTZ, RICHARD B.
Section on Neural Systems
ALKON, DANIEL L., Chief
ACOSTA-URQUIRDI, JUAN
COULTER, DOUG, Boston University
FARLEY, JOSEPH, Princeton University
GART, SERGE, University of Vermont
GOH, YASUMASA
HARRIGAN, JUNE F.
HAY, BRUCE
HILL, LENA
JACKLET, JON, State University of New York, Albany
KUZIRIAN, ALAN M.
KUZIRIAN, JEANNE
LEDERHENDLER, IZJA
LEIGHTON, STEPHEN, National Institutes of Health
LIMAN, EMILY, Princeton University
LING, LORRAINE, University of Minnesota
NEARY, JOSEPH T.
SHOUK.IMAS, JONATHAN J.
STEELMAN, JAMES, Brown University
TENGLESEN, LESLIE
WOOLF, THOMAS
LABORATORY FOR MARINE ANIMAL HEALTH
Director
LEIBOVITZ, Louis, New York State College of Veterinary Medicine
Staff
ART, DONALD A., University of Pennsylvania
RICKARD, CHARLES C., Cornell University
STONE, AMY, Cornell University
LABORATORY OF SENSORY PHYSIOLOGY
Director
MACNICHOL, EDWARD F., JR.
RESEARCH AND TRAINING PROGRAMS 87
Staff
COLLINS, BARBARA ANN
COOK, PATRICIA B.
CORNWALL, CARTER, Boston University School of Medicine
CORSON, D. WESLEY
FEIN ALAN
HAROSI, FERENC I.
HASHIMOTO, YOKO, Tokyo Women's Medical College
LEVINE, JOSEPH S.
LEVY, SIMON
LIPETZ, LEO, Ohio State University
MANSFIELD, RICHARD, Boston University School of Medicine
PAYNE, RICHARD
SZUTS, ETE ZOLTAN
WALZ, BERND, University of Ulm, West Germany
NATIONAL FOUNDATION FOR CANCER RESEARCH
Director
SZENT-GYORGYI, ALBERT
Staff
GASCOYNE, PETER R. C.
MCLAUGHLIN, JANE A.
MEANY, RICHARD A.
PETHIG, RONALD, University College of North Wales, U. K.
LABORATORY OF CARL J. BERG, JR.
Director
BERG, CARL J., JR.
Staff
ADAMS, NANCY
ALATALO, PHILIP
ORR, KATHERINE
TURNER, RUTH D., Harvard University
LABORATORY OF D. EUGENE COPELAND
Director
COPELAND, D. EUGENE
LABORATORY OF JUDITH P. GRASSLE
Director
GRASSLE, JUDITH P.
Staff
GELFMAN, CECILIA
MILLS, SUSAN
88 MARINE BIOLOGICAL LABORATORY
LABORATORY OF SHINYA INOUE
Director
INOUE, SHINYA, University of Pennsylvania/Marine Biological Laboratory
Staff
BROWN, CAROLYN, University of Pennsylvania
INOUE, THEODORE
LUTZ, DOUGLAS, University of Pennsylvania
WOODWARD, BERTHA M.
Visiting Investigators
HAMAGUCHI, YUKIHISA, Tokyo Institute of Technology
ORR, TIMOTHY, Albert Einstein School of Medicine
TILNEY, LEWIS G., University of Pennsylvania
WOODRUFF, RICHARD I., West Chester State College
LABORATORY OF ERIC KANDEL
Director
KANDEL, ERIC, Columbia University
Staff
CAPO, THOMAS, Columbia University
GREEN, ADAM, Southampton College
NADEAU, LLOYD, Boston University
PAIGE, JOHN A., Columbia University
PERRITT, SUSAN, Columbia University
SCHWARTZ, PETER
LABORATORY OF CAROL L. REINISCH
Director
REINISCH, CAROL L., Tufts University School of Veterinary Medicine
Staff
CHARLES, ANN M., Tufts University School of Veterinary Medicine
MORRIS, ELISABETH, Tufts University School of Veterinary Medicine
LABORATORY OF OSAMU SHIMOMURA
Director
SHIMOMURA, OSAMU, Princeton University
Staff
SHIMOMURA, AKEMI, Princeton University
LABORATORY OF RAYMOND E. STEPHENS
Director
STEPHENS. RAYMOND E., Marine Biological Laboratory/Boston University School of
Medicine
RESEARCH AND TRAINING PROGRAMS 89
Staff'
PORTER, MARY E., Marine Biological Laboratory/University of Pennsylvania
PRATT, MELANIE, Harvard Medical School
STOMMEL, ELIJAH, Marine Biological Laboratory/ Boston University School of Medicine
SUPRENANT, KATHY, University of Virginia
LABORATORY OF NOEL DE TERRA
Director
DE TERRA, NOEL
LABORATORY OF J. RICHARD WHITTAKER
Director
WHITTAKER, J. RICHARD, Boston University/Marine Biological Laboratory
Staff' (of Boston University)
CROWTHER, ROBERT
LOESCHER, JANE L.
MEEDEL, THOMAS H.
XII. HONORS
FRIDAY EVENING LECTURES
MARLER, PETER, Rockefeller University, January 8, "Birdsong: Nature and Nurture
Revised"
LLINAS, RODOLFO, New York University Medical Center, January 15, "Role of Calcium
in Synaptic Transmission"
GITLER, CARLOS, Weizmann Institute of Science, June 25, "Entamoeba histolytica — A
Remarkable Beast"
ATEMA, JELLA, Boston University, Marine Biological Laboratory, July 2, "To Be a
Lobster: The Biology of an Individual"
DAVIS, BERNARD, Harvard Medical School, July 9, "Is Evolution Falsifiable?"
MOSCONA, ARON, University of Chicago, July 16, Zwilling Lecture, "Embryology
Revisited: Cell Interactions in Morphogenesis and Differentiation"
AGUAYO, ALBERTO, Montreal General Hospital, McGill University, July 22, 23, Forbes
Lectures, I. "Has the Mammalian CNS Lost its Capacity for Axonal Regeneration?"
II. "Inherited Myelin Disorders of Mice and Men"
EISINGER, JOSEF, Bell Laboratories, July 30, "Lead Astray"
NEHER, ERWIN, Max Planck Institute, August 6, "Currents Flowing Through Individual
Ionic Channels in Nen>e and Muscle Membrane"
TRELSTAD, ROBERT, Rutgers University Medical School, August 13, Edds Lecture,
"Morphogenetic Alusings While Peering Through a Spiraling Collagenous Lattice"
LLINAS, RODOLFO, New York University Medical Center, August 20, Lang Lecture, "Of
Neurons, Brains and Movement"
WORCEL, ABRAHAM, University of Rochester, August 27, "Chromatin Structure of Genes'
ASSOCIATES' LECTURE
ATKINS, ELISHA, Yale University, August 14, "Bird Migration: Facts and Fancies"
90 MARINE BIOLOGICAL LABORATORY
JOSIAH MACY, JR., FOUNDATION SCHOLARS
BRYANT, VIVIAN, Benedict College
HAZELL, RHONDA, Fairleigh Dickinson University
JOHNSON, KJMBERLY, Xavier University
KELLY, HERBERT, Dillard University
MELVIN, WILLIE, Texas Southern University
WASHINGTON, SHEILA, Dillard University
STEPS TOWARD INDEPENDENCE FELLOWS
ANDERSON, PETER, Whitney Marine Laboratory
COSTELLO, WALTER, College of Osteopathic Medicine, Ohio University
DYBAS, LINDA, Knox College
HILL, SUSAN, Michigan State University
MIYAMOTO, DAVID, Seton Hall University
OBAID, ANA, University of Pennsylvania School of Dental Medicine
SCHNEIDER, E. GAYLE, University of Nebraska Medical Center
SLUDER, GREENFIELD, Worcester Foundation for Experimental Biology
TUCKER, EDWARD, Vassar College
ZIMMERMAN, ROGER, Rush University
GARY N. CALKINS MEMORIAL SCHOLARSHIP
ANMUTH, CRAIG, Oberlin College
DOETKOTT, CURT, North Dakota State University
MARZOLF, ERICH, Colorado College
FRANCES S. CLAFF MEMORIAL SCHOLARSHIP
MARLER, JENNIFER, McGill University, Canada
NICOL, DIANNE, Dalhousie University, Canada
RANKIN, CATHERINE, City University of New York
WILSON, LINDA, University of Texas at Austin
EDWIN GRANT CONKLIN MEMORIAL SCHOLARSHIP
MARZOLF, ERICH, Colorado College
LUCRETIA CROCKER SCHOLARSHIPS
ABAD, MARK, University of Chicago
BROWN, ALEXIS, California State University at Dominguez Hills
LESK, MARK, McGill University, Canada
MARLER, JENNIFER, McGill University, Canada
MARZOLF, ERICH, Colorado College
MORROW, LAURA, University of Texas at Austin
NICOL, DIANNE, Dalhousie University, Canada
SMITH, ROBERT, University of Chicago
ST. JOHNSTON. DANIEL, Harvard University
HONORS 91
FOUNDERS SCHOLARSHIPS
In 1 982, these Scholarships were given in memory of:
W. E. CARREY S. O. MAST
CASWELL GRAVE T. H. MORGAN
L. V. HEILBRUNN A. H. STURTEVANT
J. LOEB E. WITSCHI
O. LOEWI
Recipients:
ABAD, MARK, University of Chicago
BROWN, ALEXIS, California State University at Dominguez Hills
HOUGAN, LINDA, McGill University, Canada
LESK, MARK, McGill University, Canada
MARLER, JENNIFER, McGill University, Canada
MARZOLF, ERICH, Colorado College
NICOL, DIANNE, Dalhousie University, Canada
OLSEN, SCOTT, Lehigh University
REHDER, VINCENT, Free University of Berlin, Germany
REVELAS, EUGENE, State University of New York at Stony Brook
ST. JOHNSTON, DANIEL, Harvard University
WILSON, LINDA, University of Texas at Austin
ALINE D. GROSS SCHOLARSHIP
HOUGAN, LINDA, McGill University, Canada
MORROW, LAURA, University of Texas at Austin
RANKIN, CATHERINE, City University of New York
MERKEL H. JACOBS SCHOLARSHIP
STODDARD, JEFFREY, University of Wisconsin at Madison
ARTHUR KLORFEIN FUND
DAVID, SAMUEL, Montreal General Hospital, Canada
DOETKOTT, CURT, North Dakota State University
HISHINUMA, AKIRA, Columbia University
NAWROCKI, LEON, University of Oregon
PIRES, ANTHONY, Harvard College
SENIE, ALLYSON, Ithaca College
SMITH, ROBERT, University of Chicago
TREGGOR, JOSEF, Central Connecticut State College
WEISSBURG, MARC, University of California at Berkeley
JAMES S. MOUNTAIN MEMORIAL FUND, INC. SCHOLARSHIP
CONZELMAN, KAREN, Yale University
FRANCIS, RALPH, Oregon State University
MORGANELLI, CHRISTINE, Dartmouth College
RAUSCH, DIANNE, Northwestern University
92
MARINE BIOLOGICAL LABORATORY
SOCIETY OF GENERAL PHYSIOLOGISTS
HOCH, DAVID, Albert Einstein College of Medicine
ST. JOHNSTON, DANIEL, Harvard University
WILLIAMS, BENJAMIN, Yale University
XIII. INSTITUTIONS REPRESENTED
U.S.A.
Alabama, University of
Albany, Veterans Administration Hospital
Albert Einstein College of Medicine
American Cyanamid Company
American Museum of Natural History
Amherst College
Anderson Hospital and Tumor Institute,
M. D.
Argonne National Laboratory
Arizona, University of
Arizona State University
Arkansas, University of
Associated Universities, Inc.
Avon Products, Inc.
Baltimore City Hospital
Barnard College
Barrow Neurological Institute
Bates College
Battelle Marine Research Laboratory
Baylor College, School of Medicine
Bell Laboratories
Benedict College
Bio Labs
Biological Science Center
Boston College
Boston City Hospital
Boston Company, The
Boston University
Boston University, School of Medicine
Bowdoin College
Bowman Gray School of Medicine
Brandeis University
Brigham and Women's Hospital
Brock University
Brookhaven National Laboratory
Brooklyn, Veterans Administration
Hospital
Brown University
Buxton School
California Institute of Technology
California State University, Dominguez
Hills
California State University, Fullerton
California. University of, Berkeley
California, University of, Davis
California, University of, Irvine
California, University of, La Jolla
California, University of, Los Angeles
California, University of, Los Angeles,
Medical School
California, University of, Riverside
California, University of, San Francisco
California, University of, Santa Barbara
California, University of, Santa Cruz
California, University of. School
of Pharmacy
Capitol University
Cancer Research Institute
Carleton College
Carnegie Institution of Washington
Carnegie Mellon University
Case Western Reserve University
Case Western Reserve University,
School of Medicine
Catholic University of America
Central Connecticut State College
Chesapeake Bay Institute
Chicago, University of
Chicago, Veterans Administration
Medical Center
Childrens Hospital
Cincinnati, University of
Cincinnati, University of. Medical Center
City of Hope Medical Center
Claremont McKenna College
Clark University
Clarkson College
Cleveland Clinic Foundation
Colby College
Cold Spring Harbor Laboratory
Colorado College
Colorado State University
Colorado Video
Columbia University
Columbia University, College of
Physicians and Surgeons
Columbia University, Eye Institute
Connecticut, University of
Connecticut, University of, Health Center
Connecticut, University of, School
of Medicine
Cornell University
INSTITUTIONS REPRESENTED
93
Cornell University, Boyce Thompson
Institute
Cornell University Medical College
Crimson Camera Technical Sales, Inc.
DAGE-MTI
Dartmouth College
Dartmouth Medical School
Datacopy Corporation
Kenneth Dawson Company
Dayton, University of
Deep Sea Research
Delaware, University of
Dillard University
Drew University
Duke University
Duke University Medical Center
Dupont Corporation
Earlham College
Eastern Maine Medical Center
Eisenhower College
Emory University
Fairleigh Dickinson University
Fanueil Hall Associates
Federated Department Stores
Federation of American Societies for
Experimental Biology
Florida State University
Florida, University of
Florida, University of. College of Medicine
Florida, University of, Whitney
Marine Laboratories
Franklin and Marshall College
Frederick Cancer Research Facility
Gaston Snow Beekman and Bogue
General Electric Research and
Development Center
George Mason University
George Washington University
Medical Center
Georgetown University, Medical
and Dental Schools
Georgia, University of
Georgia Institute of Technology
Gerontology Research Center
Gillette Company
Gonzaga University
Goucher College
W. R. Grace Company
Grass Foundation, The
Hahnemann Medical Center
Hamamatsu Systems, Inc.
Harvard Medical School
Harvard School of Dental Medicine
Harvard School of Public Health
Harvard University
Harvard University, The Biological
Laboratories
Haskins Laboratories
Hawaii, University of
High Voltage Engineering Corporation
Hollins College
Hopkins Marine Station, Stanford
University
Houghton Mifflin Company
Howard University
Illinois Institute of Technology
Illinois, University of
Illinois, University of. College of Medicine
Immaculata College
Indiana University
Indiana University, School of
Experimental Medicine
Instar Institute
Institute for Basic Research in
Developmental Disabilities
Institute for Fundamental Studies, New
York
Institute for Medical Research, California
Iowa State University
Iowa, University of
Ithaca College
Johns Hopkins University
Johns Hopkins University, School of
Hygiene and Public Health
Johns Hopkins University, School
of Medicine
Kansas State University
Kansas, University of
Kean College of New Jersey
Keene State College
Kentucky, University of
Kentucky, University of. Medical Center
Kentucky, University of. Tobacco and
Health Research Institute
Kenyon College
Kewalo Marine Laboratory
Knox College
Kresge Eye Institute
Kutztown State College
Laboratory of Comparative Biochemistry
Lafayette College
Lazard Freres and Company
Lehigh University
94
MARINE BIOLOGICAL LABORATORY
Leitz, Inc.
Arthur D. Little, Inc.
Louisiana State University
Louisville, University of
Louisville, University of, School of
Dentistry
Lowell, University of
Luther College
Macalester College
Magee Women's Hospital
Maine, University of
Maine, University of, Farmington
Manhattanville College
Marine Research, Inc.
Marquette University
Maryland, University of
Maryland, University of. School of
Medicine
Massachusetts Eye and Ear Infirmary
Massachusetts General Hospital
Massachusetts Institute of Technology
Massachusetts, University of, Amherst
Massachusetts, University of, Boston
Massachusetts, University of. Medical
School
Mayo Clinic
Medical College of Georgia
Medical College of Ohio
Medical College of Wisconsin
Medical University of South Carolina
Richard King Mellon Foundation
Mellon Institute
Mercenene Cancer Research Hospital of
Saint Raphael
Merck, Sharp and Dome Research
Laboratories
Mercy College
Meredith and Grew, Inc.
Miami, University of
Miami, University of. School of Medicine
Miami University
Michigan State University
Michigan, University of. Dental Research
Institute
Michigan, University of
Minnesota, University of
Miriam Hospital
Mississippi, University of. Medical Center
Missouri, University of
Missouri, University of. School of Dentistry
Mohawk Carpets
Monell Chemical Senses Center
Monsanto Company
Morehouse College, School of Medicine
Mount Holyoke College
Mount Sinai School of Medicine
NINCDS, Neurological Disorders Program
National Academy of Engineering
National Eye Institute
National Institute on Alcohol Abuse and
Alcoholism
National Institute of Environmental Health
Sciences
National Institutes of Health
National Jewish Hospital and Research
Center
National Marine Fisheries Service
National Multiple Sclerosis Society
National Science Foundation
Naval Medical Research Institute
Nebraska, University of
Nebraska, University of Medical Center
Neurological Institute
New College of California
New Hampshire, University of
New Jersey Medical School
New Mexico, University of. School of
Medicine
New York Blood Center, The
New York, City University of, Brooklyn
College
New York, City University of. City College
New York, City University of. Hunter
College
New York, City University of, Herbert
Lehman College
New York Institute for Basic Research in
Mental Retardation
New York Medical College
New York State College of Veterinary
Medicine
New York, State University of, Albany
New York, State University of,
Binghamton
New York, State University of, Buffalo
New York, State University of, Potsdam
New York, State University of, Downstate
Medical Center, Brooklyn
New York, State University of. Marine
Sciences Research Center
New York, State University of, Stonybrook
New York, State University of, Syracuse
New York, State University of, Upstate
Medical Center
New York University, College of Dentistry
New York University, School of Medicine
Nikon, Inc.
North Carolina State University
North Carolina, University of, School of
Medicine
North Dakota State University
Northeastern University
Northeastern Illinois University
INSTITUTIONS REPRESENTED
95
Northeastern Ohio University, College of
Medicine
Northern Colorado, University of
Northwestern University
Northwestern University Medical School
Notre Dame, University of
Oak Ridge National Laboratory
Oakland University
Oberlin College
Ohio State University
Ohio University, College of Medicine
Ohio University, College of Osteopathic
Medicine
Oklahoma, University of
Olympus Corporation of America
Opti-Quip, Inc.
Oregon State University
Oregon, University of
Oregon, University of. Institute of
Neurosciences
Pacific University
Payne Laboratories
Pennsylvania, Medical College of
Pennsylvania State University
Pennsylvania, University of
Pennsylvania, University of, Hospital
Pennsylvania, University of. School of
Dental Medicine
Pennsylvania, University of. School of
Medicine
Pennsylvania, University of, School of
Veterinary Medicine
Pittsburgh, University of
Pittsburgh, University of, Bradford
Pittsburgh, University of, School of
Medicine
Pomona College
Population Council
Portland State University
Princeton University
Procter and Gamble Company
Public Health Research Institute
Puerto Rico, University of
Purdue University
Radcliffe College
Rainbow Babies and Children's Hospital
Reed College
Rhode Island, University of
Rice University
Rochester, University of
Rochester, University of, School of
Medicine and Dentistry
Rockefeller Foundation, The
Rockefeller University
Roosevelt University
Rush Medical College
Rush University
Russell Sage College
Rutgers — The State University of New
Jersey
Rutgers University, College of Medicine
and Dentistry
Rutgers University Medical Center
Rutgers University, Waksman Institute for
Microbiology
Saint Joseph's Hospital
Saint Jude Childrens Research Hospital
Saint Peter's College
San Francisco, Veterans Administration
Hospital
Sandoz, Inc.
Sarah Lawrence College
Savannah River Ecology Program
Science Software Systems, Inc.
Scientific American
Scripps Institute of Oceanography
Seapuit, Inc.
Seattle Pacific University
Seiler Instrument Company
Seton Hall University
Sidney Farber Cancer Center
Smith, Kline and French
Smithsonian Institution
South Florida, University of
Southampton College
Southeastern Massachusetts University
Southern California, University of, School
of Medicine
Southern Mississippi, University of
Stanford University
Stonehill College
Swarthmore College
Syracuse University
Tech Plus
Technicon Instruments Corporation
Temple University
Temple University Medical School
Tennessee, University of
Texas, University of, Austin
Texas, University of, Galveston
Texas, University of, Medical School
Texas A and M University
Texas Christian University
Texas Southern University
Toledo, University of
Tousimis Research Corporation
Trinity College
96
MARINE BIOLOGICAL LABORATORY
Tufts University
Tufts University School of Medicine
Tufts University School of Veterinary
Medicine
Tulane University
Tuskegee Institute
Union College
Union University
United States Army, Natick Research and
Development Laboratories
University Hospital of Cleveland
Upjohn Company
Vanderbilt University School of Medicine
Vassar College
Venus Scientific
Vermont, University of
Vermont, University of, College of
Medicine
Virginia Polytechnical Institute
Virginia, University of
Virginia, University of. School of Medicine
Wake Forest University
Walla Walla College
Washington State University
Washington University
Washington University School of Medicine
Wauwinet Shellfish Company
Wayne State University
Weissman Institute of Sciences
Wellington Management Company
Wesleyan University
West Chester State College
Western Washington State College
William and Mary, College of
Wisconsin, University of
Wisconsin, University of, Madison
Woodrow Wilson International Center for
Scholars
Woods Hole Oceanographic Institution
Worcester Foundation for Experimental
Biology
Xavier University
Yale University
Yale University Medical School
Carl Zeiss, Inc.
FOREIGN INSTITUTIONS
Aberdeen, University of, Scotland
Acadia, University of, Canada
Alberta, University of, Canada
Athens, University of, Greece
Basel, University of, Switzerland
Bergen, University of, Norway
Bermuda Biological Station, England
Biozentrum der Universitat, Switzerland
Calgary, University of, Canada
Cambridge, University of, England
Centre de Biofiseca, Venezuela
Centre de Investigacion y de Estudios
Avanzados, Mexico
Ciba Geigy Pharmaceuticals Division,
England
Dalhousie University, Canada
Dundee, The University of, Scotland
Ecole Normale Superieure, France
Federal University of Brasilia, Brazil
Flinders University, South Australia
Foreign and Commonwealth Office,
England
Free University of Berlin, West Germany
Freiburg, University of. West Germany
Fundacao Oswaldo Cruz, Brazil
Geneva, University of, Switzerland
Glasgow, University of, Scotland
Hebrew University, Israel
Ibadan, University of, Nigeria
Institute de Biofisica, Brazil
Institute de Investigacion Medica,
Argentina
Institut Pasteur, France
Institute Venezolano de Investigaciones
Cientificas, Venezuela
International Institute for Genetics and
Biophysics, Italy
Israel Institute of Technology, Israel
Janssen Pharmaceutical Laboratory of
Oncology, Belgium
Jeonptik Jena G. m. b. H., West Germany
Jerusalem, University of, Israel
King's College, England
Kobe University, Japan
Lancaster, University of, England
Latrobe University, Australia
Leeds, University of, England
Liverpool School of Tropical Medicine,
England
Marine Ecology Laboratory, Canada
Max-Planck Institute, West Germany
McGill University, Canada
INSTITUTIONS REPRESENTED
97
McGill University Cancer Center, Canada
McMaster University, Canada
Medical Research Council, England
Milan, University of, Italy
Montreal General Hospital, Canada
Nagoya University, Sugashima Marine
Biological Laboratory, Japan
National Institute for Basic Biology, Japan
Newfoundland, Memorial University of,
Canada
North Wales, University College of, Wales
Northwestern Regional Health Authority,
England
Odense University, Sweden
Open University Research Unit, The,
England
Ottawa, University of, Canada
Ottawa, University of. Faculty of Medicine,
Canada
Oxford, University of, England
Queen Mary College, England
Research Institute for Diseases in a
Tropical Environment, South Africa
Scarborough College, Canada
Sevilla, University of, Spain
Sheffield, University of, England
Station Marine D'Endoume, France
Stazione Zoologica, Naples, Italy
Stockholm, University of, Wenner-Gren
Institute, Sweden
Tel Aviv University, Israel
Tokyo Institute of Technology, Japan
Tokyo Metropolitan Union, Japan
Tokyo Women's Medical College, Japan
Toronto, University of, Canada
Ulm, University of. West Germany
Universidad Peruana Cayetano Heredia,
Peru
Universitats Augenklinik, West Germany
Universitats de Geneve, Switzerland
Universite Laval, Canada
University Hospital and Medical School,
England
Utrecht, University of, Netherlands
Weiterbildungsschule/Diplommittelschule
des Kantons Zug, Switzerland
Weizmann Institute of Science, Israel
Wellcome Trust Research Laboratories,
Africa
West Indies, University of. West Indies
West Indies Laboratory, Virgin Islands
XIV. LABORATORY SUPPORT STAFF
Including Persons Who Joined or Left the Staff During 1982
Controller's Office
SPEER, JOHN W., Controller
BINDA, ELLEN F.
CAMPBELL, RUTH B.
DAVIS, DORIS C.
ELLIS, NANCY L.
Director's Office
GROSS, PAUL R., President and Director
THIMAS, LISA MARIE
FERDINAND, SUSAN M.
HOBBS, ROGER W., JR
HOUGH, ROSE A.
WlCHTERMAN, MARGARET
Associate Director's Office
PALMER, DOUGLAS, W., Associate Director
ZIEMER, CAROL ANN
General Manager 's Office
SMITH, HOMER P., General Manager
BAKER, KAREN H.
BUTZ, FLORENCE S.
GEGGATT, AGNES L.
JOHNSON, FRANCES N.
ZIEMER, CAROL ANN
Development Office
SALGUERO, CAROL GANNON, Development Officer
SCARBOROUGH, BONNIE M.
WEISS, M. NAN
98
MARINE BIOLOGICAL LABORATORY
Grants and Educational Services
HOWARD, JOAN E., Coordinator of Grants and Educational Services
MASER, MORTON D., Assistant Director for Educational and Research Services
ALLEN, GENEVIEVE FOLEY, JOANNE A.
FERZOCO, SUSAN J. LEIGHTON, JANE L.
Public Relations Office
HASKELL, BARBARA, Public Relations Officer
CAMPBELL, LEE ANNE
Biological Bulletin
METZ, CHARLES B., Editor
CLAPP, PAMELA L.
CORBETT, MARGUERITE
LANG, HELEN E.
MOUNTFORD, REBECCA J.
Buildings and Grounds
GUNNING, A. ROBERT, Superintendent
ANDERSON, LEWIS B.
AVERETT, DONALD L.
BALDIC, DAVID
BERRIOS, HECTOR
BERRIOS, JOSE R.
BOURGOIN, LEE E.
BRODERICK, MADELINE
CAFARELLI, PETER A.
CARINI, ROBERT J.
COSTA, ROBERT A.
DAVIS, MARK A.
DEVEER, ROBERT L.
DUTRA, STEVEN J.
ENOS, GLENN R.
EVANS, FRANCES G.
FISH, STEPHEN
FUGLISTER, CHARLES K.
GEGGATT, RICHARD E., JR.
GONSALVES, WALTER W., JR.
IRISH, BRADFORD, D.
Gray Museum
TIFFNEY, WESLEY, N., Curator
BORETOS, C. DIANE
BUSH, LOUISE
Library
FESSENDEN, JANE, Librarian
ASHMORE, JUDITH A.
COOMBS, ROXANE
DEVEER, JOSEPH M.
GIBBONS, ROBERTO G.
GRICE, JOAN H.
HANLEY, JANICE S.
KLEINDINST, THOMAS N.
KUIL, ELISABETH
LEHY, DONALD B.
LEWIS, RALPH H.
LOCHHEAD, WILLIAM M.
LOVERING, RICHARD A.
LUNN, ALAN G.
MACLEOD, JOHN B.
MILLS, STEPHEN A.
PELLS, STANLEY
REZENDES, PATRICK M.
ST. JEAN, SIMONE
SMART, MERILYN A.
THRASHER, FREDERICK
TORRES, DAVID E.
VARAO, JOHN
WARD, FREDERICK
WEEKS, GORDON W.
WHETHAM, CHARLES
WHITTAKER, WILLIAM
MONTIERO, EVA S.
MOUL, EDWIN T.
HOUGH, NANCY L.
JOSEPH, E. LENORA
MARGOLIN, JILL
MOUNTFORD, REBECCA J.
NORTON, CATHERINE N.
SWAIN, LAUREL E.
LABORATORY SUPPORT STAFF 99
Marine Resources
VALOIS, JOHN J., Manager MURPHY, CHARLES F.
CHILD, MALCOLM SMITH, A. DICKSON
EARLY, JULIE TASSINARI, EUGENE
ENOS, EDWARD G., JR. TRAPASSO, BRUNO
ENOS, JOYCE VARAO, JOHN
LAWDAY, LEWIS M.
Research Sen'ices
MASER, Morton D., Assistant Director for Educational and Research Services
BARNES, FRANKLIN D. KERR, Louis M.
BARNES, JOHN S. MARTIN, LOWELL V.
EVANS, WILLIAM NICHOLS, FRANCIS H., JR.
COLDER, LINDA M. SILVA, MARK S.
COLDER. ROBERT J. SYLVIA, FRANK E.
Summer Support Staff
ALBERS, CHRISTINA E. LUNN, JEFFREY R.
ANDERSON, JANICE MACKEY, WILLIAM T.
ASCI, MARGUERITE M. MAXWELL, BRETT
ASHMORE, JILL M. MELLON, ARMOUR N.
ASHMORE, MICHAEL W. MICHAUD, JAYNE
BAKER, LISA B. NADEAU, LLOYD J.
BLACK, ROBERT W. PIERCE, RICHARD T.
BLAIR, RICHARD RENEK, NAOMI
BLAKE, ANN ROONEY, COLLEEN M.
BRINKMAN, PAULA ROONEY, MARK
COOMBS, GILLIAN ROONEY, MICHELE N.
COTRAN, NINA M. RUSHFORTH, LORNA A.
COURTRIGHT, MARYA J. SCUTT, DlANE H.
CRONEY, MICHAELA SEGAL, JENNIFER A.
CUSHMAN, BROOKE SENFT, VALERIE
GREENE, AMY L. SWOPE, STEPHEN P.
HAHN, ERIKA TARBELL, LESLIE
HAMMAR, KATHERINE M. VALOIS, FRANCIS X.
HANSON, ANTHONY VINITSKY, ALEXANDER
JOHNSON, JEFFREY WARNER, ANNE P.
KELLY, MICHAEL WETZEL, ERNEST D.
LAUTHER, GARY B. WHITTAKER, WILLIAM A.
LEE, JAMES M. WYTTENBACH, ROBERT
Reference: Biol. Bull. 165: 100-109. (August, 1983)
IRON ACCUMULATION IN TUNICATE BLOOD CELLS. I.
DISTRIBUTION AND OXIDATION STATE OF IRON IN THE BLOOD
OF BOLTENIA OVIFERA, STYELA CLAVA,
AND MOLGULA MANHATTENSIS
MARIA I. AGUDELO1, KENNETH KUSTIN1*, GUY C. MCLEOD2,
WILLIAM E. ROBINSON2, AND ROBERT T. WANG3
1 Department oj Chemistry, Brandeis University, Walt ham, MA 02254; 2 Harold E. Edgerton Research
Laboratory, New England Aquarium, Boston, MA 02110; 3 'Department of Chemistry,
Salem State College, Salem, MA 01970
ABSTRACT
The iron concentration, oxidation state, and distribution in blood plasma and
blood cells of three iron containing tunicates were determined. Preliminary studies
are reported on the possible role of plasma proteins in iron uptake.
Iron(II) concentration in the millimolar range was found in the blood cell cy-
toplasm of all three species; no iron(III) in solution was detected in blood cells. Over
70% of the total iron in the cells is associated with the membranes.
Although the iron concentration in S. clava blood cells is substantially greater
than that in B. ovifera cells, the iron to protein ratio by weight is similar in both
species. SDS-electrophoresis of B. ovifera blood showed two protein subunits com-
mon to both plasma and blood cells. These two subunits are most likely the major
components of the high molecular weight protein found in the plasma. This protein
was shown to bind iron(III) when iron(III) citrate was added to the plasma.
INTRODUCTION
Mechanisms of metal ion transport and accumulation in living cells are now
being investigated by new techniques (Marx and Aisen, 1981; Anderson and Morel,
1982), and new tools such as extended x-ray absorption fine structure, EXAFS
(Tullius et al, 1980). Of the essential metallic elements, iron presents one of the
most difficult systems to study in terms of elementary steps at the organism/envi-
ronment and cell/plasma barriers. Studies with bacteria (Emery, 1982) provide de-
tailed information on elementary steps in uptake, although information on com-
parable processes in animals still remain obscure. Studies with tunicates have the
potential to clarify several steps in the accumulation process.
Tunicates (class Ascidiacea) accumulate relatively high concentrations of selected
metal ions in certain blood cells. Best known is the ability of members of the order
Enterogona to accumulate vanadium (Millar, 1966; Swinehart et al, 1974). We have
identified several elementary steps in the selective vanadium uptake mechanism and
a model for this process has been constructed (Dingley et al., 198 1 ). We have recently
extended our investigations to include iron accumulating Pleurogona (Agudelo et
al., 1982; Agudelo et al., 1983). In this paper we begin our analysis of the iron
accumulation mechanism by detailing the distribution, concentration, and oxidation
Received 21 March 1983; accepted 25 May 1983.
Abbreviations. TEMED, N,N,N',N'-Tetramethylethylenediamine; SDS, sodium dodecyl sulfate; Bis,
N,N'-Methylene-bis-acrylamide.
* Author to whom all correspondence should be sent.
100
IRON IN TUNICATE BLOOD 101
state of the element in the blood of three iron-accumulating ascidians.
The iron concentration in these species is about one to two orders of magnitude
less than that of the vanadium concentration in vanadium-containing tunicates.
However, the iron concentration gradient is still very large, when compared with
the iron in the aqueous phase of sea water. Like the Enterogona, the Pleurogona
contain similar blood cell types, as well as tunichrome (Macara et al., 1979).
MATERIALS AND METHODS
Materials
NaCl, BaCl2, K3Fe(CN)6, K4Fe(CN)6, KSCN, 1,10-phenanthroline, hydrochlo-
ric acid, acetic acid, nitric acid, glycerine, and bromophenol blue dye were purchased
from Fisher Scientific Co.
Sephadex G-75 and blue dextran were purchased from Pharmacia Fine Chem-
icals.
Acrylamide, Tris buffer, TEMED, SDS-MW70 molecular weight markers kit,
albumin total protein standards and OsO4 were obtained from Sigma Chemical.
Bis, Coomassie brilliant blue, glycine, ammonium persulfate, and 2-mercapto-
ethanol were obtained from Bio-Rad Laboratories.
ACS aqueous counting scintillant was obtained from Amersham Corporation.
55FeCl3 was obtained from New England Nuclear.
2,2'bipyridine was obtained from Mallinckrodt and ascorbic acid from
Schwarz/Mann, Inc.
All chemicals were used without further purification.
Specimens
Boltenia ovifera was collected by divers off East Point, Nahant, MA at 20 m
depth. Styela clava was obtained from the Boston Harbor; Molgula manhattensis
was purchased from Marine Biological Laboratory, Woods Hole, MA. Animals were
all maintained in running sea water at 5-10°C. Blood of B. ovifera and S. clava was
extracted as described previously (Agudelo et al., 1982). Blood of M. manhattensis
was obtained by cutting the tunic at the base of the animal and allowing the blood
to drip into a test tube. Blood cell types were classified according to the criteria
summarized by Wright (1981). Blood cells were fixed and stained with osmium
tetroxide vapors (Kalk, 1963).
Blood cells were separated from the plasma by centrifuging at 1200 g for five
minutes. Plasma was frozen for later analysis; cells were used immediately.
Iron oxidation state and concentration
For the oxidation state analysis of blood, cell samples were treated with 6N HC1,
heated in a boiling water bath for five minutes and centrifuged at 18,400 g for
twenty-five minutes. For plasma, the centrifugation step was omitted.
Total iron concentration in blood cells was determined by using a Perkin Elmer
Model 305 atomic absorption spectrometer. The reduced iron concentration inside
the cells was obtained by lysing the cells in 6N HC1 and adding excess 1,10-phen-
anthroline. The Fe(Phen)32+ absorbance was measured at 510 nm using a Perkin
Elmer 5 52 A uv/vis spectrophotometer. The molar absorptivity coefficient of the
tris( 1,10-phenanthroline) iron(II) complex at 510 nm, pH 1 (HC1) and room tem-
perature was determined to be 7.6 X 103 cm"1 M"1.
The calculation of total cell volume in blood cell samples was based on the
assumption of spherical cells with mean cell diameter 16 yum, and cell counts using
102 M. I. AGUDELO ET AL.
a Levy-Hausser Hemocytometer. Total protein content was determined by the
Lowry method (Lowry et ai, 1951).
Electron paramagnetic resonance
Electron paramagnetic resonance (EPR) experiments were carried out at room
temperature on a previously described spectrometer (Dingley et al, 1981). The
fluoride method (Levanon et al., 1968) was used to gain maximum sensitivity in
the detection of iron(III). The spectrometer settings for this experiment were: 9.55
GHz, 9mW power, time constant 1 s., modular amplitude 5.0, gain 12.5. Under
these conditions, the minimum amount of iron(III) we could detect was approxi-
mately 10 micromoles. Concentrated solutions (approximately 1.3 M) of ammo-
nium fluoride (NH4F) were added to freshly drawn samples of blood producing a
dilution factor of about one-third. The final pH was 6.5. This treatment ensured
that the blood cells in the EPR tube were as intact as possible, and that the large
excess of fluoride would convert even tightly chelated intracellular iron in solution
to the FeF63~ form.
Chromatography and electrophoresis
The water soluble proteins in the plasma and blood cell cytoplasm were run
through a size exclusion chromatography column 30 cm long and 1 .5 cm in diameter
packed with Sephadex G-75. The eluent was 0.5 M sodium chloride and 0.02 M
HC1. Absorbance at 280 nm was monitored continuously using an ISCO UA5
absorbance monitor. Fractions were collected automatically and analyzed for iron
either by atomic absorption spectrometry or by adding an excess of 2,2'-bipyridine
and ascorbic acid and measuring the absorbance of the iron(II)-bipyridine complex
at 520 nm (Macara et ai, 1979). Column void volume (V0) and bed volume (Vt)
were determined using blue dextran and vitamin B12 respectively.
SDS-acrylamide gel electrophoresis of B. ovifera blood cells and plasma was
carried out by the Laemmli method (Laemmli, 1970) using vertical gel slabs; 0-
lactoglobulin ( 1 8,400 d), trypsinogen (24,000 d), egg albumin (45,000 d), and bovine
albumin (66,000 d) were used as molecular weight standards. The gels were fixed
for twelve hours with 10% trichloroacetic acid, stained with 0.25% Coomassie bril-
liant blue dye for four hours, and destained with 7% acetic acid. Independent of the
molecular weight determination, and in order to avoid denaturation, electrophoresis
of the native proteins in the plasma was run excluding SDS.
RESULTS
In the first section we report the iron concentration in tunicate blood plasma
and blood cells, the oxidation state of the iron, the iron distribution in the plasma,
cytoplasm, and the cell membranes (no differentiation between cell and intracellular
membranes was made). Protein distribution is reported in the second section.
Iron
The oxidation state of the iron present in solution can be established before a
quantitative determination of the total iron content of the plasma and blood cells
3 carried out. The advantage to this approach is that once the predominant oxidation
state of the iron is known, more than one method for total iron determination can
be employed, and the results compared.
IRON IN TUNICATE BLOOD 103
TABLE I
Oxidation state analysis of iron obtained from blood cell cytolysis
Tunicate species
Test reagent B. ovifera S. clava M. manhattensis
1,1 0-phenanthroline + + +
K,Fe(CN)6 + + +
K4Fe(CN)6
KSCN
The results of the oxidation state analysis of the iron in the cell lysates (cytoplasm)
of all three species are tabulated in Table I. The oxidation state of the iron in blood
cells is found to be in the Fe(II) form as was previously reported for Pyura stolonifera
(Endean, 1955). No precipitation occurred when barium chloride was added to
blood cell lysates, indicating the absence of sulfate. After applying the same test
reagents, shown in Table I, to plasma of S. clava acidified with concentrated HC1,
we find that both iron(II) and iron(III) are present. Addition of barium chloride
gave a white precipitate.
Osmium tetroxide (OsO4) vapors were used to localize regions of the blood cell
with reducing ability. As indicated by the staining results (Table II), it is concluded
that most of the reducing substances are found in the vacuolated cells; i.e., morula,
compartment, and signet ring cells. Amoebocytes also show some staining.
Several attempts were made to detect iron(III) by the EPR method. Since the
sensitivity depends on the total number of spins in the spectrometer, hence the cell
count, blood samples from two or three specimens were pooled. The characteristic
seven line spectrum of the FeF63~ complex was not observed in any of these ex-
periments.
Since we found no iron(III) by the available methods, total iron concentration
within the cell was determined by the 1, 1 0-phenanthroline method, which is specific
for iron(II). The accuracy of this method is limited by the accuracy of the volume
determination of the blood cells based on the estimated average cell diameter and
total cell count (10-15% error). To simplify calculations, it is assumed that all cell
types have equal amounts of iron. Total iron concentration determination using
TABLE II
Osmium tetroxide staining of blood cells
Tunicate species
Cell type
B. ovifera
S. clava M. manhattensis
Morula cell vacuoles
+++ or -*
+++ +++
Compartment cell vacuoles
**
+++ +++
Signet ring cells
++ +
NI NI
Amoebocytes
+ or -
++ +++
Lymphocytes
—
+ or -
* Some vacuoles are stained others are not.
** A few stained vacuoles; in general, clear vacuoles and stained cytoplasm.
NI not identified in blood smears.
104 M. I. AGUDELO ET AL.
TABLE III
Iron concentration in blood cells as determined in pooled samples*
Method
Species Fe(Phen)32+ AA
B. ovifera
S. clava
M. manhattensis
1 X 10 3 M
5-9 X 1(T3 M
8 X 1(T3 M
6 X 1(T3 M
7 X 10~2 M
* Relative accuracy is limited by volume determination (±10-15%) and/or iron detection.
atomic absorption spectrometry (AA) is also dependent on cell volume and cell
count determination. The results for both methods are shown in Table III. The iron
content of the plasma as determined by AA was 1.6-1.8 ppm for S. clava and B.
ovifera.
There is an order of magnitude difference in iron concentration in the blood
cells by the two methods. This difference is greater than that expected from error
in cell volume and cell count determinations. Procedural differences between iron
analysis methods account for this observation. In the phenanthroline method the
cell membranes are discarded, while in the atomic absorption analysis the whole
cells are digested and analysed. Therefore a large fraction of the iron in the cells is
associated with cell membranes.
To determine how much of the iron is found in the cytoplasm and in the cell
membranes, the following analysis was carried out: the blood cells were lysed with
distilled water, the lysate was separated from the membranes by centrifugation at
1 8,400 g for 30 min. The cell membranes were then resuspended in 0. 1 N HC1,
mixed thoroughly, and separated again by centrifugation at 18,400 g for 30 min.
The membranes were then digested with concentrated nitric acid for 3-4 hours until
a clear solution was obtained. The results of the iron analysis by atomic absorption
of the cell lysate, 0. 1 N HC1 wash and digested membranes are tabulated as percent
iron in Table IV.
As shown in Table IV, over 70% of the iron is bound to the cell membranes.
The 0. 1 N HC1 wash removes any iron that might have precipitated during cell lysis
with distilled water, as well as any loosely bound surface iron. Since most of the
iron is found in association with cell membranes, volume concentration units are
illusory. Analysis of the iron content per weight of protein was therefore carried out.
The results yield 0.05 ± 0.01 /ig Fe/mg protein in the plasma, and 1.1 ± 0.2 /xg Fe/
mg protein in blood cells of B. ovifera. In the blood of S. clava we found 0.14
±0.1 ng Fe/mg protein in the plasma and 1 . 1 1 ± 0.05 ng Fe/mg protein in
blood cells.
TABLE IV
Relative iron distribution in blood cells as determined in pooled samples
Species Cell lysate HC1 wash Cell membranes
B. ovifera ( 1 )
14%
86%
(2)
2%
7%
91%
•lava
11%
16%
73%
IRON IN TUNICATE BLOOD
105
Proteins
Size exclusion chromatography of the cell lysate (cells lysed with 0. 1 TV HC1)
gave an absorbance profile with two main peaks. One peak eluted at the exclusion
limit, V0 (molecular weight greater than 75,000). The second peak eluted at the bed
volume, Vt (molecular weight less than 3000), and is assigned to tunichrome (Macara
et al, 1979), which has a lower molecular weight. Iron was eluted with both the
high molecular weight protein and tunichrome fractions, some iron was eluted after
the tunichrome peak at a kave of 1.2.
Slightly different results were obtained in chromatography of the plasma. Only
a high molecular weight protein peak was observed. In a few cases a low molecular
weight peak, attributed to tunichrome, was also found, probably because of cell lysis
during centrifugation. The iron concentration in the plasma is very low, 1.6-1.8
ppm, and no iron was detected in the high molecular weight protein fractions. A
small amount of iron was detected at a kave of 1.2.
For a better characterization of the proteins in the plasma and the water insoluble
proteins found in the cell membranes, SDS-acrylamide gel electrophoresis of B.
ovifera blood cells and plasma was carried out using the Laemmli method. The
protein subunits found in the plasma and the cell membranes are shown in Figure
1. With 10% acrylamide gel two main bands were observed for the plasma corre-
200
Sf
'Z
t
o
I
O ("ell membrane
proteins
Plasma proteins
Protein standards
Bovine albumin
(24,000)
B-Lactoglobulin
I I
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
20
10
FIGURE 1. SDS-acrylamide gel electrophoresis (Laemmli method) of B. ovifera blood plasma (10%
and 12% gels) and blood cells (7.5%, 10%, and 12% gels).
106 M. I. AGUDELO ET AL.
spending to 31,000 and 26,000 d. The cell membrane samples showed a large
number of proteins; only the most visible and clear bands are reported: a strong
band at 130,000, two faint bands at 56,000, and 51,000; a strong band at 44,000,
a medium band at 3 1 ,000, and a strong band at 26,000 d.
With 12% acrylamide gel results showed two bands for the plasma sample at
molecular weights 28,000 and 26,500, which were also observed in the cell mem-
branes. By referring to Figure 1 , we see that these 1 2% acrylamide gel results cor-
respond to the 31,000 and 26,000 bands in the 10% acrylamide gel experiment.
These two bands are the only ones observed to occur in both the membranes and
plasma. A strong band was also observed in the membrane sample at 48,000 d
corresponding to the 45,000 d band observed in the 10% gel. The two faint bands
observed at 56,000 and 51,000 with a 10% gel, and the 130,000 band were not
observed.
To insure that the 130,000 molecular weight protein observed using a 10%
acrylamide gel was not an artifact, a 7.5% acrylamide gel was run. Since no high
molecular weight standards were available and our main concern was to determine
the existence of a high molecular weight protein, and not its exact molecular weight,
only two standards were used, bovine albumin and egg albumin. If a straight line
is assumed to pass through the two molecular weight standards used when Rf values
were plotted against log(molecular weight) in Figure 1, the following conclusions
can be drawn. The strong band observed at an Rf of 0.07 corresponds to an ap-
proximate molecular weight of 160,000, and the strong band observed at an Rf of
0.95 corresponds to a molecular weight of 45,000.
To determine the iron binding properties of the plasma proteins, we added to
1 .0 ml of B. ovifera blood plasma 0.050 ml of an 55Fe-citrate stock solution con-
taining 1,000-fold excess citrate to prevent Fe(III) precipitation. The sample was
allowed to stand for 15 minutes at 0°C. Gel electrophoresis of the native proteins
was carried out, taking care not to denature the proteins by excluding SDS from
the procedure, preventing any changes in the native protein configuration that would
alter its Fe-binding properties. After the electrophoretic separation of the proteins,
the gel was cut vertically into two pieces. One piece was fixed and stained as described
in the methods section. To prevent any radioactive iron loss into the fixative solution,
the other piece was not fixed. This gel was cut into 0.5 cm horizontal sections and
analysed for 55Fe by liquid scintillation counting. In the stained piece we observed
one band. The largest 55Fe activity was observed at the Rf value corresponding to
this protein band, indicating that the native plasma protein has the ability to
bind iron.
DISCUSSION
Oxidation state +2 predominates for iron found in the blood cell cytoplasm
of the three ascidians B. ovifera, S. clava, and M. manhattensis. No iron in oxidation
state +3 was detected. The blood plasma, however, contains both iron(II) and
iron(III). This finding is not surprising; even if only iron(II) is present in the plasma,
then as soon as the plasma is exposed to air, some of the iron will be oxidized to
the +3 oxidation state. It is also possible that iron is in the +3 form in the plasma
and is reduced as it goes into the cell. In this case, the +2 iron in the plasma arises
from cell lysis, exchange, or leaching.
The iron concentration in blood cells varies from species to species, similar to
the variation in vanadium concentration among vanadium-containing tunicates
IRON IN TUNICATE BLOOD 107
(Hawkins, personal communication). Tunicates therefore accumulate iron against
an approximately 105-106 concentration gradient (ratio of iron in tunicate blood
cells, 10 3-10"2 M, to dissolved iron in sea water, 2 X 10~8 M(Kester et al., 1975)).
Although there is considerable iron concentration in the cell cytoplasm, a large
fraction of the total iron was found in the cell membranes (over 70%). We have not
yet determined the oxidation state of the membrane-bound iron, which will require
more complicated techniques than those we report in this paper.
The OsO4 staining method has often been used to determine the metal ion
distribution in cells (Henze, 1913;Endean, 1960;Kalk, 1963;Fuke, 1979). However,
we encountered several problems in the interpretation of this method. OsO4 is sen-
sitive to many strong reducing agents. Along with Fe(II), tunicate blood cells contain
tunichrome, a relatively strong reducing agent (Macara et al, 1979) capable of
reacting with OsO4 to generate dark stains. There is also considerable variation in
the results among similar cell types. Due to these considerations, OsO4 staining leads
us to conclude that there are one or more reducing agents (iron(II), tunichrome, or
both) in the morula cell vacuoles of all three species as well as in compartment cell
vacuoles. The staining of B. ovifera compartment cell cytoplasm and not vacuoles
cannot be explained easily. Leaching of the vacuolar contents during the staining
procedure would result in uniform staining throughout the cell. On the contrary,
the vacuoles remained intact and clear, whereas the cytoplasm was stained deeply.
Metal ion content of blood cells is often given in volume-based concentration
units, such as moles/liter (e.g., Tullius et al., 1980). However, since most of the iron
is found in the cell membranes, and is probably associated with a specific protein,
we find it more useful to report the iron content as ^g Fe per mg of protein. With
this unit we find a smaller difference in the iron concentrations of B. ovifera and
5". clava; approximately 1 /ug Fe/mg of protein in the blood cells of each species.
This value is comparable to that of other iron-accumulating blood cells. For example,
it is comparable to the concentration of 3.48 ng Fe/mg of protein in human eryth-
rocytes if hemoglobin is used as the total protein content. The plasma value of 0.1-
0.05 /ug Fe/mg protein is higher than the value of 0.015 ^g Fe/mg of protein in
human plasma (Altman, 1961; Bishop and Surgenor, 1964).
Size exclusion chromatography of the cell lysate showed two main peaks: a high
molecular weight protein that elutes at V0 using Sephadex G-75, and a low molecular
weight compound, tunichrome (Macara et al., 1979). Iron was found in both peaks,
and some iron was eluted after the second peak, probably free iron, because of the
high acidity of the eluent.
Chromatography of the plasma resulted in the isolation of a high molecular
weight protein; tunichrome was also observed in some cases, probably due to cell
lysis during centrifugation. No iron was observed with the high molecular weight
protein, however some iron was observed after the tunichrome peak.
In a comparative study on the distribution of metal ions in the plasma of as-
cidians Pyura stolonifera and Ascidia ceratodes (Hawkins et al., 1980), results were
obtained which relate closely to our experiments. In common with our study, their
chromatography experiments show a protein that elutes at V0, and low molecular
weight fractions that test positive for N-acetylaminosugar and negative for protein,
and could be assigned to tunichrome. No iron was detected in the high molecular
weight fraction. Some iron was found in the low molecular weight fractions when
the eluent contained NaCl; however when distilled water was used as an eluent, no
iron was detected. Although this result is explained as iron impurities in the NaCl,
it is more likely that the iron binds to the gel due to the low ionic strength of the
eluent (namely distilled water).
108 M. I. AGUDELO ET AL.
From SDS electrophoresis of the blood cells and plasma there are two protein
components in the plasma that are found on the membranes as well (molecular
weights 31,000 and 26,000 in the 10% acrylamide gel). These two components are
probably obtained from the single denatured high molecular weight protein observed
in gel chromatography of plasma proteins. Electrophoresis without SDS of the native
plasma proteins yields one major protein band corroborating the chromatography
results. Electrophoresis of plasma doped with iron(III)-citrate shows that this protein
has an iron affinity high enough to compete with the citrate ligand. No attempt to
determine the molecular weight of this 55Fe-labeled protein was made.
Similar iron binding results were obtained by Webb and Chrystal (1981) using
blood plasma of the ascidian Herdmania momus (order Pleurogona). However, in
their experiments the iron(III) was added as iron chloride in 0.1 N HC1 and then
neutralized with bicarbonate. This procedure can cause the iron to precipitate or
form high molecular weight iron hydroxide polymers in solution that could be eluted
at the exclusion limit along with high molecular weight proteins.
ACKNOWLEDGMENTS
This research was supported in part by National Science Foundation Grant
PCM-7824782, and in part by National Institutes of Health Grant HL-24225. We
thank Mr. Paul Barrington for collecting 5". clava for us.
LITERATURE CITED
AGUDELO, M. I., K. KUSTIN, AND W. E. ROBINSON. 1982. Blood chemistry of Boltenia ovifera. Comp.
Bwchem. Physiol. 72A: 161-166.
AGUDELO, M. I., K. KUSTIN, AND G. C. McLEOD. 1983. The intracellular pH of the blood cells of the
tunicate Boltenia ovifera. Comp. Biochem. Physiol. 75A: 211-214.
ALTMAN, P. L. (comp.). 1961. Blood and Other Body Fluids; Federation of American Societies for
Experimental Biology, Washington, D. C., D. S. Dittmer, ed. 540 pp.
ANDERSON, M. A., AND F. M. M. MOREL. 1982. The influence of aqueous iron chemistry on the uptake
of iron by the coastal diatom Thalassiosira weissflogii. Limnol. Oceanogr. 27: 789-813.
BISHOP, C., AND D. M. SURGENOR. 1964. The Red Blood Cell. Academic Press, New York. 566 pp.
DINGLEY, A. L., K. KUSTIN, I. G. MACARA, AND G. C. McLEOD. 1981. Accumulation of vanadium by
tunicate blood cells occurs via a specific anion transport system. Biochim. Biophvs. Ada 649:
493-502.
EMERY, T. 1982. Iron metabolism in humans and plants. Am. Sci. 70: 626-632.
ENDEAN, R. 1955. Studies of the blood and tests of some Australian ascidians. I. The Blood of Pyura
stolonijera. Austr. J. Mar. Freshwater Res. 6: 35-59.
ENDEAN, R. 1960. The blood cells of the ascidian Phallusia mammillata. J. Microsc. Sci. 101: 177-197.
FUK.E, M. T. 1979. Studies on the coelomic cells of some Japanese ascidians. Bull. Mar. Biol. Tohoku
Univ. 16: 143-159.
HAWKINS, C. J., P. M. MEREFIELD, D. L. PARRY, W. R. BIGGS, AND J. H. SWINEHART. 1980. Com-
parative study of the blood plasma of the ascidians Pvura stolonijera and Ascidia ceratodes.
Biol. Bull. 159: 656-668.
HENZE, M. 1913. Untersuchungen uber das blut der ascidien. III. Hoppe Seyler's Z. Physiol. Chem. 86:
340-344.
KALK, M. 1963. Intracellular sites of activity in the histogenesis of tunicate vanadocytes. /. Microsc. Sci.
104: 483-493.
KESTER, D. R., R. H. BYRNE, JR., AND Y. J. LIANG. 1975. Redox reactions and solution complexes of
iron in marine systems. Pp. 56-79 in Marine Chemistry in the Coastal Environment, Thomas
M. Church, ed. ACS Symposium Series No. 18. Washington.
LAEMMLI, U. K. 1970. Cleavage of structural proteins during the assembly of the head of Bacteriophage
T4. Nature 227: 680-685.
LEV ANON, H., G. STEIN, AND Z. Luz. 1968. The electron spin resonance spectrum of (FeF6~3) in aqueous
solutions. J. Am. Chem. Soc. 90: 5292-5293.
LOWRY, O. H., N. J. ROSEBROUGH, A. L. FARR, AND R. J. RANDALL. 1951. Protein measurement with
the Folin phenol reagent. J. Biol. Chem. 193: 265-275.
IRON IN TUNICATE BLOOD 109
MACARA, I. G., G. C. McLEOD, AND K. KUSTIN. 1979. Tunichrome and metal ion accumulation in
tunicate blood cells. Comp. Biochem. Physiol. 63B: 299-302.
MARX, J. J. M., AND P. AISEN. 1981. Iron uptake by rabbit intestinal mucosal membrane vesicles.
Biochim. Biophys. Ada 649: 297-304.
MILLAR, R. H. 1966. Evolution in ascidians. Pp. 519-534 in Some Contemporary Studies in Marine
Science, H. Barnes, ed. George Allen and Unwin Ltd., London.
SWINEHART, J. H., W. R. BIGGS, D. J. HALKO, AND N. C. SCHRODER. 1974. The vanadium and selected
metal contents of some ascidians. Biol. Bull. 146: 302-312.
TULLIUS, T. D., W. O. GILLUM, R. M. K. CARLSON, AND K. O. HODGSON. 1980. Structural study of
the vanadium complex in living ascidian blood cells by X-ray absorption spectroscopy. J. Am.
Chem. Soc. 102: 5670-5676.
WEBB, J., AND P. CHRYSTAL. 1981. Protein binding of iron in blood plasma of the ascidian Herdmania
momus. Mar. Biol. 63: 107-112.
WRIGHT, R. K. 1981. Urochordates. Pp. 565-626 in Invertebrate Blood Cells, vol. 2. Arthropods and
Urochordates, Invertebrates and Vertebrates Compared, N. A. Ratcliffe and A. F. Rowley, eds.
Academic Press, London.
Reference: Biol. Bull. 165: 110-118. (August, 1983)
A CYTOLOGICAL ANALYSIS OF FERTILIZATION
IN CHAETOPTERUS PERGAMENTACEUS3
WINSTON A. ANDERSON1 AND WILLIAM R. ECKBERG'2
1 Department of Zoology, Howard University, Washington, DC 20059, and2Marine Biological
Laboratory, Woods Hole. MA 02543
ABSTRACT
We have examined sperm-egg interaction in Chaetopterus pergamentaceus by
electron microscopy. The initial contact between sperm and egg involved the mem-
brane of the unreacted acrosome and either the tips of egg microvilli which pene-
trated the vitelline layer or jelly emanating from the tips of the microvilli. This
resulted in an acrosome reaction and fusion between the inner acrosomal membrane
and the tip of the micro villus. Sperm did not produce acrosomal processes like those
of many other invertebrates, and no part of the sperm penetrated the vitelline layer
until the sperm was incorporated into the fertilization cone. The fertilization cone
was very small and was composed of egg microvilli. The sperm nucleus and mi-
tochondrion were incorporated into the fertilization cone, but a recognizable sperm
mitochondrion could not subsequently be seen in the egg cytoplasm. Although the
axoneme of the sperm tail was present in the fertilization cone at early stages of
sperm penetration, the sperm tail evidently detached in the later stages of incor-
poration because it could not be seen in the zygote cytoplasm after sperm incor-
poration. The sperm chromatin decondensed uniformly and became surrounded
by a typical nuclear envelope. The results indicate that Chaetopterus provides an
example of a previously undescribed model for sperm penetration of egg vestments
in which the sperm needs neither to produce an acrosomal process nor to liberate
vitelline layer lysins because it penetrates the vitelline layer passively after incor-
poration into the egg cytoplasm.
INTRODUCTION
Fertilization is characterized by a sequence of events. A gamete interaction trig-
gers the acrosomal reaction that initiates initial sperm-egg attachment, subsequent
gamete membrane fusion, zygote formation, and egg activation. Sperm incorpo-
ration ensues, and finally the genetic material of the two gametes combines.
In Spiralians, studies of gamete interactions have been limited to the molluscs
Barnea (Pasteels, 1965), Mytilus (Longo and Anderson, 1969), Spisula (Longo and
Anderson, 1970) and Haliotis (Lewis et ai, 1982), the annelids Hydroides (Colwin
and Colwin, 196 la, b) and Nereis (Fallen and Austin, 1967) and the echiurid,
Urechis (Tyler, 1965; Paul and Gould-Somero, 1976). Sperm-egg interaction in these
forms appears to follow several plans.
In Hydroides and Haliotis, the sperm undergoes an acrosome reaction in as-
sociation with the outer surface of the vitelline layer and penetrates the vitelline
layer with the assistance of sperm lysins which partially dissolve the vitelline layer.
In Hydroides (Colwin and Colwin, 1960), the mechanism of this penetration is not
eceived 21 March 1983; accepted 25 May 1983.
Dedicated to the memory of Professor E. E. Just on the 100th anniversary of his birth.
110
CHAETOPTERUS FERTILIZATION 1 1 1
known, but in Haliotis, the sperm lysin evidently acts by a non-enzymatic mech-
anism (Lewis et al., 1982). In Barnea (Pasteels, 1965), Urechis (Tyler, 1965; Paul
and Gould-Somero, 1976) and probably Spisula (Longo. 1976) a sperm acrosomal
filament fuses with an egg microvillus and the sperm nucleus penetrates the vitelline
layer after being incorporated into the fertilization cone.
We obtained evidence that the Chaetopterus vitelline layer played a role in
preventing polyspermy, but did so without being structurally or functionally changed
after fertilization (Eckberg and Anderson, 1983). Additionally, in preliminary stud-
ies, we did not obtain evidence for sperm lytic activity against the vitelline layer.
Therefore, we initiated a study of sperm-egg interaction in this species. The results
showed that the fertilizing sperm fuses with the tip of one or more egg microvilli
which extend beyond the vitelline layer and is surrounded by a fertilization cone.
We also found that egg microvilli retract from the vitelline layer after fertilization.
Therefore the vitelline layer of the fertilized egg can become a physical barrier to
sperm-egg fusion without being structurally or functionally altered by fertilization.
MATERIALS AND METHODS
Gametes were obtained and handled, fixed for 1 h at room temperature in 5%
glutaraldehyde, 4% paraformaldehyde, 0.1 M sodium cacodylate, pH 7.8 in artificial
sea water, and processed for light and electron microscopy as described
(Eckberg, 198 la).
Inseminated eggs were fixed at intervals after fertilization (0.5, 1, 2, 3, 5, 9, and
14 min). Male pronuclear formation was complete by 14 min. Although the eggs
examined in this study were polyspermic due to heavy insemination, sperm asso-
ciated with the vitelline layer more than 1 min after insemination were supernu-
merary because this species has a complete block to sperm penetration by this time
(Eckberg and Anderson, 1983). Polyspermic eggs develop synchronously with con-
trols up to the time of cleavage. Although they fail to divide, they undergo differ-
entiation without cleavage (Lillie, 1902; Eckberg, 1981b; Eckberg and Kang, 1981).
Therefore the events of fertilization in such polyspermic eggs are very likely to be
the same as those in monospermic eggs.
RESULTS
Oocyte surface
The Chaetopterus egg is normally inseminated at the first meiotic metaphase.
The cytoplasmic organization of the oocyte at this stage has been described (Eckberg,
198 la). Since the sperm interacts with the vitelline layer and oocyte surface, this
region will be described more fully here. The vitelline layer is fibrous and is organized
into three distinct regions: an inner region composed of a dense fibrous meshwork,
a middle region composed of fibers oriented parallel to the oocyte surface, and an
outer region of electron-dense granules interspersed with the tips of microvilli (Figs.
1, 2). This is covered by an outer diffuse "jelly" layer (Figs. 1, 2). Jelly filaments
originate from the granules and the microvillar tips.
Sperm
Mature sperm consist of a head and midpiece about 1 /um X 4 /urn and a long
flagellum (Fig. 3). Transverse sections (not shown) reveal a single mitochondrion
surrounding a centriole pair which serves as the origin of the flagellum. The aero-
112
W. A. ANDERSON AND W. R. ECKBERG
FIGURE 1 . Surface of an unfertilized egg of Chaetopterus. Note the three regions of the vitelline
layer: I = inner dense layer, M = middle layer, G = granules comprising the outer layer. Note also that
microvilli (MV) penetrate the vitelline layer completely. Also note the fibrillar jelly coat originating from
the tips of the microvilli and the granules of the vitelline layer (arrows). Bar = 1 ^m.
FIGURE 2. Tangential section of the vitelline layer and jelly coat of an unfertilized egg. Symbols
are as given in the legend to Figure 1. Note that the granules of the outer region of the vitelline layer
3 re numerous between the tips of the microvilli. Bar = 1 p.m.
FIGURE 3. Longitudinal section of a Chaetopterus sperm. A = acrosome, N = nucleus, M
= mitochondrion, C = centriole, F = flagellum. Bar = 1 nm.
FIGURE 4. Longitudinal section through the acrosomal region of a Chaetopterus sperm. Note the
CHAETOPTERUS FERTILIZATION 113
somal region consists of a cup-like acrosomal vesicle containing fibrous material
associated with its membranes and a region of granular material between the ac-
rosomal vesicle and the apex of the nucleus. The acrosomal vesicle also covers the
apical end of the sperm nucleus (Figs. 4, 5).
Gamete contact and fusion
The initial contact between sperm and egg involves the outer acrosomal mem-
brane and the jelly in association with the microvilli (Fig. 6). Sperm with reacted
acrosomes are oriented perpendicular to the oocyte surface (erect) (Fig. 7). The
acrosome reaction involves the opening of the acrosomal vesicle and results in fusion
between the inner acrosomal membrane and an egg microvillus (Fig. 8). Sperm do
not produce acrosomal processes, and the tiny membranous projections which
formed as the result of the acrosome reaction did not penetrate the vitelline layer.
Sperm incorporation
This process involves the formation of a tiny fertilization cone, barely visible
in light micrographs (Fig. 8 inset), which consists of a few thickened microvilli
surrounding the sperm (Figs. 9, 10). These microvilli contain longitudinal micro-
filament bundles (Fig. 10). The nuclear membrane of the newly-incorporated sperm
becomes vesiculated and the sperm chromatin begins to disperse (Fig. 1 1).
After incorporation of the sperm head, a sperm tail protrudes from some, but
not all residual fertilization cones (Fig. 10 insets). However, we never observed
sperm flagella within the zygote other than short segments in the fertilization cone
(Fig. 9). Nor did we observe recognizable sperm mitochondria in the zygote cyto-
plasm subsequent to sperm incorporation.
Fertilized egg surface
The vitelline layer is structurally unchanged after fertilization. All three regions
are present and structurally similar to those of the unfertilized egg. However, the
egg microvilli are generally absent from the vitelline layer (Fig. 12). Where they are
present, they are greatly reduced in number and do not penetrate to the surface of
the vitelline layer.
Formation of the male pronucleus
Sperm chromatin decondenses completely and uniformly (Fig. 13), and the nu-
clear envelope disappears (Fig. 14). Decondensed chromatin is frequently associated
with small granules similar, but not identical to, the lipid granules of the oocyte
(Fig. 14, 15). After complete decondensation, membrane vesicles surround the chro-
matin (Fig. 15) and eventually coalesce into a typical annulate pronuclear envelope
(Fig. 16).
cuplike acrosomal vesicle (AV) containing fibrous material associated with the membranes and the
granulai1 material (AG) between the acrosomal vesicle and the nucleus. N = nucleus. Bar = 0.25 ^m.
FIGURE 5. Transverse section through the acrosomal region of a Chaetopterus sperm. The plasma
membrane (PM) is clearly separated from the outer acrosomal membrane (OAM) at a few points. The
inner acrosomal membrane (IAM) is clearly separated from the nuclear membrane (NM). Fibrous ma-
terial in the acrosomal vesicle can be seen, but the acrosomal granule is out of the plane of this section.
Bar = 0.25 Mm.
114
W. A. ANDERSON AND W. R. ECKBERG
FIGURE 6. Initial interaction between a sperm and a microvillus prior to gamete fusion. Note that
the plasma membrane over the acrosome is associated with a microvillus tip (MV) via the jelly (J). Bar
= 1 ^m.
FIGURE 7. Erection of a sperm following initiation of the acrosome reaction and attachment of the
sperm to a microvillus (MV). Bar = 1 /urn.
FIGURE 8. Gamete fusion involving the sperm inner acrosomal membrane (IAM) and an egg
microvillus (MV). Bar = 1 ^m. Inset: light micrograph showing a sperm attached to an egg microvillus
which has thickened to the point where it is visible at this level of resolution and can thus be called a
fertilization cone (arrow).
FIGURE 9. Sperm incorporation into the Chaetoplerus egg. Note that the sperm nucleus (N), mi-
tochondrion (M) and base of the flagellum (F) have all been incorporated. Note also the microvilli (MV)
iiich surround the sperm and make up the small fertilization cone. Bar = 1 ^m. Inset: light micrograph
of a slightly earlier stage in fertilization cone formation showing several microvilli surrounding the
incorporated sperm.
CHAETOPTERUS FERTILIZATION
115
M
FIGURE 10. Higher magnification electron micrograph showing microfilaments (arrows) longitu-
dinally-arranged in the microvilli of the fertilization cone. Bar = 0.25 Mm. Insets: light micrographs
showing late stages in fertilization cone formation. In the upper inset, the sperm tail still protrudes from
the fertilization cone; in the lower inset, the sperm tail has been lost.
FIGURE 1 1. Nucleus of a newly-incorporated sperm. The nuclear envelope has become vesiculated
(VNE) and the chromatin appears to be beginning to decondense (DC). Bar = 0.25 nm.
FIGURE 12. Surface of a fertilized egg 9 min after insemination. Note that the egg microvilli have
shortened and no longer penetrate the vitelline layer, although all regions of the vitelline layer (I = inner,
M = middle, G = granular) and the jelly (J) remain. Bar = 1
DISCUSSION
The Chaetopterus oocyte surface was similar to that observed in other species.
The outer diffuse jelly coat was evidently the substance initially contacted by the
sperm and appeared to originate from the granules at the ends of the microvilli and
at the outer region of the vitelline layer. Similar granules appear at the initial contact
points in Urechis (Tyler, 1965), Nereis (Fallon and Austin, 1967), and Hydroides
(Colwin and Colwin, 196 la). These may originate during oogenesis as buds from
the tips of oocyte microvilli (L. E. Franklin, data presented in Metz, 1967). This
116
W. A. ANDERSON AND W. R. ECKBERG
FIGURE 13. Decondensing sperm nucleus showing a stage slightly later than that in Figure 1 1. The
sperm nuclear membrane remains vesiculated and the chromatin is decondensing uniformly throughout
the nucleus. Bar = 1 ^m.
FIGURE 14. Fully decondensed sperm nucleus, without a nuclear envelope, in association with lipid
granules (L). Bar = 0.5 nm.
FIGURE 15. Fully-decondensed sperm nucleus showing association with lipid granules (L) and a
vesiculated nuclear envelope (arrows). Bar = 0.5 /*m.
FIGURE 16. Male pronucleus showing annulate nuclear envelope (NE) and a dense nucleolus-like
body (NLB) in the pronucleus. Bar = 0.5
homology suggests similar function. We propose that these structures contain a
eptor which initiates the acrosome reaction and is therefore analogous to the
fucose-sulfate polysaccharide of sea urchin egg jelly (SeGall and Lennarz, 1979).
Additional granules may initiate the acrosome reaction in supernumerary sperm
(Eckberg and Anderson, 1983).
CHAETOPTERUS FERTILIZATION 1 1 7
Gamete fusion took place via the inner acrosomal membrane of the sperm and
the tips of egg microvilli. Sperm did not produce acrosomal processes. Other species
which do not produce acrosomal processes generally fuse with the egg with the
assistance of lysins which facilitate penetration of the vitelline layer (Colwin and
Colwin, 1960; Lewis et al, 1982). Such is evidently not the case in Chaetopterus,
because ( 1 ) in preliminary experiments we could detect no evidence for sperm lysins,
(2) sperm fused with the tips of egg microvilli which protruded through the vitelline
layer, (3) in fertilized eggs such microvilli no longer penetrated the vitelline layer,
(4) chemical disruption of the vitelline layer permitted refertilization (Eckberg and
Anderson, 1983), presumably by making the oocyte surface available again to sperm,
and (5) sperm were never observed to penetrate the vitelline layer until they were
incorporated into the fertilization cone. The preceding observations also indicate
that this microvillar retraction from the vitelline layer can provide a mechanism for
a permanent block to polyspermy in this species.
Sperm of other species which produce acrosomal processes may (Pasteels, 1965;
Tyler, 1965; Longo, 1976) or may not (Longo and Anderson, 1968) fuse with egg
microvilli, but if they fuse with microvilli preferentially, they apparently do not fuse
with the tips (Pasteels, 1965; Tyler, 1965). Chaetopterus thus provides an example
of a previously undescribed model for sperm penetration of egg vestments in which
the sperm needs neither to produce an acrosomal process nor to liberate vitelline
layer lysins because it penetrates the vitelline layer passively after incorporation into
the egg cytoplasm.
Sperm incorporation was mediated by a tiny fertilization cone (Morgan and
Tyler, 1930), shown here to be composed of slightly thickened microvilli. Since such
microvilli contained bundles of microfilaments, fertilization cone formation and
action would appear similar in mechanism to that observed in other species (Tyler,
1965; Longo, 1978).
A recognizable sperm mitochondrion could not be seen in the zygote cytoplasm
subsequent to incorporation. However, since it was present in the fertilization cone,
the sperm mitochondrion must have been incorporated into the zygote. In Mytilus,
the sperm mitochondrion reportedly becomes indistinguishable from egg mito-
chondria (Longo and Anderson, 1969). A similar situation may exist in Chaetop-
terus. This differs, however, from sea urchins, in which the sperm mitochondrion
persists as an identifiable structure during cleavage and is metabolically active (An-
derson, 1968; Anderson and Perotti, 1975).
The lack of complete incorporation of the sperm tail is similar to the situation
in other spiralians (Tyler, 1965; Longo and Anderson, 1969, 1970), but different
from that in sea urchins (Longo and Anderson, 1968) and mammals (Piko, 1969)
in which sperm tails can be seen in the zygote cytoplasm long after fertilization.
However, the sperm centriole is incorporated and sets up the first cleavage spindle
(Mead, 1895).
ACKNOWLEDGMENTS
This research was supported in part by NIH grant RR08016 to W. R. Eckberg
and a grant from the Mellon Foundation to W. A. Anderson. We thank Mr. Steven
Lindsey, Ms. Ellen Strachan and Ms. B. J. Anderson for excellent technical
assistance.
LITERATURE CITED
ANDERSON, W. A. 1968. Structure and fate of the paternal mitochondrion during early embryogenesis
of Paracentrotus lividus. J. Ultrastruct. Res. 24: 311-321.
118 W. A. ANDERSON AND W. R. ECKBERG
ANDERSON, W. A., AND M. E. PEROTTI. 1975. An ultracytochemical study of the respiratory potency,
integrity, and fate of the sea urchin sperm mitochondria during early embryogenesis. J. Cell
Biol. 66: 367-376.
COLWIN, A. L., AND L. H. COL WIN. 196 Ib. Changes in the spermatozoon during fertilization in Hydroides
hexagonus (Annelida). II. Incorporation with the egg. J. Biophys. Biochem. Cytol. 10: 255-274.
COLWIN, L. H., AND A. L. COLWIN. 1960. Formation of sperm entry holes in the vitelline membrane
of Hydroides hexagonus (Annelida) and evidence of their lytic origin. /. Biophys. Biochem.
Cytol. 7: 315-320.
COLWIN, L. H., AND A. L. COLWIN. 1 96 1 a. Changes in the spermatozoon during fertilization in Hydroides
hexagonus (Annelida). I. Passage of the acrosomal region through the vitelline membrane. J.
Biophys. Biochem. Cytol. 10: 231-254.
ECKBERG, W. R. 198 la. An ultrastructural analysis of cytoplasmic localization in Chaetopterus perga-
mentaceus. Biol. Bull. 160: 228-239.
ECKBERG, W. R. 1981 b. The effects of cytoskeleton inhibitors on cytoplasmic localization in Chaetopterus
pergamentaceus. Differentiation 19: 55-58.
ECKBERG, W. R., AND W. A. ANDERSON. 1983. Blocks to polyspermy in Chaetopterus. (Submitted for
publication).
ECKBERG, W. R., AND Y. H. KANG. 1981. A cytological analysis of differentiation without cleavage in
cytochalasin B- and colchicine-treated embryos of Chaetopterus pergamentaceus. Differentia-
lion 19: 154-160.
FALLON, J. F., AND C. R. AUSTIN. 1967. Fine structure of gametes of Nereis limbata (Annelida) before
and after interaction. J. Exp. Zool. 166: 225-242.
LEWIS, C. A., C. F. TALBOT, AND V. D. VACQUIER. 1982. A protein from abalone sperm dissolves the
egg vitelline layer by a nonenzymatic mechanism. Dev. Biol. 92: 227-239.
LILLIE, F. R. 1902. Differentiation without cleavage in the egg of the annelid Chaetopterus pergamen-
taceus. Wilhelm Roux' Arch. Dev. Biol. 14: 477-499.
LONGO, F. J. 1976. Ultrastructural aspects of fertilization in Spiralian eggs. Am. Zool. 16: 375-394.
LONGO, F. J. 1978. Effects of cytochalasin B on sperm-egg interactions. Dev. Biol. 67: 249-265.
LONGO, F. J., AND E. ANDERSON. 1968. The fine structure of pronuclear development and fusion in the
sea urchin, Arbacia punctulata. J. Cell Biol. 39: 339-368.
LONGO, F. J., AND E. ANDERSON. 1969. Cytological aspects of fertilization in the lamellibranch, Mytilus
edulis. II. Development of the male pronucleus and the association of the maternally and
paternally derived chromosomes. J. Exp. Zool. 172: 97-1 19.
LONGO, F. J., AND E. ANDERSON. 1970. An ultrastructural analysis of fertilization in the surf clam,
Spisula solidissima. II. Development of the male pronucleus and the association of the mater-
nally and paternally derived chromosomes. /. Uttrastruct. Res. 33: 515-527.
MEAD, A. D. 1895. Some observations on maturation and fecundation in Chaetopterus pergamentaceus,
Cuvier. J. Morphol. 10: 313-317.
METZ, C. B. 1 967. Gamete surface components and their role in fertilization. Pp. 1 63-236 in Fertilization:
Comparative Morphology, Biochemistry and Immunology, Vol. I, C. B. Metz and A. Monroy,
eds. Academic Press, New York.
MORGAN, T. H., AND A. TYLER. 1930. The point of entrance of the spermatozoon in relation to the
orientation of the embryo in eggs with spiral cleavage. Biol. Bull. 58: 59-73.
PASTEELS, J. J. 1965. Aspects structuraux de la fecondation vus au microscope electronique. Arch. Biol.
76: 463-509.
PAUL, M., AND M. GOULD-SOMERO. 1976. Evidence for a polyspermy block at the level of sperm-egg
plasma membrane fusion in Urechis caupo. J. Exp. Zool. 196: 105-1 12.
PIKO, L. 1969. Gamete structure and sperm entry in mammals. Pp. 325-403 in Fertilization: Comparative
Morphology, Biochemistry and Immunology, Vol. II, C. B. Metz and A. Monroy, eds. Academic
Press, New York.
SEGALL, G. K., AND W. J. LENNARZ. 1979. Chemical characterization of the component of the jelly
coat from sea urchin eggs responsible for induction of the acrosome reaction. Dev. Biol. 71:
33-48.
TYLER, A. 1965. The biology and chemistry of fertilization. Am. Nat. 99: 309-334.
Reference: Biol. Bull. 165: 119-138. (August, 1983)
LARVAL AND METAMORPHIC MORPHOGENESIS IN THE
NUDIBRANCH MELIBE LEONINA (MOLLUSCA: OPISTHOBRANCHIA)
LOUISE R. BICKELL1 AND STEPHEN C. KEMPF2
1 Department of Biology, University of Victoria, Victoria, British Columbia, Canada V8W 2Y2. and
2 Department of Zoology, University of Washington, Seattle, Washington 98195
ABSTRACT
Larval development and metamorphosis in the nudibranch Melibe leonina
(Gould) are described from observations of living animals and from one micrometer
histological sections. Larval morphogenesis is similar to that previously described
for other species of planktotrophic opisthobranch larvae except the rudiments of the
primary cerata and the oral hood of the post-metamorphic stage appear in the late
stage larva. Unlike many other opisthobranch larvae, M. leonina does not appear
to require a specific exogenous cue to induce metamorphosis. Metamorphosis in-
volves loss of the shell, operculum, velar ciliated cells, and certain components of
the larval stomach but the left and right digestive diverticula are retained. A rapid
expansion of the primary cerata and the oral hood occurs and is accompanied by
a large volume increase of the internal hemocoel of these structures and a flattening
and vesiculation of their epithelial cells. Several neuronal somata within the pleural
ganglia become notably larger than their neighbors during metamorphosis. At ap-
proximately 2.5 days after shell loss, M. leonina begins to employ the oral hood to
capture ciliates and small benthic nauplii. Morphogenesis in M. leonina is compared
to that of other opisthobranchs and the premetamorphic appearance of the cerata
and the lack of an exogenous metamorphic trigger are discussed.
INTRODUCTION
A number of histological and ultrastructural studies of opisthobranch morpho-
genesis during the larval, metamorphic, and juvenile stages have been published
during the last 25 years (Thompson, 1958; 1962; Tardy, 1970; Thiriot-Quievreux,
1970; 1977; Bonar and Hadfield, 1974; Bonar, 1976; Kriegstein, 1977a, b; Bickell,
1978; Bickell and Chia, 1979; Schacher et al, 1979a, b; Bickell et al, 1981; Kempf,
1982). Considered together, these works have elucidated a general pattern of de-
velopment through metamorphosis in opisthobranchs having a free-swimming larval
stage. However, in a comparative sense, the studies have also pointed out specific
morphogenetic differences in the ontogeny of larval opisthobranchs with different
taxonomic affinities. These include interspecific differences regarding the body size
of the metamorphically competent larva, the derivation of the adult dorsal epidermis,
the presence or absence of true detorsion of the gut, and the occurrence of char-
acteristics such as the right digestive diverticulum and the rudiments of various adult
structures (Bonar, 1978a review; Chia and Koss, 1978;Switzer-Dunlap, 1978; Bickell
and Chia, 1979; Kempf, 1982).
The information that has accumulated on opisthobranch morphogenesis suggests
several important objectives for future research. These are: 1) further histological
and ultrastructural investigations on larval development and metamorphosis to
Received 23 November 1982; accepted 25 May 1983.
119
120 L. R. BICKELL AND S. C. KEMPF
distinguish tissue and organ homologies between adults, thereby helping to solve
taxonomic questions within the subclass, 2) clarification of the phenomenon of
metamorphic induction and its ecological consequences, and 3) individual and com-
parative studies to examine neurodevelopment in both a morphological and be-
havioral sense.
Melibe leonina is a large dendronotid nudibranch that often reaches high pop-
ulation densities within eel grass and kelp beds along the west coast of North America
(Agersborg, 1923a; Hurst, 1968; Ajeska and Nybakken, 1976). Like certain other
members of the Dendronotacea, M. leonina exhibits a swimming behavior consisting
of rhythmical bending movements of the body. The most distinctive characteristic
of this species is the oral hood; a large, highly mobile expansion of cephalic tissue
that extends over and around the mouth and bears a double row of inner and outer
tentacles along its peripheral edge (Agersborg, 1923b) (Fig. 1). This oral hood ex-
pands and contracts through the action of muscles and pumped hemal fluids (Hurst,
1968) and is used to capture the small zooplanktonic organisms that comprise the
prey of adult M. leonina (Agersborg, 1923a; Ajeska and Nybakken, 1976). Various
organisms, notably crustaceans, are engulfed by the hood and subsequent cooper-
ative actions of the hood and oral lips forces the prey into the mouth (Hurst, 1968).
This novel method of prey capture is correlated with the absence of a radula in this
species (Agersborg, 1923b).
The following study of morphological development during the larval and meta-
morphic stages of M. leonina was undertaken to provide information on larval and
metamorphic morphogenesis for comparison with other opisthobranch species, to
examine metamorphic induction and survival strategies in a nudibranch that feeds
relatively non-specifically during the juvenile and adult stages, and to investigate the
potential of M. leonina as a system for studying opisthobranch neurodevelopment.
MATERIALS AND METHODS
Adult M. leonina and their egg masses were collected from a number of eel grass
and kelp beds located around the San Juan Archipelago (Washington, U. S. A.) and
the southern end of Vancouver Island (British Columbia, Canada).
Laboratory hatched larvae were cultured at an initial density of 2 to 3 larvae/
ml in bowls containing 100 ml of filtered (Millipore prefilter no. AP2004700) natural
sea water with 104 cells/ml of the alga Pavlova (Monochrysis) lutheri (Carolina
Biological Supply). The larvae were transferred to fresh culture medium at 1 or 2
day intervals and the antibiotics streptomycin sulfate (50 ^m/ml) and penicillin G
(60 Mm/ml) (Switzer-Dunlap and Hadfield, 1977) were added at 2 to 6 day intervals.
Cultures were maintained at a temperature of 12 to 14°C.
Young juveniles of M. leonina were fed a mixture of unidentified ciliates har-
vested from various types of decomposing animal tissue (sea urchin eggs, crushed
limpets, chunks of sea pen). This diet was supplemented with nauplii of harpacticoid
copepods.
Ten developmental stages were processed for histological examination. Larvae
were fixed at hatching, mantle fold retraction, onset of mantle fold hypertrophy,
and full development of the propodium. Metamorphic stages were fixed at the time
of velum loss, at shell loss, and at 5, 10, 24, and 48 hours after loss of the larval
shell. Primary fixation was accomplished in 2.5% glutaraldehyde and post-fixation
in 2% osmium tetroxide as described previously (Bickell and Chia, 1979). Larval
stages were anaesthetized prior to fixation by placing them in an incubation vessel
ontaining 3 ml of sea water and 7 drops of 2% procaine. After 15 min at room
MORPHOGENESIS IN A NUDIBRANCH 121
temperature, 0.5 ml of a saturated solution of chlorobutanol in sea water was added
and the incubation vessel placed on ice for 10 min. Anaesthetized animals were
placed in primary fixative for 30 min, followed by a 1 h treatment in a mixture of
equal parts primary fixative and 10% ethylenediaminotetraacetic acid (disodium
salt) to decalcify the larval shells (Bonar and Hadfield, 1974). Metamorphic stages
were anaesthetized for 5 min in 1 part saturated chlorobutanol solution and 9 parts
filtered sea water on ice and transferred to primary fixative for 1 h. All larval and
metamorphic stages were post-fixed for 1 h. Fixed animals were dehydrated in
ethanol and embedded in a plastic prepared by substituting Poly/Bed 812 (Poly-
sciences) for Epon 812 in the recipe of Luft (1961). Embedded specimens were
serially sectioned at 1 micrometer thickness and stained with Richardson's stain
(Richardson et al, 1960).
RESULTS
Structure of the larva at hatching
The veliger larvae of Melibe leonina hatch from the benthic egg mass approx-
imately 10 days after oviposition and are structurally similar to the young plank-
totrophic veligers of other opisthobranchs. At hatching, the larval body is small and
morphologically simple relative to the size and complexity that is achieved by the
end of the obligatory larval stage (compare Figs. 2 and 3).
The veliger has two major body regions: a cephalopedal mass and a visceropallial
mass. The cephalopedal mass consists of the two ciliated lobes of the velum that
effect swimming and capture of food particles, and a small pointed foot that bears
a circular operculum on its posterior face (Figs. 2, 4). A ciliary tract extends down
the midventral surface of the foot and transports rejected particles away from
the mouth.
The visceropallial mass includes a functional digestive tract, the so-called larval
kidney complex, and the larval shell with its underlying perivisceral epithelium (Figs.
2, 4). The digestive tract is composed of an esophagus, a stomach, a large left and
much smaller right digestive diverticulum, and an intestine (Figs. 2, 4, 5). The
intestine leaves the postero-dorsal region of the stomach and recurves anteriorly to
terminate at the anus located on the floor of the right mantle cavity (Fig. 5). The
larval stomach has two major divisions that Thompson (1959) termed the ventral
and dorsal stomach. The ventral stomach consists of a ciliated region that receives
the openings of the esophagus and digestive diverticula and an area lined by a gastric
shield (Fig. 4). The dorsal stomach is lined on three sides by a band of densely
packed, transversely beating cilia (Fig. 6). A sparsely ciliated groove extends down
the upper wall of the dorsal stomach (Fig. 6). The band of densely-packed cilia and
the sparsely ciliated groove are structurally similar to the style sac ciliation and
intestinal groove, respectively, of lamellibranch and some prosobranch molluscs
(Graham, 1941). The manner in which food particles are transported and digested
by the gut of opisthobranch veligers has been described previously (Thompson,
1959; Bickell et al, 1981; Kempf, 1982). The anterior deflection of the intestine is
evidence of partial torsion of the larval digestive tract.
The larval kidney complex is a cluster of distinctive yet heterogenous cells located
adjacent to the anus on the right side of the veliger (Figs. 2, 4). The function of
these cells, which degenerate at metamorphosis, is not clear. The two nephrocysts
are located on either side of the esophagus (Fig. 6). They are uniquely larval struc-
tures whose function is enigmatic but may involve storage or excretion of waste
material (Bonar, 1978a).
122
L. R. BICKELL AND S. C. KEMPF
50
EY,
SH
\
m
.
P
"» .«•'
SH
'LD
ST
FIGURE 1. Juvenile of Mel 'i be leonina at 2.5 months after metamorphosis showing the foot (f),
double row of lobate cerata (C) containing dendritic branches of the digestive diverticula, and oral hood
(OH) surrounding the mouth. The oral hood bears peripheral hood tentacles (HT) and a pair of rhi-
nophores (R) mounted on a rhinophoral process (RP). The arrowhead indicates the position of the anus.
FIGURE 2. Larva of M. leonina immediately after hatching showing the velum (VE), foot (F), and
statocyst (S) of the cephalopedal mass and the stomach (ST), right and left digestive diverticula (RD and
LD, respectively), intestine (I), larval kidney complex (LK), mantle fold (MF), and shell (SH) of the
visceropallial mass. The arrowheads indicate the tuft of long, stiff cilia at the apex of the foot.
FIGURE 3. Late stage larva of M. leonina showing the right eye (EY), propodium (P), enlarged
stomach (ST), left digestive diverticulum (LD), and the shell (SH), statocyst (S), and velum (VE). The
arrowheads indicate the rudiments of the primary cerata.
At the aperture of the shell, the associated perivisceral epithelium is termed the
mantle fold and its cells are specialized for secretion of shell material (Fig. 7). The
emainder of the mantle extends from the aperture of the shell to the cephalopedal
mass and thus demarcates a shallow mantle cavity in the newly hatched veliger.
MORPHOGENESIS IN A NUDIBRANCH
123
PV
RD
20pm
VE
D 20pm
FIGURE 4. Oblique sagittal section of a newly hatched larva of M. leonina that passes through the
foot (F), operculum (O), statocyst (S), and velum (VE) of the cephalopedal mass and the stomach (ST),
left digestive diverticulum (LD), intestine (I), larval kidney complex (LK), perivisceral epithelium (PV),
and mantle fold (MF) of the visceropallial mass. The section shows the vestibule (V) and gastric shield
(arrowheads) of the larval stomach.
FIGURE 5. Frontal section through a newly hatched larva showing the small right and large left
digestive diverticula (RD and LD, respectively) flanking the stomach <ST). Also note the torted intestine
(I) that terminates at the anus (A) in the mantle cavity on the right side. Mantle fold (MF).
FIGURE 6. Frontal section through a newly hatched larva showing the sparsely ciliated groove (CIG)
and band of dense cilia (arrowheads) within the dorsal part of the larval stomach (style sac). The cerebral
ganglia (CG) are connected over the esophagus (E) by a commissure and an apical tuft of cilia (arrow)
arises from the cephalic epithelium between the velar lobes (VE). Also note the nephrocyst (N) and
intestine (I).
FIGURE 7. Detail of the mantle fold on the right side of a newly-hatched larva. The mantle fold
(MF) is a continuation of the perivisceral epithelium (PV) and elaborates shell material (SH) at the shell
aperture. A shallow mantle cavity (MC) is demarcated by the mantle and velar (VE) epithelia.
124 L. R. BICKELL AND S. C. KEMPF
The muscle systems of the veliger of M. leonina extend through both the cephalo-
pedal and visceropallial portions of the larval body. The base of the large larval
retractor muscle is attached to the posterior end of the shell via specialized cells of
the perivisceral epithelium (Bonar, 1978b) and branches extend anteriorly into the
tissues of the foot and velum. A bundle of accessory pedal retractor muscles orig-
inates on the pedal epithelium underlying the operculum and extends over the
ventral lip of the shell to insert on the perivisceral epithelium immediately ventral
to the anus. Contraction of the larval retractor and accessory pedal retractor muscles
pulls the larval body and operculum into the shell cavity. In addition, a diffuse
system of slender visceral muscles are associated with the digestive tract and with
the mantle fold and perivisceral epithelium.
At the time of larval hatching in M. leonina, the only central ganglia that are
clearly recognizable in one micrometer sections are a pair of small cerebral ganglia;
these are connected dorsally over the esophagus by the cerebral commissure (Fig.
6). Sensory structures include a pair of statocysts within the base of the foot (Figs.
2, 4), a tuft of stiff cilia extending from the apex of the foot (Fig. 2), and an apical
organ that bears a long tuft of cilia and is located within the cephalic epidermis
overlying the cerebral commissure (Fig. 6). Bonar (1978c) described the ultrastruc-
ture of the apical organ in larvae of the nudibranch Phestilla sibogae and suggested
that it may be chemosensory.
Larval morphogenesis
The sketches in Figure 8 portray three stages of the larval development ofMelibe
leonina: the hatching stage, the eyespot — mantle retraction stage ( 16 to 20 days post-
hatching), and the stage at which the larvae become capable of settlement and
metamorphosis (30 to 48 days post-hatching). The shell increases in length from
149 /urn (S.D. 9 Mm) at hatching to 250 ^m (S.D. 3 urn) (Fig. 8). As described below,
morphogenetic events occur throughout the larval phase but tend to be concentrated
within the latter half of development.
The mantle fold of M. leonina veligers undergoes a series of major morphogenetic
changes during the larval stage. After secreting shell material during the initial por-
tion of larval development, the mantle fold epithelium detaches from the rim of the
shell and is pulled posteriorly (Figs. 8b, 9), presumably by slender muscles that
extend from the mantle fold and larval kidney complex to various sites on the
viscera and perivisceral epithelium. The cells of the retracted mantle fold epithelium
subsequently proliferate and hypertrophy and cells of unknown origin accumulate
along the hemal side of the retracted epithelium. Eventually, the mantle fold becomes
composed of closely packed columnar cells and assumes the form of two large
protuberances projecting from the postero-dorsal surface of the visceral mass (Figs.
3, 8c, 10). These structures are the rudiments of the primary cerata of the juvenile-
adult stage. Hypertrophied mantle fold cells also extend a short distance over the
latero-dorsal side of the large left digestive diverticulum and along the right side
towards the anus.
The foot of the larva is enlarged considerably by proliferation of the pedal ep-
ithelial cells (compare Figs. 4, 9, 10). During the latter half of development, foot
growth is accompanied by the differentiation of intrinsic pedal muscles and of large
pedal glands that expand within the pedal hemocoel as they become filled with
secretory product. These events ultimately result in the development of the pro-
podium, a large swelling on the proximal, ventral surface of the foot (Figs. 3, 10).
The full development of the propodium and the concurrent growth of a dense
MORPHOGENESIS IN A NUDIBRANCH
125
MF
CE
260.
240
220
SHELL 20°
LENGTH
180
160
140
120
11
ill
\
H
8
10 15 20 25
DAYS POSTHATCHING
30 35
FIGURE 8. Growth rate of the shell during the larval development of M. leonina. The points indicate
the mean length of a minimum of five larvae and the vertical bars show the standard deviation. Diagrams
of the newly hatched stage (a), the mantle retraction stage (b), and the late larval stage (c) correspond
in approximate size and age to the sites indicated by arrows on the graph. Abbreviations: C, ceras; CE,
hypertrophied cephalic epithelium; EY, eye; F, foot; I, intestine; LD, left digestive diverticulum; MF,
mantle fold; P, propodium; S, statocyst; SH, shell; ST. stomach; VE, velum.
covering of cilia over the ventral surface of the foot enables crawling behavior; a
phenomenon that provides a convenient marker for the recognition of metamorphi-
cally competent opisthobranch veligers.
The cephalic epithelium that lies immediately dorsal and lateral to the velar
lobes also exhibits proliferation and hypertrophy during the latter part of the larval
development of M. leonina (compare Figs. 9, 10). This band of columnar cephalic
epithelium will form the epidermis of the post-metamorphic oral hood.
The basic structure of the gut is preserved throughout the larval phase, although
the digestive tract grows considerably and the cells of the stomach and left digestive
diverticulum accumulate lipid deposits (Fig. 10). In late stage larvae, a vestigial
radular rudiment becomes evident as a slight evagination of the ventral wall of the
distal esophagus (Fig. 10), but neither radular teeth nor muscles differentiate in
association with this outpocketing as typically occurs during the development of
other opisthobranch larvae (Bonar, 1978a).
126
L. R. BICKELL AND S. C. KEMPF
CG
CE
EY
PLG
ST
FIGURE 9. Sagittal section through a larva of M. leonina in which the mantle fold (MF) has retracted
from the aperture of the shell. Note the eye (EY), statocyst (S), cerebral ganglion (CG), pedal ganglion
(PG), and pleural ganglion (PLG) of the larval nervous system and the thin cephalic epithelium (CE),
the elongate but low profile of the foot (F), and the gonadal rudiment (G). The section also passes through
the stomach (ST), right digestive diverticulum (RD), and the intestine (I).
FIGURE 10. Mid-sagittal section through a larva of M. leonina just prior to the onset of meta-
morphosis showing the hypertrophied cephalic epithelium (CE), a ceratal rudiment (C), the propodial
swelling (P) on the ventral surface of the foot, the gonadal rudiment (G), and the many large lipid deposits
(arrowheads) within the walls of the stomach (ST) and left digestive diverticulum (LD). A vestigial radular
rudiment (asterisk) has evaginated from the ventral wall of the esophagus (E) at the level of the cerebral
commissure (CC).
The nervous system of M. leonina becomes extensively elaborated during larval
development. By the time of mantle retraction, the pedal and pleural ganglia are
clearly recognizable, the cerebral ganglia have enlarged, and a pair of eyespots have
differentiated (Fig. 9). The pedal ganglia differentiate adjacent to the statocysts and
are connected to each other by a pedal commissure and to their respective ipsilateral
cerebral ganglion by a cerebropedal connective. Each pleural ganglion extends from
the ipsilateral cerebral ganglion via a broad cerebropleural connective. Between the
stages of mantle retraction and the onset of metamorphosis, the buccal and rhino-
phoral ganglia differentiate.
Three additional developments that occur during the larval phase of M. leonina
are the development of the pulsatile larval heart soon after mantle retraction, the
appearance of the adult kidney rudiment adjacent to the larval kidney complex and
intestine, and the enlargement of the rudiment of the gonad (Fig. 10).
Metamorphosis
Larvae of Melibe leonina do not appear to require a specific, external chemical
cue for the induction of metamorphosis. After full development of the propodium,
the larvae of this species settle onto the foot, exhibit a brief period of crawling, and
ommence metamorphosis.
The events of metamorphosis that are seen during external inspection of this
process are shown in Figures 11 through 16. The first superficial indication that
75pm
13
14
75pm
FIGURE 1 1 . Dorsal view of a larva of M. leonina that has settled onto the foot in preparation for
metamorphosis. The stomach (ST), left digestive diverticulum (LD) and eyes (EY) are visible through
the transparent larval shell (SH). The velar lobes (VE) are retracted but still intact.
FIGURE 1 2. Onset of metamorphosis. The slurry of cells indicated by the arrow are dissociated velar
cells. Inset: lateral view of a post-larva after loss of the ciliated velar cells.
FIGURE 13. Post-larva withdrawing the visceral mass from the shell (SH).
FIGURE 14. Post-larva immediately after shell loss. Note the left ceras (C).
FIGURE 15. Post-larva at 10 hours after shell loss showing the initial expansion of the cephalic
epithelium to form the oral hood (OH). Inset: lateral view of a post-larva showing a ceras (C) and the
oral hood (OH).
FIGURE 16. Post-larva at approximately 36 hours after shell loss showing the cerata (C), the dramatic
enlargement of the oral hood (OH), and the buds of the initial hood tentacles (HT). The developing
rhinophores (R) appear as two crescent-shaped ridges on the dorsal surface of the oral hood.
127
128 L. R. BICKELL AND S. C. KEMPF
metamorphosis is irreversibly underway is the dissociation of the ciliated velar cells
(Fig. 12). Many of these cells are ingested but some escape into the surrounding
environment. Dissociation of the velar cells is followed by the loss of the operculum
and the larval shell (Figs. 13, 14). The time interval between settlement and shell
loss is variable but is usually between 1 2 and 24 hours. During and following shell
loss, the cerata and the oral hood undergo a period of rapid and pronounced en-
largement (Figs. 15, 16).
Serial sections of M. leonina fixed at various stages after settlement reveal that
much structural reorganization and tissue morphogenesis occurs during metamor-
phosis. Some of these changes are illustrated schematically in Figure 17.
Beginning soon after the dissociation of the velar cells, the trunk of the larval
retractor muscle becomes detached from the posterior wall of the shell. Subsequent
contractions of this muscle appear to pull the visceral mass out of the shell in a
manner similar to that described for other nudibranchs (Bonar and Hadfield, 1974;
Bonar 1976; Bickell et ai, 1981). The larval retractor and accessory pedal retractor
muscles degenerate following shell loss.
During and immediately following shell loss, the hypertrophied mantle fold
epithelium spreads posteriorly and laterally over the stomach and digestive diver-
ticula and anteriorly toward the hypertrophied epithelium of the presumptive oral
hood (Figs. 1 7a-c). As the migrating edge of the mantle fold epithelium reaches the
gonadal rudiment, the latter tissue invaginates and the converging margin of the
spreading mantle tissue eventually fuses over the site of this internalization at the
posterior extremity of the visceral mass (Figs. 17b, 18). The fate of the perivisceral
epithelium is not apparent; it may be sloughed into the environment or overgrown
and subsequently phagocytized. Nevertheless, the lateral and posterior margins of
the mantle fold epithelium are continuous with the pedal epithelium by 5 hours
after shell loss.
The loss of the shell and operculum at metamorphosis permits a broadening of
the connection between the visceral mass and the foot (Figs. 17d-g). In M. leonina,
this process appears to be facilitated by a large increase in the volume of the hemal
space within the foot and surrounding the viscera. Inspection of living animals and
histological sections of metamorphosing M. leonina give the impression that a large
volume of external fluid has been pumped through the body wall and into the
hemolymph. A similar but much more pronounced expansion of the hemal space
accompanies the rapid enlargement of the cerata (compare Figs. 19 and 20) and the
oral hood during and following shell loss. As these structures expand, the surface
area of their covering epithelia is increased by conversion from a columnar to
squamous epithelial type. A marked increase in the vesiculation of the epithelial
cells occurs concurrently with their shape change (Figs. 19, 20, 21). Each ceras
contains longitudinal muscle fibres and tufts of stiff cilia are distributed along the
length of these structures (Figs. 20, 21).
The anus of M leonina is displaced posteriorly following shell loss, presumably
by the posterior migration of the mantle fold epithelium that surrounds the anus
and by the broadening of the connection between the foot and visceral mass (Figs.
17a, b). Subsequently, the anus moves dorsally along the postero-lateral side of the
post-larva. The latter movement appears to be effected by a dorsal shifting of mantle
epithelium resulting from the inflation of the cerata and from a dorsally directed
spread of pedal epithelium (Fig. 17c). Although the anus is moved posteriorly and
orsally, its definitive location is slightly to the right of the mid-sagittal plane of the
post-metamorphic stage. Furthermore, the proximal end of the intestine continues
to exit from the dorsal side of the posterior end of the stomach (Figs. 17g, 23). These
MORPHOGENESIS IN A NUDIBRANCH
129
17 a
FIGURE 17. Sketches of successive stages during the metamorphosis of M. leonina drawn from
reconstructions of serial, one micrometer sections. Figures 1 7a, 1 7b, and 1 7c show post-larva at the time
of velum loss, and at 5 and 24 hours after shell loss, respectively. These three diagrams illustrate the
migratory movements of the hypertrophied mantle fold and cephalic epithelia (the borders of these
epithelia are demarcated by broken lines), the invagination of the gonadal rudiment, and the postero-
dorsal displacement of the anus. The arrows indicate specific movements of the mantle fold epithelium.
Figures 17d, 17e, 17f, and 17g show post-larvae at velum loss, and at 5, 24, and 48 hours after shell loss,
respectively. These four diagrams illustrate the size and positional changes undergone by the component
organs of the digestive system during metamorphosis. Abbreviations: A, anus; AK, adult kidney rudiment;
C, ceras; CE, hypertrophied cephalic epithelium; CG, cerebral ganglion; E, esophagus; EY, eye; F, foot;
G, gonadal rudiment; HT, hood tentacle; I, intestine; LD, left digestive diverticulum; LK, larval kidney
complex; O, operculum; OH, oral hood; PG, pedal ganglion; PLG, pleural ganglion; R, rhinophore; RD,
right digestive diverticulum; RG, rhinophoral ganglion; S, statocyst; SH, shell; ST, stomach.
observations indicate that the digestive tract of M. leonina undergoes partial, but
not complete detorsion at metamorphosis.
As shown diagramatically in Figures 1 7a-c, the larval kidney complex and the
20 7
jjm
FIGURE 18. Cross section through the posterior portion of the foot (F) and visceral mass at 5 hours
after shell loss. The epithelium of the mantle fold (arrowheads) has spread over the visceral mass so as
to completely cover the large left digestive diverticulum (LD), the adult kidney (AK), and the invaginated
rudiment of the gonad (G). The section also passes through the posterior margin of the anus (A).
FIGURE 1 9. Section through a primary ceras of a late stage larva of M. leonina. Occasional unicellular
mucous glands (arrow) are embedded in the pseudostratified columnar epithelium of the ceratal (mantle
fold) epithelium (C). The interior of the structure is packed with cells, some of which contain prominent
vacuoles.
FIGURE 20. Section through the apical portion of a primary ceras at 5 hours after shell loss. The
eratal epithelium (C) is composed of highly vacuolated, squamous cells and occasional ciliated sensory
:dls (arrow). A transverse muscle fiber (MU) traverses the expanded hemocoel (H) of the ceras.
FIGURE 2 1 . Photomicrograph using Nomarski differential interference optics of a primary ceras
of M. leonina during metamorphosis showing the extension of the left digestive diverticulum (LD) into
the ceratal hemocoel and patches of stiff cilia (arrows) arising from the ceratal epidermis (C). The pho-
tomicrograph indicates that the large vacuoles within the ceratal epidermis are not fixation artifacts.
130
MORPHOGENESIS IN A NUDIBRANCH 131
rudiment of the adult kidney move posteriorly with the anus and distal end of the
intestine. The larval kidney complex subsequently degenerates within the post-larval
body, whereas the cells of the adult kidney rudiment begin to proliferate and the
internal lumen enlarges (Fig. 18).
The diagrams shown in Figures 1 7d-g illustrate the positional changes exhibited
by the organs of the larval digestive system during metamorphosis of M. leonina.
Unlike the process of gut metamorphosis in the dorid nudibranch Doridella stein-
bergae (Bickell et al, 1981), the stomach of M leonina does not undergo additional
torsional displacement at metamorphosis, nor does it shift to the mid-dorsal surface
of the large left digestive diverticulum. Although the left digestive diverticulum
continues to reside beside the stomach, a dramatic enlargement of the right digestive
diverticulum gradually displaces the stomach to a central position within the visceral
mass (Figs. 1 7g, 22). Soon after shell loss, both the left and right digestive diverticula
begin to extend into the expanded hemocoel of their respective ceras (Figs. 2 1 , 22).
The conversion of the phytoplanktotrophic larva to the carnivorous juvenile-
adult necessitates extensive changes of the tissues comprising the larval gut. The
cells of the densely ciliated band (style sac) have completely dissociated by the time
the post-larva has lost the shell and the gastric shield subsequently peels away from
its underlying cells (Fig. 24). Soon thereafter, the cells that produced the larval gastric
shield and the cells of the vestibule begin to produce the cuticular material that lines
the stomach of the post-metamorphic animal (Agersborg, 1923b) (Fig. 25).
As previously stated, the enlargement of the oral hood is accompanied by the
same types of events that occur during expansion of the primary cerata. The hy-
pertrophied cells of the cephalic epithelium convert from a columnar to squamous
shape, numerous intracellular vesicles appear, and the enclosed hemocoel becomes
inflated. The initial hood tentacles appear as 8 small papillae distributed around the
periphery of the hood (Fig. 16). In living animals, particularly after the onset of
feeding, prominent nerve tracts extend from the cerebral ganglia to a small cluster
of cells underlying each of the hood tentacles (Figs. 26, 27). Transmission electron
microscopy has confirmed that these tracts are nerves rather than muscle bundles
(Bickell, unpublished observations). The epithelium of each hood tentacle gives rise
to several tufts of stiff cilia (Fig. 27) and additional ciliary tufts appear on the ventral
surface of the hood during metamorphosis.
Differentiation of muscles within the periphery of the oral hood enables it to
close (compare Figs. 26 and 28) if a tactile stimulus is applied to the ventral surface
of the hood. Melibe leonina is able to capture and ingest ciliates using the oral hood
and oral lips at approximately 2.5 days after shell loss.
Several morphological changes in the nervous system of M. leonina can be
resolved in one micrometer sections of metamorphic stages. The parapedal com-
missure can be resolved at 10 hours after shell loss as a slender tract just posterior
to the pedal commissure. The cerebrobuccal connectives also become distinguishable
at this time and the pleuropedal connectives become distinct from the cerebropedal
connectives. By 24 hours after shell loss, a lengthening of the pedal and parapedal
commissures and of the cerebrobuccal connectives has occurred. The neuropile
region of all the central ganglia enlarges during metamorphosis.
During the period of velum dissociation, a neuronal soma located medio-dorsally
within the right pleural ganglion, at the level of the pleuropedal connective, becomes
notably larger (10 nm diameter) than the surrounding ganglionic cell bodies (3 to
5 ^m diameter) (Figs. 29, 30). By virtue of its size, position, and large nucleus
containing a prominent nucleolus, this neuron can be re-identified in all subsequent
metamorphic stages. Several other neuronal somata within the right and left pleural
132
L. R. BICKELL AND S. C. KEMPF
FIGURE 22. Cross section of M. leonina at 5 hours after shell loss that passes through the left and
right digestive diverticula (LD and RD, respectively) where they enter the stomach (ST). Both diverticula
are beginning to project into their respective ceras (C).
FIGURE 23. Cross section of M. leonina at 5 hours after shell loss showing the emergence of the
intestine (I) from the dorsal side of the posterior end of the stomach (ST). The left digestive diverticulum
(LD) and degenerating larval kidney complex (LK) are also shown.
FIGURE 24. High magnification of the stomach area of Figure 22. The larval gastric shield (large
arrows), which can be recognized by the presence of small hyaline rods (small arrowheads) embedded
the shield matrix, is sloughing into the lumen of the stomach (ST).
"FIGURE 25. High magnification of the wall of the stomach (ST) at 24 hours after shell loss. The
rowhet.cls indicate the cuticle that lines the inner side of the gastric epithelium in post-metamorphic
animals.
MORPHOGENESIS IN A NUDIBRANCH 133
ganglia become notably larger than their neighbors during the period of metamor-
phosis.
At the time of shell loss, the rhinophoral ganglia are closely apposed to the
antero-dorsal surface of their respective cerebral ganglion and the cells of the cephalic
epithelium that directly overlie each of the rhinophoral ganglia are taller and more
lightly staining than the surrounding epithelial cells. These patches of thickened
epithelium, the presumptive rhinophores, and their associated rhinophoral ganglia
are carried anteriorly as the cephalic epithelium expands to form the oral hood. As
each rhinophoral ganglion moves away from its ipsilateral cerebral ganglion, the
two remain connected by a thick rhinophoral nerve (Fig. 31). The epithelial cells
of the presumptive rhinophores proliferate so as to form prominent bulges on the
dorsal surface of the enlarging oral hood. Cells bearing tufts of stiff cilia differentiate
within the rhinophoral epithelium by 5 hours after shell loss and patches of motile
cilia appear during the following 2 days.
DISCUSSION
Although the developmental events that occur during the larval stage of opis-
thobranchs are similar in kind and sequence, various differences often occur between
species. In some cases, these differences can be interpreted as ontogenic anticipation
of unique structural features of the post-metamorphic stage or special features to
facilitate the success of settlement and metamorphosis or the survival of young
juveniles in the adult habitat (Chia and Koss, 1978; 1982; Switzer-Dunlap, 1978;
Bickell and Chia, 1979). This phenomenon is illustrated by three unusual features
of the late stage larva of Melibe leonina. These are: the absence of radular teeth, the
appearance of presumptive oral hood tissue, and the precocious development of the
primary cerata.
The almost complete omission of the radula — odontophore complex from the
sequence of developmental events in M. leonina eliminates an unnecessary energy
expenditure as this structure has no larval function and is not required for food
capture or ingestion in the adult. However, the small size of newly metamorphosed
nudibranchs may preclude feeding on the same type of prey or in the same manner
as the adults of their species. At least one species of nudibranch utilizes its radula
to graze on an organic surface film until sufficiently large to exploit the preferred
prey of the adult stage (Perron and Turner, 1977). Juveniles of M. leonina cannot
employ this type of interim feeding due to the lack of a radula. Instead, metamor-
phosis in M. leonina involves a rapid differentiation of the oral hood, thereby per-
mitting young juveniles to capture small prey in a manner similar to that employed
by the adult. Selective pressures acting to promote the rapid formation of the oral
hood during metamorphosis may have resulted in the preliminary development of
this structure during the final part of the larval stage. Furthermore, non-specific
metamorphic induction and the active nature of the juvenile prey (e.g., ciliates)
confront newly metamorphosed M. leonina with the problems inherent in feeding
on organisms having a patchy distribution in time and space. This challenge may
have resulted in selection for the greater activity and tactile — positional awareness
observed in newly metamorphosed juveniles of M leonina. In response to ciliates
colliding with various parts of their body, the juveniles can rapidly turn the anterior
body, expand the oral hood, and make a directed and effective capture of the or-
ganism. The active prey searching behavior of young M. leonina juveniles and their
high degree of responsiveness to tactile environmental stimuli contrasts with the
behavior of recently metamorphosed juveniles of other opisthobranchs, which tend
134
L. R. BICKELL AND S. C. KEMPF
OH
PLG
PPC
FIGURE 26. Ventral view of M. leonina at 5 days after shell loss showing the extended oral hood
(OH). Nerve tracts (arrowheads) extend from the cerebral ganglia (CG) to the buds of the peripheral hood
tentacles (HT). A ceras (C) is also visible.
FIGURE 27. Photomicrograph using Nomarski differential interference optics of a portion of the
i.ral hood margin showing the terminal region of a cerebral nerve (CN) extending to a peripheral hood
ganglion (HG) that underlies a hood tentacle bud. A tuft of stiff cilia (arrow) extends from the epithelium
of the tentacle bud.
MORPHOGENESIS IN A NUDIBRANCH 135
to be sluggish grazers on the prey organism that induced the metamorphosis of the
preceding larval stage (Thompson, 1958; 1962; 1964; Tardy, 1970; Bonar and Had-
field, 1974; Kempf and Willows, 1977; Switzer-Dunlap and Hadfield, 1977; Chia
and Koss, 1978; Bickell, 1978; and others).
The hypertrophy of the larval cephalic epidermis has not been noted in pre-
metamorphic veligers of dorid nudibranchs, which tend to lack a large oral veil over
the mouth, but is shown in drawings by Thompson (1962) of premetamorphic
veligers of Tritonia hombergi (Dendronotacea) and by Tardy (1970) of Aeolidiella
alderi (Aeolidacea). The juveniles and adults of both these species have a prominent
oral veil that is derived from this hypertrophied cephalic tissue. These observations
confirm that the oral hood of M. leonina and the oral veil of other nudibranchs are
homologous structures.
The appearance of ceratal rudiments in the larval stage of nudibranchs has not
been reported previously, although many aeolids and dendronotids have been reared
in the laboratory. It has been suggested that the thin-walled cerata of nudibranchs
provide an increased surface area for gas exchange with the environment (see Mor-
ton, 1958). This hypothesis is strengthened by the fact that the metabolically active
digestive diverticula often extend into the cerata. Ajeska and Nybakken (1976) found
that oxygen consumption/gm body weight was an inverse function of animal size
in M. leonina. They suggested that the higher metabolic rate of young juveniles
reflects the fact that they must actively seek-out their benthic prey, whereas larger
animals simply extend their hood into the surrounding waters to intercept passing
zooplankton. Of the 10 species of newly metamorphosed opisthobranchs that the
present authors have observed, M. leonina young juveniles are most active. Together,
these observations suggest that the development of ceratal rudiments during the
larval stage of M. leonina and their rapid expansion and invasion by the digestive
diverticula during metamorphosis may be necessary to sustain a high oxygen demand
resulting from an active juvenile life style.
The present study of larval development and metamorphosis of M. leonina
provides the second histological description of gut metamorphosis in a plankto-
trophic nudibranch veliger. As in the dorid nudibranch, Doridella steinbergae (Bick-
ell el al., 1981), the morphologically complex stomach of M leonina veligers is
transformed to the post-metamorphic stomach by dissociation of the cells com-
prising the ciliated band (style sac) and loss of the gastric shield. In Doridella stein-
bergae, Bickell et al. ( 1 98 1 ) speculated that the gastric shield was lost by dissociation
of the underlying cells. Observations made in the present study indicate that the
gastric shield is simply sloughed from the gut wall; the underlying cells are retained
and subsequently secrete a portion of the cuticle that lines the stomach of the post-
metamorphic stage.
FIGURE 28. Same animal as that in Figure 26 showing closure of the oral hood by contraction of
muscles extending along the hood periphery.
FIGURE 29. Slightly oblique cross section through the esophageal region (E) of M. leonina at 24
hours after shell loss. Note the left statocyst (S), buccal ganglia (BG), pleural ganglia (PLG), and pedal
ganglia (PG). The arrowhead indicates the distinctive neuronal soma (see Fig. 30) that is situated dorso-
medially within the right pleural ganglion at the level of the pleuro-pedal connective (PPC).
FIGURE 30. Enlargement of the pleural ganglion (PLG) from Figure 29 indicating the large neuronal
soma (arrow) containing a prominent nucleolus.
FIGURE 31. Slightly oblique cross section through the base of the oral hood (OH) and the anterior
end of the foot (F) at 24 hours after shell loss showing the developing rhinophore (R) and its underlying
rhinophoral ganglion (RG) on the left side. A rhinophoral nerve (arrowhead) extends between the rhin-
ophoral ganglion and the cerebral ganglion (CG). The section also passes through the left eyespot (EY).
136 L. R. BICKELL AND S. C. KEMPF
The stomach of M. leonina does not undergo additional torsional displacement
during metamorphosis, as observed in D. steinbergae (Bickell et al, 1981), nor does
it exhibit complete detorsion, as described for the aeolid nudibranch Phestilla si-
bogae (Bonar and Hadfield, 1974). In M. leonina, the dorso-lateral position of the
anus and the fact that the intestine emerges from the dorsal aspect of the stomach
are post-metamorphic vestiges of the torted larval digestive tract.
As is typical of most opisthobranch veligers, those of M. leonina possess a large
left and a much smaller right digestive diverticulum. The few histological investi-
gations that have considered gut metamorphosis in opisthobranch larvae indicate
that the right diverticulum 'disappears' at or soon after metamorphosis in the dorids
Adalaria proxima (Thompson, 1 958) and Doridella steinbergae (Bickell et al., 1981).
Thompson (1962) reported the persistence of this organ for a period of time after
metamorphosis in the dendronotid Tritonia hombergi but noted that it eventually
became impossible to differentiate the right diverticulum from the left. Nevertheless,
on the basis of adult morphology, the right diverticulum appears to persist in the
Dendronotacea, Arminacea, and Aeolidacea (see Hyman, 1967, p. 443). Our study
of morphogenesis in larvae and juveniles of M. leonina shows that both the right
and left diverticula are retained during metamorphosis. Each diverticulum prolif-
erates into its ipsilateral ceras and opens separately into the stomach. This feature
persists into the adult stage, although the main duct of the left digestive diverticulum,
but not the right, eventually branches at its point of exit from the stomach ( Agers-
borg, 1923b).
Larval settlement and metamorphosis has been observed in three species of
dendronotid nudibranchs. Tritonia hombergi is typical of many opisthobranchs (see
Hadfield, 1978) in that metamorphosis will occur only in the presence of its post-
metamorphic prey, Alcyonium digitatum (Thompson, 1962). Metamorphosis of the
larvae of Tritonia diomedia is promoted by the preferred pennatulacean prey of the
adults, but metamorphosis will also occur without this external inducer. Kempf and
Willows (1977) suggested that the absence of absolute dependence on an external
metamorphic trigger in T. diomedia may relate to the fact that adults will also feed
on several other pennatulaceans. In M. leonina, the presence of a substratum appears
to be the only requirement for the onset of larval settlement and metamorphosis.
The prey of young juveniles (which was benthic ciliates and crustacean nauplii in
this study and benthic crustaceans and bivalve spat in the field study of Ajeska and
Nybakken, 1976) probably occurs ubiquitously on marine substrates, and the zoo-
planktonic prey of larger juveniles and adults is continuously transported through
coastal waters. Therefore, the need for specific metamorphic induction to ensure
a benthic food source (Thompson, 1964) seems unnecessary in this species.
Despite the apparent absence of environmental induction of metamorphosis,
populations of M. leonina are consistently found in eel grass and kelp beds located
in protected waters (Agersborg, 1923a; Hurst, 1968; Ajeska and Nybakken, 1976).
Pelagic individuals of this species, which include the larvae and post-metamorphic
animals that have become dislodged from a surface (Hurst, 1968), may become
passively concentrated in areas of reduced water flow. The buoyant fronds of eel
grass and certain large kelp species that are typical of these locations might be
expected to promote the survival of M. leonina because the plants provide a sub-
merged, tidal adjusting attachment substratum (M. leonina cannot withstand at-
mospheric exposure) that is suspended within the upper levels of the water column
where the flow of plankton-carrying currents is greatest.
Melibe leonina offers considerable potential for studies on opisthobranch neu-
rodevelopment. Unlike many other species, reproductive adults and egg masses can
MORPHOGENESIS IN A NUDIBRANCH 137
be collected throughout the year (Hurst, 1967). Furthermore, the successful rearing
of juveniles on ciliates followed by commercially available Anemia nauplii simplifies
the problem of obtaining a continuous supply of food for the post-metamorphic
stage. The central ganglia of M. leonina include many large, identifiable neurons
(Hurst, 1968) and the present study has shown that several neuronal cell bodies
become morphologically distinct during metamorphosis. Finally, the rapid forma-
tion of cerebral nerve tracts innervating the oral hood and their visibility through
the transparent epithelium of this structure may allow investigation of axonal guid-
ance during neurodifferentiation.
ACKNOWLEDGMENTS
The authors thank R. O. Brinkhurst for providing some of the microscope
equipment used during the preparation of this manuscript. G. O. Mackie offered
comments on the manuscript and provided NSERC grant support for the research.
LITERATURE CITED
AGERSBORG, H. P. VON WOLD KJERSCHOW. 1923a. A critique on Professor Harold Heath's Chioraera
dalli, with special reference to the use of the foot in the nudibranchiate mollusk, Melibe leonina
Gould. Nautilus 36: 86-96.
AGERSBORG, H. P. VON WOLD KJERSCHOW. 1923b. The morphology of the nudibranchiate mollusk
Melibe (syn. Chioraera) leonina (Gould). Q. J. Microsc. Sci. 67: 507-592.
AJESKA, R. A., AND J. W. NYBAK.K.EN. 1976. Contributions to the biology of Melibe leonina (Gould,
1852) (Mollusca: Opisthobranchia). Veliger 19: 19-26.
BICK.ELL, L. R. 1978. Larval Development, Metamorphosis, and Juvenile Feeding of Doridella steinbergae
(Lance) (Opisthobranchia: Nudibranchia). M.Sc. thesis. University of Alberta. 226 pp.
BICK.ELL, L. R., AND F. S. CHIA. 1979. Organogenesis and histogenesis in the planktotrophic veliger of
Doridella steinbergae (Opisthobranchia: Nudibranchia). Mar. Biol. 52: 291-313.
BICKELL, L. R., F. S. CHIA, AND B. J. CRAWFORD. 1981. Morphogenesis of the digestive system during
metamorphosis of the nudibranch Doridella steinbergae (Gastropoda): conversion from phy-
toplanktivore to carnivore. Mar. Biol. 62: 1-16.
BONAR, D. B. 1976. Molluscan metamorphosis: a study in tissue transformation. Am. Zool. 16: 573-
591.
BONAR, D. B. 1978a. Morphogenesis at metamorphosis in opisthobranch molluscs. Pp. 177-196 in
Settlement and Metamorphosis of Marine Invertebrate Larvae, F. S. Chia and M. E. Rice, eds.
Elsevier/North-Holland, New York.
BONAR, D. B. 1978b. Fine structure of muscle insertions on the larval shell and operculum of the
nudibranch Phestilla sibogae (Mollusca: Gastropoda) before and during metamorphosis. Tissue
Cell 10: 143-152.
BONAR, D. B. 1978c. Ultrastructure of the cephalic sensory organ in larvae of the gastropod Phestilla
sibogae (Aeolidacea, Nudibranchia). Tissue Cell 10: 153-165.
BONAR, D. B., AND M. G. HADFIELD. 1974. Metamorphosis of the marine gastropod Phestilla sibogae
Bergh (Nudibranchia: Aeolidacea). I. Light and electron microscopic analysis of larval and
metamorphic stages. /. Exp. Mar. Biol. Ecol. 16: 227-255.
CHIA, F. S., AND R. Koss. 1978. Development and metamorphosis of the planktotrophic larvae of
Rostanga pulchra (Mollusca: Nudibranchia). Mar. Biol. 46: 109-119.
CHIA, F. S., AND R. Koss. 1982. Fine structure of the larval rhinophores of the nudibranch, Rostanga
pulchra, with emphasis on the sensory receptor cells. Cell Tissue Res. 225: 235-248.
GRAHAM, A. 1941. The molluscan stomach. Trans. R. Soc. Edinb. 61: 737-778.
HADFIELD, M.G. 1978. Metamorphosis in marine molluscan larvae: an analysis of stimulus and response.
Pp. 165-175 in Settlement and Metamorphosis of Marine Invertebrate Larvae, F. S. Chia and
M. E. Rice, eds. Elsevier/North-Holland, New York.
HURST, A. 1967. The egg masses and veligers of thirty north-east Pacific opisthobranchs. Veliger 9: 255-
288.
HURST, A. 1968. The feeding mechanism and behavior of the opisthobranch Melibe leonina. Symp.
Zool. Soc. Land. 22: 151-166.
HYMAN, L. 1967. The Invertebrates: Mollusca I, Vol. VI. McGraw-Hill, New York. 792 pp.
KEMPF, S. C. 1982. Acquisition, Storage and Utilization of Nutrients by the Embryos and Larvae of
Opisthobranch Molluscs. Ph.D. thesis. University of Hawaii. 287 pp.
138 L. R. BICKELL AND S. C. KEMPF
KEMPF, S. C., AND A. O. D. WILLOWS. 1977. Laboratory culture of the nudibranch Tritonia diomedia
Bergh (Tritonidae: Opisthobranchia) and some aspects of its behavioral development. J. Exp.
Mar. Biol. Ecol. 30: 261-276.
KRIEGSTEIN, A. R. 1977a. Development of the nervous system of Aplysia californica. Proc. Nat. Acad.
Sci. 74: 375-378.
KRIEGSTEIN, A. R. 1977b. Stages in the post-hatching development ofAp/vsia californica. J. Exp. Zool.
199: 275-288.
LUFT, J. H. 1961. Improvements in epoxy resin embedding methods. J. Biophys. Biochem. Cytol. 9:
409-414.
MORTON, J. E. 1958. Molluscs: An Introduction to Their Form and Function. Harper and Bros., New
York. 232 pp.
PERRON, F. E., AND R. D. TURNER. 1977. Development, metamorphosis, and natural history of the
nudibranch Doridella obscnra Verrill (Corambidae: Opisthobranchia). J. Exp. Mar. Biol. Ecol.
27: 171-185.
RICHARDSON, K. C., L. JARRETT, AND E. H. FINKE. 1960. Embedding in epoxy resins for ultrathin
sectioning in electron microscopy. Stain Technol. 35: 313-323.
SCHACHER, S., E. R. KANDEL, AND R. WOOLLEY. 1979a. Development of neurons in the abdominal
ganglion of Aplysia californica. I. Axosomatic synaptic contacts. Dev. Biol. 71: 163-175.
SCHACHER, S., E. R. KANDEL, AND R. WOOLLEY. 1979b. Development of neurons in the abdominal
ganglion of Aplysia californica. II. Nonneural support cells. Dev. Biol. 71: 176-190.
SWITZER-DUNLAP, M. 1978. Larval biology and metamorphosis of aplysiid gastropods. Pp. 197-206 in
Settlement and Metamorphosis of Marine Invertebrate Larvae, F. S. Chia and M. E. Rice, eds.
Elsevier/North-Holland, New York.
SWITZER-DUNLAP, M., AND M. G. HADFIELD. 1977. Observations on development, larval growth and
metamorphosis of four species of Aplysiidae (Gastropoda, Opisthobranchia) in laboratory cul-
ture. J. Exp. Mar. Biol. Ecol. 29: 245-26 1 .
TARDY, J. 1970. Contribution a 1'etude des metamorphoses chez les nudibranches. Ann. Sci. Nat. Zool.
12: 299-370.
THIRIOT-QUIEVREUX, C. 1970. Transformations histologiques lors de la metamorphose chez Cymbulia
peroni de Blainville (Mollusca, Opisthobranchia). Z. Morphol. Tiere 67: 106-1 17.
THIRIOT-QUIEVREUX, C. 1977. Veligere planctotrophe du doridien Aegires punctilucens (D,Orbigny)
(Mollusca: Nudibranchia: Notodorididae): description et metamorphose. J. Exp. Mar. Biol.
Ecol. 26: 177-196.
THOMPSON, T. E. 1958. The natural history, embryology, larval biology, and post-larval development
of Adalaria proximo (Alder and Hancock) (Gastropoda, Opisthobranchia). Phil. Trans. R. Soc.
Lond. B242: 1-57.
THOMPSON, T. E. 1959. Feeding in nudibranch larvae. J. Mar. Biol. Assoc. U. K. 38: 239-248.
THOMPSON, T. E. 1962. Studies on the ontogeny of Tritonia hombergi Cuvier (Gastropoda, Opisthob-
ranchia). Phil. Trans. R. Soc. Lond. B245: 171-281.
THOMPSON, T. E. 1964. Grazing and the life cycles of British nudibranchs. Pp. 275-297 in Grazing in
Terrestrial and Marine Environments, D. J. Crisp, ed. Blackwell Scientific Publ., Oxford.
H
Reference: Biol. Bull. 165: 139-153. (August, 1983)
VERTICAL MIGRATION RHYTHMS OF NEWLY HATCHED LARVAE
OF THE ESTUARINE CRAB, RHITHROPANOPEUS HARRISII
THOMAS W. CRONIN1 AND RICHARD B. FORWARD, JR.
Duke University Marine Laboratory, Beaufort, NC 28516 and Department
of Zoology, Duke University, Durham, NC 27706
ABSTRACT
Zoea larvae of the estuarine crab Rhithropanopeus harrisii were maintained in
constant conditions in the laboratory, and their vertical migrations were followed
for two or more days. Larvae which hatched in the laboratory, but which underwent
embryonic development in an estuary having semidiurnal tides, often expressed
circatidal rhythms in vertical migration. However, first-stage zoea larvae collected
by plankton net in the same estuary had circatidal vertical migration rhythms of
much greater amplitude and with a constant phase with respect to the natural tidal
cycle. Laboratory-hatched larvae of crabs from an estuary with aperiodic tides had
more variable vertical migrations, and field-caught larvae from the same habitat
never expressed clear migration rhythms. When reared to the third zoeal stage in
the laboratory under a diel light:dark cycle, larvae from both estuaries usually mi-
grated arhythmically under constant conditions. Vertical migration rhythms of lar-
vae of this species appear to be strongly predisposed to entrainment by natural tidal
cues. Such migrations probably contribute to estuarine retention of the developing
larvae.
INTRODUCTION
Estuaries are characterized by rapidly changing environmental conditions which
often stress the organisms inhabiting them. In spite of this, larvae of the estuarine
crab Rhithropanopeus harrisii are capable of remaining within an estuary near parent
crab populations throughout development (Sandifer, 1973, 1975; Cronin, 1982).
Retention is assisted by means of vertical migrations between the landward-flowing
and seaward-flowing layers of the estuary; these migrations are partly under endog-
enous control. Cronin and Forward (1979) showed that R. harrisii larvae from an
estuary with strong semidiurnal tides continued tidal vertical migrations in con-
stant laboratory conditions, whereas laboratory-reared larvae from an estuary with
irregular tides expressed a weak circadian rhythm. The tidal vertical mi-
gration was probably due to a circatidal rhythm in activity (Forward and Cronin,
1980).
R. harrisii passes through 4 zoeal stages before molting to the postlarva (Con-
nolly, 1925). In our previous study of rhythmicity in vertical migration, experiments
began with the stage III zoea, which were collected directly from the plankton and
thus had had several days in which to become entrained to the estuarine tidal cycle.
Yet the first-stage larvae of this species also migrate vertically under natural tidal
conditions (Cronin, 1982). We therefore initiated a series of experiments to learn
Received 24 January 1983; accepted 25 May 1983.
' Current address: Department of Biological Sciences, University of Maryland, Baltimore County,
Catonsville, MD 21228.
139
140 T. W. CRONIN AND R. B. FORWARD. JR.
whether newly hatched larvae also possess vertical migration rhythms. Migrations
of larvae which hatched in the laboratory from crabs collected just before larval
release occurred were compared to those of first stage larvae collected in the field.
To understand better the origins of rhythmic behavior, we compared larvae from
a population of crabs living in an estuary having semidiurnal tides with those from
an estuary with irregular tides. Finally, the behavior of these newly hatched larvae
was compared with that of later-stage larvae entrained under laboratory or field
conditions. We found that newly hatched larvae could perform rhythmic migrations,
but that the particular pattern of the migration varied with larval age and habitat.
MATERIALS AND METHODS
Preparation of larvae
Larvae from the estuarine crab Rhithropanopeus harrisii (Gould) were used
exclusively in these experiments. Larvae were obtained from two populations of
crabs in North Carolina, one population occurring in the Newport River estuary
and the other in the Neuse River estuary. Conditions in these two estuarine systems
differ strikingly (Forward et ai, 1982). The Newport River is strongly tidal with
equal semidiurnal tides. It has extremely dark-colored water in its upper reaches
(Cronin, 1982), and light intensities on the bottom, where R. harrisii adults live,
are below the crabs' threshold (Forward et ai, 1982). The Neuse River has aperiodic
tides (Roelofs and Bumpus, 1953) and contains rather transparent water.
Field-caught larvae were taken at each site by towing plankton nets within 1 m
of the water's surface. These larvae were taken directly to the laboratory where the
desired stage(s) of zoea larvae were identified using a dissecting microscope and
placed into newly prepared water of identical salinity to that in which they were
collected. Water of desired salinity was made by mixing filtered sea water with
distilled water. In the Newport River, larvae were collected at high tide during the
daytime by towing the sampling net from an outboard motor boat. Neuse River
larvae were caught in plankton nets as we walked in shallow water at night. Tem-
peratures at both collecting sites were between 25° and 30°C.
Laboratory-hatched larvae were obtained from recently collected crabs. Crabs
were maintained in salinities similar to those of the collection site, in constant
temperature (28 ± 1 °C) and low-level light. These conditions are identical to those
previously used for crabs originally entrained in natural environments (Forward et
al., 1982). In order that prior possible entrainment of larval rhythms be altered as
little as possible, only hatches occurring within 4 days of collection were used. In
the majority of cases (11 of 1 5 laboratory hatches), larvae hatched within 24 hours
of collection of the mother crab. In a few experiments, larvae were reared to the
third stage zoea before experiments were begun. These larvae were changed daily
to clean water of appropriate salinity and fed newly hatched Anemia salina nauplii
until the day of the molt to stage III.
In all cases, larvae were placed in clean water of identical salinity to that of
previous field or laboratory exposure and allowed to feed on newly hatched Artemia
nauplii for at least l/i hour before experiments began. Larvae were then transferred
once more to clean water, and the desired number of individuals (usually 100, but
occasionally fewer for field-caught larvae; see Table I) were added to a vertical lucite
3lumn. The dimensions of the enclosed water column were 190 cm tall X 5.0 cm
5 cm.
MIGRATION RHYTHMS OF LARVAL CRABS
141
TABLE I
Basic information about vertical migration experiments
Experiment
Number Starting date
Larval source
estuary
Length of
Hatch Larval stage experiment
location at start (h)
Initial
number
of larvae
791
Aug 13, 1979
Newport
Laboratory
83
100
792
Aug 17, 1979
Newport
Laboratory
70
100
793
Aug 29, 1979
Newport
Field
77
100
794
Sept 7, 1979
Neuse
Laboratory
49
100
801
Aug 7, 1980
Neuse
Laboratory
52
100
802
Aug 13, 1980
Newport
Field
42
100
803
Aug 15, 1980
Neuse
Laboratory
69
100
804
Aug 22, 1980
Neuse
Laboratory
67
100
805
Aug 27, 1980
Newport
Laboratory
54
100
811
Aug 11, 1981
Newport
Laboratory
77
100
812
Aug 14, 1981
Neuse
Laboratory
71
100
813
Aug 19, 1981
Newport
Laboratory
62
100
814
Aug 26, 1981
Neuse
Field
54
100
815
Aug 31, 1981
Neuse
Field
45
100
816
Sept 3, 1981
Neuse
Field
46
100
817
Sept 10, 1981
Neuse
Field
51
100
821
July 7, 1982
Neuse
Field IV 51
37
822
July 10, 1982
Neuse
Field IV 56
24
823
July 15, 1982
Neuse
Field III & IV 77
80
824
July 18, 1982
Neuse
Laboratory III 83
100
825
July 22, 1982
Neuse
Field III & IV 76
100
826
Aug 10, 1982
Neuse
Laboratory III 105
100
827
July 26, 1982
Newport
Laboratory III 97
100
828
Aug 4, 1982
Newport
Laboratory III 108
100
829
Aug 31, 1982
Newport
Laboratory III 102
100
Monitoring of larval vertical distributions
Once placed in the experimental column, larvae were maintained in constant
darkness and temperature (experiments 791 and 792, 21 ± 1°C; all others, 25
± 1 °C). Because larvae were not fed again, the total length of each experiment was
limited by the ability of each larval population to resist starvation. Experiments were
usually permitted to run until larval mortality and deterioration left fewer than 20%
of the original number of larvae in the water column; occasionally, experiments
were terminated before this point if greater than 50 h of data had been obtained.
Distributions of larvae were determined by the method of Cronin and Forward
(1979), the only difference being that the experimental column was backlit with
diffuse infrared light passing through a Kodak Wratten 87 filter (50% transmission
wavelength, 790 nm). Briefly, a closed-circuit TV camera equipped with a silicon-
target vidicon vertically scanned the entire height of the lucite column once each
half hour, and the camera's output was stored on videotape for later analysis. The
infrared backlight was switched on only during the 2 min scan time; the camera
also passed a clock on each scan to record the time of day.
Videotapes were analyzed during replay by counting the number of larvae in
each 10 cm segment of the water column. Larvae actually on the bottom were not
counted, since our experience has been that over 80 percent of well-fed larvae remain
142 T. W. CRONIN AND R. B. FORWARD, JR.
in the water column. The two counts obtained each hour for each segment were
summed and an hourly depth-weighted mean calculated. All analyses were per-
formed on the time series of mean depth values.
Data analysis
These experiments were designed to reveal rhythmic behavior in crab larvae
following entrainment in specific embryonic or early larval environments. The re-
sulting data challenged straightforward time series analysis for several reasons (see
the Figures). Records were short in length, usually less than 72 h. Data represented
output from groups of larvae whose individuals did not necessarily have highly
synchronous behavior. There commonly were long-term vertical trends. Larval ver-
tical migrations were often of low amplitude and included considerable noise. Of
benefit to data analysis, however, was the fact that we restricted our interest to
rhythms of circatidal or circadian periods, since these were the periods observed in
the earlier study (Cronin and Forward, 1979). Following consultation with J. Har-
tigan, Yale University Department of Statistics, we decided to analyze each data set
using three statistical techniques and a visual evaluation of the data. Each statistical
method of analysis approached the data from an independent point of view, and
it was common for one method to indicate rhythmicity where the others did not.
All analytical methods required complete time series. In the two cases in which
some data were missing due to equipment failure, missing values were replaced with
values calculated by linear interpolation between adjacent measured values. Because
the use of the statistical analyses was useful for extracting information about larval
rhythms from the raw data, they are described in some detail below.
Fisher's Periodogram Test. To perform a rough linear detrend, a regression line
was fitted to the raw data and subtracted from all points. Detrended data were
subjected to Fisher's periodogram test (Fuller, 1976). This test only applies to har-
monics of the total time series; periods of significant cycles can be compared to the
tidal or diel period, but because the tested cycles are harmonics of the series they
do not always fall very near the precise environmental period. Fisher's test has the
further limitation of examining only the frequency of largest amplitude in a time
series.
Multiple Autoregression. Each data set was multiply regressed on itself at three
lags. For the circatidal rhythm analysis, the 3 independent variables were the mea-
sured mean depth values at 1 h, 2 h, and 12 h or 13 h prior to the value at a given
hour. (Both 12 h and 13 h were tested in order to bracket the average natural tidal
period, which was 12.4 h in the Newport River.) For circadian rhythm analysis, the
lags were 1 h, 2 h, and 24 h. Periodicity in the data was taken to be significant if
the regression coefficient of the 3rd independent variable (lag of 12, 13, or 24 h)
was significantly greater than 0. Multiple autoregression was relatively inefficient in
finding rhythms in these experiments.
Analysis of Variance (ANOVA). ANOVA is not a traditional statistical tool for
time series analysis. We were able to apply it because we restricted our analytical
effort to periods approximating the natural diel and tidal cycles. Each data set was
broken into a whole number of segments; for tidal analysis these segments were 1 2
h or 1 3 h in length, while for diel analysis they were 24 h long. To minimize the
effects of long-term trends, the mean value in each segment was calculated and
removed from all values in that segment. Next, a 1-way ANOVA was performed
on the 12, 13, or 24 hourly values, with the number of replicates in each hour being
the total number of segments in the data set. This analysis therefore tested whether
MIGRATION RHYTHMS OF LARVAL CRABS 143
there was significant hour-to-hour variation in the data within blocks 12 h, 13 h,
or 24 h in length. A significant result could occur when the averaged segments
contained a monotonic trend, when high-frequency noise was present, when repeated
smaller cycles fell within the total segment length, or when cycles of the total segment
length occured. Since only the last case was of interest, the sequence of hourly means
was examined for rejection of misleading significance due to trends, noise, or internal
cycles. ANOVA proved to be a powerful method for determining rhythms in our
data, probably because the method of removing the mean from each section of data
was an effective way to minimize the contributions of irregular long-term variations.
Subjective Evaluation oj Data. Because of the nature of the process under study,
it is probably at present impossible to obtain data which are completely amenable
to statistical treatment. We have relied on the techniques described above to provide
an objective base for drawing conclusions, but occasionally we also turned to a
subjective evaluation of the data in hopes of increasing our understanding of larval
rhythmic behavior. We encourage readers to inspect thoroughly the data we present
so that they can decide whether to accept our conclusions.
RESULTS
Essential information about each experiment is given in Table I. When classed
according to larval source (Newport vs. Neuse River), hatch location (laboratory vs.
field), and larval stage at the beginning of the experiment (zoea I vs. zoea III or IV),
a total of 7 types of experiments was performed. [Ideally, there should have been
8 possible combinations of categories, but results obtained with late-stage larvae
collected in the Newport River have been reported earlier (Cronin and Forward,
1979)]. Experiments are grouped by type in Figures 1-7, and results of statistical
analyses are given in Table II. For convenience in presenting results, each type of
experiment will be described separately.
Newport River: Laboratory-hatched stage I zoea larvae
This series of experiments investigated rhythmic vertical migration behavior in
newly hatched larvae which had been entrained as embryos in the strongly tidal
conditions of the Newport River estuary, but which hatched in the laboratory. Five
replicates were performed (Fig. 1), and of these, 4 revealed significant evidence of
circatidal rhythmicity in larval vertical migration (Table II). The 5th experiment
also illustrated circatidal periodicity after an initial 24 h rise in the water column
(Table II, Fig. 1 ). Visual examination of Figure 1 reveals that low points in the larval
migration were not particularly well synchronized with the time of low tide at the
site of collection of the parent crab. No evidence was found in any analysis for
circadian rhythmicity, nor is any circadian variation suggested in the individual
graphs of Figure 1 . Experiment 813 is a possible exception, since alternately deeper
low points occurred near midnight (Fig. 1).
Neuse River: Laboratory-hatched stage I zoea larvae
These experiments were similar to those of the previous group except that prior
entrainment occurred in the nontidal, well lit environment of the Neuse River
estuary. All larvae hatched near the time of sunset on the night the experiment
began, as is typical of larval hatches of crabs from this location (Forward et «/.,
1982). The results were more varied than those obtained with Newport River larvae
(Fig. 2). Data analysis revealed significant circatidal periodicity in 2 cases (experi-
144 T. W. CRONIN AND R. B. FORWARD, JR.
TABLE II
Results of statistical tests for periodicity performed on the time series of each experiment's data*
Experiment type
Experiment
number
Fisher's Multiple
test autoregression
Analysis of
variance
Newport River, First
791
+ +
+
Stage, Lab-Hatched
792
11.7 h
12 h, 13 h
805
— —
12 h
811
— —
12 h
813
13 h
—
Neuse River, First Stage,
794
— —
—
Lab-Hatched
801
13.0 h —
12 h, 13 h
803
34.5 h
—
804
— —
24 h
812
— —
12 h, 24 h
Newport River, First
793
12.8 h 12 h, 24 h
12 h, 13 h
Stage, Field-Caught
802
14.0 h 12 h, 13 h, 24 h
12 h, 13 h
Neuse River, First Stage,
814
— —
—
Field-Caught
815
— —
—
816
— —
—
817
— —
—
Newport River, Late
827
— —
—
Stage, Lab-Hatched
828
— —
12 h, 13 h
829
— —
—
Neuse River, Late Stage,
824
— —
—
Lab-Hatched
826
— —
—
Neuse River, Late Stage,
821
25.5 h —
—
Field-Caught
822
18.7 h
—
823
— —
—
825
— —
—
* Included are all results of statistical tests giving P < 0.05. Significant results are given for Fisher's
Periodogram Test only if the significant period was not equal to the entire length of the time series.
+ In experiment 79 1 , no test yielded a significant result for the entire 83 h of the time series. However,
if the first 24 h of data (during which there was a continuous rise) were eliminated, Fisher's Test indicated
a significant period of 1 1.8 h, and ANOVA gave significant results for periods of 12 h and 13 h.
ments 801 and 812, Table II); circadian periodicity was indicated in 2 experiments
as well (804 and 812). In spite of the heterogeneous mixture of vertical migration
patterns, one migration feature was consistently observed. All groups of larvae ini-
tially migrated downward until near midnight (near dawn in experiment 794), when
they reversed their course and rose for the succeeding several hours (Fig. 2). This
pattern is not simply a response to being placed in the experimental column, since
it was not observed in Newport River larvae (Fig. 1 ). Subsequent pre-dawn rises are
also visible in several cases (Experiments 801, 803, 804, and 812).
Newport River: Field-caught stage I zoea larvae
In these experiments, larvae were taken from the plankton and thus had an
opportunity to experience conditions in the Newport River as free-living individuals
>r some time prior to being placed under constant conditions. Only two experiments
; ere performed since the results were very clearcut. Larvae were strongly circatidally
nic (Table II), reaching the low points of their migrations just after the time
MIGRATION RHYTHMS OF LARVAL CRABS
145
00T
I05
.c
Q.
& 10
c
D
Exp 791
T 1 T T
T T
50
Hour
100
00T
.§05
10
15
Exp 805
t T
5O
Hour
KDO
00
.§05
10
15
Exp 813
00
.c
"o.
a
c
o
10
Exp 792
00
05
a
§ I0
T T T T
T T T
50
Hour
100
Exp 81 1
T T
T T T T T T
50
Hour
OO
T T I t T T T T
50
Hour
100
FIGURE 1 . Hourly positions of the mean depths of populations of first-stage Rhithroponopeus
harrisii larvae maintained in constant darkness and at constant temperature. Each panel represents the
results obtained with larvae of a single hatch from a crab collected in the Newport River estuary. Dark
and light bands on the abscissa indicate the times of natural night and day, respectively. Arrows indicate
times of low tide at the collection site. Gaps correspond to missing data due to equipment failure.
of low tide at the collection site (Fig. 3). Such a pattern duplicates results already
obtained for field-entrained late-stage Newport River R. harrisii larvae (Cronin and
Forward, 1979). Autoregression analysis also showed that larval depths were sig-
nificantly predicted by those 24 h earlier in the record (Table II), but we feel that
this is actually a correlation with the second previous tidal cycle. The other statistical
tests found no 24 h rhythmicity, and inspection of Figure 3 shows no sign of circadian
activity.
Neuse River: Field-caught stage I zoea larvae
Here, larval sample populations were taken in nightime plankton tows in the
Neuse River. The majority of the larvae almost certainly hatched on the same night
146
T. W. CRONIN AND R. B. FORWARD, JR.
00T
'05
§
Exp 794
50
Hour
100
00
.c
-*-
a.
I
c 10
o
a>
Exp 801
50
Hour
100
00T
05
'a.
§ I0
15
Exp 803
50
Hour
100
00
-§05
.c
'a.
&
Exp 812
OOi
§05.
.c
c 10
15
Exp 804
5O
Hour
100
50
Hour
KX>
FIGURE 2. Hourly positions of the mean depths of populations of first-stage Rhithropanopeus
harrisii larvae maintained in constant darkness and at constant temperature. Each panel represents the
results obtained with larvae of a single hatch from a crab collected in the Neuse River estuary. Dark and
light bands on the abscissa indicate the times of natural night and day, respectively.
on which they were collected (see Discussion). Four replicates were completed.
Results revealed a variety of irregular migration patterns (Fig. 4), but in no case
was there any significant circadian or circatidal rhythmicity (Table II). However,
the pattern described earlier for laboratory-hatched Neuse River larvae was again
present. Larvae usually moved downward until late in the first dark phase, at which
time they reversed course and rose in the water column. In these experiments,
however, there was little evidence for repeated cycling (Fig. 4).
'export River: Laboratory-hatched and reared late-stage zoea larvae
We have previously reported that Neuse River larvae reared in constant tem-
(erature and in a 12 h light: 12 h dark cycle performed a low-amplitude circadian
MIGRATION RHYTHMS OF LARVAL CRABS
147
oo
•05
Q.
£
c 10
o
0)
15
Exp 793
00'
'05
Q.
O)
Q
T T T I
t t
1.5
Exp 802
T T
T t T T
50
Hour
100
50
Hour
100
FIGURE 3. Hourly positions of the mean depths of populations of first-stage Rhithropanopeus
harrisii larvae maintained in constant darkness and at constant temperature. Each panel represents the
results obtained with larvae caught in plankton net tows taken in the Newport River estuary. Dark and
light bands on the abscissa indicate the times of natural night and day, respectively. Arrows indicate times
of low tide at the collection site.
rhythm of vertical migration, reaching their lowest position near midnight and their
greatest height near midday (Cronin and Forward, 1979). We were interested to
learn whether the same result would obtain with laboratory-reared Newport River
larvae; therefore, on 3 occasions we reared laboratory-hatched Newport River larvae
to zoea III before placing them in experimental conditions. Larvae were maintained
in a 14 h light: 10 h dark cycle which closely matched the actual times of sunrise
O.OT
'05
o.
OJ
Q
1.5
Exp 814
50
Hour
100
00T
E
— 05
15
Exp 815
50
Hour
100
00
IDS
o.
<D
Q
10
o
0)
1.5
Exp 816
50
Hour
100
O.Oi
-§0.5
o.
o>
Q
c 10
o
0)
Exp 817
50
Hour
100
FIGURE 4. Hourly positions of the mean depths of populations of first-stage Rhithropanopeus
harrisii larvae maintained in constant darkness and at constant temperature. Each panel represents the
results obtained with larvae caught in plankton net tows taken in the Neuse River estuary. Dark and
light bands on the abscissa indicate the times of natural night and day, respectively.
148
T. W. CRONIN AND R. B. FORWARD, JR.
oo
05
_c
Q.
s
§10
15
Exp 827
t T T ! T T 1 T
oo
Q.
&
§ 10
o>
15
Exp 828
5O
Hour
(00
T T T T T ! T T T
50 100
Hour
00
.c
-*-
Q.
05
1.5
Exp 829
T T T T T t T
50
Hour
100
FIGURE 5. Hourly positions of the mean depths of populations of late-stage Rhithropanopeus harrisii
larvae maintained in constant darkness and at constant temperature. Each panel represents the results
obtained with larvae of a single hatch from a crab collected in the Newport River estuary, which were
reared to zoeal stage III before the experiment began. Light and dark bands on the abscissa indicate the
light:dark cycle to which the larvae were exposed during rearing. Arrows indicate times of low tide at
the collection site.
and sunset. Migration patterns were largely random (Fig. 5), but in one case (ex-
periment 828) ANOVA indicated the presence of circatidal rhythmicity (Table II).
Neuse River: Laboratory-hatched and reared late stage zoea larvae
For these 2 experiments, Neuse River larvae were prepared under identical con-
ditions to the Newport River larvae just described. Once more, the light:dark cycle
00
:05
Q.
Ol
Q
1.5
Exp 824
50
Hour
100
OOi
o.
11
Q
15
Exp 82
50
Hour
100
FIGURE 6. Hourly positions of the mean depths of populations of late-stage Rhithropanopeus harrisii
' ;rvae maintained in constant darkness and at constant temperature. Each panel represents the results
htained with larvae of a single hatch from a crab collected in the Neuse River estuary, which were reared
?oeal stage III before the experiment began. Light and dark bands on the abscissa indicate the light:dark
> hich the larvae were exposed during rearing.
MIGRATION RHYTHMS OF LARVAL CRABS
149
consisted of 14 h of light alternating with 10 h of dark, instead of the 12 h:12 h
cycle previously used (Cronin and Forward, 1979). In contrast to the results of that
work, no rhythmicity was indicated, either statistically or by visual inspection of the
data (Fig. 6, Table II). Essentially random movements of the center of the larval
population occurred.
Neuse River: Field-caught late-stage zoea larvae
Larvae were collected in an identical way to the first-stage Neuse River larvae.
Four replicates were performed (Fig. 7); there was some statistical evidence in two
of them of circadian rhythmicity (experiments 821 and possibly 822, Table II).
However, the form of the circadian pattern is difficult to recognize in the figures,
and all the graphs reveal considerable random movement.
DISCUSSION
Larvae of the estuarine crab Rhithropanopeus harrisii are known to possess
endogenous rhythms of vertical migration by the time they attain the third zoeal
stage (Cronin and Forward, 1979). The period lengths of these rhythms can ap-
proximate the period of either the tidal cycle or the diel cycle, depending on the
prior entrainment regime and the larval source. We designed the experiments de-
scribed in this paper to answer the questions of whether these rhythms are expressed
early in larval life and whether it is possible for larvae to become entrained to
oo
05
-C
"o.
Q
15
Exp 821
5O
Hour
100
00
05
.
fi
15
Exp 822
50
Hour
100
00
.
<i>
Q
g 10
O)
15
Exp 823
00
50
Hour
OO
'05
o.
o>
Q
10
1.5
Exp 825
50
Hour
100
FIGURE 7. Hourly positions of the mean depths of populations of late-stage Rhithropanopeus harrisii
larvae maintained in constant darkness and at constant temperature. Each panel represents the results
obtained with stage III and/or stage IV larvae collected by plankton net in the Neuse River estuary. Dark
and light bands on the abscissa indicate the times of natural night and day, respectively. Gaps correspond
to missing data due to equipment failure.
150 T. W. CRONIN AND R. B. FORWARD, JR.
environmental cycles during their embryonic development. The results, although
more equivocal than the ones obtained previously, suggest that both questions may
be answered in the affirmative. However, conditions which vary between estuaries,
and which also differ in their effects on larvae before and after hatching, strongly
modify the rhythmic aspects of larval migrations. Furthermore, there probably exist
differences among larval cohorts or larval populations which also affect expression
of larval rhythms.
Stage I zoea larvae from the Newport River performed vertical migrations in
the constant conditions of the laboratory, even if they had never experienced the
strong tidal influences of this estuary as free-living individuals. The apparent cir-
catidal rhythms of laboratory-hatched larvae were probably entrained during em-
bryonic development. While the larvae are developing, the parent crab with the egg
mass remains at depths where the diel light:dark cycle is imperceptible (Forward et
al, 1982), so the lack of circadian rhythmicity is not surprising. The observed
rhythmicity could be a product of embryonic entrainment by tidal cycles of pressure
and/or salinity, as both are known to induce circatidal rhythmicity in crustaceans
(Naylor and Atkinson, 1972; Taylor and Naylor, 1977). Pressure receptors have not
been described in larval brachyurans, much less in their embryos; but R. harrisii
larvae are highly responsive to pressure changes (Bentley and Sulkin, 1977; Wheeler
and Epifanio, 1978). They also respond to small salinity changes (Latz and Forward,
1977; Harges and Forward, 1982).
The mother crab could assist in entraining larval rhythms by manipulating the
egg masses at a specific phase in the tidal cycle, since mechanical stimulation is
effective in entraining circatidal rhythmicity (Enright, 1963, 1965). We have never
observed such behavior in crabs in the laboratory. Hatching itself could be a "one-
shot" synchronizer since larval release is precisely timed with respect to environ-
mental cycles in R. harrisii (Forward et al, 1982) as well as in other estuarine crabs
(DeCoursey, 1979; Bergin, 1981; Saigusa, 1981). This seems unlikely since crabs
from the Neuse River have a larval release rhythm, but their larvae do not always
reveal circatidal rhythms after hatching. Furthermore, unpublished observations
suggest that the time of hatching is more likely controlled by the embryos themselves
than by the mother crab.
In any case, much stronger rhythms are expressed by stage I larvae from the
Newport River after a short time in the plankton. Our field-caught larvae were
probably less than 3 days old, since the first zoeal stage is passed in 2-3 days at
environmental temperatures in the laboratory (Costlow and Bookhout, 1971). In
the few tidal cycles after hatching, the larval rhythms became enhanced in amplitude
and, probably, coherency (seen as reduced noise, c.f. Fig. 1 and 3), and thus became
very similar to rhythms of late-stage R. harrisii larvae from the same location (Cronin
and Forward, 1979). The entraining stimuli, whether the same or different, are
clearly much more effective upon free-living larvae than on developing embryos.
Compared to Newport River larvae, stage I zoea larvae from the Neuse River
were much more variable in their expression of vertical migration rhythms. Two
experiments showed statistical evidence of circadian rhythmicity and 2 of circatidal
rhythmicity, while one group of laboratory-hatched larvae and all field-caught larval
groups were arhythmic. The statistical results agree reasonably well with the sub-
jective appearance of the results (Fig. 4). Circadian rhythmicity is not unexpected
in this population since hatching time is under circadian rhythmic control (Forward
/ al., 1982) and late-stage larvae of crabs from this estuary can show circadian
hms of vertical movement (Cronin and Forward, 1979). The crabs were collected
from shallow depths in quite transparent water, so entrainment by the daily
MIGRATION RHYTHMS OF LARVAL CRABS 151
light:dark cycle was possible. In all these experiments there was an initial nightime
sinking phase followed by an upward migration early the next day (experiment 816
did not have the initial descent, but this experiment began just before dawn). The
results thus resemble the single-cycle "hourglass" timing of the vertical migrations
of some marine zooplankton, which require resetting by external inputs each day
(Enright and Hamner, 1967). It appears that newly-hatched larvae from the Neuse
River can express a weak, rapidly-damping circadian rhythm in vertical migration.
Evidence that at least some cohorts of Neuse River larvae have circatidal rhythms
is surprising. Tides in this estuary are reportedly aperiodic (Roelofs and Bumpus,
1953). We measured changes in salinity and depth at one of our collection sites in
the Neuse River for 24 h and found no evidence of regular tidal influence. However,
R. harrisii from this estuary have circatidal hatching rhythms once placed in natural
tidal conditions (Forward et ai, 1982). Until further experiments are done, it will
be impossible to know whether the 12-13 h rhythms observed here are expressions
of this innate tidal clock.
Only laboratory hatches of Neuse River larvae had significant rhythms. This
probably reflects the fact that in each experiment, larvae had identical developmental
histories and a single hatching time. In the field samples, first-stage R. harrisii larvae
were much more abundant than later stages, and were most easily obtained soon
after dark. Therefore, the ones used in our experiments had most likely hatched on
the night they were collected, but they had experienced a range of developmental
conditions and probably had hatched over a period of hours (see Forward el al,
1982). The irregular vertical movements of these larval groups evidently result from
the lack of synchrony among the individual larvae of the experiment. This contrasts
with the greater synchrony seen in the migrations of Newport River larvae entrained
in the field. Evidently, tidal variables of latter estuary are much more effective
synchronizers than the diel variations in the Neuse River environment.
In their essential features, the migration patterns of laboratory-hatched stage I
zoea larvae from each estuary were similar to those of first-stage larvae collected
from the plankton of that estuary. In contrast, late-stage larvae reared in the lab-
oratory had vertical migrations which resembled neither those of field-caught larvae
of similar age nor those of the first-stage larvae. Newport River larvae which lived
in natural field conditions until the third zoea had dramatic migration rhythms
when placed in constant conditions (Cronin and Forward, 1979). When reared to
the same stage in the laboratory in an imposed lightdark cycle, these larvae were
usually arhythmic (Fig. 5). In one case, statistical analysis indicated that a cycle of
circatidal period was present (Table II). It therefore remains possible that an initial
circatidal rhythm can continue throughout development, though it is not clear how
coherency among larval individuals could be maintained.
In neither laboratory-reared nor field-caught groups of stage III Neuse River
larvae do vertical migration patterns commonly show significant rhythmicity. To
the eye, these migrations appear essentially random. Larvae from this source are
capable of expressing circadian migration rhythms when reared on a 12 h:12 h
light:dark cycle (Cronin and Forward, 1979). The present results suggest that the
circadian tendency is rather weak.
Taken as a whole, the results of this study indicate that entrainment of vertical
migration rhythms may occur during embryonic development. Larvae ofR. harrisii
seem strongly biased towards circatidal, rather than circadian, rhythmicity. Entrain-
ment to the tidal cycle is dramatically enhanced once larvae become free living,
whereas entrainment to the diel cycle is clearly no more effective on free-living than
on embryonic larvae. Our failure to find strong circadian rhythms after rearing
152 T. W. CRONIN AND R. B. FORWARD, JR.
larvae in an imposed lightdark cycle is especially impressive when compared to
results of a previous experiment which studied vertical migration of late-stage R.
harrisii larvae when exposed to an external light:dark cycle (Cronin and Forward,
1982). In this case, exogenously driven diel vertical migrations occurred over most
of the 1.9 m height of the experimental column. Such migrations have not been
observed in field populations of these larvae (Cronin, 1982), and the importance of
the diel light:dark cycle in controlling larval behavior of R. harrisii remains to be
understood.
Several other species of estuarine crabs have circatidal rhythms of larval release
(DeCoursey, 1979; Bergin, 1981; Saigusa, 1981), and field sampling by Christy and
Stancyk (1982) strongly suggested that virtually all crab species in a South Carolina
estuary release larvae near local high tide times. One might therefore expect circatidal
rhythms in the larvae of these crabs. Yet R. harrisii stands alone among species yet
studied in having highly effective mechanisms for larval retention in estuaries
(Cronin, 1982); the other larval species all apparently undergo rapid export from
estuaries, perhaps to reduce predation pressure (Christy, 1982). Estuarine retention
ofR. harrisii larvae is thought to be assisted by their tidal vertical migrations (Cronin
and Forward, 1979, 1982; Cronin, 1982). Studies of the circatidal rhythms of other
species of estuarine crab larvae, as well as of the endogenous and exogenous controls
on these rhythms, should prove highly informative to our understanding of the bases
of larval ecology.
ACKNOWLEDGMENTS
We thank W. Hunnings for assistance with design, construction, and mainte-
nance of the monitoring system; K. Lohmann and M. Butka for helping with the
collection of animals; and J. Hartigan for patient advice on the use of statistical
techniques. R. Cole and T. Seeley provided useful comments on the manuscript.
This material is based on research supported by the National Science Foundation
under grant no. OCE-8007434.
LITERATURE CITED
BENTLEY, E., AND S. D. SULK.IN. 1977. The ontogeny of barokinesis during zoeal development of the
xanthid crab Rhithropanopeus harrisii (Gould). Mar. Behav. Physiol. 4: 275-282.
BERGIN, M. E. 1981. Hatching rhythms in Uca pugilator (Decapoda: Brachyura). Mar. Biol. 63: 151-
158.
CHRISTY, J. H. 1982. Adaptive significance of semilunar cycles of larval release in fiddler crabs (genus
Uca): test of an hypothesis. Biol. Bull. 163: 251-263.
CHRISTY, J. H., AND S. E. STANCYK. 1982. Timing of larval production and flux of invertebrate larvae
in a well-mixed estuary. Pp. 489-503 in Estuarine Comparisons, V. S. Kennedy, ed. Academic
Press, New York.
CONNOLLY, C. J. 1925. The larval stages and megalopa of Rhithropanopeus harrisii (Gould). Contr. Can.
Biol. 2: 329-334.
COSTLOW, J. D., JR., AND C. G. BOOKHOUT. 1971. The effect of cyclic temperatures on larval devel-
opment in the mud-crab Rhithropanopeus harrisii. Pp. 211-220 in Fourth European Marine
Biology Symposium, D. J. Crisp, ed. Cambridge Univ. Press, Cambridge.
CRONIN, T. W. 1 982. Estuarine retention of larvae of the crab Rhithropanopeus harrisii. Estuarine Coastal
Shelf Sci. 15: 207-220.
CRONIN, T. W., AND R. B. FORWARD, JR. 1979. Tidal vertical migration: an endogenous rhythm in
estuarine crab larvae. Science 205: 1020-1022.
CRONIN, T. W., AND R. B. FORWARD, JR. 1982. Tidally timed behavior: effects on larval distributions
in estuaries. Pp. 505-520 in Estuarine Comparisons. V. Kennedy, ed. Academic Press, New
York.
DECOURSEY, P. 1979. Egg hatching rhythms in three species of fiddler crabs. Pp. 399-406 in Cyclic
Phenomena in Marine Plants and Animals, E. Naylor and R. G. Hartnoll, eds. Proceedings of
the 1 3th European Marine Biology Symposium, Pergamon Press, Oxford.
MIGRATION RHYTHMS OF LARVAL CRABS 153
ENRIGHT, J. T. 1963. The tidal rhythm of activity of a sand-beach amphipod. Z. Vgl. Phvsiol. 46: 276-
313.
ENRIGHT, J. T. 1965. Entrainment of a tidal rhythm. Science 147: 864-867.
ENRIGHT, J. T., AND W. M. HAMMER. 1967. Vertical diurnal migration and endogenous rhythmicity.
Science 157: 937-941.
FORWARD, R. B., JR., AND T. W. CRONIN. 1 980. Tidal rhythms of activity and phototaxis in an estuarine
crab larva. Biol. Bull. 163: 295-303.
FORWARD, R. B., JR., K. LOHMANN, AND T. W. CRONIN. 1982. Rhythms in larval release by an estuarine
crab (Rhithropanopeus harrisii). Biol. Bull. 163: 287-300.
FULLER, W. A. 1976. Introduction to Statistical Time Series. J. Wiley, New York. 470 pp.
HARGES, P., AND R. B. FORWARD, JR. 1982. Salinity perception by larvae of the crab Rhithropanopeus
harrisii (Gould). Mar. Behav. Physiol. 8: 31 1-331.
LATZ, M. I., AND R. B. FORWARD, JR. 1977. The effect of salinity upon phototaxis and geotaxis in a
larval crustacean. Biol. Bull. 153: 163-179.
NAYLOR, E., AND R. J. A. ATKINSON. 1972. Pressure and the rhythmic behaviour of inshore marine
animals. Symp. Soc. Exp. Biol. 26: 395-415.
ROELOFS, E. W., AND D. F. BuMPUS. 1953. The hydrography of Pamlico Sound. Bull. Mar. Sci. Gulf
Canb. 3: 181-205.
SAIGUSA, M. 1981. Adaptive significance of a semilunar rhythm in the terrestrial crab Sesarma. Biol.
Bull. 160: 311-321.
SANDIFER, P. A. 1973. Distribution and abundance of decapod crustacean larvae in the York River
estuary and adjacent lower Chesapeake Bay, Virginia, 1968-1969. Chesapeake Sci. 14: 235-
257.
SANDIFER, P. A. 1975. The role of pelagic larvae in recruitment to populations of adult decapod crus-
taceans in the York River estuary and adjacent lower Chesapeake Bay, Virginia. Estuarine
Coastal Mar. Sci. 3: 269-279.
TAYLOR, A. C., AND E. NAYLOR. 1977. Entrainment of the locomotor rhythm of Carcinus by cycles of
salinity change. J. Mar. Biol. Assoc. U.K. 57: 273-277.
WHEELER, D. E., AND C. E. EPIEANIO. 1978. Behavioral response to hydrostatic pressure in larvae of
two species of xanthid crabs. Mar. Biol. 46: 167-174.
Reference: Biol. Bull. 165: 154-166. (August, 1983)
CONTROL OF EGG HATCHING IN THE CRAB
RHITHROPANOPEUS HARRISII (GOULD)
RICHARD B. FORWARD, JR. AND KENNETH J. LOHMANN
Duke University Marine Laboratory, Beaufort, NC 28516 and Department of
Zoology, Duke University, Durham, NC 27706
ABSTRACT
Ovigerous females of the crab Rhithropanopeus harrisii were collected from an
estuary having irregular tides. When monitored under constant conditions in the
laboratory, the crabs have a circadian rhythm in larval release. Eggs removed from
the female within 2 days of hatching hatched at about the same time as larvae were
released by the female. Hatching became increasingly desynchronized with longer
removal times. Upon exposure to water in which the larvae hatched, ovigerous
females diplayed increased abdomen pumping, a behavior observed at the time of
larval release. The active substance was released at the time of egg hatching but not
by newly hatched larvae. Homogenized eggs of different ages and homogenized
larvae induced similar behavior. There was no change in female sensitivity with
clutch age or time of day. Active pumping by the female only induced hatching at
times predicted by the larval release rhythm, not at other times during the solar day.
These results indicate that an interaction between the eggs and female is responsible
for synchronized development while the actual timing of hatching is controlled by
the embryo. At this time an active substance is released. This substance induces
abdomen pumping by the female which serves to synchronize larval release.
INTRODUCTION
Precisely-timed rhythms in larval release are common among crustaceans. The
timing may be related to lunar phase (Christy, 1978, 1982; Saigusa and Hidaka,
1978; Wheeler, 1978; Saigusa, 1981), time of day (Ennis, 1973, 1975; Branford,
1978; Moller and Branford, 1979), or phase of the tide (DeCoursey, 1979; Bergin,
198 1 ). Detailed studies of fiddler crabs (DeCoursey, 1979) and lobsters (Ennis, 1973)
indicate that larval release lasts only a few minutes and is associated with rapid
movements of the ovigerous female's abdomen and pleopods. Since larval release
is a short, precisely-timed event, an important question is whether the timing is
controlled by the female or the developing embryos.
Previous studies produced divergent results. DeCoursey's (1979) work with the
fiddler crab Uca minax suggests that physical stimulation of hatching by the female
is necessary for larval release. For the lobster Homarus gammarus, Branford's (1978)
results indicate that larval release is regulated by the female and that her role in the
hatching process is under endogenous control. Alternatively, other investigators
(Pandian, 1970; Ennis, 1973) have suggested that for this lobster species the clock
which sets the hatching time is in the egg itself.
The estuarine crab Rhithropanopeus harrisii also releases its larvae over a short
time interval (Forward et al., 1982). In the laboratory under constant conditions,
arval release by crabs from an estuary lacking regular tides occurs mainly in the
Received 28 March 1983; accepted 23 May 1983.
154
EGG HATCHING CONTROL 155
2 h interval after the time of sunset, which suggests the presence of a circadian
rhythm. In contrast, releases in the laboratory by crabs from an estuary with semi-
diurnal tides generally begin at the time of high tide in the field and continue for
2 h, suggesting a circatidal rhythm. The present study examines whether the female
or the embryo is controlling the timing of hatching.
Using crabs with a circadian rhythm, we first determined whether the detached
eggs hatch at the same time as eggs attached to the female. Since the timing was
similar, we then experimented to determine if a chemical cue from the hatching
eggs induced the female to undergo the behavioral sequence observed during larval
release.
MATERIALS AND METHODS
The crab Rhithropanopeus harrisii (Gould) was collected from the Neuse River
estuary (North Carolina). Ovigerous females were obtained either from the field or
from a breeding population in a laboratory habitat (described in Forward et al.,
1982). Females were maintained in an environmental chamber (Sherer Gillett Co.,
Model CEL 4-4) at 26-27 °C in 8 ppt sea water which was filtered to remove particles
larger than 5 p.. A 14 h light: 10 h dark photoperiod was employed. Under these
conditions, females from both the field and the laboratory habitat release larvae in
the interval beginning at the end of the light phase and concluding about 2 h later
(Forward et al., 1982). The general experimental procedures are described below
while specific modifications are explained in the Results section. The term "eggs"
refers to the combination of outer covering, enclosed non-living material, and the
developing embryo before hatching.
The first series of experiments was designed to determine whether detached eggs
can hatch and if so, whether the time of hatching is similar to that of larval release
by the female. Hatching was monitored under 3 conditions: detached eggs in still
water, detached eggs on a mechanical shaker, and eggs attached to the female. Crabs
with eggs which would hatch within 1 day (based on eye development and yolk
consumption) were used. For monitoring hatching by detached eggs in still water
100-200 eggs were removed from each female about 3 h before the end of the light
phase and placed in a 7.9 cm diameter finger bowl containing 8 ppt sea water filtered
to remove particles larger than 5 n. Preliminary observations indicated that most
of the eggs hatched in a several hour interval just after the beginning of the dark
phase. To quantify this sequence, hatching was monitored over a 5 h sampling time
beginning 1 h before the end of the light phase. At 0.5 h intervals, the number of
free swimming zoeae was counted and removed by pipette from the finger bowl.
Care was taken not to swirl the bowl water during removal. After the day phase
ended the eggs were placed in constant low level light (photographic safe light con-
taining a 1 5 W bulb and fitted with a Kodak OA filter; wavelength maximum = 573
nm, half band pass 37 nm, intensity = 10 2 W/m2) at 27° ± 1°C. A microscope
illumination lamp (American Optical Co.) interference filtered to 660 nm (Ditric
Optics Inc., half band pass 1 1 nm) was used briefly to see the larvae each half hour.
Larvae are very insensitive to this wavelength (Forward and Cronin, 1979). If eggs
remained at the end of the 5 h sampling interval, then they were maintained under
constant conditions, and the observation procedure was repeated at the next mon-
itoring time, 19 h later.
In order to simulate the pumping action of the female a similar group of eggs
was placed in a finger bowl on a mechanical shaker (Eberbach Corp.). Eggs were
shaken at a rate of about 1 cycle/s. Shaking began 1 h before the beginning of the
156 R. B. FORWARD, JR. AND K. J. LOHMANN
dark phase and continued until hatching occurred. Thus eggs which hatched at the
time of the second night in constant conditions had actually been shaken contin-
uously for at least the preceding 24 h. The bowl was briefly removed from the shaker
to count swimming zoeae.
Simultaneously, we also determined the time of larval release by the parent
female with attached eggs. Crabs were subjected to the same conditions as the eggs.
During the 5 h sampling interval, the female was transferred every 30 min to a new
7.9 cm diameter finger bowl. At the end of the sampling period, she was placed in
a 10.4 cm diameter finger bowl and if eggs remained, the procedure was repeated
at the next monitoring time. The number of larvae released within each 30 min
interval was recorded. Crabs and detached eggs were monitored only for 2 consec-
utive nights in constant conditions.
The detached eggs hatched over several hours (see Figs. 1A and 1C for typical
profiles). For the females with eggs, most larvae are released within a 30 min interval,
though a few commonly appear in the intervals immediately preceding and following
the peak (Fig. IB). The mean time of hatching by the eggs and larval release by the
female was calculated by multiplying the number of larvae observed in each 30 min
interval by that interval, taking the sum of these products over all intervals and
dividing this sum by the total number of larvae. In this way a single 30 min interval
was designated as the time of hatching/larval release.
The next series of experiments was designed to test for the presence of chemical
communication between the eggs and the female. At the time of larval release the
female elevates her body upon her walking legs, then repeatedly flexes her abdomen.
Larvae are released with each "pump."
The frequency of abdomen pumping was used to quantify the inclination of the
female to undergo larval release behavior. The procedure was to first place the female
in a 7.9 cm diameter finger bowl containing 40 ml of 8 ppt sea water filtered to
remove particles larger than 0.2 n and at 27 ± 1 °C. The number of pumps in the
initial 2 min interval was counted by a stationary observer. The crab was then placed
in 40 ml of the test solution (e.g., water in which hatching had occurred) and the
number of pumps in the initial 2 min period was similarly recorded. The control
consisted of placing previously untested crabs sequentially in clean 8 ppt sea water.
The control level did not change with embryo development and is reported for
females with eggs that would hatch within 3 days. Each crab was used only once
in a particular test solution. Initial experiments were conducted during the 5 h
interval after the beginning of the dark phase, because this is the time of normal
larval release. In this case the crabs were observed under red light. Since we sub-
sequently found that responsiveness by the crabs does not vary over the day, later
trials were conducted under room lighting during the day.
RESULTS
Hatching by separated eggs
A typical hatching profile of eggs removed from the female and kept in still
water is shown in Figure 1A. Hatching of detached eggs is not as synchronized as
larval release by the female (Fig. IB). It usually begins shortly before the time of
greatest release by the female and continues over about the next 3 to 4 hours.
Most of the crabs (95%) both released their larvae and had their detached eggs
hatch shortly after the beginning of the night phase. For all of the eggs monitored
till water (Fig. 2), an average of 96% (SE = 0.8%) of each group hatched within
the 5 h sampling interval. These results clearly indicate hatching can occur inde-
EGG HATCHING CONTROL
157
eggs in
still water
n=!3l
ovigerous
female
n=2!23
eggs on
shaker
n=!74
Time (h)
FIGURE 1. The percentage of eggs hatching (ordinate) from one crab over time (abscissa) as related
to the time of the normal end of the light phase (lights out). The eggs in still water (A), left upon the
ovigerous crab (B), and eggs shaken continuously (C), were placed under constant conditions about 24
h before hatching. The arrows indicate the mean time of hatching, n is the number of hatched eggs.
25
15
tn
O>
O
O
10-
Ovigerous crabs
N I
n=30
Eggs in still water
N I
n-30
Ovigerous crabs
N 2
n=24
Eggs in still water
N 2
n = 24
Time (h)
FIGURE 2. Distribution of mean times of larval release by ovigerous females and of hatching by
their detached eggs for different numbers of crabs (ordinate) relative to time (abscissa) of end of the light
phase. After "lights off' on the first night (N-l), crabs and separated eg^s were maintained in constant
conditions and monitored again at the time of the next night (N-2). n indicates the number of hatches
or releases measured for each condition.
158
R. B. FORWARD, JR. AND K. J. LOHMANN
pendently of the female and is not randomly occurring over the solar day. Thus it
is possible to compare the mean time of hatching by detached egg to the mean time
of larval release by the female. On all nights and conditions (Fig. 2) hatching times
are not uniformly distributed over the 5 h sampling interval (P < .05, Kolmogorov-
Smirnov test for goodness of fit). The greatest number of females released larvae
between 30-60 min after the beginning of the dark phase, while for the detached
eggs the time is about 1 h later (Fig. 2). This relationship was further verified by
specifically comparing the mean hatch time of the detached eggs to the mean time
of larval release by the parent female. The modal time of hatching by detached eggs
was 1 h later. In conclusion: eggs detached within 2 days of hatching hatch at about
the same time as those attached to the female, but the former do so more variably
and about 1 h later.
The differences between the two situations may result from the females' behavior,
that is, vigorous female pumping may assist the opening of the egg covering, resulting
in the release of most of the larvae over a shorter period of time. To test this
hypothesis the previous experiment was expanded to include a group of detached
eggs which were placed on a mechanical shaker. The eggs were shaken continuously
to determine whether mechanical agitation alone causes hatching at times other
than the interval just after the end of the day phase. The female does not pump her
abdomen continuously.
A typical hatching sequence for this group is shown in Figure 1C. Hatching
occurred earlier in the night, as compared to eggs in still water (Fig. 1 A). However,
larval release by the female was still more synchronized. Most of the crabs (91%)
both released their larvae and had their detached eggs on the shaker hatch shortly
after the end of the day phase. For these detached eggs (Fig. 3), an average of 99%
(SE == 0.4%) of each group hatched within the 5 h sampling interval. Therefore,
hatching is not occurring randomly and agitation assists hatching only during the
time interval at the beginning of the dark phase.
10-
O)
<v
.a
E
3
10 J
Eggs on shaker
N I
n = !4
Eggs on shaker
N 2
n = !7
-I Lights
out
-I Lights +1
out
Time (h)
FIGURE 3. Distribution of larval releases and of hatching by eggs on shaker relative to the end of
the light phase. Symbols, as in Figure 2.
EGG HATCHING CONTROL 159
On all nights and conditions (Fig. 3), hatching times are not uniformly distributed
over the 5 h sampling interval (P < .05). In general, hatching by detached eggs on
the shaker occurs near the times of larval release by the females (Fig. 3). If the mean
hatching time of the shaken eggs is compared to the mean time of larval release by
the parent female, the modal difference in timing is zero (n = 30). Thus the shaken
eggs hatch at about the same time as eggs attached to the female, and mechanical
agitation seems to mimic abdomen pumping.
For all but two crabs, larval release occurred on a single night. In the two
exceptions the crabs released one group of larvae on the first night and the rest at
the time of the second night. This is not unusual (Forward et ai, 1982). For these
crabs, detached eggs were also monitored in still water and on the shaker. In both
cases some of the eggs hatched on the first night with the remaining eggs hatching
about 24 h later. These results further support the conclusion that hatching in the
detached eggs occurs at about the same time as larval release by the female.
Although this conclusion is consistent for eggs removed within 2 days of hatch-
ing, a further question is whether eggs, which are removed from the female earlier,
hatch in synchrony with eggs attached to the female. We selected 3 crabs with eggs
which would hatch in about 9 days and entrained them for 4 days to a 14 h light: 10
h dark photoperiod having the beginning of the dark phase at 1200. The time of
end of the day phase was shifted so that hatching would occur at a convenient time.
Four days is sufficient to shift the timing of the rhythm (Forward et al., 1982).
Beginning 5 days before the expected time of hatching, a group of approximately
the same number of eggs (average difference == 26%) was removed from each crab
at daily intervals. The eggs and female were maintained in aerated 8 ppt sea water
filtered to remove particles larger than 0.2 p. and to which the antibiotic chloram-
phenicol was added (5 mg/1). Extremely clean water containing the antibiotic was
necessary to permit viable embryo development. Only 1% of the removed embryos
died. Chloramphenicol at this concentration does not affect biological rhythms in
eukaryotes (e.g., Goodenough et al., 198 1 ). The water was changed every other day.
The eggs and female were maintained at 27° ± 1°C on the LD cycle throughout
the experiment. On the day when hatching was expected, larval release by the females
and hatching by the free eggs was monitored in still water at 1 h intervals beginning
2 h before the end of the light phase.
The hatching cycles are shown in Figure 4. Since the number of eggs removed
each day from each female was similar, the total numbers of eggs hatched in all 3
broods were combined, and the absolute numbers presented for each 1 h interval.
For the ovigerous females the number of released larvae differed greatly. To combine
these data, the percentages of larvae released in each 1 h interval were averaged.
Eggs removed on the final day of development hatched at about the same time as
larvae were released by the females. For eggs removed for longer times, hatching
was never uniform (P < .05, Kolmogorov-Smirnov test for goodness of fit) over the
32 hours that hatching was monitored. However, hatching became increasingly
desynchronized with longer removal times. Since the eggs were maintained on the
LD cycle, synchrony does not result only from entrainment on a LD cycle. This
result suggests that some aspect of the female-egg interaction is important in estab-
lishing synchronized hatching.
Cues from the eggs
The next series of experiments was designed to answer several questions. First,
is there a chemical cue released at the time of hatching which induces pumping
160
R. B. FORWARD, JR. AND K. J. LOHMANN
O>
100-
0
2?
5 60
o
c
o
20 1
300
100
40
20]
& X>i
$
<D
JD
3
c
ovigerous
crab
eggs removed on
day of hatching
n=36l
eggs removed
-I day
n = 420
* ._
eggs removed
-2 days
n=456
^ — •
eggs removed
-3 days
~ n=425
eggs removed
-4 days
= 36l
eggs removed
-5 days
= 204
10 14
18 22 02
Time (h)
FIGURE 4. Distribution of larval releases by 3 females (A) and of hatching of their detached eggs
(B-G) relative to time in the LD cycle. Eggs from only 2 crabs were used for G. The cross hatched bars
indicate the time of the dark phase. In all cases greater than 94% of the eggs hatched during the observation
time, n indicates the number of eggs hatched for the different conditions.
behavior by ovigerous females? If so, does receptivity to the cue change with age
of the female's embryos?
After a female released her larvae into a volume of clean water (filtered initially
to 0.2 n), she was quickly removed. The water was filtered to remove the larvae,
and the larvae were counted. The water was diluted so that there was 1 ml for each
40 larvae released. This concentration of "larval water" was selected because pilot
experiments showed that it induced a strong pumping response (Fig. 5). Upon ex-
posure to this larval water, ovigerous females showed an initial period of agitated
movement (0.25 to 1.5 min), after which they elevated their bodies on the walking
EGG HATCHING CONTROL
161
1*60-
"o.
I
"55
o
(D
b
c
20
49
20
20/
C
20
20 40 60 80 100
Concentration (larvae released/ml)
FIGURE 5. The percentage of crabs increasing their pumping rate (ordinate) upon exposure to water
in which different concentrations of larvae hatched (abscissa). The number near each point is the total
number of crabs tested at that concentration. C is the control and indicates the percentage of crabs
increasing their pumping rate upon sequential exposure to clean 8 ppt sea water.
legs and vigorously pumped their abdomens. Pumping by ovigerous crabs with
different age embryos (determined by eye development and yolk consumption) was
monitored first in clean water and then in the larval water.
The percentage of crabs which showed an increase in pumping rate in the larval
water was significantly greater (P < .05, Z statistic for comparing proportions) for
all ovigerous crabs, as compared to non-ovigerous females (Table I). There was also
a significant increase in the mean pumping rate (Student's / test, P < .05) when
exposed to larval water. However, the rates were not significantly different between
crabs with different age embryos (One-way ANOVA, model I).
These results indicate that the water in which the larvae were released contained
a chemical which induced behavior observed during larval release by the female.
In addition, even though the largest response occurred among crabs with the oldest
embryos, there was little change in responsiveness throughout embryonic devel-
TABLE I
Variation in female pumping response with embryonic development
Estimated time until
egg hatching (days)
Number of pumps/2 min
clean water
hatch water
crabs increasing
pumping
m
SE
m
SE
0-1
18
61
1.1
0.5
8.3
2.1
2-3
31
42
1.5
0.5
4.5
1.4
4-5
19
53
1.2
0.5
4.6
1.4
6-7
33
33
0.6
0.3
3.6
1.4
>7
35
46
0.4
0.2
5.4
1.4
NOF
22
14
1.2
1.2
2.5
1.5
The percentages of crabs displaying an increase in pumping in the hatch water as compared to
pumping in clean water and the mean (m) number of pumps/crab are shown. NOF indicates non-
ovigerous females, n is the sample size and SE is standard error.
162
R. B. FORWARD, JR. AND K. J. LOHMANN
opment. For uniformity in future experiments, tests were run with crabs having
embryos which were expected to hatch within 3 days.
To test for a change in female responsiveness to different concentrations of the
chemical cue, the water in which the larvae were released was diluted to a range of
concentrations (Fig. 5). Responsiveness varied with concentration. Concentrations
of 1 and 5 larvae/ml induced responses indistinguishable from the controls. Re-
sponses to concentrations of 10 larvae/ml or higher were significantly greater (P
< .0 1 ) but were not significantly different from another.
Is the chemical cue released at the time of egg hatching or is it emitted by newly
hatched larvae? To answer this question larvae, immediately upon hatching, were
twice transferred to clean water in finger bowls. This served to wash them and dilute
any chemical cue in the hatch water from which the larvae were transferred. The
larvae were then placed in clean water (concentration 40 larvae/ml) for 2 h, then
removed by filtration through clean plankton netting. This water was tested against
clean water for its ability to induce increased pumping by ovigerous crabs. Only
15% (n ---- 20) of the crabs tested showed an increase in pumping rate in the larval
water. The expected response at this concentration is 49% (Fig. 5), which is signif-
icantly greater (P < .02). Furthermore the per cent response is not significantly
different from that of control crabs tested in clean sea water (5% response; n = 20).
The results indicate that the active chemical is not emitted by the larvae but rather
is released at the time of egg hatching.
To learn if the response can be elicited by crushed eggs, and if so, whether there
is a difference in effectiveness with embryo age, we removed eggs which would hatch
within 1 day (oldest embyros) or within about 8-10 days (youngest embryos). The
eggs were homogenized in clean water, and the homogenate was then diluted to the
appropriate concentration. There was an increase in pumping response with egg
concentration (Fig. 6). The percent response was significantly greater (P < .05) than
100
o>
5. 80
Q.
E 60
0)
t/>
o
0)
40
20
C
20
O.I
0.5 1.0 5.0 10 20
Egg concentration (eggs/ml)
40
FIGURE 6. The percentage of crabs increasing their pumping rate (ordinate) upon exposure to
different concentrations of homogenized eggs (abscissa). Solid line, responses to eggs expected to hatch
within 1 day; dashed line, responses to eggs expected to hatch in 7 or more days. Numbers of females
tested are shown adjacent to each point. C, controls tested sequentially with clean (8 ppt) sea water.
EGG HATCHING CONTROL 163
the control level at all but the lowest test concentrations (0. 1 egg/ml older embyro;
1.0 egg/ml young embryo). At all concentrations the crabs were more responsive
to the older embryos (P < .04), which indicates that the amount of the active
chemical increases with embryonic age.
The potency of the crushed older eggs may be a cumulative result of the embryo,
its embryonic fluid, and egg membrane. This suggestion is supported by two ob-
servations. First, the levels of response to different concentrations of the larval water
(Fig. 5) were below levels shown in response to comparable concentrations of crushed
eggs (Fig. 6). Secondly, if newly hatched zoeae were homogenized and the resulting
mixture diluted to a concentration of 10 larvae/ml, the response level was 40% (n
= 25), which is significantly (P < .03) lower than that for 10 eggs/ml (67%; Fig. 6).
Thus, the egg parts produce responses which were below those of the eggs themselves.
Is there a rhythm in female responsiveness to the chemical cue over the day?
This was tested by collecting crabs expected to release larvae in 2 days and main-
taining them on a 14 h light: 10 h dark cycle in phase with field LD cycle. Starting
the morning after collection, the crabs were tested every 4 h in clean water and then
in a solution having a concentration of 10 homogenized eggs/ml. The test solution
was prepared from eggs which would probably hatch within 1 day. This concentra-
tion was used because it induces a substantial but not a maximal response (Fig. 6).
Crabs were maintained in clean water between trials. At night, pumping was mon-
itored under red light. Preliminary tests indicated that the test solution retained its
activity for at least 36 h if refrigerated. Therefore a stock solution was prepared. A
40 ml aliquot was removed from the refrigerator 3 h before each test and allowed
to warm to experimental temperatures (27 °C). Responsiveness did not change over
the day (Fig. 7), as neither the percentage of crabs showing an increase in pumping
nor the average pumping rates varied significantly.
To test if induced female pumping can cause the eggs to hatch, we maintained
crabs on a 14 h light: 10 h dark cycle in the laboratory until the expected day of
larval release. Crabs were then tested at 0.5 h intervals beginning shortly before the
predicted time of larval release. These crabs were sequentially tested in clean water
and in a solution having a concentration of either 20 or 40 homogenized eggs/ml.
The eggs were predicted to hatch in 1 day. These concentrations were used because
they induce strong responses (Fig. 6). After testing, crabs were rinsed in clean water
and then returned to a new holding bowl also containing clean water. The number
of pumps in 2 min for each trial and the number of larvae released into each bowl
were recorded.
Even though the egg solution induced strong repetitive pumping in all tests, in
no case did this action cause an early mass release of the larvae (Table II). Only an
occasional larva appeared. This result indicates that egg hatching is not induced by
vigorous pumping, except at times when the larvae are predicted to hatch.
DISCUSSION
The embryos of R. harrisii can complete development and hatch as viable larvae
even when removed from the parent female. If eggs are removed within 2 days of
hatching, they hatch at approximately the same time as larval release by the females.
Forward et al. (1982) showed that crabs maintained under the constant conditions
used for the present study have a circadian rhythm in larval release. The similarity
in the timing suggests that the detached eggs also have a circadian rhythm. The
conditions of the experiment, however, do not fully meet the requirements for
demonstrating the presence of a circadian rhythm. The cycle in hatching was only
164
R. B. FORWARD, JR. AND K. J. LOHMANN
80
£40
to
Q.
O
o
20-
16-
12-
8
4
12 16 20
24 04 08 12
Time (h)
16 20 24
FIGURE 7. The percentage of crabs (N = 8, used throughout) increasing their pumping rate (A)
and the average number of pumps/2 min for all tested crabs (B) upon exposure to a solution having a
concentration of 10 crushed eggs/ml at different times of day (abscissa). Brackets are SE. The time of
the dark phase is indicated by the heavy black bar.
measured in eggs maintained under constant conditions for the last two days of
embyro development, and hatching time was measured as the mean time for a
population of eggs from one female. The requirements for the endogenous rhythm
to persist for 5-10 cycles in a single individual under constant conditions cannot
be fulfilled. Thus the results only suggest the presence of a circadian rhythm in the
detached eggs.
The results with R. harrisii eggs differ from those of Branford ( 1 978) for hatching
in the lobster Homarus gammarus. In that case detached eggs hatched rhythmically
under a LD cycle but arrhythmically in constant light or darkness. His procedure
may contribute to these results, since the eggs were removed and held under constant
conditions for 3 days before hatching was monitored. Eggs removed from R. harrisii
for longer than 2 days (Fig. 4) become progressively desynchronized in their hatching,
even when exposed to a LD cycle.
Ovigerous R. harrisii show rhythmic larval releases after up to 5 days in constant
conditions (Forward et ai, 1982). The difference between the persistence of the
larval release rhythm by ovigerous females in constant conditions and the loss of
hatching synchrony by detached eggs in a LD cycle (Fig. 4) suggests that some
unknown aspect of the maternal environment is responsible for the establishment
or maintenance of developmental synchrony, apparent during the last 2 days of
embryonic development.
Although detached eggs hatched in still water, they were not as synchronized
as the larval release by the female (Fig. 1 ). The rapid, vigorous pumping of the
fen ies' abdomen during hatching must enhance synchrony, since groups of larvae
are released with each pump. This suggestion is supported by data (Figs. 1,3) showing
that hatching synchrony in detached eggs was improved by shaking.
Since pumping improves synchrony it is important to know whether egg hatching
EGG HATCHING CONTROL 165
induces pumping. Females placed in water in which larval release occurred showed
an increase in pumping. This indicates chemical and not mechanical cues are in-
volved. For R. harrisii, the active chemical is apparently released at the time of egg
hatching, since newly hatched larvae did not excrete a substance which induced
pumping. Responsiveness is confined to females carrying eggs, as nonovigerous
females showed a very low level of response. These nonovigerous females were newly
collected, and it was possible that the few responsive individuals had either recently
released larvae or were about to oviposit.
Responsiveness did not vary in the female. There was no diel rhythm or change
in responsiveness with embryonic development. Pumping was induced by hatch
water, crushed eggs of different ages, or crushed larvae. Nevertheless, the females
were fairly sensitive to the chemical cue, as the lowest concentrations to induce
significant responses were 10 larvae/ml of the hatch water and 0.05 crushed eggs/
ml. The identity of the substance which induces pumping is currently under in-
vestigation.
Our initial question was whether the time of hatching is controlled by the female
or the developing embryo. The time of hatching of a clutch depends upon those
events which synchronized development of the embryos and those which control
the actual hatching. An interaction between the eggs and female is responsible for
synchronized development, while the embryo controls the actual event of hatching.
The latter conclusion is supported by observations that the eggs hatched rhythmically
independently of the female and upon hatching released a substance which induced
pumping by the female. Pumping alone, however, did not cause hatching (Table
II). Ennis (1973) similarly found that shaking the pleopods of the lobster H. gam-
mams did not induce egg hatching. Thus the conclusion reached by Pandian (1970)
for this lobster, that the eggs control the actual event of hatching, also applied for
R. harrisii.
The foregoing considerations allow us to speculate about the sequence of events
during larval release. Some unknown aspect of maternal care is important in syn-
chronizing embryo development. At the appropriate time the eggs are easily broken
open. Initially, a few eggs hatch either by themselves or due to breakage by the
female's normal body movements. A chemical cue is released which induces the
female to move into position for larval release and to pump her abdomen. Pumping
causes more eggs to hatch, which increases the concentration of the chemical cue,
TABLE II
Pumping response of crabs to clean water and a solution of homogenized eggs at concentrations
of 20 eggs/ml (A) and 40 eggs/ml (B)
# Pumps/2
min in clean
water
# Larvae
# Pumps/2
min in egg
water
# Larvae
Time before
released in
released in
hatching (h)
m
SE
clean water
m
SE
egg water
A
-1.5
1.2
0.6
0
14
4.0
1
-1
0.4
0.3
4
16.3
3.5
2
-0.5
0.6
0.3
1
10.1
2.7
2
B
-1.0
0.75
0.5
0
10.5
5.2
0
-0.5
2.25
0.75
0
19.0
9.5
0
The mean (m) number of pumps and the total number of larvae released in all experiments in the
test bowls (clean water or egg homogenate) are shown. The number of crabs tested in A was 12 while
4 were tested in B. Tests were performed at 0.5 h intervals before the actual time of larval release, which
occurred in the holding bowls shortly after the end of the day phase.
166 R. B. FORWARD, JR. AND K. J. LOHMANN
thereby causing further pumping. This sequence continues until all eggs which are
ready have hatched.
A final consideration concerns the functional significance of having the actual
time of hatching controlled by the embryo. Nocturnal larval release probably lowers
mortality of larvae and adults due to predators which visually sight and actively
pursue their prey (Ennis, 1975; Branford, 1978; DeCoursey, 1979; Bergin, 1981).
Nevertheless, the female must expose herself at or near the entrance of her burrow
during larval release 'making her still somewhat vulnerable to predation. By re-
sponding only when the appropriate chemical cue is present, the female does not
try to release larvae at inappropriate times but rather concentrates her efforts on the
times when the greatest number of larvae will be released. The consequence is
synchronized hatching. In addition to inducing hatching, abdomen pumping also
serves the function of oxygenating the embryos. If the female alone controlled larval
release by regulation of pumping activity, then normal pumping during oxygenation
could potentially release undeveloped embryos.
ACKNOWLEDGMENTS
This material is based on research supported by the National Science Foundation
under Grant No. OCE80-07434. We thank M. Budka for his technical assistance
and Dr. T. Cronin for his critical comments on the manuscript.
LITERATURE CITED
BERGIN, M. E. 1981. Hatching rhythms in Uca pugilator (Decapoda: Brachyura). Mar. Biol. 63: 151-
158.
BRANFORD, J. R. 1978. The influence of daylength, temperature and season on the hatching rhythm of
Homarus gammarus. J. Mar. Biol. Assoc. U. K. 58: 639-658.
CHRISTY, J. H. 1978. Adaptive significance of reproductive cycle in the fiddler crab Uca pugilator: A
hypothesis. Science 197: 453-455.
CHRISTY, J. H. 1982. Adaptive significance of semilunar cycles of larval release in fiddler crabs (Genus
Uca): Test of an hypothesis. Biol. Bull. 163: 251-263.
DECOURSEY, P. 1979. Egg hatching rhythms in three species of fiddler crabs. Pp. 399-406 in Cyclic
Phenomena in Marine Plants and Animals, Proceeding of the 1 3th European Marine Biological
Symposium, E. Naylor and R. G. Hartnoll, eds. Pergamon Press, Oxford.
ENNIS, G. P. 1973. Endogenous rhythmicity associated with larval hatching in the lobster Homarus
gammarus. J. Mar. Biol. Assoc. U. K. 53: 531-538.
ENNIS, G. P. 1975. Observations on hatching and larval release in the lobster Homarus americanus. J.
Fish. Res. Board Can. 32: 2210-2213.
FORWARD, R. B., JR., ANOT. W. CRONIN. 1979. Spectral sensitivity of larvae from intertidal crustaceans.
J. Comp. Physiol. 133: 311-315.
FORWARD, R. B., JR., K. LOHMANN, AND T. W. CRONIN. 1982. Rhythms in larval release by an estuarine
crab (Rhithropanopeus harrisii). Biol. Bull. 163: 287-300.
GOODENOUGH, J. E., V. G. BRUCE, AND A. CARTER. 1981. The effects of inhibitors affecting protein
synthesis and membrane activity on the Chlamydomonas reinhardii phototactic rhythm. Biol.
Bull. 161: 371-381.
MOLLER, T. H., AND J. R. BRANFORD. 1979. A circadian hatching rhythm in Nephrops norvegicus
(Crustacea: Decapoda). Pp. 391-397 in Cyclic Phenomena in Marine Plants and Animals, Pro-
ceedings of the 13th European Marine Biological Symposium, E. Naylor and R. G. Hartnoll,
eds. Pergamon Press, Oxford.
PANDIAN, T. J. 1970. Ecophysiological studies on the developing eggs and embryos of the European
lobster Homarus gammarus. Mar. Biol. 5: 154-167.
SAIGUSA, M. 1981. Adaptive significance of a semilunar rhythm in the terrestrial crab Sesarma. Biol.
Bull. 160: 311-321.
SAIGUSA, M., AND T. HIDAKA. 1978. Semilunar rhythms in zoeae-release activity of the land crab.
Sesarma. Oecologia (Berl.) 37: 163-176.
WHEELER, D. E. 1978. Semilunar hatching periodicity in the mud fiddler crab Uca pugnax (Smith).
Estuaries 1: 268-269.
Reference: Biol. Bull. 165: 167-181. (August, 1983)
ENERGY METABOLISM PATHWAYS OF HYDROTHERMAL VENT
ANIMALS: ADAPTATIONS TO A FOOD-RICH AND
SULFIDE-RICH DEEP-SEA ENVIRONMENT
STEVEN C. HAND AND GEORGE N. SOMERO
Department of Biology. University of Southwestern Louisiana. P.O. Box 42451, Lafayette,
Louisiana 70504, and Marine Biology Research Division, A-002. Scripps Institution of
Oceanography. University of California, San Diego, La Jolla, California 92093
ABSTRACT
The activities of enzymes of the major pathways of energy metabolism (glycol-
ysis, the citric acid cycle, and the electron transport system) were measured in tissues
of animals from the deep-sea hydrothermal vent site at 21°N latitude. Enzymic
activities of related shallow-living marine animals were assayed for comparison.
Vent species studied were the large pogonophoran tube worm, Rift i a pachyptila, the
clam, Calyptogena magnified, the crab Bythograea thermydron, the polychaete
worm, Alvinella pompejana, and an unidentified zoarcid fish, in general, the enzymic
activities found in the tissues of the vent animals were qualitatively and quantita-
tively similar to those of phylogenetically related shallow-living marine species, sug-
gesting that the types of energy metabolism pathways, and the potential flux rates
through these pathways, are similar in both groups. The enzymic activities of the
vent zoarcid fish were much higher than those of all other deep-sea fishes studied
to date. Despite the occurrence in the vent waters of high concentrations of hydrogen
sulfide (HS ), a potent inhibitor of the cytochrome c oxidase system, most of the
vent animals possessed cytochrome c oxidase activities comparable to those of re-
lated shallow-living species. The cytochrome c oxidase systems of the vent species
and shallow-living species so examined were half-inhibited by HS~ concentrations
in the nanomolar to micromolar range. The mechanisms by which the vent animals
avoid poisoning of respiration by HS" are discussed. Calyptogena magnifica was the
only vent species that appeared to have a minimal capacity for aerobic respiration,
as judged by extremely low activities of the cytochrome c oxidase system and citrate
synthase in its tissues compared to other bivalves. We propose that C. magnifica
may rely largely on anaerobic pathways of energy metabolism.
INTRODUCTION
The unusual water chemistry and biological characteristics of the deep-sea hy-
drothermal vent habitats may favor a number of adaptations in the energy metab-
olism pathways of the vent animals. Unlike typical deep-sea regions, the hydro-
thermal vents have a dense biomass (Spiess et al, 1980) which appears to be sup-
ported by primary production by chemolithotrophic bacteria, especially sulfide
oxidizing species. These bacteria are free-living in the sea water (Karl et al, 1980;
Tuttle et al, 1983) and symbionts of dominant members of the vent fauna, including
the large pogonophoran tube worm, Riftia pachyptila (Cavanaugh et al, 1981; Fel-
beck, 1981; Felbeck and Somero, 1982), the clam, Calyptogena magnifica (Felbeck
etal, 1981; Cavanaugh, 1983), and the unnamed vent mussel (Felbeck et al, 1981).
The presence at the vents of a rich food supply, by deep-sea standards, may permit
Received 1 1 April 1983; accepted 25 May 1983.
167
168 S. C. HAND AND G. N. SOMERO
a relatively high rate of energy metabolism in the vent animals compared to animals
from the typical deep sea. The latter animals may have oxygen consumption rates
that are only a few percent of those of related shallow-living species (Childress, 1971,
1975; Smith and Hessler, 1974; Smith, 1978; Torres et al, 1979; Somero et al,
1983), and these extremely low rates of metabolism may reflect adaptations to the
low food availability in non-vent deep-sea habitats.
Despite the occurrence of a rich food supply in the vent habitats, however, the
presence of high (up to 1 mM; Edmond et al., 1982) concentrations of hydrogen
sulfide (HS~) in the vent waters could potentially block the abilities of vent animals
to metabolize aerobically at high rates. HS is a potent inhibitor of the cytochrome
c oxidase (CO) system and, therefore, of aerobic respiration (Hydrogen Sulfide, 1979;
Powell and Somero, 1983). Thus it is of interest to determine if the energy metab-
olism pathways utilized by vent animals include the same types of reactions found
in marine animals from habitats with low HS~ concentrations, or if the vent animals
are unusually dependent on anaerobic mechanisms of energy metabolism.
The present studies examined several animals thought to be endemic to the
deep-sea hydrothermal vents, including R. pachyptila; C. magnified; the brachyuran
crab, Bythograea thermydron; the polychaete worm, Alvinella pompejana (Pompeii
worm); and an unidentified fish of the family Zoarcidae. We sought answers to the
following questions. First, are the types of aerobic and anaerobic energy metabolism
pathways used by the vent animals similar to those found in phylogenetically related
shallow-living marine animals? Second, if the vent animals do utilize aerobic res-
piration, as judged by the presence of the CO system, is this enzyme system less
sensitive to poisoning by HS~ than the homologous systems of animals from habitats
where HS is not present in high concentrations? Third, are the quantities of enzymic
activity similar in tissues of vent and non-vent animals? An answer to this question
bears directly on the point concerning metabolic rates in the vent animals, since
enzymic activity measurements have proven to be a useful means for obtaining
estimates of respiration rates of shallow- and deep-living marine animals (cf. Chil-
dress and Somero, 1979).
MATERIALS AND METHODS
The hydrothermal vent animals were collected at the 21°N latitude vent site on
the East Pacific Rise (Spiess et al., 1 980). Except for the two specimens of the zoarcid
fish, which were generously provided by Dr. Harmon Craig following the Pluto
Expedition to this site in late 1981, all specimens were collected during the Oasis
Expedition in April-May 1982. The fish were frozen (-20°C) shortly after recovery
at the surface, and were held frozen until the enzymic activity measurements were
made. The enzymes studied in the fish are all known to be stable during freezing
(Childress and Somero, 1979). All enzymic activities in the invertebrates were made
using tissues from live, freshly collected adult animals. The tissues sampled in the
different species are given in the legend to Figure 1 . In most cases activities were
measured within a few hours of retrieval of the specimens, which were collected at
a depth of approximately 2600 m by the DSRV Alvin. The specimens were trans-
ported from the collection site to the surface in an insulated box, and were judged
all cases to be in healthy condition. When specimens were maintained alive
aboard ship (RV New Horizon), they were held in circulating sea water (2-5 °C) at
a pre^ure of 120 atmospheres and used within 2 days. The animals survived for
at least everal days under these holding conditions.
Live specimens of animals from non-vent habitats were obtained as follows. The
stone crab, A fenippe mercenaria, and the hardshell cockle, Chione undatella, were
ENERGY METABOLISM OF VENT ANIMALS 169
collected subtidally off La Jolla, California. Mercenaria mercenaria (the quahog
clam) were collected on the East Coast of the U. S. and purchased from a local
seafood supplier. Solemya reidi, a gutless bivalve found in sulfide-rich habitats, were
collected at depths of approximately 120 m near the Hyperion sewage outfall off
Los Angeles, California, using the RV Velero. Specimens of 5. reidi were maintained
in aquaria in the presence of 1 mM HS until analyzed.
Enzymic activity determinations
For all invertebrates, tissue samples taken from live specimens were homoge-
nized immediately in ice-cold buffer (20 mM potassium phosphate, pH 7.4). In the
case of the vent species, motion of the ship prevented accurate measurement of
tissue weights, so precise dilutions of the tissue samples with homogenization buffer
could not be made. Consequently, enzymic activities for the invertebrates are ex-
pressed in terms of international units (^moles substrate converted to product per
min) per mg protein in the supernatants. The tissues were homogenized using a
Duall-23 ground glass surfaced homogenizer (Kontes Glass Co., Vineland, NJ)
driven by hand. The homogenates were centrifuged at 2500 g for 10 minutes, and
the supernatants were saved and used without further purification for the activity
assays. Enzymic activities were measured immediately at a temperature of 20
± 0.2°C, using Varian-Techtron 634 or 635 spectrophotometers. The activities pre-
sented were all determined at 1 atm pressure. A survey of pressure effects on these
enzymes from vent organisms showed that in situ pressures (approximately 260
alms) had only minimal effects on activities under our assay conditions using sat-
urating substrate concentrations. Maximal inhibition noted was 7%, and maximal
activation was 9%. Thus, the use of 1 atm pressure in these studies is not likely to
have led to artifacts.
The enzymic activities in muscle of the vent zoarcid fish were measured in La
Jolla, California, using tissue samples from two deep frozen specimens. Muscle
samples were removed from the area just behind the operculum and above the
lateral line; these samples appeared to be entirely of white muscle. Samples were
homogenized in 10 mM Tris/HCl buffer (pH 7.5 at 10°C), and the homogenates
were centrifuged at 2500 g for 10 minutes. Enzymic activities were measured at 10
± 0.2°C, and are expressed as international units per g fresh (wet) weight of tissue.
This normalization of activity on a fresh weight basis for the fish enzymes was done
to enable comparisons to be made with data gathered under identical experimental
conditions in studies of other deep- and shallow-living marine fishes (Childress and
Somero, 1979; Sullivan and Somero, 1980; Siebenaller and Somero, 1982; Sieben-
allertf a/., 1982).
The following enzymes were studied in some or all of the species: L-lactate
dehydrogenase (LDH, EC 1.1.1.27; L-lactate: NAD+ oxidoreductase); pyruvate ki-
nase (PK, EC 1.7.1.40; ATP: pyruvate phosphotransferase); phosphofructokinase
(PFK, EC 2.7.1.11; ATP: D-fructose-6-phosphate 1 -phosphotransferase); L-malate
dehydrogenase (MDH, EC 1.1.1.37; L-malate: NAD+ oxidoreductase); citrate syn-
thase (CS, EC 4.1.3.7; citrate: oxaloacetate lyase (CoA-acetylating); and cytochrome
c oxidase (CO, EC 1.9.3.1; ferrocytochrome c oxygen oxidoreductase).
Measurements of LDH, PK, MDH, and CS activities were performed following
the protocols given in Somero and Childress (1980). PKF activities were measured
in an assay medium containing 33 mM Tris-acetate buffer (pH 8.0 at 20°C), 2 mM
Mg-acetate, 2 mM ATP, 2 mM fructose-6-phosphate, 40 mM KC1, 4 mM NH4C1,
0.16 mM NADH, 400 ng of aldolase, 20 ^g of triose phosphate isomerase, and 50
Mg of glycerol-3-phosphate dehydrogenase, as described by Hand and Somero ( 1 982).
170 S. C. HAND AND G. N. SOMERO
CO activities were measured using the protocol of Yonetani and Ray (1965).
The assay solution contained 0. 1 M potassium phosphate buffer (pH 6.0), 1 mM
EDTA, and 0. 1 mM reduced cytochrome c, in a total volume of 2.0 ml. The reaction
was followed by recording the decrease in absorbance at 550 nm, using an extinction
coefficient for cytochrome c of 18.5 mM'1 cm"1 (reduced minus oxidized). Reduced
cytochrome c (horse heart, Type III, Sigma Chemical Co., St. Louis, Missouri) was
prepared as follows. A stock solution of cytochrome c (final concentration of 1 mM)
was prepared in 10 mM potassium phosphate buffer, pH 7.0, containing 1 mM
EDTA. The buffer stock was saturated with N2 and stored tightly-capped. The cy-
tochrome c solution was reduced by adding trace amounts of sodium dithionite. A
change in solution color from reddish-brown to bright red-orange indicated quan-
titative reduction of cytochrome c. Excess dithionite and its breakdown products
were removed by gel sieving with Sephadex G-25, utilizing the centrifugation method
of Helmerhorst and Stokes (1980). Sephadex G-25 was hydrated with distilled water
and then equilibrated with 10 mM potassium phosphate buffer, pH 7.0, containing
1 mM EDTA. Next, 3 ml syringes (tips plugged with glass wool) were filled with
hydrated G-25 and placed into conical centrifuge tubes. The syringes were centri-
fuged for 2 minutes at approximately 1 900 g. The liquid that collected in the bottom
of the tubes was discarded, and the reduced cytochrome c solution was added to
the syringes. For a syringe with a 3 ml bed volume, about 0.4 ml of solution can
be added per syringe. The syringes were then centrifuged as above, and the liquid
at the bottoms of the centrifuge tubes was collected. Prepared in this fashion, the
cytochrome c is at least 95% reduced. The rate of autooxidation is only about 1-2%
per day when the solution is stored tightly stoppered at 2°C.
In view of the occurrence of CO activities in most of the tissues of the vent
animals so examined (Fig. 1), it was important to determine if this enzyme system
was resistant to inhibition by HS~ in these species. HS concentrations in the mi-
cromolar range or below typically are strongly inhibitory of CO (Hydrogen Sulfide,
1979; Powell and Somero, 1983).
Except for R. pachyptila, the CO activities were determined using the crude
supernatant fractions prepared as described above. For the CO of R. pachyptila
additional tests were run using partially purified CO prepared by sequential acid
precipitation of the enzyme system ("once acid precipitated" and "twice acid pre-
cipitated"). In this case, the crude supernatant was titrated to pH 5.6 with cold 1.0
M acetic acid and then centrifuged at 2500 g for 10 minutes. Four to nine concen-
trations of HS were used to determine each Kj value. The stock solution of HS
was prepared by dissolving freshly washed crystals of Na2S in deoxygenated distilled
water. The data for CO ofR. pachvptila are derived from data in Powell and Somero
(1983).
Protein concentration measurements
For all of the invertebrate tissues the protein concentration of the supernatant
fractions was measured using the technique of Peterson (1977).
RESULTS
Enzymic activities of invertebrates from vent and non-vent habitats
Figure 1 presents the activities of the glycolytic, citric acid cycle, and electron
transport system enzymes that were analyzed in the different invertebrate species.
As a broad generalization the types of pathways and the flux potentials through
ENERGY METABOLISM OF VENT ANIMALS
171
CYTOCHROMEc OXIDASE
J
006
.
O . .
-
O
~0
0.04
-
1
O
..
i 1 0 1 1
uJ 002
-
o
0 O
i —
o
0 n
O
1
i
O
cr
r—
r
u
- • • r-
0 %
v
'<•
\*<
r \
% \ \ \
2 CALYPTOGENA
MERCENARY CHIONE SOLEMYA
w 0.8
O
_
• o
5 0.6
04
—
0.2
O
1
-
Q
Q
r~
s
\
\
\ X
^ ' *^- ^
•V
^ <
RIFTIA
BYTHOGRAEA
MENIPPE ALVINELLA
CITRATE SYNTHASE
0.32
-
o
.
0.24
.
0
016
.
o
o
0
— 1
o
JTi
-
0 0
n
0
o
o
Q
O
(J
\- 008
.
o
o
o
O O
O
' ' I
(X
(——,——1
CL
n r-H-rt-h
« % X \ -%> ^
** % \ \ *4j. "«<, \ \ \>( *'<< \
?>. CALYPTOGENA
MERCENARIA CHIONE SOLEMY/
± 0.20
0.12
004
RIFTIA
BYTHOGRAEA
MENIPPE
FIGURE 1 . Enzymic activities in different tissues of invertebrates from hydrothermal vent and
shallow marine habitats. Activities are expressed as international units (^moles substrate converted to
product per minute) per mg protein in the supernatant fractions used as sources of enzyme. The heights
of the bars indicate the average values for each tissue; the open circles indicate the measured values. In
most cases two individuals of a species were measured. The tissues are abbreviated on the abscissa of
each graph as follows: add. (adductor muscle), hrt. (heart), mntl. (mantle), tent, (tentacle), vest, (vesti-
mental muscle), troph. (trophosome), chel. (cheliped), bd. wl. (body wall). The habitats of the species are
given in Materials and Methods.
172
S. C. HAND AND G. N. SOMERO
LACTATE DEHYDROGENASE
0.4
•
0.2
z
£2.2
tr
o_
^
^
}__,;
O
j_ . n 1 o Ltj-i o i G • i o i r> i fj i o i i r, . n i
CALYPTOOENA
% > '<< %% \ '<< °o,% \ °0>
MERCENARIA CHIONE SOLEMYA
1 1.4
o
h-
O
5 i.o
0.6
0.2
vs
o
O
0
o
O
Cj ~~
O O
\
% V \
RIFTIA
8YTHOGRAEA MENIPPE
MALATE DEHYDROGENASE
6.0
4.0
n 20
\-
o
CALYPTOOENA
MERCENARIA
CHIONE
SOLEMYA
CO
|6.0
-
-
4.0
•
0
2.0
o
o
o
o
0
o
^1
, °*1
pj
o
0
0
\ V \
^
V
-,<
•v s \
RIFTIA BYTHOGRAEA MENIPPE
FIGURE 1. (Continued)
these pathways appear similar in related vent and non-vent species. For the two
crabs, B. thermydron and M. mercenaria, the types and quantities of enzymic ac-
tivities found in the tissues studied (cheliped, heart, and gill) were strikingly similar.
In both crabs heart tissue displayed the highest aerobic capacity, as judged by ac-
ities of CO, a direct indicator of potential for aerobic respiration, and CS, a strong
indicator of citric acid cycle flux potential. Activities of these two enzymes were
lower in cheliped and gill. The activity of PFK, an indicator of total (aerobic plus
ENERGY METABOLISM OF VENT ANIMALS
173
0.25
0.15
UJ
O
cc
Q.
0.05
0.30
0.20
0.10
PHOSPHOFRUCTOKINASE
MERCENARIA
RIFTIA BYTHOGRAEA MENIPPE
< Of-
CHIONE
L=LZL
ALVINELLA
SOLEMYA
PYRUVATE KINASE
1.00
0.60
^020
O
cr
CL
CALYPTOOENA
MERCENARIA
^ 1.8
0
V,
CO
o
t |.4
— 1
^^
i r\
-
o
o
o
1 U
0.6
•
i — |
0.2
' r^n
0
~J ' ' 1 ' 1
-frl
*<<
e\
v-v
°'<(
RIFTIA BYTHOGRAEA MENIPPE
FIGURE 1. (Continued)
CHIONE
9, /•
SOLEMYA
anaerobic) glycolytic flux potential, was highest in cheliped, as was the activity of
PK, another indicator of glycolytic potential. LDH activity, an indicator of a lo-
comotory muscles' capacity for anaerobic glycolysis, also was highest in cheliped.
Thus, based on these enzymic activity measurements, there would appear to be a
174 S. C. HAND AND G. N. SOMERO
similar capacity for energy metabolism in the vent crab and subtidal crab, a con-
clusion that is consistent with oxygen consumption determinations of B. thermydron
and shallow-living crustaceans under laboratory conditions (Mickel and
Childress, 1982).
For the pogonophoran tube worm, R. pachyptila, no phylogenetically similar
species was available for comparisons. The enzymic activities measured in tissues
of R. pachyptila do allow, however, for conclusions to be drawn about the abilities
of the animal to conduct different types of energy metabolism. The occurrence of
CO and CS activities at levels similar to those found in the two crabs suggests that,
despite living continuously in the presence of high concentrations of HS , R. pa-
chyptila is capable of sustaining aerobic respiration. Tentacle (plume) tissue dis-
played the highest activities of these two enzymes. The tentacle is highly vascularized,
and serves as the major site of gas and nutrient exchange between the animal and
its environment (Jones, 198 1). The aerobic poise of metabolism in tentacle is further
suggested by the relatively low levels of activity of the glycolytic enzymes, PFK, PK,
and LDH, compared to CO and CS activities. Vestimental muscle, which functions
to hold the worm in its tube and to power withdrawal of the tentacle, displayed
lower aerobic capacities than tentacle, but it had higher levels of glycolytic activity.
The high activities of MDH found in vestimental muscle may be indicative of a
high capacity for the type of anaerobic scheme found in many invertebrates, which
involves the channeling of phosphoenolpyruvate towards succinate production via
the intermediates, oxaloacetate, malate, and fumarate (Hochachka, 1980; see Dis-
cussion). The trophosome of R. pachyptila is a soft, highly vascularized tissue that
fills much of the animal's coelom. The trophosome is a complex tissue, containing
high densities of bacterial symbionts (up to approximately 109 bacteria per g fresh
weight; Cavanaugh, 1983; Cavanaugh et al., 1981). Bacterial enzymes may have
made the dominant contribution to the enzymic activities measured in trophosome.
Like the tentacle and vestimental muscle, trophosome displayed capacities for both
glycolytic and electron transport functions.
For the polychaete worm, A. pompejana, which grows abundantly on the walls
of white smoker chimneys and may be exposed to very high concentrations of sulfide
(Desbruyeres and Laubier, 1980; Spiess et al, 1980), limitations in specimen avail-
ability precluded making an extensive enzyme survey. However, the Pompeii worm
exhibited both PFK and CO activities, suggesting that both glycolysis and aerobic
respiration occur in this animal.
Among the four bivalve molluscs we studied, some interesting similarities and
differences were noted. The activities of enzymes associated with glycolysis in bi-
valves, PFK, PK, and MDH, were generally the highest of all enzymic activities,
and the capacities for glycolytic flux seemed generally similar in a given tissue among
species. LDH activity was very low, in keeping with the fact that MDH, rather than
LDH, is the major reaction of glycolytic redox balance in the anaerobic metabolic
scheme of bivalves.
Although as a group, bivalves' CO values were considerably lower than those
of other species, the most striking difference among the bivalves was the apparently
very low capacity for aerobic respiration in C. magnifica. CO activities were ex-
tremely low in all tissues examined, and were barely measureable in foot. CS ac-
tivities also were extremely low compared to the other bivalves studied, suggesting
that C. magnifica has a low capacity for aerobically poised citric acid cycle function.
It is noteworthy that another clam from a sulfide-rich habitat, S. reidi, which was
collected in a sewage outfall habitat where HS concentrations of up to 25 mM have
been measured (J. J. Childress, personal communication) had CO and CS activities
ENERGY METABOLISM OF VENT ANIMALS 175
similar to those of C. undatella and M. mercenaria, two bivalves that do not en-
counter such high HS concentrations in their habitats. Thus a variety of metabolic
strategies may be present in bivalves that occur in sulnde-rich environments (See
Discussion). In C. magnified and S. reidi the gills contain high densities of bacterial
endosymbionts (Felbeck et ai, 1981; Cavanaugh, 1983; Felbeck, 1983). Thus, as in
the case of trophosome tissue of R. pachyptila, a significant fraction of the enzymic
activities measured in the gills of these two bivalves may be of bacterial origin.
Enzymic activities of the vent zoarcidfish
In keeping with the trends noted for the crustacean and molluscan species ex-
amined, the enzymic activities in the vent zoarcid fish were very similar to activities
found in many shallow-living fishes. Activities of LDH, PK, and MDH in white
muscle of the vent zoarcid were 216 (185, 246), 36 (28, 43), and 41 (19, 62) units
per g fresh weight at 10°C, respectively (mean and values for two fish are given).
For LDH and PK these activities are compared with data gathered using identical
protocols on a number of other marine teleost fishes having different minimal depths
of occurrence (Fig. 2). The LDH and PK activities of the vent zoarcid are the highest
found for any deep-sea fish, i.e., for any fish having a minimal depth of occurrence
greater than approximately 200-300 m, and these activities are within the range
noted for many shallow-living, demersal species (cf. Sullivan and Somero, 1980, for
discussion of the other species indicated in Fig. 2). MDH shows a similar trend (cf.
Sullivan and Somero, 1980).
Sensitivities ofcytochrome c oxidase systems to HS~
Using crude supernatant fractions and, for R. pachyptila, partially purified CO,
we determined the sensitivities of the CO systems of several animals (Table I). In
all cases, half-inhibition (K,) concentrations of HS were in the range of 10"9 to 10~5
M. Even though the CO system of tentacle tissue of R. pachyptila appears less
sensitive to HS~ than the other CO systems studied (however, see Discussion), in
all cases the CO systems of the vent animals were inhibited by HS~ concentrations
that were much lower than environmental levels and, in R. pachyptila, were vastly
lower than the HS concentrations found in the animal's blood, where HS con-
centrations up to 1.1 mM have been measured (Arp and Childress, 1983). The bases
for the interspecific differences in CO sensitivity to HS", and possible mechanisms
for resistance to poisoning by HS~ are discussed below.
DISCUSSION
The major conclusion resulting from these comparisons of enzymic activities
of animals from the hydrothermal vents habitat and shallow marine habitats is that,
in almost all cases, the tissues of the vent animals have similar types of energy
metabolism pathways, and similar potentials for flux through these pathways, to
tissues of shallow-living marine species of similar phylogenetic status. These qual-
itative and quantitative similarities in the energy metabolism pathways of these two
groups of organisms merit discussion in terms of the physical, chemical, and bio-
logical characteristics of the hydrothermal vent habitats.
The generally similar activities of the diagnostic enzymes of glycolysis, the citric
acid cycle, and electron transport in the tissues of vent animals and shallow-living
animals suggest that these two groups of organisms have very similar metabolic
rates. Childress and Somero (1979) showed that activities of enzymes of energy
176
S. C. HAND AND G. N. SOMERO
2000
1000
500
9-300
I-
100
0 20
£
1 m
LJ W
O
g 6
u 4
T
0 200 400 6OO 800 1000 1200 1400 I6OO I8OO 2000 2200 2400 2600
MINIMAL DEPTH OF OCCURRENCE (m)
200
100
I 60
I 40
^ 20
H
b o
UJ
% 6
& 2
Q.
0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 24OO 2600
MINIMAL DEPTH OF OCCURRENCE (m)
FIGURE 2. Activities of lactate dehydrogenase and pyruvate kinase assayed at 1 atm in white skeletal
muscle of marine teleost fishes having different minimal depths of occurrence. Values for the vent zoarcid
ire indicated by the arrow above the point at 2600 m minimal depth of occurrence. Each point
represents a different species, and is based on from one to several individuals. Data are from Childress
and Somero (1979), Sullivan and Somero (1980), Siebenaller and Somero (1982) and Siebenaller et al.,
(1982).
ENERGY METABOLISM OF VENT ANIMALS 177
metabolism correlate well with rates of oxygen consumption in marine fishes, and
the generality of this relationship is further suggested by several other studies of
activities of enzymes of energy metabolism in organisms having widely different
metabolic capacities (cf. Simon and Robin, 1972; Sugden and Newsholme, 1973;
Alp et al, 1976; Zammit et al, 1978; Somero and Childress, 1980; Siebenaller and
Somero, 1982). The similarities in amounts of activity of enzymes of energy me-
tabolism in the vent animals and related shallow-living species were noted for the
crustaceans, molluscs, and fishes we compared, and although no shallow living
pogonophorans were available for comparison (most members of this phylum are
endemic to the deep sea; Southward and Southward, 1982), the enzymic activities
found in R. pachyptila also indicate a substantial capacity for energy metabolism.
The vent animals thus contrast sharply with deep-sea animals from non-vent hab-
itats. Animals from non-vent regions in the deep sea have been shown to have
extremely low metabolic rates (Childress, 1975; Smith and Hessler, 1974; Smith,
1978; Torres et al., 1979) and very low amounts of activity of enzymes of energy
metabolism in their tissues (Fig. 2; Childress and Somero, 1979; Sullivan and So-
mero, 1980; Siebenaller and Somero, 1982; Siebenaller et al, 1982). For example,
the activities of LDH in fish locomotory muscle differ by almost three orders of
magnitude between highly active, shallow-living fishes and sluggish deep-sea fishes
(Fig. 2).
The finding that animals from the hydrothermal vent habitat have a high po-
tential for energy metabolism is further evidence that the low temperatures and
elevated hydrostatic pressures of the deep sea are not, in and of themselves, important
factors in selecting for low metabolic rates in deep-sea organisms. At the 21°N site
where the vent species used in this study were collected, pressure was approximately
260 atms (depth of 2600 m), and the temperature of the water in the immediate
vicinity of the animals was below approximately 20°C and, in almost all cases, was
probably within one or two degrees of the ambient bottom water's temperature (near
2°C) (J. J. Childress, personal communication). The Pompeii worm was the only
species likely to experience temperatures much above 2-5 °C, since this polychaete
forms burrows on the sides of white smoker chimneys (Desbruyeres and
Laubier, 1980).
Waters issuing from the vents are rich in HS~, methane, and hydrogen (Edmond
et al, 1982), all of which are energy-rich compounds that can be oxidized by chem-
olithotrophic bacteria. The base of the food chain at the vents is thought to be
bacteria, e.g., sulfide-oxidizing chemoautotrophic bacteria, that occur free-living in
the sea water (Karl et al, 1980), on the surfaces of rocks and animals, and within
certain tissues of R. pachvptila, C. magnifica, and the vent mussel (Cavanaugh et
al, 1981; Felbeck, 1981; 'Felbeck et al, 1981; Felbeck and Somero, 1982; Cavan-
augh, 1983). The existence of primary production by bacteria at the vents may
preclude the vent animals from having to rely significantly on reduced carbon and
nitrogen compounds descending from the surface, a conjecture supported by stable
carbon and nitrogen isotope ratios (Rau and Hedges, 1979; Rau, 198 la, b; Williams
et al, 1981). Although this point remains to be proven, the rates of primary pro-
duction at the vents may be high enough to allow the vent animals to sustain
metabolic rates comparable to those found for animals in food-rich, shallow marine
habitats. High metabolic capacities are noted for vent animals containing bacterial
endosymbionts (R. pachyptila and C. magnifica), and for species that graze on
bacteria or prey on the vent animals. It bears mentioning that one of the two zoarcid
fishes used in this study contained fresh trophosome tissue of R. pachyptila in its
gut (Somero, personal observations).
178 S. C. HAND AND G. N. SOMERO
TABLE I
Inhibition by HS~ of the cytochrome c oxidase systems of vent and non-vent marine invertebrates
Species Inhibition constant (Kj)([HS~]
[enzyme preparation] yielding 50% inhibition)
Bythogrea thermydron
[heart supernatant] 2.0 X 10~9
Riftia pachyptila
[tentacle supernatant] 1.4 X 10 5
[once acid precipitated] 3.5 X 10~6
[twice acid precipitated] 1.8 X 10 6
Mercenaria mercenaria
[heart supernatant] 1.4 x 10 7
Menippe mercenaria
[heart supernatant] 2.0 X 10~7
In all of the vent animals examined except C. magnified the levels of CO activity
present in different tissues suggested a significant capacity for aerobic respiration.
The occurrence of the CO system in animals exposed to HS" concentrations known
to be adequate to completely inhibit respiration (Hydrogen Sulfide, 1979; Powell
and Somero, 1983) suggest that the vent animals, as well as species like S. reidi that
live in other sulfide-rich marine habitats, may have evolved mechanisms for pre-
vention of poisoning by HS of aerobic respiration. We found no evidence of sulfide-
insensitive variants of the CO system in these species. Thus, half-inhibiting con-
centrations of HS~ for the vent species ranged between 2 X 10~9 M (B. thermydron)
and 1.4 X 10~5 M (crude supernatant of tentacle of/?, pachyptila). Concentrations
of HS~ in the vent waters can approach 1 mM (Edmond et al, 1982), albeit HS"
concentrations are much lower in the waters immediately surrounding the animals,
and blood sulfide levels in R. pachyptila of up to 1.1 mM have been found (Arp
and Childress, 1983). Thus, in the absence of mechanisms for preventing HS from
coming into contact with the CO system, there would appear to be a strong likelihood
that aerobic respiration would be sulfide poisoned in the vent animals. In R. pa-
chyptila one possible mechanism for prevention of poisoning of aerobic respiration
by HS" entails essentially quantitative binding of HS to blood-borne sulfide binding
(transport) proteins (Arp and Childress, 1983; Powell and Somero, 1983). The in-
crease in sensitivity of the CO system of tentacle of R. pachyptila to HS with
successive acid precipitation purification steps (Table I) reflects the removal of these
sulfide binding proteins from the system. Thus, even though the CO system of R.
pachyptila displays a somewhat reduced sensitivity to HS~ compared to the other
CO systems studied, we predict that the inherent sensitivities of completely purified
CO systems from all of these animals would be essentially equal.
In addition to sulfide binding proteins that may function both in protection of
respiration and in sulfide transport to bacterial endosymbionts (Arp and Childress,
1983), systems for oxidizing HS" to less toxic, or non-toxic, sulfur metabolites may
be present in the cells of animals from sulfide-rich environments. For example, we
have found high activities of these types of reactions in foot of S. reidi (Powell and
Somero, in prep.). In assays of CO activity that use crude supernatant fractions that
contain sulfide oxidizing enzyme systems as well as CO activity, the K, value obtained
may be artifactually high due to the removal of HS" from the assay solution by the
sulfide oxidizing system. Thus, the Kj values listed in Table I should be viewed as
upper limits to the K, values that would be found in the absence of sulfide binding
ENERGY METABOLISM OF VENT ANIMALS 179
proteins or sulfide oxidizing systems, both of which can effectively reduce the amount
of free HS~ present in the assay medium.
Calyptogena magnified was the only vent species to show marked differences in
metabolic potentials relative to the shallow-living comparison species. Although
tissues of C. magnified had activities of PFK, PK, and MDH that were comparable
to, and often higher than, the corresponding activities in the other bivalve molluscs
examined, levels of CS and CO were extremely low in the vent clam. Thus, the
enzyme profiles of C. magnified are suggestive of a very high reliance on anaerobic
metabolism. In certain marine bivalves a substantial fraction of energy metabolism
occurs via anaerobic pathways even in the presence of oxygen (DeZwann and Wijs-
man, 1976). The diagnostic enzymes for high potentials for the types of anaerobic
schemes common in marine bivalves include MDH, the enzyme showing the highest
activity in adductor and heart muscle of C. magnifica. The basis for this species'
reliance on anaerobic metabolism may be the nature of the microhabitat in which
the clam is found. Calyptogena magnifica at the 21°N study site were almost in-
variably found along cracks in the basaltic seafloor through which sulfide-rich waters
issued (personal observations). The large foot of the clam was sometimes extended
deeply into the crack, and thus was exposed to high concentrations of HS . The
steady flux of high quantities of HS into the clam may preclude the possibility of
detoxifying HS~ by the mechanisms discussed above, and without the means for
preventing contact between HS and the CO system, aerobic respiration is not
possible. It is important to point out, however, that C. magnifica does "respire" in
the sense that the intact symbiosis consumes oxygen at an appreciable rate (Kenneth
L. Smith, Jr., personal communication), as has recently been reported for Calyp-
togena pacified (Childress and Mickel, 1982), which also harbors bacterial endo-
symbionts in its gills (Felbeck el al, 1981). As Childress and Mickel (1982) em-
phasize, caution must be exercised in attempts to attribute specific fractions of
oxygen uptake to the animal's tissues, on the one hand, and the sulfide oxidizing
bacterial endosymbionts, on the other. The very low CO activities found in C
magnifica suggest that by far the larger share of oxygen consumption by the intact
symbiosis may be due to the sulfide oxidizing activities of the endosymbionts.
ACKNOWLEDGMENTS
These studies were supported by National Science Foundation grants OCE80-
2425 1 to GNS, and OCE80-24895 to Dr. Kenneth L. Smith, for support of the
Oasis Expedition. Travel funds to SCH were provided by the University of South-
western Louisiana. We gratefully acknowledge the assistance of expedition leader,
Dr. K. L. Smith, Ms. R. Baldwin, Ms. G. Niles, and the captains and crews of the
research vessels RV New Horizon, RV Lulu, RV Melville, and DSRV Alvin. Dis-
cussions of this work with Drs. J. J. Childress and H. Felbeck are gratefully ac-
knowledged. This is contribution number 1 1 of the Oasis Expedition.
LITERATURE CITED
ALP, P. R., E. A. NEWSHOLME, AND V. A. ZAMMIT. 1976. Activities of citrate synthase and NAD+-
linked and NADP+-linked isocitrate dehydrogenase in muscle from vertebrates and inverte-
brates. Biochem. J. 154: 689-700.
ARP, A., AND J. J. CHILDRESS. 1983. Sulfide binding by the blood of the hydrothermal vent tube worm
Riftia pachyptila. Science 219: 295-297.
CAVANAUGH, C. M. 1983. Symbiotic chemoautotrophic bacteria in marine invertebrates from sulphide-
rich habitats. Nature 302: 58-61.
180 S. C. HAND AND G. N. SOMERO
CAVANAUGH, C. M, S. GARDINER, M. K. JONES, H. W. JANNASCH, AND J. B. WATERBURY. 1981.
Prokaryotic cells in the hydrothermal vent tube worm Riftia pachyptila Jones: possible che-
moautotrophic symbionts. Science 213: 340-342.
CHILDRESS, J. J. 197 1 . Respiratory rate and depth of occurrence of midwater animals. Limnol. Oceanogr.
16: 104-106.
CHILDRESS, J. J. 1975. The respiratory rates of midwater crustaceans as a function of depth of occurrence
and relation to the oxygen minimum layer off southern California. Comp. Biochem. Phvsiol.
50A: 787-799.
CHILDRESS, J. J., AND T. J. MICKEL. 1982. Oxygen and sulfide consumption rates of the vent clam,
Calyptogena pacifica. Mar. Biol. Lett. 3: 73-79.
CHILDRESS, J. J., AND G. N. SOMERO. 1979. Depth-related enzymic activities in muscle, brain and heart
of deep-living pelagic marine teleosts. Mar. Biol. 52: 273-283.
DESBRUYERES, D., AND L. LAUBIER, 1980. Alvinella pompejana gen. sp. nov., ampharetidae aberrant
des sources hydrothermales de la ride Est-Pacifique. Oceanol. Ada 3: 267-274.
DEZWAAN, A., AND T. C. M. WusMAN. 1976. Anaerobic metabolism in bivalvia (Mollusca) Charac-
teristics of anaerobic metabolism. Comp. Biochem. Physiol. 54B: 3 1 3-324.
EDMOND, J. M., K. L. VON DAMM, R. E. McDuFF, AND C. I. MEASURES. 1982. Chemistry of hot
springs on the East Pacific Rise and their effluent dispersal. Nature 297: 187-191.
FELBECK, H. 1981. Chemoautotrophic potential of the hydrothermal vent tube worm, Riftia pachyptila
Jones (Vestimentifera). Science 213: 336-338.
FELBECK, H. 1983. Sulfide oxidation and carbon fixation by the gutless clam Solemya reidi: An animal-
bacteria symbiosis. /. Comp. Physiol. in press.
FELBECK, H., AND G. N. SOMERO. 1982. Primary production in deep-sea hydrothermal vent organisms:
Roles of sulfide-oxidizing bacteria. Trends Biochem. Sci. 7: 201-204.
FELBECK, H., J. J. CHILDRESS, AND G. N. SOMERO. 1981. Calvin-Benson cycle and sulphide oxidation
enzymes in animals from sulphide-rich habitats. Nature 293: 291-293.
HAND, S. C., AND G. N. SOMERO. 1982. Urea and methylamine effects on rabbit muscle phosphofruc-
tokinase. Catalytic stability and aggregation state as a function of pH and temperature. /. Biol.
Chem. 257: 734-741.
HELMERHORST, E., AND G. B. STOKES. 1980. Microcentrifuge desalting: A rapid, quantitative method
for desalting small amounts of protein. Anal. Biochem. 104: 130-135.
HOCHACHKA, P. W. 1980. Living Without Oxygen, Harvard, Cambridge. 181 pp.
Hydrogen Sulfide. 1979. Subcommittee on hydrogen sulfide, Division of Medical Sciences, National
Research Council, University Park Press, Baltimore. 183 pp.
JONES, M. L. 1 98 1 . Riftia pachyptila Jones: Observations on the vestimentiferan worm from the Galapagos
Rift. Science 213: 333-336.
KARL, D. M., C. O. WIRSEN, AND H. W. JANNASCH. 1980. Deep-sea primary production at the Galapagos
hydrothermal vents. Science 207: 1345-1347.
MICKEL, T. J., AND J. J. CHILDRESS. 1982. Effects of temperature, pressure, and oxygen concentration
on the oxygen consumption rate of the hydrothermal vent crab Bvthograea thermydron (Brachy-
ura). Physiol. Zool. 55: 199-207.
PETERSON, G. L. 1977. A simplification of the protein assay method of Lowry el al. which is more
generally applicable. Anal. Biochem. 83: 345-356.
POWELL, M. A., AND G. N. SOMERO. 1983. Blood components prevent sulfide poisoning of respiration
of the hydrothermal vent tube worm Riftia pachyptila. Science 219: 297-299.
RAU, G. H. 198 la. Low 15N/14N in hydrothermal vent animals: ecological implications. Nature 289:
484-485.
RAU, G. H. 1981b. Hydrothermal vent clam and tube worm I3C/I2C: Further evidence of nonphoto-
synthetic food sources. Science 213: 338-340.
RAU, G. H., AND J. I. HEDGES. 1979. Carbon- 13 depletion in a hydrothermal vent mussel: Suggestion
of a chemosynthetic food source. Science 203: 648-649.
SIEBENALLER, J. F., AND G. N. SOMERO. 1982. The maintenance of different enzyme activity levels in
congeneric fishes living at different depths. Physiol. Zool. 55: 171-179.
SIEBENALLER, J. F., G. N. SOMERO, AND R. L. HAEDRICH. 1 982. Biochemical characteristics of macrourid
fishes differing in their depths of distribution. Biol. Bull. 163: 240-249.
SIMON, L. M., AND E. D. ROBIN. 1972. Relative anaerobic glycolytic capacity and pyruvate kinase activity
of rabbit tissues. Int. J. Biochem. 3: 329-332.
SMITH, K. L., JR. 1978. Metabolism of the abyssopelagic rattail Coryphaenoides armatus measured in
situ. Nature 274: 362-364.
SMITH, K. L., JR., AND R. R. HESSLER. 1974. Respiration of benthopelagic fishes: In situ measurements
at 1230 meters. Science 184: 72-73.
SOMF.RO, G. N., AND J. J. CHILDRESS. 1980. A violation of the metabolism-size scaling paradigm:
ENERGY METABOLISM OF VENT ANIMALS 181
Activities of glycolytic enzymes in muscle increase in larger-size fish. Physiol. Zool. 53: 322-
337.
SOMERO, G. N., J. F. SlEBENALLER, AND P. W. HoCHACHKA. 1983. Biochemical and physiological
adaptations of deep-sea animals. (In press) in The Sea, Vol. 8, G. T. Rowe, ed. Wiley, New
York.
SOUTHWARD, A. J., AND E. C. SOUTHWARD. 1982. The role of dissolved organic matter in the nutrition
of deep-sea benthos. Am. Zool. 22: 647-659.
SPIESS, F. N., K. C. MACDONALD, T. ATWATER, R. BALLARD, A. CARRANZA, D. CORDOBA, C. Cox,
V. M. DIAZ GARCIA, J. FRANCHETEAU, J. GUERRERO, J. HAWKINS, R. HAYMON, R. HESSLER,
T. JUTEAU, M. KASTNER, R. LARSON, B. LUYENDYK, J. D. MACDOUGALL, S. MILLER, W.
NORMARK, J. ORCUTT, AND C. RANGIN. 1980. East Pacific rise: Hot springs and geophysical
experiments. Science 207: 1421-1433.
SUGDEN, P. H., AND E. A. NEWSHOLME. 1973. Activities of hexokinase, phosphofructokinase, 3-oxo acid
coenzyme A-transferase and acetoacetylcoenzyme A thiolase in nervous tissue from vertebrates
and invertebrates. Biochem. J. 134: 97-101.
SULLIVAN, K. M., AND G. N. SOMERO. 1980. Enzyme activities of fish skeletal muscle and brain as
influenced by depth of occurrence and habits of feeding and locomotion. Mar. Biol. 60: 91-99.
TORRES, J. J., B. W. BELMAN, AND J. J. CHILDRESS. 1979. Oxygen consumption rates of midwater fishes
off California. Deep-Sea Res. 26A: 185-197.
TUTTLE, J. H., C. O. WIRSEN, AND H. W. JANNASCH. 1983. Microbial activities in the emitted hydro-
thermal waters of the Galapagos rift vents. Mar. Biol. 73: 293-299.
WILLIAMS, P. M., K. L. SMITH, E. M. DRUFFEL, AND T. W. LINICK. 1981. Dietary carbon sources of
mussels and tubeworms from Galapagos hydrothermal vents determined from tissue I4C activity.
Nature 292: 448-449.
YONETANI, T., AND G. RAY. 1965. Studies on cytochrome oxidase. VI. Kinetics of the aerobic oxidation
of ferrocytochrome c by cytochrome oxidase. J. Biol. Chern. 240: 3392-3398.
ZAMMIT, V. A., I. BEIS, AND E. A. NEWSHOLME. 1978. Maximum activities and effects of fructose
bisphosphate on pyruvate kinase from muscles of vertebrates and invertebrates in relation to
the control of glycolysis. Biochem. J. 174: 989-998.
Reference: Biol. Bull. 165: 182-196. (August, 1983)
EFFECTS OF FEEDING, FEEDING HISTORY, AND FOOD
DEPRIVATION ON RESPIRATION AND EXCRETION RATES OF
THE BATHYPELAGIC MYSID GNATHOPHAUSIA INGENS
PAGE HILLER-ADAMS AND JAMES J. CHILDRESS
Marine Science Institute, Department of Biological Sciences, University of California,
Santa Barbara, Santa Barbara, California 93106
ABSTRACT
Groups of the large bathypelagic mysid Gnathophausia ingens were fed at dif-
ferent frequencies for at least three months in the laboratory, then starved for five
weeks or alternately fed and starved over shorter periods of time. Oxygen con-
sumption and ammonia excretion rates were determined before and after feeding
and during starvation. Prolonged differences in the amount of food eaten prior to
starvation affected the animals' initial responses to starvation. In the first 3 weeks,
animals which had been more frequently fed maintained higher respiration and
ammonia excretion rates relative to rates after this time. Animals fed less frequently
maintained stable rates throughout the 5 week period of starvation. After a maxi-
mum of 3 weeks, starved individuals relied largely on nonnitrogenous energy stores,
presumably lipids, regardless of feeding frequency prior to starvation. The high lipid
content of G. ingens and the low metabolic rate of starved individuals are advan-
tageous for life in the energy-poor deep-sea.
We have observed transient postfeeding increases in respiration and excretion
rates. Excretion rate (E, in micromoles NH3/h) increased with amount eaten (F, in
mg ash-free dry weight of food) (E == 0.038F). Respiration rate (R, in micromoles
O2/h) increased with excretion rate (R == 1 .40 + 1 .03E). Measurements of respiration
and excretion rates using postdigestive individuals of G. ingens therefore underes-
timate average field rates by an amount proportional to food intake. The energetic
effects of feeding on the metabolism of G. ingens are not negligible. We estimate
that about 29% of the energy in the laboratory ration ingested by G. ingens is
expended in the postfeeding increase in respiration.
INTRODUCTION
The dramatic decrease in biomass of the world's oceans with increasing depth,
and the generally patchy distribution of animals living in the water column, suggest
that food scarcity is one of the most physiologically important characteristics of the
deep sea. Food scarcity may in part be responsible for characteristics of the chemical
composition (Childress and Nygaard, 1973, 1974), metabolic rates (Childress, 1971a,
1975; Smith and Hessler, 1974; Torres et al, 1979), and life histories (Childress and
Price, 1978; Childress et al., 1980) of deep-sea animals. Information on the metabolic
responses of deep-sea animals to feeding and to food deprivation should therefore
add to our knowledge of the physiological and energetic adaptations of deep-sea
animals to their environment. The responses may also be compared to those of
shallow water animals which live in an environment which is physically more vari-
able. This paper presents the results of an investigation of the effects of feeding, food
Received 21 June 1982; accepted 25 May 1983.
182
METABOLISM OF A BATHYPELAGIC MYSID 183
deprivation, and feeding history on the respiration and excretion rates of the large
bathypelagic mysid Gnathophausia ingens Dohrn. We have investigated the effects
of feeding frequency prior to prolonged periods of starvation on respiration and
excretion rates during starvation and on the substrates metabolized during starva-
tion. We have also quantified the relationships between food intake and transient
postfeeding increases in respiration and excretion by this species, and estimated the
energetic importance to G. ingens of these increases.
G. ingens is well-suited to such a study since individuals can be obtained in
relatively large numbers off the coast of southern California, and may be maintained
for relatively long periods in the laboratory (up to 2.5 years: Childress and Price,
1978). The life history of G. ingens is well known (Childress and Price, 1978).
Females brood their young at depths between 900 and 1400 m. Newly released
young ascend to depths of about 175-300 m. On reaching the fifth instar (carapace
length between 14.2 and 17.7 mm: Childress and Price, 1978) they descend to depths
of 650-750 m, dispersing at night to depths of 400-900 m. After the fifth instar,
individuals live permanently beneath the photic zone.
MATERIALS AND METHODS
Animal capture and maintenance
Individuals of Gnathophausia ingens were captured in San Clemente and San
Nicholas Basins off the coast of southern California during January and April 1979,
using an opening and closing 3.3 m X 3.3 m Tucker trawl equipped with a thermally
protecting cod-end (Childress et ai, 1978). We removed live individuals from the
cod-end as soon as it arrived on deck, wrapped them loosely in nylon mesh, placed
them in 1 gallon jars filled with sea water, and maintained them at approximately
5°C. On our return to the laboratory, each animal was unwrapped and put in a
container of about 700 ml of chilled sea water, and placed in the laboratory cold
room (5.5°C). The room was dark except for occasional short (several minutes)
periods during the day when laboratory personnel entered. After two weeks in cap-
tivity, each animal was assigned a number and all the animals were alternately fed
salmon muscle and ridgeback shrimp (Sicyonia incertus) tails according to the feed-
ing regimes described below. Maintenance water was changed once every 2-3 weeks,
and after each feeding.
All animals were maintained at atmospheric pressure. We believe that this does
not bias the results since this species can live and grow in the laboratory for periods
of up to 2.5 years at atmospheric pressure (Childress and Price, 1978). Further,
research in this laboratory has shown that the respiration rates and activity of this
and other midwater species are relatively unaffected by a pressure of 1 atm (Meek
and Childress, 1973; Mickel and Childress, 1982). Our studies have also shown that
G. ingens swim continuously at a rather fixed rate which does not decline in captivity
(Quetin et ai, 1978; Quetin and Childress, 1980; Mickel and Childress, 1982). Our
experience with this mysid therefore leads us to believe that our results are reasonably
representative of the field situation.
Feeding
Individuals were fed either 6X/mo for 3.5 months 0'6X/mo, 1st series"), 6X/
mo for 5 months (tl6X/mo, 2nd series"), 1 X/mo for 3-4 months (ct 1 X/mo"), or 6X/
mo for 2 months followed by 3 months of feeding 2X/mo ("2X/mo"). Individuals
were not offered food for two weeks after they had molted since they generally do
184 P. HILLER-ADAMS AND J. J. CHILDRESS
not accept food prior to this time. Feedings within each group were evenly spaced
over time. Experiments in which respiration and excretion rates were determined
began at the end of these feeding periods. G. ingens were fed shrimp during the
experiments, except for two 1 X/mo individuals which were fed salmon at the start
of the experiment using 1 X/mo animals. Only shrimp meals are considered in the
analysis of postfeeding increases in respiration and excretion since all but two of
the metabolic measurements were made after the animals had fed on shrimp.
To feed the animals, and to determine the amount eaten by each individual, a
small preweighed piece of food was held near each animal until it grasped the piece
with its pereiopods. Four to six hours later the remaining food was removed with
forceps and with a perforated spoon which facilitated the removal of smaller pieces
of food. All visible pieces of food were removed, and the water in each animal's
container was replaced with fresh, chilled (5.5°C) sea water. The food which had
been removed was placed in tared pans, dried to constant weight at 60 °C and
weighed (dry weight), then ashed to constant weight at 500°C and weighed again
(ash weight). The difference between the dry weight and ash weight constituted the
ash-free dry weight (AFDW) of the uneaten food. For each feeding, two preweighed
pieces of food were placed in sea water without an animal for the duration of the
feeding period and were similarly dried and ashed. These pieces served as controls
for the loss of material from the food due to immersion. The AFDW available to
the animals was estimated by multiplying the fresh weight of each piece fed to an
animal by a conversion factor which was the average ratio of AFDW/fresh weight
determined from the two control pieces. The ash-free dry weight of the food eaten
by each individual could then be calculated as the difference between the "available"
AFDW and final AFDW of its food.
We chose AFDW for quantification of food eaten because it is a measure of total
organic matter and as such is a better estimator of food value than wet or dry weight,
each of which include substantial amounts of inorganic material. Use of AFDW
also avoids the complication of variable amounts of salt water on the surface of the
left-over food. The method which we have used is a way to approximate the actual
ingestion since some additional material may be leaked from the food during external
chewing. The error from this source is probably minor since these animals typically
ingest small pieces of food immediately after removing them from the main chunk.
The salmon and shrimp meals were the only significant sources of food for G.
ingens since the species is not suited for filtering fine particles from the water (setae
on the pereiopods are sparse), and since several studies have failed to detect signif-
icant uptake of dissolved amino acids by aquatic crustaceans (Stephens and Schinske,
1961; Stephens, 1972; Ferguson, 1982).
Protocol for respiration and excretion measurements
In a typical experiment, animals were removed from their open maintenance
containers and placed in individual one-liter flasks containing sea water (5.5°C) to
which 50 mg/1 each of streptomycin and neomycin had been added. The sea water
had been filtered either through a 0.45 micron membrane filter or through glass
wool. The flasks, including two control flasks which contained only sea water and
antibiotics, were closed with rubber stoppers. Care was taken to exclude all air
bubbles. At the end of the experiment water samples were removed for oxygen and
ammonia analyses and the animals were replaced in their maintenance containers.
A>! experiments were conducted in the dark at 5.5°C.
Each experiment lasted from 8.5 to 10 hours. The duration was adjusted to
METABOLISM OF A BATHYPELAGIC MYSID 185
obtain measurable decreases in oxygen contents of the flasks without allowing the
oxygen content to decrease below the level at which respiration rates become de-
pendent on the partial pressure of oxygen (Childress, 1971b; Mickel and
Childress, 1978).
Oxygen
Oxygen was analyzed using standard Winkler techniques (Strickland and Par-
sons, 1972). Flasks were unstoppered at the end of each experiment. A water sample
for analysis of oxygen content was carefully siphoned from each newly opened one-
liter flask into a 125 ml glass-stoppered flask, and Winkler reagents were added.
Sample concentrations of oxygen were corrected for changes in the oxygen content
of the control flasks. These changes were always less than 3% of the starting con-
centrations, which varied between 497 and 692 micromoles O2/l. Oxygen con-
sumption rates determined from duplicate titrations of single samples differed by
an average of less than 2%.
Ammonia
The ammonia content of a 50 ml subsample from each one-liter flask was de-
termined using an ammonia electrode (Orion Research, Inc., Cambridge, MA).
Electrode potentials were converted to ammonia concentrations using an average
of two standard calibration curves, one made immediately before and one imme-
diately after sample analysis. To make each curve, known amounts of a standard
ammonium chloride solution were added successively to 50 ml of sea water made
basic (pH about 1 1 ) with sodium hydroxide. The sea water was continously mixed
with a magnetic stirrer at low speed. After each addition, the electrode was allowed
to stabilize before the electrode potential was recorded. The two curves for an ex-
periment generally differed by less than 0.5 micromole/1 for a given electrode po-
tential in the sample concentration range. Consecutive measurements were repeat-
able to ±4%. Electrode drift was minimized by placing the electrode in sea water
adjusted to pH 1 1 for 30 minutes prior to use. Concentrations of ammonia in the
control flasks did not change detectably during any of the experiments.
Analysis of results
In comparing the physiological responses to food deprivation by animals on
different feeding regimes, we considered only data for those individuals which sur-
vived the initial feeding period and the experimental period in which respiration
and excretion rates were determined, and did not molt during the experimental
period. There were four such 6x/mo (1st series) individuals (6.5-18.6 g), five 1X/
mo individuals (5.5-1 1.8 g), four 6X/mo (2nd series) individuals (5.2-9.2 g), and
two 2x/mo individuals (5.2 and 9.4 g) of an initial 9, 5, 6, and 4 individuals,
respectively. Six of the original 24 animals (25%) died during the initial feeding
period or during the experiment, and 3 (13%) molted and were not considered.
It is important to note that in each experiment in which individuals were starved
we have compared and contrasted the trends in, rather than the levels of, respiration
and excretion since the number of individuals available was small. Consistent dif-
ferences in individual rates which in a small sample might unduly affect comparisons
of means between groups, will not affect comparisons of individuals followed over
a long period of time.
186 P. HILLER-ADAMS AND J. J. CHILDRESS
For analysis of short-term responses to feeding (responses occurring within a few
days after feeding), we considered data from those individuals which did not molt
or die within one week of respiration and excretion measurements (n = 28).
In all cases, data are expressed as a mean ± one standard error of the mean.
RESULTS
Experiments A
Two series of experiments were conducted to determine the effects of a long
(35 day) period of starvation on oxygen consumption and ammonia excretion by
individuals of G. ingens, and to determine the influence of feeding history on these
effects. In the first experiment, oxygen consumption and ammonia excretion rates
of the five 1 X/mo animals were determined prior to feeding. The animals were then
fed on shrimp (n : 3) or salmon (n = 2). Respiration and excretion rates were
determined again 12 hours after the food was removed (Day 1), and periodically
throughout the 35 days of starvation. After 35 days, the animals were fed shrimp,
and respiration and excretion rates were determined 1 2 hours (Day 1 ) and 60 hours
(Day 3) after the food had been removed. In the second series of experiments, food
was withheld for 35 days from animals previously fed 6X/mo (1st series). Respiration
and excretion rates were determined periodically. After 35 days the animals were
fed shrimp. Respiration and excretion rates were determined a final time 12 hours
after removing the food.
Results oj Experiments A
For the reasons stated in "Materials and Methods — Analysis of results", we have
compared and contrasted the trends in rates in respiration and excretion, rather
than the rates themselves.
The mean oxygen consumption rate of animals previously fed 1 X/mo did not
change significantly during 35 days of starvation (Fig. 1): the highest (Day 35) and
lowest (Day 14) mean respiration rates are not significantly different (P > 0.90,
paired /-test).
The mean ammonia excretion rate of these animals increased after the initial
feeding and returned within 5-9 days to the prefeeding level of 0.253 ± 0.043
micromoles NH3/g wet weight/h. By the fourteenth day the mean excretion rate had
stabilized at a lower rate (0.114 ± 0.021 micromoles NH3/g wet weight/h) and
remained stable through the 35th day of starvation. Subsequent feeding again pro-
duced a large increase in mean ammonia excretion rate. The geometric mean atomic
O:N ratio decreased on Day 1 as a result of the large increase in ammonia excretion,
then increased gradually through the third week. Feeding again produced a sharp
decrease in the O:N value.
The mean oxygen consumption rate of animals previously fed 6X/mo was stable
through the twenty-first day of starvation, then decreased through the 35th day (Fig.
1). The final mean rate of 1.28 ± 0.20 micromoles O2/g wet weight/h was 64% of,
and significantly lower than, the initial post-digestive (Day 3) rate of 1.87 ± 0.17
micromoles O2/g wet weight/h (P < 0.05, paired Mest). Subsequent feeding increased
the mean oxygen consumption rate measured on Day 1 to 1.51 ±0.18 micromoles
O2/g wet weight/h.
Ammonia excretion rates (Fig. 1 ) dropped rapidly in the first 5 days after feeding.
The mean rate dropped again between the third and fourth weeks, and did not
change significantly in the fifth week. Subsequent feeding produced a large increase
METABOLISM OF A BATHYPELAGIC MYSID
187
3.00
2.50
2.00
1.50
0
1.00
N
0.500
0
O:N
60 -
40 -
20 -
0
6x/mo
1x/mo
L_L
1 3 5
t
Feeding
_L
_L
10
15 20
25 30 35 1 4
t
Feeding
DAYS OF STARVATION
FIGURE 1. Oxygen consumption rates (R), ammonia excretion rates (N) and atomic O:N ratios of
individuals previously fed either 6x/mo (n = 4) or ix/mo (n = 5), during five weeks of starvation.
Oxygen consumption and ammonia excretion rates are expressed in Mmoles/g wet weight h~'. Mean ±
standard error.
in mean ammonia excretion rate on Day 1 . The geometric mean atomic O:N ratio
was stable at values of 24-26 from Day 3 through the third week of starvation and
increased significantly in the fourth and fifth weeks (P < 0.002 for both weeks, when
compared with Day 2 1 , paired /-test).
The early response to starvation therefore differed between the two groups. The
respiration and excretion rates of animals previously fed 6X/mo were higher in the
first three weeks of starvation relative to rates after 35 days of starvation. The
respiration rates of animals previously fed 1 X/mo were more stable, and ammonia
excretion rates stabilized within 2 weeks. The slower stabilization of the excretion
rates of the 1 X/mo animals, which caused a gradual increase in the O:N ratio, is
probably due to the fact that, on the average, IX/mo animals ate 2.5 times more
per gram of animal weight at the initial meal than did animals fed 6x/mo.
188 P. HILLER-ADAMS AND J. J. CHILDRESS
Experiments B
Two series of experiments were conducted to determine the effects of feeding,
feeding history, and food deprivation over shorter periods of time, on respiration
and excretion rates. Respiration and excretion rates of animals previously fed 6X/
mo (2nd series) or 2X/mo were measured periodically during two sequential periods
of food deprivation, the first lasting 12 days and the second for 10 days. The ex-
periments in each period began with individual measurements of rates made a few
days before feeding. At the end of the 10 day fast, the animals were fed again and
respiration and excretion rates determined 12 hours (Day 1) and 60 hours (Day 3)
after food removal.
Results of Experiments B
The mean oxygen consumption rates of the two 2X/mo individuals were higher
than those of the 6X/mo individuals in the first 12 day period (Fig. 2). The two
groups did not otherwise differ in their responses to feeding or to this period of
starvation, although a larger sample size is needed before conclusions can be drawn
concerning differences between effects of these feeding frequencies on responses to
short-term starvation.
The data indicate, however, that feeding induces transient increases in respiration
and excretion rates in individuals of both groups. Mean oxygen consumption rates
and ammonia excretion rates increased immediately after each feeding, then de-
creased to or below pre feeding levels sometime between Day 1 and Day 3 to Day
7. Feeding restored respiration and excretion rates to pre-starvation levels. However,
the feeding "peaks" only briefly interrupted the continuing decreases in ammonia
excretion rates in both groups. Consequently, the O:N ratios for individuals in both
groups generally decreased on Day 1, indicating protein metabolism, and increased
between feedings, indicating increasing reliance on non-nitrogenous energy sources.
The more rapid increase in O:N ratios during the second period of starvation suggests
that the metabolic shift to non-nitrogenous compounds occurred more quickly.
Meal size and animal weight
The amount of shrimp eaten at each feeding increased with increasing animal
weight (Fig. 3). Meals eaten by an individual within a week before molting or dying
and the first meal eaten after molting are omitted since the former are often small
due to softening of the exoskeleton, and the latter are sometimes atypically large,
probably due to the two week fast imposed after molting. The regression includes
only data on the average meal sizes of animals which had been fed at least two
quantified shrimp meals and were not affected by molting or death. There was no
significant difference between the regressions for the three feeding regimes (P > 0.75,
F-test). The overall regression of average meal size on animal wet weight is M
= 33.59 + 1 1.32W (r = 0.70, n = 14), where M = average mg AFDW eaten/meal,
and W = wet weight of G. ingens, in grams.
Feeding "peaks"
We determined the correlation between increases in ammonia excretion rates
and oxygen consumption rates which often followed feeding, and between the in-
creases in each rate and the amount of food eaten, using the 28 data sets which
satisfied the following criteria: ( 1 ) oxygen consumption and ammonia excretion rates
METABOLISM OF A BATHYPELAGIC MYSID
189
o 6x/mo
0 2x/mo
1.00
0
^r | i i
1.00
N
0.500
O:N
0
60
40
20
0
i i i
i MI
i i i
M I
J I
1 3
9 121 3 7 101 3
t t i
Feeding Feeding Feeding
DAYS OF STARVATION
FIGURE 2. Oxygen consumption rates (R), ammonia excretion rates (N) and atomic O:N ratios of
individuals which were alternately fed and starved. Individuals had previously been fed 6X/mo (n = 4)
or 2x/mo (n = 2). Oxygen consumption and ammonia excretion rates are expressed in ^m/g wet weight
h~'. Mean ± standard error.
were determined within two days before the animal fed (prefeeding rates), and again
on Day 1(12 hours) and Day 3 or 4 (60 and 84 hours, respectively) after feeding
ceased (postfeeding rates); (2) the individual fed on shrimp, and (3) did not molt
or die within one week after the meal. A "peak" was considered to have occurred
in respiration or excretion if the rate measured on Day 1 was greater than both the
prefeeding and postfeeding rates. The magnitude of the peak was calculated as the
difference between the rate measured on Day 1 and the average of the pre- and post-
190
P. HILLER-ADAMS AND J. J. CHILDRESS
250
O)
g>
"<D
£
T>
0)
200
150
w 100
LU
N
c/) 50
III
0
o 6x/mo
A 1x/mo
0 2x/mo
4 8 12
ANIMAL SIZE (wet weight, g)
FIGURE 3. Meal size as a function of animal weight. Individuals were fed ix/mo, 2X/mo or 6X/
mo. Numbers in parentheses indicate number of shrimp meals contributing to the data point (mean
± standard error). Meals eaten by an individual within a week before molting or dying, and the first
meal eaten after molting, are omitted, y = 33.59 + 1 1.32X (r = 0.70).
feeding rates. Peaks in ammonia excretion occurred in all 28 data sets, which in-
volved 12 animals in the 6X/mo (2nd series), IX/mo and 2X/mo groups. Peaks
occurred in oxygen consumption in 71% of the sets (20 of 28).
The linear regressions of increase in ammonia excretion rate (E, in micromoles
NH3/h) on amount eaten (F, in mg AFDW) did not differ between the three regimes
(Fig. 4; P > 0.25, F-test). The regression is E = 0.038F if it is constrained to pass
through the origin, i.e., if one assumes that the peaks in excretion are due to feeding.
This regression does not differ significantly from the unconstrained regression (P
> 0.10, F-test). Since 75% of the ash-free dry weight of the shrimp fed to G. ingens
is protein (Childress, unpublished data), the regression of the increase in ammonia
excretion (12 hours after feeding) on protein ingested (P, in mg) is E = 0.05 IP. The
correlation between animal wet weight (y, in grams) and the magnitude of the
postfeeding ammonia peak (x, in micromoles NH3/h) is y = -4.98 + 10.44X (P
< 0.00 1 , Mest for significant slope). This is consistent with the positive correlation
between meal size and the weight of G. ingens.
Although there was no significant linear correlation between the magnitudes of
post-feeding oxygen peaks and the amount of food eaten (P > 0.10, t-test for sig-
nificant slope), the probability of a peak in oxygen consumption occurring after
feeding was much higher than the probability of a peak occurring in a set of three
sequential measurements not separated by a meal (P < 0.005, X2 test on a 2 X 2
contingency table). It is likely that measurements made over a longer period of time,
or at a different point in the time course of the change in respiration rate, would
have revealed a correlation between the magnitude of the peak in respiration rate
and the amount of food eaten, since there is a significant positive correlation between
the increases in rates of respiration and ammonia excretion (Fig. 5). The least-
METABOLISM OF A BATHYPELAGIC MYSID
191
squares linear regression of the increase in oxygen consumption (R, in micromoles
OVh) on the increase in ammonia excretion (E, in micromoles NH3/h) is R =: 1.396
+ 1.027E(r == 0.74, n == 20).
DISCUSSION
The data obtained for individuals starved for 35 days indicate that the respiration
rates of previously well-fed individuals are stable for several weeks of food depri-
vation before beginning to decrease. This suggests that activity levels of animals
which have fed frequently prior to food deprivation may also be maintained during
this period. A similar transient "plateau" in respiration rate, followed by a rate
decrease, has previously been observed for two benthic, shallow water crustacean
species during starvation (Wallace, 1973; Regnault, 1981). The stable mean respi-
ration rate of the IX/mo animals suggests that the respiration rates of the 6x/mo
individuals would have stabilized at a lower level.
9.0 i-
o 6x/mo
A 1x/mo
0 2x/mo
LLJ
rr
O
X
111
Z
O
LLJ
C/)
<
LU
cc
O
Z
150 200 250
MEAL SIZE
(ash-free dry weight, mg)
FIGURE 4. Postfeeding increase in ammonia excretion rate as a function of meal size. Shrimp were
used for all meals. The increase in excretion rate is the difference between the rate measured 12 hours
after feeding and the average of the rates measured before and 3-4 days after feeding. The regression is
constrained to pass through the origin, y = 0.038X.
192
P. HILLER-ADAMS AND J. J. CHILDRESS
_ 11.0
.c
CO
"o
E
g
a.
CO
Z
O
o
z
LU
x
O
UJ
CO
<
111
rr
o
r- O 6x/mo
A 1x/mo
0 2x/mo
9.0
7.0
5.0
3.0 -
1.0-
>
0
I
I
0 1.0 3.0 5.0 7.0 9.0
INCREASE IN NH3 EXCRETION (umoles/h )
FIGURE 5. Postfeeding increase in oxygen consumption rate as a function of the postfeeding increase
in ammonia excretion rate. The increase in each rate is the difference between the rate measured 12 hours
after feeding and the average of the rates measured before and 3-4 days after feeding, y = 1.40 + 1.03X
(r = 0.74).
The mean O:N ratios of the 6x/mo (1st series) individuals suggest that, when
first deprived of food, previously well-fed individuals of G. ingens metabolize a large
proportion of protein or other nitrogen-containing compounds relative to lipids and
carbohydrates. The increase in the mean O:N ratios of starved individuals, which
was also observed by Quetin el al. (1980), indicates that more lipid and/or carbo-
hydrate is oxidized by the animals as the length of time without food increases.
Lipids are probably a more important energy source for G. ingens during starvation
than are carbohydrates since lipid typically comprises about 45%, and carbohydrate
only about 0.6%, of the ash-free dry weight of an individual (Childress and
Nygaard, 1974).
The responses of G. ingens to starvation are not unusual. Rates of respiration
(Wallace, 1973; Mayzaud, 1976;Regnault, 1981) and ammonia excretion (Mayzaud,
1976) of a number of shallow water benthic and epipelagic crustacean species also
decrease during starvation. This is not surprising since reduced energy expenditures
are potentially advantageous during a prolonged period without food. Many inver-
tebrate species appear to rely primarily on large lipid and/or carbohydrate reserves
during starvation (Conover, 1964; Chaisemartin, 1971), while some species metab-
ize protein concurrently with lipid, as G. ingens appears to do (Schafer, 1968;
Chaisemartin, 1971; Ikeda, 1971). A few species soon rely primarily on protein
METABOLISM OF A BATHYPELAGIC MYSID 193
metabolism when their smaller lipid and carbohydrate reserves have been exhausted
(Cowey and Corner, 1963; Regnault, 1981).
Two factors suggest that G. ingens is better able to survive prolonged periods
without food than are many shallow water species. The metabolic rates even of
freshly-captured individuals of G. ingens and of other midwater animals are lower
than those of shallower-dwelling marine animals (Childress 197 la, 1975), and the
lipid content of G. ingens and of many midwater crustaceans is higher than that of
many species living in shallower water (Childress and Nygaard, 1974). Considered
together, the high lipid content, low metabolic rate, and similarity in response of
individuals of G. ingens to starvation regardless of feeding history suggest that G.
ingens and perhaps other midwater crustaceans are able to survive relatively long
periods of food deprivation. This capability may be essential in a food-poor or patchy
environment.
The similarities in the responses to starvation by all four groups of individuals
suggest that the responses of field individuals to artificially imposed starvation will
be similar in all seasons, and that the responses of individuals fed and starved in
the laboratory are similar to the responses of field individuals confronted with natural
variations in food availability.
The elevation in respiration rate after an animal has ceased feeding has been
referred to as the "specific dynamic action," "calorigenic effect," and "heat incre-
ment" of food. Although protein ingestion appears to produce a greater increase in
respiration rate than does ingestion of lipid or carbohydrate, the effect of each varies
with the total composition of the food (Forbes and Swift, 1944). The importance
of postfeeding increases in oxygen consumption to a valid estimation of a species'
energy budget depends on the magnitude of the increases, which, for G. ingens,
depends on meal size and feeding frequency. An estimate of the potential energetic
importance of these peaks to G. ingens may be obtained by combining data on the
probable average daily caloric intake by an individual with data on the caloric
content of the ridgeback shrimp on which G. ingens was fed (0.515 cal/mg AFDW:
Childress, unpublished data). Hiller- Adams (1982) has estimated that a 2.8 g instar
7 individual of G. ingens requires about 32 calories/day for growth and metabolism.
This is equivalent to 1 56 mg AFDW of ridgeback shrimp every 60 hours. [The data
on which the caloric requirement is based were obtained using individuals in instar
7; 2.8 grams is the mean weight of individuals in this instar (Childress and Price,
1978).] This would produce an ammonia peak of 5.9 micromoles/h (Fig. 4) and an
increase in respiration of 7.5 micromoles 02/h (Fig. 5). Since the height of the
postfeeding peak was calculated as the difference between the 1 2-hour postfeeding
rate and the average of the prefeeding and 60-hour (or 84-hour) postfeeding rates,
we assume that the additional oxygen respired may be approximated by the area
of the triangle bounded by the peak and the average of the prefeeding and postfeeding
rates. The "base" of the triangle is then 60 hours long, and the area is (30 hours)
X (peak height, in micromoles O2/h), or in this case, 225.3 micromoles O2. If one
assumes that a mmole of oxygen consumed represents an energy expenditure of
103.7 calories (Brett and Groves, 1979), 23.4 calories are expended in the postfeeding
increase in respiration. This represents 29% of the calories ingested, and an 85%
increase in the estimated average respiration rate of 4.4 micromoles O2/h for an
animal of this size. (The average respiration rate is calculated from the emperical
relationship between size and respiration rate in Childress, 197 Ib). Since the protein
content of fish and crustaceans which live in shallower water (and the shrimp on
which G. ingens were fed in the laboratory) tends to be higher than that of pelagic
midwater species (Childress and Nygaard, 1937, 1974), and since protein may cause
194 P. HILLER-ADAMS AND J. J. CHILDRESS
a relatively large portion of the postfeeding increases, we expect that the increases
which occur in nature are somewhat smaller than those we have measured in the
laboratory.
The percent increase in respiration which we have observed appears to be con-
siderably larger than the 1 7-37% increase reported for juvenile lobsters (Capuzzo
and Lancaster, 1979), and the 7-40% increase reported for juvenile Macrobrachium
rosenbergii (Nelson et al, \ 977). The two species were fed several diets which differed
in composition. These are the only other crustaceans of which we are aware for
which quantitative data have been published. However, we determined the duration
of the postfeeding increases. This was not determined in the previous two studies
(postfeeding measurements were made within 24 hours of feeding), and might well
affect estimates of the energetic importance of the postfeeding increases. Addition-
ally, the percent increase for G. ingens is calculated relative to a postdigestive met-
abolic rate which is quite low relative to that of other crustaceans (Childress, 197 la).
Among fish, "specific dynamic activity" accounted for 4-45% of the energy ingested
by young coho salmon (Averett, quoted by Warren, 1971), 14 ±4% (mean
± standard deviation) for largemouth bass (Beamish, 1974), and 5-24% for bluegill
sunfish (Pierce and Wissing, 1974).
Data regarding the magnitude of the postfeeding increase in oxygen consumption
by animals have often been expressed as a percent of the postdigestive metabolic
rate in determining correlations with other parameters. However, since the increase
in respiration is due to the food eaten by the animal it is not dependent on, and
so should not be expressed (as many workers have) as a function of an animal's
postdigestive metabolic rate in determining correlations. In addition, the data are
more useful expressed as increases in the absolute amount of oxygen consumed, as
we have done, rather than as a percent of postdigestive metabolic rate since the
usual size-dependency of metabolic rates (larger animals tend to respire at a lower
weight-specific rate) prevents the extrapolation of percent data to animals and meals
of other sizes.
It should also be noted that our data indicate that metabolic measurements
made within a day or two after feeding may not represent postdigestive rates, as has
often been assumed for other species. Up to a week may be required before G.
ingens achieves predigestive rates after having fed. Other investigators, working with
teleosts (Beamish, 1974, Pierce and Wissing, 1974) have also found that several days
may be required before animals are postdigestive.
The transient increase in ammonia excretion rates of recently fed individuals
suggests that a significant portion of the amino acids in protein assimilated by G.
ingens is rapidly deaminated. If one assumes that ( 1 )75% of the AFDW of the shrimp
food is protein (Childress and Price, unpublished data), (2) the protein is 16% ni-
trogen by weight (Kleiber, 1961), and (3) the total quantity of ammonia excreted
as a result of feeding is the area of the triangle bounded by the postfeeding peak
and the average of the prefeeding and 60-hour postfeeding rates, then G. ingens
deaminated about 1 1% of the shrimp protein ingested within 60 hours of feeding.
The remainder of the ingested protein apparently is incorporated into the
animal's body.
The pronounced increases in respiration and excretion by G. ingens after feeding
indicate that average respiration and excretion rates of individuals in the midwater
environment depend strongly on ingestion. Post-digestive respiration and excretion
rates may considerably underestimate average field rates of a species when food
intake in the ocean is high.
METABOLISM OF A BATHYPELAGIC MYSID 195
In summary, G. ingens appears to rely largely on protein or other nitrogen-
containing compounds when first deprived of food. Lipid reserves become more
important to previously well-fed individuals after about three weeks without food.
Respiration and excretion rates begin to decrease at this time. The stability of the
respiration rate of 1 X/mo individuals suggests that respiration stabilizes at a lower
level. Transient postfeeding increases in respiration and excretion indicate that
measurements of respiration and excretion rates using postdigestive animals un-
derestimate the average rates in nature by an amount directly proportional to food
intake. The energy expended in increased respiration is not negligible: about 29%
of the caloric value of the ingested laboratory ration may be expended in postfeeding
increases in respiration by an instar 7 individual.
ACKNOWLEDGMENTS
The smooth operation of the RV Velero IV by its captain and crew facilitated
capture of the animals. The research was made possible by grants OCE 78-08933
and OCE 81-10154 to Dr. J. J. Childress.
LITERATURE CITED
BEAMISH, F. W. H. 1974. Apparent specific dynamic action of largemouth bass, Micropterus salmoides.
J. Fish. Res. Board Can. 31: 1763-1769.
BRETT, J. R., AND T. D. D. GROVES. 1979. Physiological energetics. Pp. 279-351 in Fish Physiology,
Vol. VIII, W. S. Hoar and D. J. Randall, eds. Academic Press, New York.
CAPUZZO, J. M., AND B. A. LANCASTER. 1979. The effects of dietary carbohydrate levels on protein
utilization in the American lobster. Proc. World Maricul. Soc. 10: 689-700.
CHAISEMARTIN, C. 1971. Mobilisation des reserves de metabolites chez les Astacidae: influences com-
parees de la stabulation a jeun et de certains etats pathologiques. C R. Soc. Biol. Paris 165:
671-676.
CHILDRESS, J. J. 1971a. Respiratory rate and depth of occurrence of midwater animals. Limnol. Oceanogr.
16: 104-106.
CHILDRESS, J. J. 197 lb. Respiratory adaptations to the oxygen minimum layer in the bathypelagic mysid
Gnathophansia ingens. Biol. Bull. 141: 109-121.
CHILDRESS, J. J. 1975. The respiratory rates of midwater crustaceans as a function of depth of occurrence
and relation to the oxygen minimum layer off southern California. Comp. Biochem. Physiol.
50A: 787-799.
CHILDRESS, J. J., AND M. H. NYGAARD. 1973. The chemical composition of midwater fishes as a function
of depth of occurrence off southern California. Deep-Sea Res. 20: 1093-1 109.
CHILDRESS, J. J., AND M. H. NYGAARD. 1974. Chemical composition and buoyancy of midwater crus-
taceans as a function of depth of occurrence off southern California. Mar. Biol. 27: 225-238.
CHILDRESS, J. J., AND M. H. PRICE. 1978. Growth rate of the bathypelagic crustacean Gnathophausia
ingens (Mysidacea: Lophogastridae). I. Dimensional growth and population structure. Mar.
Biol. 50: 47-62.
CHILDRESS, J. J., A. T. BARNES, L. B. QUETIN, AND B. ROBISON. 1978. Thermally protecting cod ends
for the recovery of living deep-sea animals. Deep-Sea Res. 25: 419-422.
CHILDRESS, J. J., S. M. TAYLOR, G. M. CAILLIET, AND M. H. PRICE. 1980. Patterns of growth, energy
utilization and reproduction in some meso- and bathypelagic fishes of southern California. Mar.
Biol. 61: 27-40.
CONOVER, R. J. 1964. Food relations and nutrition of zooplankton. Occ. Pubi, Narragansett mar. lab
2: 81-91.
COWEY, C. B., AND E. D. S. CORNER. 1963. Amino acids and some other nitrogenous compounds in
Calanus finmarchicus. J. Mar. Biol. Assoc. U. K. 43: 483-493.
FERGUSON, J. C. 1982. A comparative study of the net metabolic benefits derived from the uptake and
release of free amino acids by marine invertebrates. Biol. Bull. 162: 1-17.
196 P. HILLER-ADAMS AND J. J. CHILDRESS
FORBES, E. B., AND R. W. SWIFT. 1944. Associative dynamic effects of protein, carbohydrate and fat.
J. Nutr. 27: 453-468.
HILLER-ADAMS, P. 1982. Effects of feeding and food deprivation on the bathypelagic mysid
Gnathophausia ingens. PhD dissertation, University of California, Santa Barbara.
IK.EDA, T. 1971. Changes in respiration rate and in composition of organic matter in Calanus cristatus
(Crustacea Copepoda) under starvation. Bull. Fac. Fish. Hokkaido Univ. 21: 280-298.
KLEIBER, M. 1961. The Fire of Life. John Wiley and Sons, New York. 454 pp.
MAYZAUD, P. 1976. Respiration and nitrogen excretion of zooplankton. IV. The influence of starvation
on the metabolism and the biochemical composition of some species. Mar. Biol. 37: 47-58.
MEEK., R. P., AND J. J. CHILDRESS. 1973. Respiration and the effect of pressure in the mesopelagic fish
Anoplogaster cornutus (Beryciformes). Deep-Sea Res. 20: 1 1 1 1-1 1 18.
MICKEL, T. J., AND J. J. CHILDRESS. 1978. The effect of pH on oxygen consumption and activity in the
bathypelagic mysid Gnathophausia ingens. Biol. Bull. 154: 138-147.
MICKEL, T. J., AND J. J. CHILDRESS. 1982. Effect of pressure and pressure acclimation on activity and
oxygen consumption in the bathypelagic mysid Gnathophausia ingens. Deep-Sea Res. 29: 1293-
1301.
NELSON, S. G., A. W. KNIGHT, AND H. W. Li. 1977. The metabolic cost of food utilization and ammonia
production by juvenile Maerobrachium rosenbergii(Cruslacea: Palaemonidae). Comp. Biochem.
Physiol. 57A: 67-72.
PIERCE, R. J., AND T. E. WISSING. 1974. Energetic cost of food utilization in the bluegill (Lepomis
macrochirus). Trans. Am. Fish. Soc. 103: 38-45.
QUETIN, L. B., T. J. MICKEL, AND J. J. CHILDRESS. 1978. A method of simultaneously measuring the
oxygen consumption and activity of pelagic crustaceans. Comp. Biochem. Physiol. 59A: 263-
266.
QUETIN, L. B., AND J. J. CHILDRESS. 1980. Observation on the swimming activity of two bathypelagic
mysid species maintained at high hydrostatic pressures. Deep-Sea Res. 27A: 383-391.
QUETIN, L. B., R. M. Ross, AND K. UCHIO. 1980. Metabolic characteristics of midwater zooplankton:
ammonia excretion O:N ratios, and the effect of starvation. Mar. Biol. 59: 201-209.
REGNAULT, M. 1981. Respiration and ammonia excretion of the shrimp Crangon crangon L.: metabolic
response to prolonged starvation. J. Comp. Physiol. 141: 549-555.
SCHAFER, H. J. 1968. Storage materials utilized by starved pink shrimp, Penaeus duorarus Burkenroad.
Proc. World Sci. Conf. Biol. Cult. Shrimps. FAO Fish. Rep. 57: 393-403.
STEPHENS, G. C. 1972. Amino acid accumulation and assimilation in marine organisms. Pp. 155-184
in Nitrogen Metabolism and the Environment, J. W. Campbell and L. Goldstein, eds. Academic
Press, New York.
STEPHENS, G. C., AND R. A. SCHINSK.E. 1961. Uptake of amino acids by marine invertebrates. Limnol.
Oceanog. 6: 175-181.
SMITH, K. L., AND R. R. HESSLER. 1974. Respiration of benthopelagic fishes: in situ measurements at
1230 meters. Science 184: 72-73.
STRICKLAND, J. D. H., AND T. R. PARSONS. 1972. A practical handbook of seawater analysis. Fish. Res.
Board Can. Bull. 167: 310 pp.
TORRES, J. J., B. W. BELMAN, AND J. J. CHILDRESS. 1979. Oxygen consumption rates of midwater fishes
as a function of depth of occurrence. Deep-Sea Res. 26: 185-197.
WALLACE, J. C. 1973. Feeding starvation and metabolic rate in the shore crab Carcinus meanas. Mar.
Biol. 20: 277-281.
WARREN, C. E. 1971. Biology and Water Pollution Control. W. B. Saunders, Philadelphia. 434 pp.
Reference: Biol. Bull. 165: 197-208. (August, 1983)
GRAZING AND PREDATION AS RELATED TO ENERGY
NEEDS OF STAGE I ZOEAE OF THE TANNER CRAB
CHIONOECETES BAIRDI (BRACHYURA, MAJIDAE)*
LEWIS S. INCZE1 •** AND A. J. PAUL2
[School of Fisheries WH-IQ, College oj Ocean and Fishery Sciences, University of
Washington, Seattle, Washington 98195, and2 Institute of Marine Science,
University of Alaska, Seward Marine Center, Seward, Alaska 99664
ABSTRACT
The ability of first-feeding stage I zoea larvae of Chionoecetes bairdi to obtain
energy from phytoplankton was investigated using a range of phytoplankton cell
sizes and cell densities. An early first stage zoea requires approximately 6.8 X 10
calories or 0.60 ^g carbon (approximately 4% body C) per day for metabolic needs
at 5°C. Experiments with dinoflagellates and large centric diatoms demonstrated
that the larvae are capable of capturing and ingesting these cells. However, the zoeae
grazed at rates which satisfied less than 15% of basal metabolic energy requirements
at cell concentrations similar to those prevailing in coastal and shelf sea environ-
ments where the crabs are found. Grazing on smaller cells, including chain-forming
species common in nature, was not detected. In the laboratory, first-feeding zoeae
were capable of consuming zooplankton prey at rates which provided up to 308%
of basal metabolic requirements.
INTRODUCTION
Laboratory studies have demonstrated that availability and nutritional adequacy
of food are among the most important factors affecting survival of crab larvae
(Roberts, 1974; Sulkin, 1975, 1978; Sulkin and Epifanio, 1975; Christiansen and
Yang, 1976; Sulkin and Norman, 1976; Anger and Nair, 1979). Generally, labo-
ratory diets consisting primarily of zooplankton have provided the highest survival
rates (Brick, 1974; Roberts, 1974; Sulkin, 1975, 1978; Bigford, 1978). There is a
high degree of morphological similarity of the feeding appendages of crab larvae and
numerous reports of their attacking single zooplankton prey (Sato and Tanaka,
1949; Knudsen, 1960; Herrnkind, 1968; Gonor and Conor, 1973). This evidence
has led to the widely held belief that phytoplankton is of limited dietary importance.
However, there is evidence that phytoplankton may be a common component of
the diet of some larvae in nature (e.g., LeBour, 1922, 1927). Laboratory studies with
the larvae of a brachyuran crab (Hartman and Letterman, 1978) and a pandalid
shrimp (Stickney and Perkins, 1981) indicated that phytoplankton diets can signif-
icantly prolong the life of these larvae compared to unfed control animals, even
though both larvae showed markedly better survival on zooplankton diets. Both
studies noted that specimens collected at sea contained phytoplankton in the stom-
achs. Roberts (1974) and Sulkin (1975) reported that the larvae of crabs used in
their experiments (an anomuran and brachyuran crab, respectively) consumed phy-
Received 21 January 1983; accepted 25 May 1983.
* Contribution No. 618 of the School of Fisheries, University of Washington and No. 523 of the
Institute of Marine Science, University of Alaska.
** Present address: National Marine Fisheries Service, 2725 Montlake Blvd. East, Seattle, WA 98112.
197
198 L. S. INCZE AND A. J. PAUL
toplankton in laboratory studies and that this prolonged survival slightly compared
to unfed zoeae. In contrast, Atkins (1955) and Bousquette (1980) were able to rear
the larvae of pinnotherid (brachyuran) crabs through all zoeal stages to the megalops
stage with phytoplankton alone, but the authors did not specify survival times of
unfed controls. Despite the large body of evidence that zooplankton are the primary
prey for larvae of most species of crab, it appears that a functional role for phyto-
plankton cannot be ruled out for all species.
In the genus Chionoecetes, both zoeal stages of C. opilio have been successfully
cultured when fed the nauplii of Anemia sp. alone (Motoh, 1973) and combined
with rotifers, Brachionus plicatilis (Kon, 1970, 1979). Using natural prey, Paul et
al. (1979) reported that stage I zoeae of C. bairdi can consistently capture copepods,
copepodids, and copepod nauplii when these prey are offered at densities of 20-40
per liter. However, the requirement of prey concentrations above 20 per liter for
consistent prey capture suggested that there may be times when zooplankton prey
are not present in sufficient numbers to ensure successful feeding by the zoeae.
An alternate, or supplementary, food source might be phytoplankton, which are
more abundant than zooplankton and are perhaps more easily captured. Bright
(1967) reported that the principal stomach contents of Tanner crab zoeae collected
from Cook Inlet, Alaska, were unidentified diatoms; little zooplankton material was
reported. In the southeastern Bering Sea, thecate dinoflagellates were found along
with parts of copepods, pteropods, tintinnids, and other zooplankters in the stomachs
of stage I and stage II zoeae of Chionoecetes spp. (K. O. Coyle, Univ. of Alaska,
pers. comm.). Examining specimens from the same region in later years. Incze and
Armstrong (unpubl. observations) found little evidence of zooplankton in the stom-
achs of stage I and stage II zoeae, but frequently found solitary and chain-forming
centric diatoms. Although these observations suggest that phytoplankton may be
an important component in the diet of these zoeae, the relative value of this material
to these larvae is unknown.
Currently, there exists little information on relationships between type and avail-
ability of prey and feeding success of crab zoeae in the ocean. Consequently, we
know little about this major determinant of larval survival. The objective of this
study is to evaluate the relative value of phytoplankton in supplying the energy
(calories) or material (carbon) needed for maintenance metabolism and growth of
the first feeding stage I zoeae of C. bairdi (for a description of this stage, refer to
Haynes, 1973). First-feeding zoeae were used because they (1) are unaffected by
previous feeding experience and (2) co-occur with the spring phytoplankton bloom
which precedes development of the zooplankton community in Alaskan waters.
MATERIALS AND METHODS
Several egg-bearing female crabs captured near Kodiak, Alaska, were held in
circulating sea water tanks at 4-5 °C. Zoeae hatched continually and tanks were
drained each day before an experiment so that only freshly hatched, actively swim-
ming zoeae were used. All respiration and feeding experiments were conducted at
5°C. Stage I zoeae normally encounter temperatures of 4 to 6°C in Alaskan waters.
Respiration rates of 1 2-hour old stage I zoeae were measured in a glass differential
syringe manometer (Umbreit et al., 1972). The 15 ml respirometer vessel held 4 to
6 unfed zoeae, 6.0 ml of 1 .0 ^m filtered sea water, and 0. 1 ml 20% KOH to absorb
CO2. The active zoeae were acclimated to vessel temperature for one hour before
the manometers were sealed. Observations of oxygen uptake were made after a
minimum of five hours. Shaking of the respirometers was restricted to the last 10
minutes of the final observation. There were eight replications for respiration rate
FEEDING AND ENERGETICS OF CRAB LARVAE 199
(VO2) measurement. Oven dried weights (60°C) of the zoeae from each observation
of VO2 were determined with an electrobalance. Values of VO2 were converted to
calories using the conversion 4.73 X 10"3 cal-/il 02 ' (Brody, 1945). Carbon equiv-
alents of VO2 were calculated using a respiratory quotient (RQ) of 0.90 for phy-
toplankton prey and 0.75 for zooplankton prey and unfed zoeae. Values of RQ
(Giese, 1973) were assigned based on the approximate proportion of carbohydrate,
lipid and protein in phytoplankton (Parsons et al, 1961) compared to zooplankton
(Corner and Cowey, 1968) and Chionoecetes spp. zoeae (Incze, 1983). Adjusting
RQ values to account for the possible range of substrate proportions in phytoplank-
ton and in zooplankton produces minor changes in the estimated carbon equivalents
ofVO2.
Ingestion rates of one day old zoeae were measured using animal and plant prey
in several ways. Rates of ingestion of newly hatched Anemia sp. nauplii (San Fran-
cisco Bay variety) by one day old zoeae were determined by placing 50 nauplii in
500 ml of 1 jim filtered sea water in a lightly aerated 550 ml black plastic beaker
containing five zoeae. The zoeae were allowed to feed for 24 hours in a 12 hour
light, 300 lux: 12 hour dark cycle at 5°C. Three hundred lux was approximately 2%
of light intensity at sea surface during the experiments. Nauplii remaining after 24
hours were counted under a microscope. Forty replicate prey consumption mea-
surements were made. A caloric value of 8.7 X 10~3 cal (calculated from data of
Levine and Sulkin, 1979) and a carbon value of 1 .2 ^g C (present study) per nauplius
were used to estimate the value of the ingested ration. Carbon was measured on a
Perkin-Elmer Model 240 elemental analyzer. Six beakers containing only nauplii
were used to demonstrate that all prey were recovered during subsequent recounts.
An assimilation efficiency of 0.70 (Conover, 1966) was assumed.
Rates of ingestion of various phytoplankton cells by one day old zoeae were
investigated by (1) comparing chlorophyll concentrations in initial, grazed, and
control containers using cultured algae, and (2) counting cells in initial, grazed, and
control containers for selected algal species. Phytoplankton cells representative of
the shape and size of those found in the planktonic environment and stomachs of
C. bairdi zoeae were used. Mono-specific cultures of Phaeodactylum tricornutum,
Chaetoceros compressus, Gonyaulax grindleyi (Prorocentrum reticulatum), and an
unidentified thecate dinoflagellate (referred to here as F16) were among the algae
used. Grazing experiments were also conducted with large centric diatoms (Cosci-
nodiscus spp. and Thalassiosira spp.) removed from plankton samples collected
with a vertical tow using a 44 j*m mesh net. All experiments were conducted at 5°C
in a 12 hour light, 300 lux: 12 hour dark cycle. Experimental containers were placed
on a slowly rotating wheel. The size and shape of phytoplankton cells used in these
experiments are provided in Table I.
Different container sizes were used for cultured versus sorted, natural phyto-
plankton. Grazing experiments and controls using the cultured algae were conducted
in 250 ml translucent polyethylene bottles. Approximately 100 zoeae were placed
in each bottle; accurate counts of zoeae were made at the end of each experiment.
Four algal food concentrations corresponding to chlorophyll concentrations of ap-
proximately 2, 10, 50, and 100 ^g chl a per liter were used. Replicates of all ex-
perimental conditions were run. Gut flourescence of zoeae and chlorophyll con-
centrations of initial, grazed and control media were measured using a Turner Model
1 1 1 fluorometer, media sample volumes up to 200 ml, and extraction volumes of
10 ml (Strickland and Parsons, 1968).
From a separate series of cell count experiments with G. grindleyi cultures, cell
count samples (40-60 ml) were preserved from initial, experimental and control
vessels with 0.2 ml Lugol's solution. Cells were counted using a settling chamber
200 L. S. INCZE AND A. J. PAUL
TABLE I
Approximate size and shape of phytoplankton cells used in grazing experiments with the first zoeae of
Chionoecetes bairdi
Phaeodactylum tricornutum pennate, 7X21 pm
Chaetoceros compressus individual centric cells, 10 ^m diameter; average chain length, 8-1 1
cells, 90-120 ^m; average chain width (with spines), 90
dinoflagellate F16 10 X 15 ^m
Gonyaulax grindleyi 35-45 ^m, roughly symmetrical but irregularly shaped
Conscinodiscus spp. centric, 71 jzm (height) x 222 ^m (diameter)
Thalassiosira spp. centric, 100 nm (height) X 168 nm (diameter)
and an inverted microscope (Utermohl, 1958). A minimum of 400 cells was counted
for each estimate of cell density (Lund el ai, 1 958) and three to four replicate counts
were made of each sample (initial, control, and experimental) at the highest and
lowest cell densities. To determine carbon content of phytoplankton cells, samples
of culture were collected with a 500 n\ Oxford pipette and dispensed in a carefully
measured volume of filtered sea water for cell counts and on pre-combusted Gelman
Type A/E glass fiber filters for CHN analysis. Carbon content of the cells was mea-
sured using a Perkin-Elmer Model 240 elemental analyzer.
Grazing experiments and controls using Coscinodiscus and Thalassiosira cells
removed from natural phytoplankton were conducted in 60 ml transluscent plastic
bottles containing five zoeae each. Cells were individually counted before and after
a 24 hour feeding period. Cell carbon content was estimated from cell volume
according to the method of Strathmann (1967).
Average cell concentrations during each experiment were calculated according
to the method prescribed by Frost (1972) which assumes that the number of cells
during an experiment changes at a constant exponential rate. The equation corrects
for the growth of phytoplankton measured in control containers. All relationships
between ingestion and cell concentration were calculated using the average cell
concentration value. The functional response of zoeae grazing on dinoflagellates at
various cell densities was plotted using the Holling "disc" equation (Holling, 1959)
which treats each capture of a food particle as an independent event.
Carnivorous feeding activity of zoeae on Anemia sp. nauplii was compared in
the presence and absence of a natural spring bloom phytoplankton assemblage col-
lected in Resurrection Bay, Alaska. A 10 liter Niskin bottle, cast at approximately
1 m depth, was used to collect the phytoplankton. A sample from the bottle was
concentrated by pouring the sea water through a 40 ^m mesh sieve resting in an
over-flow vessel. All conspicuous zooplankton and most conspicuous micro-zoo-
plankton were removed from this sample under a microscope. The remaining phy-
toplankton was diluted with filtered sea water to return the sample to its original
volume for use in feeding observations. Two subsamples were examined to obtain
the initial phytoplankton concentration (Lund et ai, 1958). Large solitary cells and
long chain-forming species were retained at approximately their original concen-
trations by the procedure; these are the cell types seen in the stomachs of zoeae from
plankton collections. The assemblage contained Chaetoceros spp. chains at approx-
imately 5.5 X 104 cells 1 ' and large centric diatoms at approximately 1.2 X 10' T1.
Four groups of five zoeae each were placed in 250 ml vessels containing the following
FEEDING AND ENERGETICS OF CRAB LARVAE 201
prey assemblages: 60 nauplii 1" ' , 96 nauplii 1" ' , 60 nauplii 1" ' and bay phytoplankton,
and 96 nauplii 1 "' and bay phytoplankton. There were 10 replicates for each prey
assemblage. At the end of the experiments, surviving nauplii were counted and zoea
stomachs were examined under a microscope for phytoplankton cells.
RESULTS
Oxygen consumption rates (VO2) of one day old stage I C. bairdi zoeae averaged
1.3 ^1 O2 • mg dry wt ' h ' at 5°C (Table II). The average dry weight of an individual
zoea within 24 hours of hatching was 47.8 ± 7.6 ng, so the corresponding value of
VO2 was 0.06 n\ O2 zoea ' h '. The energy required for respiratory metabolism (R)
of first feeding zoeae at this temperature was estimated to be 6.8 X 10 3 cal zoea"1
d '. Carbon equivalents of respiration were 0.59 and 0.71 ;ug C zoea~' d~' using
respiratory quotients (RQ) of 0.75 and 0.90, respectively. Stage I zoeae were ap-
proximately 33% C (based on dry weight); therefore, respiratory energy needs ranged
from 3.7 to 4.5% body C zoea ' d'1.
The average daily feeding rate of one day old zoeae was one Anemia nauplius
per zoea (range: 0.2 to 2.2) at 5°C when prey were offered at initial densities of 100
per liter (Table II). The average assimilated carbon values of prey consumed in
individual experiments ranged from 28 to 308 percent of the mean respiratory energy
needs of each zoea.
The Student /-test comparing predation rates in the presence and absence of
phytoplankton showed no significant differences (P < 0.05) in the number of nauplii
consumed in the two groups. At 60 prey per liter, the average daily consumption
rates per zoea were 1.0 ± 0.5 and 0.9 ± 0.4 nauplii in the presence and absence of
phytoplankton, respectively. Corresponding values at 96 prey per liter were 1.4
± 0.5 and 1.0 ± 0.2 nauplii zoea ' d '. No phytoplankton cells were found in the
stomachs of the zoeae at the end of the experiment.
Chlorophyll a measurements (ng liter"1) of replicate initial, control and exper-
imental grazing bottles showed no detectable difference for P. tricornutum, C. com-
pressus, or F16 experiments at any of the chlorophyll concentrations employed.
Likewise, no discernible difference in gut flourescence between fed and unfed zoeae
was found in any of the above experiments. A slight decrease was found in some
of the grazed bottles of G. grindleyi cells, and some measurements of gut flourescence
of zoeae from these experiments showed higher levels in fed than unfed animals.
However, all measurements were at the limits of sensitivity of the methods and,
when the experiments were repeated, results were not consistent between trials or
between replicates.
Data from the cell count experiments with G. grindleyi, Coscinodiscus spp. and
Thalassiosira spp. are summarized in Table III, along with a comparison of ingested
ration (carbon) and respiratory requirements of the zoeae. No phytoplankton growth
was observed in control vessels. Over the range of cell concentrations used, zoeae
obtained an average of 1.4 to 14% of their respiratory energy requirements from
phytoplankton. The relationship between ingestion (grazing) rate of zoeae and av-
erage cell concentration is shown in Figure 1 for dinoflagellates and Figure 2 for
large centric diatoms. The functional response of zoeal ingestion rate to increases
in dinoflagellate abundance demonstrated a leveling at cell concentrations above 8.0
X 104 1 ' (Fig. 1).
The coefficient of variation of replicate cell counts of samples from the G. grin-
dleyi experiments was less than 9% at the lowest cell densities used and less than
5% at the highest. No subsampling variability was associated with the Coscinodiscus
202
L. S. INCZE AND A. J. PAUL
oo
U
O 00 ^
~" ~~ oo
<N
c
*
4
__
ON <O "^
u
o
OO OO \
OO
o
^H
"1
a
ei,
^r
2
U
r, ._ fN
1
00
_; \
1
r^i
K
d
"5.
a
E
\D
\
o
p
r~ oo —
C
\
<&
X
00 ^ i-i
d
O
1>
•—
r-
&o
N
u
d
.s:
'^!
o
Si
^
fa.
u
1
o!
O ON <N
i
N
Z 2
-1 0 fN
u
">3
c
d
c
*
o
3
1
~ **•
P.
•7
LU ,
u
O m oo
s
oo
^ ~ oo
c
*t ~~^
.T~^ •3-
^ ^ •*$•
ca
H
T3
'
u
B
_.
E
~^* *"^S
3
a fij
S
^ a
o r
eS
"a S
_N 0
°^.
•o
^ 'Ci
Q/
OO -~* |
C
Rj ^J
X
so ^^ i —
rt
f .2
15
i/-;
c
o
1
§£
'3
Of, s:
zr*
p
S"
• \
II
^ f^j.
?*• °
Q oo
OO SO u~>
r-' r-' ^
oo
0
2
^ S^
^3.
r*1
S *-
'5.
S"-i
fi
^ ~s
c
S' a
i
nj
p
OJ
U
E
.^ s
O !
O O |
!• s
N jr
dog
-C
C s:
"a.
d
c
. .0
o
5 5
6
1
1 2
$~
ON
OO
1
15
IT
-25
P
0
s o
"^ ^
^H
C^
to .
~^-
^2
s: c
5*
° -2
^*
ed
^ -§
U
60 'C
00
"""
*~. *^,
c
•*' B
fS ^
*v* ^3 ^
>J O
'rS t/i O-i
FEEDING AND ENERGETICS OF CRAB LARVAE
203
TABLE III
Ingestion rate (I) of 24 hour old zoeae grazing on phytoplankton fGonyaulax grindleyi, Coscinodiscus
spp., Thalassiosira spp.) at various cell concentrations ((€}) at 5°C, and percent contribution to
respiratory requirement (% R)
Cell type
Carbon <C>
(MgcelP1) (cells r1)
I (zoea '
d-')
%R'
No. cells
Mg C
G. grindleyi
2.9 x ID'3 7.3 x 103
4.3
1.2 x 10"2
1.4
3.1 x 104
19.0
5.0 x 10 -
5.8
6.5 x 104
33.6
9.7 X 10 :
11.3
1.2 X 105
39.9
1.2 X 10-'
14.0
Coscinodiscus spp.
2.66 x 10 7.8 X 102
2.6
6.8 X 10"2
7.9
8.40 X 102
2.9
7.6 X 10~2
8.9
8.80 X 102
2.7
7.0 X 10"2
8.2
9.40 X 102
2.4
6.3 X 10 2
7.4
9.83 X 102
2.3
6.0 X 10"2
7.0
9.83 X 102
2.3
6.0 X 10 2
7.0
1.68 x 103
2.2
5.7 X 10 2
6.6
1.68 X 103
2.5
6.6 X 10 2
7.7
1.68 x 103
4.2
1.1 X 10"'
12.8
Thalassiosira spp.
2.30 x 10 6.26 x 102
1.5
3.6 X 10 2
4.2
1.66 X 103
2.7
6.4 X 1Q-2
7.5
1 Calculation is based on a mean respiratory requirement of 0.6
an RQ of 0.9 and an assimilation efficiency of 0.70.
C zoea ' d ' (from Table II),
or Thalassiosira experiments since all cells were individually added to, and removed
from, the experimental and control vessels and counts were double-checked.
DISCUSSION
Respiration rates measured at 5°C in this study are lower than those reported
for many decapod larvae that inhabit warmer environments (see Mootz and Epi-
fanio, 1974; Schatzlein and Costlow, 1978; Levine and Sulkin, 1979). However, the
hourly weight-specific rates for stage I C. bairdi zoeae (X = 1 .3 ^1 O2 • mg dry wt~' •
IT1) are similar to those of first stage zoeae of Cancer borealis (1.3 ^1 O2) and C.
CO
LJ
o
4.0 6.0
<C> (cells x 10'
14.0
FIGURE 1. Relationship between ingestion (I) and average cell concentration ((C)) of G. Grindleyi.
The curve was fit using the Holling equation (Holling, 1959). Area of I/(C) observations for large centric
diatoms (Fig. 2) shown by box in lower left corner.
204
L. S. INCZE AND A. J. PAUL
1
5.0
4.0
o
CL>
° 3.0
_co
" 2.0
h- 1.0
LJ
O
- o
0.6
I I I I
I I I I
II
II
0.8 1.0 1.2 1.4 1.6
<C> (cells x 103 ,T1)
1.8
2.0
FIGURE 2. Relationship between ingestion (I) and average cell concentration ((C)) of large centric
diatoms: Coscinodiscus spp. (O); Thalassiosira spp. (A).
irroratus ( 1 .7 ^l O2) measured at 5°C by Sastry and McCarthy (1973). Furthermore,
the respiration rates measured in this study are corroborated by estimated in situ
stage I growth rates and by VO2 measurements of stage II zoeae captured at sea.
The estimated growth rates of C. bairdi first stage zoeae from the southeastern Bering
Sea (3.9 to 4.7% body C zoea ' d ': Incze, 1983) are almost identical to the carbon
equivalents of respiration measured in this study (3.7 to 4.5% body C zoea"1 d~').
This agrees with the findings of laboratory studies of other crab species where zoeal
respiration and growth of early stages were approximately equal (Mootz and Epi-
fanio, 1974;LevineandSulkin, 1979). In another study (Incze, 1983), measurements
of VO2 of 550 to 590 ng dry weight stage II C. bairdi zoeae captured with a plankton
net were obtained with a Radiometer blood-gas analyzer following methods of
Laughlin et al. (1979) and using incubation volumes of 10 and 20 ml. The allometric
equation relating respiration (R) to dry body weight using the results of the stage
I and stage II measurements provides a weight exponent of 0.72 (R = 1.198 W072:
Incze, 1983), a value similar to those reported for other decapod larvae (Schatzlein
and Costlow, 1978) and for animals in general (McMahon, 1973). Estimated in situ
growth rates of stage I zoeae and respiration rates of stage II zoeae therefore sub-
stantiate the results obtained from the small-volume manometric methods used on
first stage larvae in this study.
The ability to feed on a wide variety of prey particles is one adaptation for
procuring food in a diverse and dispersed community. Many planktonic organisms
employ this strategy of omnivorous feeding, though they may do so to different
degrees (Marshall, 1973; Landry, 1981). Despite its advantages, omnivory may in-
volve compromises in structure and function of feeding appendages which decrease
feeding performance on certain types of prey. For instance, Robertson and Frost
(1977) found that Aetidius divergens could feed efficiently on large diatoms and
Anemia nauplii, but was inefficient at ingesting small diatoms when compared with
herbivorous calanoid copepods.
Measurements made in this study demonstrate that C. bairdi zoeae are omni-
vorous and consume some phytoplankton. Since large phytoplankton cells were
captured and ingested in the absence of zooplankton prey, directed grazing activity
is indicated. However, in these experiments, grazing rates were too low to meet
respiratory energy requirements, even assuming that variations in individual zoeal
FEEDING AND ENERGETICS OF CRAB LARVAE 205
grazing rates existed. The functional response (Fig. 1) of zoeae grazing on dinofla-
gellates indicates that ingestion rates would not increase substantially at cell con-
centrations higher than 2 X 105, presumably because feeding ability is saturated.
Consequently, cells of this size could not sustain the zoeae. Even the large diatoms,
which contained about ten times as much carbon per cell as the dinoflagellates,
could not sustain zoeae under most natural conditions. It would require approxi-
mately 2 X 105 of these large cells per liter to satisfy the respiratory requirements
of C. bairdi first-feeding zoeae, assuming the same functional feeding response to
large diatoms and dinoflagellates at high cell concentrations. However, this would
be an extraordinarily high concentration for diatoms of this size in the upper 20 m
of the ocean where most of the larvae are found (Incze, 1983). When growth re-
quirements averaging about 4.3% body C zoea~' d~' (see above) are added to re-
spiratory requirements, the contribution of phytoplankton to total energy needs is
further diminished. However, a nutritional role for phytoplankton, such as providing
micro-nutrients, is not ruled out by these findings.
The chlorophyll method employed with the grazing experiments which used
cultured algae was useful as a screening process to see if measurable chlorophyll
depletion (cell consumption) occurred with any of the cells used. The technique
verified that there were no instances where rates of consumption satisfied basal
metabolic needs, since these rates would have been detected. However, the method
was clearly not sensitive enough to measure the low grazing rates observed in cell
count experiments using G. grindleyi.
The measurements of carnivorous feeding rates in the presence and absence of
a natural phytoplankton assemblage collected at the time that C. bairdi zoeae were
hatching in Resurrection Bay indicated that zoeal predation was unaffected by the
phytoplankton. Since none of the conspicuous chain-forming diatoms from the
assemblage were found in the stomachs of the zoeae at the end of the experiments,
it appears that little or no grazing occurred. However, it is not known whether the
apparent lack of grazing in these experiments reflects specific predatory behavior
directed at the larger zooplankton or simply a high probability of predator-nauplius
encounter at the prey densities used. It may be that diatoms and dinoflagellates
appear frequently in the stomachs of zoeae collected from the plankton because
more time is spent grazing when zooplankton prey are less available than in these
experiments. When phytoplankton cells are consumed, they may only appear quan-
titatively important because their thecae are conspicuous and may not be digested
rapidly. Softbodied prey, such as nauplii, have few hard parts which can be identified
in zoea stomachs, and the hard parts of other zooplankters may be rejected during
feeding (Fowler et ai, 1971; Mauchline, 1980). Their stomachs could also contain
phytoplankton if zoeae proved to be coprophagous feeders (see data on fecal pellet
sizes and contents presented by Urrere and Knauer, 1981). Additional laboratory
observations on the ability of zoeae to consume specific species and developmental
stages of zooplankton, as well as fecal pellets, are necessary to complement the
stomach analysis of individuals feeding in situ.
In the zooplankton prey consumption rate experiments, it is likely that some
non-feeding zoeae were included, since only individuals no more than 24 hours old
and with no previous feeding experience were used. Work by Kon (1979) suggests
that a three day "critical" period exists during which first stage zoeae must initiate
feeding before subsequent mortality increases markedly. Thus it is probable that
newly molted stage I zoeae of this genus have some stored energy that can be used
to meet metabolic energy requirements. The occurrence of some non-feeding in-
dividuals in experiments may explain why, on the average, only enough energy was
206 L. S. INCZE AND A. J. PAUL
obtained to meet metabolic requirements. We stress, however, that there were some
groups of zoeae which consumed zooplankton prey at rates which provided over
300% of daily carbon needs. It is reasonable to assume that similar differences occur
during predation on natural prey. The relative abundance of first-feeding larvae
which are competent predators under various planktonic conditions may be an
important aspect of year-class survival.
Although appropriate zooplankton prey must be rare compared with phyto-
plankton, they are probably a major component of the zoeal diet of C. bairdi. Zoeae
of this species do not appear to be well adapted to handling a large number of small
prey. These experiments used only 24 hour old first-feeding zoeae, but the larvae
demonstrated competence at feeding by capturing and ingesting Anemia nauplii
(present study) and active zooplankton prey (Paul et al., 1979). Larval growth should
increase the relative disadvantage of the predator-prey size relationship with respect
to phytoplankton. Unless behavioral or morphological changes occur which favor
grazing, phytoplankton should remain a comparatively minor source of energy in
larval development of this species.
ACKNOWLEDGMENTS
We thank J. M. Paul and J. Erickson for technical assistance, S. Hall for providing
dinoflagellate cultures, and K. O. Coyle for sharing unpublished data with us. M.
Landry, C. Greene, M. Ohman, S. Smith, and R. Strathmann provided helpful
discussions during the course of this work, and J. Vidal, D. Armstrong, and R.
Lasker offered constructive comments on the manuscript. The use of laboratory
facilities of the Institute of Marine Science, University of Alaska Seward Marine
Station, is gratefully acknowledged. Adult crabs were provided by the Alaska De-
partment of Fish and Game. This study was supported by the National Science
Foundation (PROBES, Grant DPP-76-23340) and the National Marine Fisheries
Service (Grant 82-ABH-1082).
LITERATURE CITED
ANGER, K., AND K. K. C. NAIR. 1979. Laboratory experiments on the larval development of Hyas
aranem (Decapoda, Majidae). Helgol. Wiss. Meeresunters. 32: 36-54.
ATKJNS, D. 1955. The post embryonic development of British Pinnotheres. Proc. Zoo/. Soc. Land. 124:
687-715.
BIGFORD, T. E. 1978. Effect of several diets on survival, development time, and growth of laboratory-
reared spider crab, Libinia emarginata larvae. Fish. Bull. 76: 59-64.
BOUSQUETTE, G. D. 1980. The larval development of Pinnixia longipes (Lockington, 1877) (Brachyura:
Pinnotheridae), reared in the laboratory. Biol. Bull. 159: 592-605.
BRICK, R. W. 1974. Effects of water quality, antibiotics, phytoplankton and food on survival and de-
velopment of larvae of Scylla serrata (Crustacea: Portunidae). Aquaculture 3: 231-244.
BRIGHT, D. B. 1967. Life histories of the king crab, Paralithodes carntschatica, and the Tanner crab.
Chionoecetes bairdi. in Cook Inlet, Alaska. Ph.D. thesis, Univ. Southern California, Los Angeles.
265 pp.
BRODY, S. 1945. Biogenergetics and growth. Rheinhold, NY. 1023 pp.
CHRISTIANSEN, M. E., AND W. T. YANG. 1976. Feeding experiments on the larvae of the fiddler crab,
Uca Pugilator (Brachyura, Ocipodidae), reared in the laboratory. Aquaculture 8: 91-98.
CONOVER, R. J. 1966. Factors affecting the assimilation of organic matter by zooplankton and the
questions of superfluous feeding. Limnol. Oceanogr. 11: 346-354.
CORNER, E. D. S., AND C. B. COWEY. 1968. Biochemical studies on the production of marine zooplank-
ton. Biol. Rev. 43: 393-426.
FOWLER, S. W., G. BENAYOUN, AND L. F. SMALL. 1971. Experimental studies on feeding, growth and
assimilation in a Mediterranean emphausiid. Thalassia Jugosl. 7: 35-47.
FROST, B. W. 1972. Effects of size and concentration of food particles on the feeding behavior of the
marine planktonic copepod Calanus pacificus Limnol. Oceanogr. 17: 805-815.
FEEDING AND ENERGETICS OF CRAB LARVAE 207
GIESE, A. C. 1973. Cell Physiology. W. B. Saunders. 741 pp.
CONOR, S. L., AND J. J. GONOR. 1973. Feeding, cleaning and swimming behavior in larval stages of
porcellanid crabs (Crustacea: Anomura). Fish. Bull. 71: 225-234.
HARTMAN, M. C., AND G. R. LETTERMAN. 1978. An evaluation of three species of diatoms as food for
Cancer magister larvae. Pp. 271-276 in Ninth Ann. Meeting, World Mariculture Soc. J. Avault,
ed.
HAYNES, E. 1973. Description of the prezoea and Stage I zoea of Chionoecetes bairdi and C. opilio
(Oxyrhyncha, Oregoniinae). Fish. Bull. 71: 769-775.
HERRNKIND, W. F. 1968. The breeding of Uca pugilator (Bosc) and mass rearing of the larvae with
comments on the behavior of the larval and early crab stages (Brachyura, Ocypodidae). Crus-
taceana, Suppl. 2: 214-223.
HOLLING, C. S. 1959. Some characteristics of simple types of predation and parasitism. Can. Entomol.
91: 385-398.
INCZE, L. S. 1983. Larval life history of Tanner crabs, Chionoecetes bairdi and C. opilio, in the south-
eastern Bering Sea and relationships to regional oceanography. Ph.D. dissertation, Univ. Wash-
ington, Seattle. 192 p.
KNUDSEN, J. W. 1960. Reproduction, life history and larval ecology of the California xanthidae, the
pebble crabs. Pac. Sci. 14: 3-17.
KON, T. 1970. Fisheries biology of the Tanner crab. IV. The duration of planktonic stages estimated by
rearing experiments of larvae. Bull. Jpn. Soc. Sci. Fish. 36: 2 1 5-224.
KON, T. 1979. Ecological studies on crabs belonging to the genus Chionoecetes. I. The influence of
starvation on the survival and growth of the zuwai crab. Bull. Sci. Fish. 45: 7-9.
LANDRY, M. R. 1981. Switching between herbivory and carnivory by the planktonic marine copepod
Calanus pacificus. Mar. Biol. 65: 77-82.
LAUGHLIN, R. B., H. W. WOFFORD, AND J. M. NEFF. 1979. Simple potentiometric method for the rapid
determination of respiratory rates of small aquatic organisms. Aquaculture 16: 77-82.
LEBOUR, M. V. 1922. The food of plankton organisms. J Mar. Biol. Assoc. U. K. 12: 644-677.
LEBOUR, M. V. 1927. Studies of the Phymouth Brachyura. I. The rearing of crabs in captivity with a
description of the larval stages of Inochus dorsettansis, Macropodia longirostris and Maia squi-
nado. J. Mar. Biol. Assoc. U.K. 14: 795-821.
LEVINE, D. M., AND S. D. SULK.IN. 1979. Partitioning and utilization of energy during the larval de-
velopment of the xanthid crab, Rithropanopeus harrisii. J. Exp. Mar. Biol. Ecol. 40: 247-257.
LUND, J. W. G., C. KIPLING, AND E. D. LECREN. 1958. The inverted microscope method of estimating
algal numbers, and the statistical basis of estimation by counting. Hydrobiologia 11: 143-170.
MARSHALL, S. M. 1973. Respiration and feeding in copepods. Adv. Mar. Biol. 11: 57-120.
MAUCHLINE, J. 1980. The biology of mysids and euphausiids. Adv. Mar. Biol. 18: 373-677.
McMAHON, T. 1973. Size and shape in biology. Science 179: 1201-1204.
MOOTZ, C. A., AND C. E. EPIFANIO. 1974. An energy budget for Mennipe mercenaria larvae fed Anemia
nauplii. Biol. Bull. 146: 44-55.
MOTOH, H. 1973. Laboratory-reared zoeae and megalopae of zuwai crab from the Sea of Japan. Bull.
Jpn. Soc. Sci. Fish. 39: 1223-1230.
PARSONS, T. R., K. STEPHENS, AND J. D. H. STRICKLAND. 1961. On the chemical composition of eleven
species of marine phytoplankters. J. Fish. Res. Board Can. 18: 1001-1016.
PAUL, A. J., J. M. PAUL, P. A. SCHOEMAK.ER, AND H. M. FEDER. 1979. Prey concentrations and feeding
response in laboratory reared stage-one zoeae of king crab, snow crab, and pink shrimp. Trans.
Am. Fish. Soc. 108: 440-443.
ROBERTS, M. H. 1974. Larval development of Pagunis longicarpus Say reared in the laboratory. V. Effect
of diet on survival and molting. Biol. Bull. 146: 61-11 .
ROBERTSON, S. B., AND B. W. FROST. 1977. Feeding by an omnivorous planktonic copepod, Aetidius
divergens Bradford. J. Exp. Mar. Biol. Ecol. 29: 231-244.
SASTRY, A. N., AND J. F. MCCARTHY. 1973. Diversity in metabolic adaptation of pelagic larval stages
of two sympatric species of brachyuran crabs. Neth. J. Sea Res. 7: 434-446.
SATO, S., AND S. TANAK.A. 1949. Study on the larval stages of Paralithodes camtschatica (Tilesius). II.
On the rearing. Sci. Pap. Hokkaido Fish. Sci. Inst. 3: 1 8-30.
SCHATZLEIN, F. C., AND J. D. CoSTLOW. 1978. Oxygen consumption of the larvae of the decapod
crustaceans, Emerita talpoida (Say) and Libinia emarginata Leach. Comp. Biochem. Physiol.
61: 441-450.
STICK.NEY, A. P., AND H. C. PERKINS. 1981. Observations on the food of the larvae of the northern
shrimp, Pandalus borealis Kroyer (Decapoda, Caridae). Crustaceana 40: 36-49.
STRATHMANN, R. R. 1967. Estimating the organic carbon content of phytoplankton from cell volume
or plasma volume. Limnol. Oceanogr. 12: 411-418.
208 L. S. INCZE AND A. J. PAUL
STRICKLAND, J. D. H., AND T. R. PARSONS. 1968. A practical handbook of seawater analysis. Fish. Res.
Board Can. Bull. 167: 310 pp.
SULKJN, S. D. 1975. The significance of diet in the growth and development of larvae of the blue crab,
Callinectes sapidus Rathbun, under laboratory conditions. /. Exp. Mar. Biol. Ecol. 20: 1 1 5-
135.
SULKIN, S. D. 1978. Nutritional requirements during larval development of the portunid crab, Callinectes
sapidus Rathbun. /. Exp. Mar. Biol. Ecol. 34: 25-41.
SULKJN, S. D., AND C. E. EPIFANIO. 1975. Comparison of rotifers and other diets for rearing early larvae
of the blue crab, Callinectes sapidus Rathbun. Estuarine Coastal Mar. Sci. 3: 109-1 13.
SULKJN, S. D., AND K. NORMAN. 1976. A comparison of two diets in the laboratory culture of the zoeal
stages of the branchyuran crabs, Rithropanopeus harrisii and Neopanope spp. Helgol. Wiss.
Meeresunters. 28: 183-190.
UMBREIT, W. W., R. H. BURRIS, AND J. F. STAUFFER. 1972. Manometric and Biochemical Techniques.
Burgess Publ. Co., Minneapolis. 308 pp.
URRERE, M. A., AND G. A. KNAUER. 1981. Zooplankton fecal pellet fluxes and vertical transport of
paniculate organic material in the pelagic environment. J. Plankton Res. 3: 369-385.
UTERMOHL, H. 1958. Zor Vervollkommnung der quantitativen phytoplankton methodik. Mitt. Int. Ver.
Theor. Angew. Limnol. 9: 1-38.
Reference: Biol. Bull. 165: 209-220. (August, 1983)
COORDINATION OF COMPOUND ASCIDIANS BY EPITHELIAL
CONDUCTION IN THE COLONIAL BLOOD VESSELS!
G. O. MACKIE AND C. L. SINGLA
Biology Department, University of Victoria. Victoria, British Columbia, V8W 2Y2
ABSTRACT
Bursts of electrical potentials propagate at ca. 2.0 cm s"1 within the colonial
vascular system ofBotryllus, Botrylloides, and Metandrocarpa, serving to coordinate
contractions of the vascular ampullae and mediating protective closure and ciliary
arrest in zooids. Nerves are absent from the vessels and ampullae. Impulses are
presumed to propagate from cell to cell in the vascular epithelium via gap junctions,
shown to be present by electron microscopy.
INTRODUCTION
Colonial ascidians of the Family Styelidae produce zooids by budding as de-
scribed for Botryllus (Berrill, 1941; Sabbadin, 1955) and for Botrylloides (Berrill,
1947). The buds, instead of separating as in most ascidians, remain permanently
attached via the vascular system which consists of an elaborate colony-wide network
of blood vessels, details of which are given by Brunetti and Burighel (1969). Blind-
ending swellings, or ampullae, are produced at various points, chiefly around the
edges of the colony (Figs. 1, 2). Regular cycles of expansion and contraction occur
in these ampullae causing a tidal ebb and flow of blood within the connecting vessels,
as first described by Bancroft (1899). Bancroft noted that as many as 50 ampullae
within an area of about 4-5 mm2 can exhibit coordinated contractions. Contractions
occur not only in the ampullae, but in the blood vessels as well (Mukai et ai, 1978).
Contraction is attributable to bundles of microfilaments in the epithelial cells form-
ing the vascular walls (DeSanto and Dudley, 1969) but the mechanism responsible
for coordination has not been satisfactorily explained. Bancroft's observations sug-
gested variations in blood pressure as the principal means of coordination, as did
those of DeSanto and Dudley (1969). However, Mukai et al. (1978) found that
coordination could best be explained on the assumption that signals of some kind
are conducted along the vessels. As nerves have never been described in the vessels,
these workers proposed epithelial conduction as the signalling mechanism. This
possibility was also advanced by V. L. Scofield (pers. comm.) on the basis of her
observation that ampullar rhythms of recenty settled larvae immediately come into
synchrony when the larvae undergo fusion. Torrence and Cloney (1981) favored
epithelial conduction as the probable basis for coordination of ampullae in Molgula,
and showed that gap junctions are present between the living cells.
The present study demonstrates that electrical impulses propagate throughout
the vascular network and ampullae, and shows that these signals are not only re-
sponsible for coordination of ampullar rhythms but also serve to coordinate pro-
tective responses of the interconnected zooids.
Received 21 February 1983; accepted 23 May 1983.
t Dedicated to N. J. Berrill on the occasion of his 80th birthday, April 6th, 1983, in appreciation
of his many important contributions to tunicate biology.
209
210
G. O. MACKIE AND C. L. SINGLA
FIGURE 1. Edge of a colony of Botrylloides. a, ampullae; c, common cloacal siphon. Bar represents
1 mm.
MATERIALS AND METHODS
During February and March 1982 colonies of Botryllus sp. (the "Monterey Bo-
tryllus" of Scofield el al, 1982) and Botrylloides diegensis were collected from rocks
in Monterey Bay, California and maintained in the sea water system at the Hopkins
FIGURE 2. Scheme showing the circulatory system and zooid interrelationships in a young colony
of Botryllus. a, ampullae; B1-B4, first four generations of blastozooids; v, connecting vessels (from
Brunetti and Burighel, 1969).
COORDINATION OF COMPOUND ASCIDIANS 2 1 1
Marine Station. Larvae were settled and grown on glass or plastic sheets. Dr. D. P.
Abbott made available a colony of Metandrocarpa taylori found growing in a water
table in the basement of the Agassiz Building.
Botryllus specimens were fixed in 4% glutaraldehyde in 0.2 M cacodylate buffer
(pH 7.4) for one hour at room temperature, rinsed in the cacodylate buffer and post
fixed in 1% osmium tetroxide in the same buffer for one hour at 4°C. The material
was rinsed with distilled water, dehydrated progressively in acetone, transferred to
propylene oxide, and embedded in Epon 812. Thin sections were stained wth uranyl
acetate and lead citrate and were examined in Philips EM 300.
Botrylloides larvae were fixed while attached to acetate sheets. The acetate dis-
solved in the acetone used for dehydration.
For the electrophysiological recordings, fine polyethylene suction electrodes (30-
50 nm I.D.) were used. Recorded potentials were amplified and displayed on a
Tektronix storage oscilloscope and on a Brush chart recorder. A thermistor flow
meter was used to detect variations in siphonal currents.
RESULTS
Fine structure
We have examined Botryllus and Botrylloides to 1 ) determine if nerves are
present in the vascular vessels and ampullae, and 2) to confirm that the junctional
specializations reported by Torrence and Cloney (1981) for Molgula are also to be
found in our species.
Except for the specialized glandular cells of the ampullar tips ("pad cells", Katow
and Watanabe, 1978) the cells forming the vascular lining are of a single type
equivalent to the "parietal cells" of Torrence and Cloney (1981). These cells may
be columnar or squamous depending on the state of contraction at the moment of
fixation. Their inner borders become somewhat folded in the contracted state (Fig.
3a), and a layer of micro-filaments can be seen running close beside the luminal
border, as first noted by DeSanto and Dudley (1969). Nerves have not been seen
in any part of the system.
Tight junctions (zonulae occludentes) are located at the outer ends of the line
of apposition between adjacent cells (Fig. 3a, c). These junctions are of the punctate
type common in tunicates. No such junctions occur on the inner (luminal) side.
Gap junctions are seen at intermediate points along the line of apposition (Fig. 3b).
The intercellular space along most of the line of apposition is about 15-20 nm wide,
but in the gap junction, the two membranes are separated by about 2 nm.
General activity and responses to stimulation
The following comments refer equally to Botryllus and Botrylloides. Colonies
maintained in clean water pump water continuously through the branchial sac,
showing occasional arrests of the branchial cilia along with siphon contractions when
large particles strike the siphons. No regular pattern of ciliary arrests or muscular
contractions was observed in the zooids. The hearts of the different zooids beat
rhythmically at their own individual frequencies, reversing periodically. Peristaltic
waves pass along the gut, and feces are eliminated at regular intervals.
Movement of blood is not confined to the zooids but takes place throughout
the colonial vascular network. The vascular ampullae associated with these vessels
can be seen swelling and contracting rhythmically. In newly settled zooids and in
larger colonies as noted by Bancroft (1899) contractions of the ampullae are syn-
212
G. O. MACKIE AND C. L. SINGLA
FIGURE 3. Fine structure of parietal cells forming the ampullar wall in Botrylloides. (A) shows
several partially contracted cells. On the inner side, contractile folds (cf) are seen, and microfilaments
(mf) lie against the membrane. On the outer side, next to the tunic (t), the cells are joined by zonulae
occludentes (zo). (B) shows a gap junction between two parietal cells and (C) shows detail of the zonula
occludens. Bar in (A) is 1 ^m, in (B) and (C) 0.1
chronized. Blood flow through the colonial vascular network appears to be due
chiefly to the contractions of the ampullae, and is affected only locally by the actions
of the hearts of the zooids. Bancroft (1899) showed that the rhythmic flushing of
blood through the vascular network continued in anaesthetized colonies after the
hearts of the zooids had stopped beating, and that coordinated ampullar contractions
persisted in regressive colonies after the zooids had degenerated. In the present study,
regular, synchronized ampullar contractions were seen in strips cut from the edge
COORDINATION OF COMPOUND ASCIDIANS
213
of a colony which contained no zooids. Torrence and Cloney (1981) showed that
individual ampullae, when isolated, continued to pulsate rhythmically.
In intact colonies, contractions of the ampullae drive blood into the zooids,
causing them to swell (DeSanto and Dudley, 1969). The slow rising and falling of
the surface of the colony allows results to be recorded with flow meters placed over
the oral or cloacal siphons. Each time the colony swells, the siphons are brought
nearer to the sensor and a surge in flow rate is recorded (Fig. 4A). Short term changes
in flow rate due to siphon contractions or ciliary arrests do not affect the overall
rhythm, which is manifested throughout the whole colony, or large parts of it.
Coordination was demonstrable in one colony over a distance of 1 cm.
When the surface of a zooid is touched lightly with a needle, the zooid contracts
its oral siphon and its portion of the common cloacal passage, while the branchial
cilia show a brief arrest. Slightly stronger stimulation causes contractions and ciliary
arrests in adjacent zooids within the same group ("system") sharing a common
cloacal opening. Spread to adjacent systems takes place with still stronger or re-
petitive stimulation. Spread appears to occur more readily within a system than
between systems, regardless of proximity to the site of stimulation. When excitation
reaches a new system several zooids usually contract together almost synchronously,
rather than in a wavelength-like sequence. These observations suggest that contrac-
B
5min
FIGURE 4. Ampullar rhythms monitored with a flow meter (A) and electrically (B-F). (A) Botryl-
loides: variations in water flow past sensor placed over cloacal siphon as colony rises and falls (retouched
to remove artefacts). (B) Botryllus: electrical record of multiple NP bursts from two ampullae, 2 mm
apart. Although contractions were perfectly syncrhonized the number of NB bursts at each contraction
varied at the two sites. (C) Botrylloides: NP burst record from two points 1.5 cm apart, in a large colony.
Most but not all contractions are synchronized at the two sites. (D) Botrylloides: NP bursts from an
ampulla of a recently settled larva (oozooid). (E) Metandrocarpa: NP bursts from an ampulla. (F) Bo-
trylloides: ampullar NP burst record. At the arrow, a scalpel blade was drawn along near the edge of the
colony separating a group of about 200 ampullae, including the one with recording electrode attached,
but excluding all zooids. Activity is shown immediately after the cut, and after a break in the record
lasting 18 h. The 5 min bar applies to all records.
214 G. O. MACKIE AND C. L. SINGLA
tions are probably not mediated mechanically, but instead are due to impulse con-
duction. This conclusion is further supported by the observation that zooids can be
made to contract by stimulation of nearby ampullae. The responding ampulla moves
so little that a mechanical effect on adjacent zooids is scarcely conceivable.
The effects of electrical stimulation resemble those produced by tactile stimu-
lation. Colonies adapt readily to maintained stimulation. A stimulus which would
cause a spreading response in a rested colony may cause only a small local response
after the colony has been stimulated for a period of time.
Electrical monitoring ofampullar rhythms
Suction electrodes attached to ampullae or zooid walls pick up a rapid burst of
potentials, or several such bursts, each time the ampullae contact. Monitoring from
the ampullae is preferable, as an electrode on the zooid wall picks up ciliary arrest
potentials as well as the events correlated with ampullar contraction. These events
are termed network potentials (NPs) as they propagate throughout the vascular
network interconnecting the ampullae and zooids. NPs characteristically occur in
short bursts. Recorded at slow chart speeds, these bursts appear as single events
(Figs. 4B-F). This method of monitoring ampullar rhythms is very simple, and
causes minimal disturbance to the colony. The NP burst coincides with the start
of the ampullar contraction phase. As many as five bursts (each consisting of several
individual NPs) may accompany a single contraction (Fig. 4B). The larger numbers
are typical of colonies subjected to damage or overstimulation. Rested colonies
maintained in slowly running water usually show only one NP burst at each con-
traction. Recordings from two ampullae within the same colony show coordination
ofampullar rhythms, although in large colonies (<1.0 cm wide, as in Fig. 4C) some
loss of coordination may be apparent. In six colonies of Botrylloides, contractions
occurred at mean intervals of 2.2-4.0 minutes (X :: 2.6, SD == 0.6 min overall).
Values for Botryllus fell within the same range (Fig. 4D). Records from Metandro-
carpa gave a mean value of 4.2 min (Fig. 4E).
While the NP burst lasts less than a second, the contraction phase of the ampullar
cycle lasts for more than a minute. It appears that the NP burst is essentially a
triggering event serving to initiate contractions simultaneously throughout the net-
work. Electrical or mechanical stimulation, or damage, evokes NP bursts and may
reset the ampullar rhythm, and alter its pattern, as seen in Figure 4F.
Generation and conduction of network potentials
The composite nature of the NP burst is readily observed when the burst is
displayed at higher sweep speeds on the oscilloscope, but the component potentials
are not well resolved in suction electrode recordings, but merge into an irregular
wave. In the clearest recordings the event can be broken down into about 6-8
separate potentials, 50-70 ms apart. No two bursts are the same, and the same burst
may show different time relationships when recorded at two different places (Figs.
5 A, B). Initial attempts by the first author to insert glass microelectrodes into the
epithelial cells forming the wall of the ampullae were unsuccessful, but later A. N.
Spencer succeeded in obtaining an intracellular recording of an action potential of
duration ca. 50 ms, rising from a 48 mv resting potential (Fig. 5C). The thinness
of the epithelium and the lack of firmness of the tissue generally makes microelec-
trode work difficult in this material. The one successful recording was of brief du-
ration, and probably provides a somewhat attenuated version of the spike. Though
COORDINATION OF COMPOUND ASCIDIANS 215
-^
j
FIGURE 5. Network potentials (NPs) recorded from ampullae ofBotryllus (A, B, D) and Botrylloides
(C). (A), NP burst recorded at two different ampullae in a 2-zooid colony. (B), same preparation as in
(A), another NP burst. (C), intracellular recording of a single NP. (D), time relationships within an NP
burst. Triggering from the lower trace reveals two sets of NPs within a single burst in terms of wave form
and arrival time on upper trace. Bars represent 500 ^v, 200 ms in (A) and (B), 20 mv, 200 ms in (C)
and 1 mv, 50 ms in (D).
preliminary in nature, this result does confirm that the ampullar wall is the site of
impulse conduction. Taking into account the absence of nerves, and the presence
of gap junctions we can conclude that the excitable elements are the cells of the
vascular epithelium itself.
Measurement of conduction velocities is hampered by the irregular configuration
of the vascular network along which the signals are conducted and by the composite
nature of the NP itself. Values ranging from 0.5-1.9 cm s'1 (X == 0.9. SD == 0.59)
have been obtained. The distances between the electrodes in these experiments were
measured directly and no allowance was made for deviousness in the actual con-
duction pathways. The highest velocity values were obtained from strips near the
margin of the colony where a major vessel runs circumferentially (Fig. 2), and the
electrodes were placed along this line. Thus, the 'true' conduction velocity is probably
close to 2 cm s"1.
As noted above, simultaneous recordings from two ampullae in the same general
vicinity show differences in the time relationships of bursts arriving at the two sites
(Figure 5A, B) as well as in the numbers of potentials comprising the bursts, and
the intervals between them (Fig. 4B). It seems likely that bursts are produced by
interaction of many different pacemaker sites. Figure 5D shows an example of an
apparent shift in the pacemaker site during the course of a single burst.
Coordination of protective responses in the zooids
Like a number of other ascidians, both compound and solitary (Mackie, 1974,
Mackie et ai, 1974) botryllid zooids show characteristic, large potentials when stim-
ulated, termed ciliary arrest potentials (CAPs). The CAP system is under nervous
control from the brain. Muscles in the siphons and mantle usually contract con-
currently with ciliary arrest, but the potentials due to muscle contraction are small.
Botryllids differ from other ascidians by showing a second major type of electrical
signal in recordings from their zooids. These events appear indistinguishable from
NPs recorded from the ampullae, and it is concluded that NPs are conducted through
the vascular network, to the zooids. Here they may merge giving a composite ir-
regular sort of electrical event (Fig. 6A) or remain recognizable as discrete events
(Fig. 6B). Although the distances are small, conduction velocities have been mea-
sured in isolated strips of tissue cut from the surface of Botryllus zoooids between
the oral and cloacal siphons. Such strips give NP conduction velocities in the order
of 2.0 cm s"1. The NP conducting tissues in such strips have not been identified.
216 G. O. MACKIE AND C. L. SINGLA
ZOOID
/A AMPULLA
1 1111
J
ZOOID
AMPULLA
D
FIGURE 6. Two way transmission of NPs (arrow heads) between ampullae and zooids
(A, B, E, F) and Botrylloides (C, D). (A), a tactile stimulus (needle prick) to an ampulla evokes a NP
burst which propagates to a zooid causing it to contract and to another ampulla. (B), same as in (A), but
spontaneous NP burst, for comparison. The zooid did not contract. (C), following a shock (asterisk) on
a zooid an NP burst is recorded on another zooid. (D), same as in (C), several sweeps superimposed. The
NP burst fails to reach the recording site on some occasions. NP bursts which do arrive trigger CAPs
(spots) after variable latencies. (E), stimulating and recording on same zooid. With 4v shock, only a CAP
is evoked. Stronger shocks caused repetitive firing of CAPs and triggered NP bursts. (F), same as in (E),
but with a second electrode on an ampulla. The CAP does not spread to the ampulla, but the NP burst
does. Bars represent 500 ^v, 50 ms in (A) and (B), 500 juv, 100 ms in (C) and (D), and 1 mv, 50 ms in
(E) and (F).
They could be the blood vessels of the mantle, or the inner or outer mantle epithelia.
No other organs were in the strips in question.
Stimulation of a zooid can evoke a NP burst which propagates to other zooids
(Fig. 6C, D) and may trigger the usual effector responses (CAPs and muscle twitches)
in them. There is a delay of at least 100 ms between the arrival of the NP burst and
the production of the triggered events, which suggests that excitation passes into and
through the central nervous system of the zooid before entering the ciliary and
muscular effectors. It is not known how NPs, as epithelial events, enter the nervous
system, but epithelio-neural transmission steps have been identified in other tuni-
cates, e.g., Oikopleura (reviewed by Bone and Mackie, 1982). Stimulating the surface
of a zooid at low voltage may evoke one or a series of CAPs, with muscle twitches,
but slightly stronger shocks can evoke NPs as well (Fig. 6E). Both of these electrical
events, CAPs and NPs, are evoked after delays indicating passage through the ner-
vous system rather than being due to the direct action of stimulating current across
the body wall. The delay is greater in the case of NPs, which might reflect the greater
length of the motor pathway involved. NPs evoked in this way can propagate to
other zooids and to ampullae (Fig. 6F). CAPs never spread outside the zooid they
are evoked in, but they can be elicited indirectly in other zooids by propagated NPs
(Fig. 6D).
NPs cannot spread between zooids by way of the upper surface of the colony,
even though the mantles of the adjacent zooids are closely applied to one another
COORDINATION OF COMPOUND ASCIDIANS 217
in this region (Fig. 1 ). The routes in and out of zooids must lie deeper, and are
presumably the vascular connections.
NPs do not always spread beyond the confines of a stimulated zooid, and they
do not always enter zooids when spreading through the vascular network. When
they do enter, they may or may not cause CAPs. Conduction barriers must exist.
It seems likely, but has not been proven, that the time relationships and number
of NPs in a burst is critical in overcoming these barriers. There is some evidence
that conduction velocity declines within a burst. If so, the intervals between the
pulses in a burst would increase with distance from the site of stimulation, which
might account for activation of effectors in zooids near the stimulus, while distant
ones remain unaffected. Better evidence is needed on this point.
DISCUSSION
The findings reported here are of interest first because they throw new light on
the question of how contractions of the vascular ampullae are coordinated, secondly
because they reveal the existence of a capability for coordination of protective re-
sponses hitherto unrecognized in ascidian colonies, and finally for the interest at-
taching to a new case of a conducting epithelium.
Coordination of ampullae
The findings make it clear that the ampullae are coordinated by electrical im-
pulses. NP bursts always accompany ampullar contractions, and are phase locked
to the contraction cycle. Alterations in the ampullar rhythm due to stimulation or
injury, for example, are faithfully mirrored in the changed pattern of electrical
impulses. The rhythm, and accompanying NP burst pattern, is shown by small
groups of ampullae isolated from parts of the colony containing zooids, showing
that the peripheral vascular system generates the rhythm and provides the coordi-
nating pathway for the contractions. The observed conduction velocity of ca. 2 cm
s"1 though slow compared with most nervous and non-nervous conduction suffices
to keep large colonies up to and beyond 1.0 cm in diameter well coordinated.
Conduction in the vascular epithelium may take place by direct current flow through
the gap junctions shown to be present between the cells. The observation that the
ampullar rhythms of two colonies come into synchrony at the moment of fusion
suggests that a critical step in the self-recognition process is the ability to form gap
junctions with cells derived from another oozooid. While the present observations
clarify the problem of how the ampullae are coordinated, they still do not tell us
how the ampullar rhythm is generated. The evidence suggests multiple dispersed
pacemaker sites, but the identity of the cells generating the rhythm remains to be
determined.
Coordination of zooid responses
A number of functions have been associated with the vascular network and
ampullae of botryllids and other ascidians including respiration, circulation of me-
tabolites, secretion of tunic and substrate adhesive, and elimination of cells liberated
by degenerative processes (Abbott, 1953; Mukai et al, 1978; Katow and Watanabe,
1978; Torrence and Cloney, 1981). During asexual reproduction in Metandrocarpa
the vascular ampullae "withdraw the buds from the parental mantle and pull them
over the substrate to points some distance from the parent" (Abbott, 1953). Whether
218
G. O. MACKIE AND C. L. SINGLA
or not two botryllid colonies fuse during growth is determined by processes of self-
nonself discrimination at the ampullae, which are the only contact points between
the two colonies (Scofield et al, 1982; Watanabe and Taneda, 1982).
Another function can now be added to this list: the ampullae and vascular vessels
provide a conduction pathway mediating defensive behavior in the zooids. Any
sharp or damaging stimulus to the ampullae or connecting vessels causes siphonal
retraction and closure along with ciliary arrest in nearby zooids, equivalent to the
well known protective squirting of solitary sea squirts. Similar systems exist in other
colonial forms, both sessile and pelagic. Most animal colonies are coordinated by
nervous or non-nervous conduction pathways, sometimes by both. The closest par-
allels to the botryllid system are to be found in certain hydroid colonies where non-
nervous impulses are conducted along the stolons interconnecting the polyps. The
polyps retract protectively on receiving the excitation (reviewed by Spencer and
Schwab, 1982). Colonies of some other hydroids and of bryozoans and corals are
coordinated by nerves, but the responses are again protective in character (reviewed
by Shelton, 1982; Thorpe, 1982).
Conducting epithelia can provide an adequate pathway for simple impulse con-
duction over considerable distances, but the responses they mediate are complex
and labile, and at the effector end they are nearly always organized by nerves (e.g.,
Anderson and Bone, 1980; Mackie and Carre, 1983). In the botryllid system, there
is no reason to suspect the involvement of nerves in the coordination of ampullar
rhythms, but the effector responses of muscle contraction and ciliary arrest in the
zooids are almost certainly organized by nerves. In every zooid two-way epithelio-
m
ne
FIGURE 7. Wiring diagram of an idealized botryllid colony. Nerves are shown as solid lines, con-
ducting epithelia as broken, a, ampulla; b, brain; c, ciliated cell of branchial sac; en, epithelio-neural
transmission step; m, muscle in mantle wall; ne, neuro-epithelial step; s, sensory cell; v, vascular network.
Exact locations of the transmission steps between nerves and epithelia are unknown. This diagram is
designed to be understood in terms of Figure 1 1.23 in Bone and Mackie (1982).
COORDINATION OF COMPOUND ASCIDIANS 219
neural transmission links must exist by which NPs can excite impulses in afferent
nerves and be excited in turn by efferent nerves (Fig. 7). Nothing is known about
these links except that they exist. Another poorly understood feature is the mech-
anism responsible for incremental spread of excitation through the system. Such
spread is also known in corals and hydrozoan colonies and various mechanisms
have been proposed for it, but the evidence from botryllids is still too rudimentary
to justify further discussion here.
A new tunicate conducting epithelium
The skins of some larvaceans and of one ascidian tadpole (Dendrodod) are known
to be capable of impulse conduction, and conducting epithelia are widespread in
salps (reviewed by Anderson, 1980; Bone and Mackie, 1982). In each case, the
epithelium is a covering layer, and a 'pure' conducting epithelium, not a myoepi-
thelium. The botryllid system described in this paper is a system of blood vessels,
and it is contractile in at least some regions, due presumably to the actin-like mi-
crofilaments in the lining cells. Thus it more closely resembles the tunicate heart,
a myoepithelium which conducts the impulses for its own contraction (Kriebel,
1970). Like the heart, the vascular ampullae generate a rhythm. Nerves do not
appear to be involved in either case. It might appear then that the contractility,
rhythmicity and conducting ability are inherent properties of tunicate vascular tissue.
However it is worth noting that the vascular ampullae and vessels are of epidermal
origin, and that the epidermis of at least one tunicate tadpole larva is excitable.
Perhaps then, the excitability of the botryllid vascular network is ontogenetically or
phylogenetically linked to excitability in the larval epidermis. Whatever its origins,
the system does not fit neatly into any existing category and must be classed as a
new type of conducting epithelium. A study of the compound ascidian Distaplia
failed to reveal any conduction between the zooids (Mackie, 1974) and the zooids
in Pyrosoma colonies are not coordinated, except by serial photic auto-excitation
(see Bone and Mackie, 1982). Botryllids differ from these organisms in being truly
colonial: their zooids are interconnected by the vascular network. It may prove to
be the case that zooid coordination by propagated impulses is only developed in
those ascidian colonies which, like the botryllids, have direct vascular intercon-
nections.
ACKNOWLEDGMENTS
The physiological work reported here was carried out during a visit by the first
author to the Hopkins Marine Station, Stanford University, at Pacific Grove, Cal-
ifornia, with assistance in the form of an International Collaborative Research Grant
from the National Sciences and Engineering Research Council of Canada (NSERC).
This help, and the cooperation of the director and staff of the station is gratefully
acknowledged. D. P. Abbott, A. Harrington, and V. L. Scofield collaborated in
various phases of the work. We thank A. N. Spencer, University of Alberta, for
contributing Figure 5C. The electron microscopy was carried out by the second
author at the University of Victoria, with support from a NSERC operating grant.
LITERATURE CITED
ABBOTT, D. P. 1953. Asexual reproduction in the colonial ascidian Metandrocarpa taylori Huustman.
Univ. Cal. Publ. Zool. 61: 1-78.
ANDERSON, P. A. V. 1980. Epithelial conduction: its properties and functions. Prog. Neurobiol. 15: 161-
203.
220 G. O. MACKIE AND C. L. SINGLA
ANDERSON, P. A. V., AND Q. BONE. 1980. Communication between individuals in salp chains II. Phys-
iology. Proc. Roy. Soc. B. 210: 559-574.
BANCROFT, F. W. 1899. A new function of the vascular ampullae in the Botryllidae. Zoo/. Anz. 22: 450-
462.
BERRILL, N. J. 1941. The development of the bud in Botryllus. Biol. Bull. 80: 169-184.
BERRILL, N. J. 1947. The developmental cycle of Botrylloides. Q. J. Microsc. Sci. 88: 393-408.
BONE, Q., AND G. O. MACKIE. 1982. Urochordata. Pp. 473-535 in Electrical conduction and behaviour
in 'simple' invertebrates, G. A. B. Shelton, ed. Clarendon Press, Oxford.
BRUNETTI, R., AND P. BURIGHEL. 1969. Sviluppo dell'apparato vascolare coloniale in Botryllus schlosseri
(Pallas). Publ. Staz. Zoo/. Napoli 31: 137-148.
DESANTO, R. S., AND R. L. DUDLEY. 1969. Ultramicroscopic filaments in the ascidian Botryllus schlosseri
(Pallas) and their possible role in ampullar contractions. /. Ultrastr. Res. 28: 259-274.
KATOW, H., AND H. WATANABE. 1978. Fine structure and possible role of ampullae on tunic supply
and attachment in a compound ascidian, Botryllus primigenus Oka. J. Ultrastr. Res. 64: 23-34.
KRIEBEL, M. E. 1970. Wave front analyses of pulses in tunicate heart. Am. J. Physiol. 218: 1 194-1200.
MACKIE, G. O. 1974. Behaviour of a compound ascidian. Can. J. Zoo/. 52: 23-27.
MACKJE, G. O., D. H. PAUL, C. L. SINGLA, M. A. SLEIGH, AND D. E. WILLIAMS. 1974. Branchial
innervation and ciliary control in the ascidian Corella. Proc. Roy. Soc. B 187: 1-35.
MACKJE, G. O., AND D. CARRE. 1983. Coordination in a diphyid siphonophore. Mar. Behav. Physiol.
9: 139-170.
MUKAI, H., SUGIMOTO, K., AND TANEDA, Y. 1978. Comparative studies on the circulatory system of
the compound ascidians Botryllus, Botrylloides and Symplegma. J. Morphol. 157: 49-78.
SABBADIN, A. 1955. Osservazione sulla sviluppo, Taccrescimento, e la riproduzione di Botryllus schlosseri
(Pallas) in condizioni di laboratorio. Boll. Zoo/. 22: 243-263.
SCOFIELD, V. L., J. M. SCHLUMPBERGER, AND I. L. WEISMANN. 1982. Colony specificity in the colonial
tunicate Botryllus and the origins of vertebrate immunity. Am. Zoo/. 22: 783-794.
SHELTON, G. A. B. 1982. Anthozoa. Pp. 203-242 in Electrical Conduction and Behaviour in 'Simple'
Invertebrates, G. A. B. Shelton, ed. Clarendon Press, Oxford.
SPENCER, A. N., AND W. E. SCHWAB. 1982. Hydrozoa. Pp. 73-148 in Electrical Conduction and Behaviour
in 'Simple' Invertebrates, G. A. B. Shelton, ed. Clarendon Press, Oxford.
THORPE, J. P. 1982. Bryozoa. Pp. 393-439 in Electrical Conduction and Behaviour in 'Simple' Inver-
tebrates, G. A. B. Shelton, ed. Clarendon Press, Oxford.
TORRENCE, S. A., AND R. A. CLONEY. 1981. The rhythmic contractions of the ampullar epidermis
during metamorphosis of the ascidian Molgula accidentalis. Cell Tissue Res. 216: 293-312.
WATANABE, H., AND Y. TANEDA. 1982. Self and non-self recognition in compound ascidians. Am. Zoo/.
22: 775-782.
Reference: Biol. Bull. 165: 221-240. (August, 1983)
\SCIDI\N-PROCHLORON SYMBIOSIS: THE ROLE OF
LARVAL PHOTOADAPTATIONS IN MIDDAY LARVAL
RELEASE AND SETTLEMENT
RICHARD RANDOLPH OLSON
Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts 02138
ABSTRACT
Colonies of the algal-ascidian symbiosis Didemnum molle at Lizard Island, Aus-
tralia, release more than 95% of their larvae daily between 1 1:00 and 14:00 with a
peak around 12:30, shortly after meridian passage of the sun. In shallow-water
habitats, larvae are photoadapted to lower light environments than are adult col-
onies. Unlike adult colonies, larvae lack spicules and brown pigmentation in their
tunic. They also have a lower chlorophyll a/b ratio than do their parent colonies.
In the field, larvae seek a light intensity of approximately 100 fj-E m"2s"1 and settle
preferentially on dark or shaded substrata. Settled larvae that were transplanted into
full sunlight perished after 4 days. Larvae observed in the field swam for less than
10 minutes before settling. When denied a shaded substrate, larvae swam for up to
1.5 hours and eventually settled in full sunlight (an unsuitable habitat). Larvae in
total darkness swam for at least 2 hours before settling. The larval photoadaptations,
settlement behavior, and mortality of D. molle juveniles in full sunlight suggest that
the release of larvae at midday, when sunlight is greatest, enables larvae to search
for settlement sites when conditions are most severe, minimizing the chance they
will settle in unsuitable habitats.
INTRODUCTION
The availability of suitable habitats for the settlement of larvae of sessile marine
invertebrates is known to vary spatially (Grosberg, 1981; Palmer and Strathmann,
1981; Sebens, 1981; Keough and Downes, 1982) as well as temporally (Grosberg,
1982). Although considerable research has been conducted on factors that induce
larvae to settle (Meadows and Campbell, 1972), very little is known about the
ecological significance of the time of day that larvae are released. Many species of
sessile invertebrates have larvae that swim for less than an hour before settling [e.g.,
some ascidians (Crisp and Ghobashy, 1971), bryozoans (Ryland, 1974), and corals
(Lewis, 1974)]. Such a short time between larval release and settlement potentially
enables the parent to control the time of day its larvae will settle.
Colonial ascidians are commonly members of fouling communities in temperate
waters (Millar, 1971) and cryptic communities on coral reefs (Jackson, 1977). Eigh-
teen species of one family (Didemnidae) possess symbiotic unicellular algae (Kott,
1980). These species are found only in the tropics and commonly occur, not in
cryptic communities, but in fully sunlit areas on coral reefs. Although considerable
research has been conducted on the symbiotic algae, there is little known about the
ecology of the animals or their larvae.
Of the eighteen species of ascidian-algal associations, two species possess algae
of the genus Synechocystis (Lafargue and Duclaux, 1979; Olson, 1980), a cyanophyte
Received 17 February 1983; accepted 25 May 1983.
221
222
R. R. OLSON
which contains phycobilin pigments. The other species contain algae of the recently
discovered genus Prochloron (Newcomb and Pugh, 1975). The algae are unique in
that they have a cell structure (Whatley, 1977), cell wall (Moriarty, 1979), and
genome which resemble procaryotes (Seewaldt and Stackebrandt, 1982), but contain
chlorophyll b and lack phycobilin pigments, typical eucaryotic features (Lewin and
Withers, 1975). This contradiction has led to their designation as a new genus and
provisionary new division, the Prochlorophyta (Lewin, 1976).
Colonial ascidians studied to date, which do not have symbiotic algae, have been
reported to release their larvae primarily at dawn or upon first light after a period
of darkness. Of the thirteen species listed in Table I, nine release their larvae in the
morning, two release at midday, and two release larvae throughout the day-night
cycle. Duyl et al. (198 1 ) (Table II) reported the first case of a species in which larval
release takes place only during the midday hours. The larvae of this species, Tri-
didemnum solidum, possess symbiotic algae and are released between 10:15 and
14:00. Here I report on another species of colonial ascidian with symbiotic algae
which releases 95 percent of its larvae between 1 1:00 and 14:00. Experimental field
TABLE I
Larval release times for colonial ascidians without symbiotic algae
Species
Time of release
Location
Reference
Aplidium constellation
Botrylloides mutabilis
Botrylloides nigrum
Botryllus schlosseri
Cystodytes lobatus
Diplosoma listerianum
Distaplia occidentalis
Ecteinascidia turbinata
Leptoclinum mitsukurii
Metandrocarpa taylori
Perophora viridis
Polyandrocarpa tincta
Symplegma viride
dawn
all morning
dawn
morning
morning
all day with a peak at
midday
all day with a peak at
midday
3-4 hours after dawn, all
day
all day, peak at midday
morning
morning
morning
Woods Hole, MA
Woods Hole, MA
Woods Hole, MA
Tokyo Bay, Japan
Puerto Rico
Woods Hole, MA
Woods Hole, MA
Pacific Grove, CA
Menai Bridge, North
Wales
Friday Harbor, WA
Puerto Rico
Tokyo Bay, Japan
continuous over day/night Pacific Grove, CA
cycle
morning Friday Harbor, WA
early morning
8:00-11:00
morning
Woods Hole, MA
Woods Hole, MA
Tortugas, FL
continuous over day/night Tortugas, FL
cycle
Mast (1921)
Scott (1924)
Costelloand Henley (1971)
Yamaguchi (1975)
Morgan (1977)
Grave and Woodbridge
(1924)
Grave (1937)
Lambert (1979)
Crisp and Ghobashy
(1974)
Watanabe and Lambert
(1973)
Morgan (1977)
Yamaguchi (1975)
Abbott (1955)
Watanabe and Lambert
(1973)
Grave and McCosh (1924)
Costello and Henley (1971)
Grave (1936)
Grave (1937)
ASCIDIAN-ALGAL LARVAL ECOLOGY 223
TABLE II
Larval release times of colonial ascidians with symbiotic algae
Species
Time of release
Location
Reference
Didemnum molle
11:00-14:00
midday
Lizard Island, Australia
Palau, Caroline Islands
This paper
Olson, unpublished data
Diplosoma similis
midday
Lizard Island, Australia
Olson, unpublished data
Lissoclinum patella
midday
Palau, Caroline Islands
Lewin, pers. comm.
Lissodinum voeltzkowi
11:45-13:30
Lizard Island, Australia
Olson, unpublished data
Trididemnum solidum
10:15-14:00
Curacao
Duyl el al. (1981)
evidence is presented showing that light intensity of the habitat in which a larva
settles can be very important to its eventual growth and survival.
The colonial ascidian Didemnum molle Herdman, lives on coral reefs throughout
the Indo-West Pacific (Kott, 1980). All colonies contain symbiotic algae of the genus
Prochloron, which are extracellularly attached to the walls of the cloacal chambers
of the ascidian. The algae are shielded from full sunlight by the ascidian tunic which
contains spherical calcareous spicules (40-80 //m diameter) and a dark brown pig-
ment (Fig. la). The larva of/), molle (Fig. 2a) is relatively large (2.5 mm length)
and can be seen easily underwater. Its large size, midday release, short swimming
time, and relatively large amount of algae (0.39 ^g chlorophyll a/larva, s.d. = 0.09)
enabled me to study aspects of its larval ecology in the field which have not been
examined previously in an algal-invertebrate symbiosis.
MATERIALS AND METHODS
All experiments reported, unless otherwise noted, were conducted at a depth of
2 m on a patch reef approximately 200 m directly offshore of the Lizard Island
Research Station, Lizard Island, Australia (14 40' S. lat, 145 28' N. long.) from
August to December 1981. Few laboratory experiments were performed because D.
molle colonies will seldom survive for more than a day in aquaria, and experiments
conducted underwater on the reef more closely resemble the light and temperatures
to which the larvae are acclimated. All times reported are local mean time (LMT)
which is zone time corrected for longitude and the equation of time. LMT means
that the sun is directly overhead at exactly 12:00.
According to Kott (1980), D. molle colonies may be brown or white in color.
Recent findings (Olson, in prep.) suggest that the two color morphs are different
species. Brown colonies and their larvae were used for all experiments reported in
this paper.
"Settling panels" were 20 cm X 20 cm squares of 3 mm thick asbestos fiberboard
with 3 cm long wooden legs at each corner. The legs supported the panels slightly
above the substrate so that larvae could settle on the shaded undersides. "Juveniles"
were sexually immature (less than 0. 1 g wet weight) colonies that have a transparent
tunic and are not heavily spiculated. Such colonies appear green due to the algae
they contain (Figs. 2b, 2c). Sexual maturation occurs at 0.5 g wet weight (unpublished
data). "Edge distance" is the distance between a newly settled larva on the underside
of a settling panel and the nearest edge of the panel.
224
R. R. OLSON
FIGURE 1 . A — Adult colonies of D. molle fully expanded showing their single large exhalent opening
and many, small inhalent openings of individual zooids. Note hair-like projections from edge of lower
colony. These are extensions of the test, used for whole colony movement. B — D. molle habitat. This
aggregation of approximately 1 50 colonies at 2 m depth was used for larval release observations. Colonies
are in full sunlight. Juvenile colonies could be found on the underside of the boulder to which the colonies
are attached.
Field observations of larval release
The timing of larval release was studied by observing one clump of approximately
150 colonies, closely aggregated on a piece of coral rubble approximately 0.5 m
X 0.5 m across. Colonies were observed continuously from 1 1:00 to 15:00 for three
consecutive days every two weeks (on the full and new moons) from August to
December, 198 1 . As larvae were released from the colonies their time of release was
recorded. The data were grouped into 15 minute intervals.
ASCIDIAN-ALGAL LARVAL ECOLOGY
225
B
FIGURE 2. A — Larva of D. molle. Note three adhesive papillae (P). Prochloron algae (A) are attached
to hair-like projections from the posterior end of the larval body according to Kott (1980). Distance from
the base of the tail (T) to the tip of the middle adhesive papilla is approximately 1 mm. Larva contains
three blastozooids. B — 3 day-old juvenile. The colony has three zooids. White areas are spicules. Note
that spicules are located around zooids (Z). The rest of the colony is green from the algae. Individual
algal cells can be seen in this photo. Protrusions of test at base of colony are used for whole colony
locomotion. Bar equals 0.25 mm. C — 12 day-old juvenile. The colony has approximately 9 zooids. This
photo was taken in the laboratory so colony is tightly contracted. The colony has no brown pigmentation.
Bar equals 0.5 mm. D — Field photo of colony of approximately 30 zooids fully expanded. This colony
was found on the underside of a coral rubble boulder at 2 m depth. In the original color photo, the colony
can be seen to have a small amount of brown pigmentation on its topside.
Chlorophyll determinations
Chlorophyll content of whole colonies was measured by macerating individual
colonies in the dark in 20 to 40 ml (depending on colony size) of 90 percent acetone
buffered with MgCO3. Samples were extracted overnight at 10°C, then centrifuged
for 5 minutes at 2200 g. Samples were analyzed on a Varian spectrophotometer for
absorbance at 647 and 664 nm wavelengths. Chlorophyll a and b concentrations
were calculated with the equations of Jeffrey and Humphrey (1975).
Chlorophyll content of larvae was measured by the same methods. Larvae were
collected as they were released from freshly collected colonies in aquaria. 5 ml of
90 percent acetone were used for extraction of groups of 30 to 40 larvae.
Light intensity measurements
Light intensities were measured using a Li-Cor light meter with a cosine corrected
submersible quantum sensor. All measurements were taken on a calm, cloudless
day in February between 12:00 and 13:00. To estimate light levels beneath the
settling plates, holes were drilled at different distances from the edge. The holes were
the same diameter as the light probe so that the plate could be placed on the bottom
226 R. R- OLSON
with the light probe inserted upside down. The holes were drilled at 8, 20, and 35
mm distances from the edge. The first two distances correspond to the mean dis-
tances of larval settlement at 4 and 2 m depths, respectively. Light readings were
taken at 2 and 4 m depths.
Survivorship experiment
The importance of settling in a shaded habitat was investigated by comparing
survivorship of juveniles placed in a variety of light conditions. Larvae were allowed
to settle on the undersides of settling panels at 2 m depth. The panels were inverted
and subjected to one of the following three treatments: 1 ) shade — panel was covered
by another panel of identical dimensions mounted 3 cm above it, 2) full sunlight,
3) clear plexiglas roof — a control for alterations in sedimentation and flow in the
shade treatment. Juveniles on the undersides of uninverted panels served as controls
for the inversion. Survivorship was recorded after four days. The experiment was
replicated three times.
Swimming experiments
To determine how long larvae are capable of swimming, two experiments were
conducted — one in the lab, the other in the field. In the field experiment, larvae
were captured underwater with a 10 ml plastic syringe just as they were released
from their parent colony. Each replicate consisted of ten larvae captured within 2
minutes time to assure that they were all at approximately the same developmental
stage. The larvae were injected into plastic boxes measuring 1 2 cm X 8 cm X 6 cm.
In the dark treatment, the entire box was painted black on the outside with latex
paint and wrapped with two layers of aluminum foil. The dark roof treatment was
a clear box with the outside of the roof of the box painted black. The shade treatment
was a clear box placed 20 cm beneath a 80 cm X 60 cm white plastic shade. The
clear treatment was a clear box left in full sunlight. Each treatment was checked
every ten minutes for one hour except for the dark treatment which was checked
only at the end.
In the lab experiment, larvae were obtained from freshly collected colonies in
aquaria and placed into 500 ml beakers which were then wrapped in black plastic
and placed in a darkroom. After two hours the beakers were uncovered and the
larvae were censused.
Larval swimming observations
Larval swimming times were recorded for 89 larvae by visually tracking the
larvae underwater using scuba gear. Although eighty-four percent of the larvae fol-
lowed were lost as they swam among corals or when they reached the surface water
where surface waves would toss them around, 14 larvae were followed all the way
to settlement. The longest that any larvae were followed was 1 5 minutes, with the
exception of one larva which was swept off of the reef and swam for over 25 minutes
over the sand flats.
RESULTS
Larval release observations
The larvae of D. molle are easily visible underwater. They are approximately
2.5 mm total length with the main body approximately 1 mm in length (Fig. 2a).
25 AUG 25
n
-TH
AUG 29
,-r|Th"l-»-n
AUG
25
SEPT 27
U nfl
-i SEPT 28
SEPT 29
jt
ttk
o
UJ
<
UJ
25
OCT II
r
OCT 12
OCT 15
n-rw-rf]^
UJ
OCT 25
OCT 26
UJ
nff^
Jl-fl h
cr
< 25
NOV 9
NOV 10
u_
0
cc o
nT
iffn
dflr
UJ
CD
m 25
NOV 25
NOV 27
0
4
1
fit
OCT 27
.r-TWl
NOV ||
Jlh
tnl
J
""
-
-
^Wn
n oec 10
0 I -
1100 1200 J100 1400
OEC II
OEC 12
1100 1200 1300 1400 1100 1200 1300 1400
TIME OF DAY (LMT)
FIGURE 3. Daily larval release observations of 1981 showing the consistency of midday release
times between days. The ordinate is number of larvae released during each 1 5 minute interval from the
group of approximately 150 colonies shown in Figure 2B. The total number of colonies observed varied
from day to day due to colonies dividing, migrating, and mortality.
227
228
R. R. OLSON
The clump of bright green Prochloron algae attached to the larva make it easily
discernable as a dark spot against a light background. Larvae release themselves by
rupturing through the wall of the common cloaca where they have been developing,
then swim vigorously out of the large common cloacal aperture (Kott, 1980). Similar
to other colonial ascidian larvae (Millar, 1971), D. molle larvae are attracted to
bright light during the beginning of their swimming stage. Larvae generally swim
towards the surface, then drift back downward. After a short period of time (1-10
minutes) the larvae begin to seek dark surfaces. When visually following larvae, I
had to maintain a distance of at least 0.5 m to prevent them from swimming towards
my black wet suit.
A total of 88 larvae were followed in the field from their time of release. Of
these, 14 were followed all the way to settlement. Their swimming times ranged
from 40 seconds to 370 seconds, with a mean of 201 seconds (s.d. = 121). This
value is, of course, skewed to the lower end since longer swimming larvae have a
lesser chance of being followed all the way to settlement. However, it does show
that many larvae swim for a very short period of time.
Of the 1 4 larvae that were followed all the way to settlement, twelve settled on
the undersides of coral rubble and two settled on polyps of Porites coral. Fourteen
observations were made of pomacentrid fish ingesting D. molle larvae. In all in-
stances, the larvae were immediately egested and continued to swim, apparently
unharmed. One larva disappeared into the inhalent siphon of a solitary ascidian
(Polycarpa sp.). Several larvae were observed temporarily snagged on coral tentacles
(acroporids and poritids), but they managed to free themselves. It thus appears that
there are no major predators on the swimming stage of D. molle.
Larvae are released near midday. During a two week period larval settlement
on settling plates was monitored every day at 1 1:00 and 16:00. Ninety-three percent
of the recruitment took place during the midday interval. During hundreds of hours
underwater, no larvae were ever seen before 10:30 or after 15:00. Figure 3 shows
clear peaks in the daily time of larval release. The compilation of this data (Fig. 4)
Q
UJ
CO
UJ
_J
UJ
or
t-
z
Ul
tr
UJ
CL
25r
20
15--
10"
^rh-f-
1100
1200 1300
TIME OF DAY (LMT)
1400
1500
FIGURE 4. Compiled larval release times for all days of observation in which more than 40 larvae
were released (N == 15 days, 1126 larvae). Bars represent 95% confidence limits. Note that 95% of all
larvae were released between 1 1:00 and 14:00. Mean time of larval release is 12:54 (S.E. = 12.7 minutes).
ASCIDIAN-ALGAL LARVAL ECOLOGY
229
gives a mean time of release of 12:54 (SE : 12.5 minutes). The data do not differ
significantly from a normal distribution [Kolmogorov-Smirnoff, dmax = 0.053, P
> 0.1 (Sokal and Rohlf, 1969)].
The symbiotic algae of D. molle are extracellular to the animal host. In an adult
colony, the algae line the walls of the common cloaca. In a larva, the algae are
attached to small hairlike projections at the posterior end of the larval body (Kott,
1980). Although no data exist on the physiological importance of the algae to the
larva, it is doubtful that they contribute much to the larva's nutrition, considering
their external location.
Photoadaptations of the algae
The ratio of chlorophyll a to chlorophyll b is generally regarded as a relative
indicator of the light levels to which a plant is photoadapted (Boardman, 1977).
Chlorophyll b is an accessory photosynthetic pigment, absorbing light primarily
around 470 nm and 650 nm. It is usually produced in higher quantities in lower
light environments. This appears to hold true for the D. molle colonies analyzed
(Fig. 5). Adult colonies living deeper have greater amounts of chlorophyll b relative
to chlorophyll a. However, examination of larvae collected from shallow water
colonies (2 m depth) shows that their chl a/chl b ratio is less than that of their parent
colonies. The larvae thus appear to be photoadapted to lower light regimes than the
habitat of the parent colonies.
How can the larval algae have a lower chl a/chl b ratio than the parent from
which it was released? Although current research suggests that some phytoplankton
can alter their chlorophyll ratio in very short periods of time (Falkowski, 1980), it
is doubtful that this is the case for D. molle larvae since they have such a brief
swimming period and are exposed to a wide range of light intensities. The values
presented in Figure 5 represent averages for extracts from whole colonies. There is
undoubtedly a great deal of self-shading within a colony so that much less light
o
10-
CL
O
tr
O
O
5 I 0
DEPTH (m)
FIGURE 5. Chlorophyll a/b ratios of adult colonies (unlabeled points, N = 5 for each point), larvae
(L) (N = 8 extractions of 30-40 larvae), and 3 colonies bisected into top (T) and bottom (B) halves.
Error bars are standard deviations.
230 R. R. OLSON
reaches the bottom of the colony than the top. Three colonies from 2 m depth were
bisected with a razor blade into upper and lower halves. Each portion was analyzed
for chlorophylls. The results are points T (top) and B (bottom) in Figure 5. The
lower half has a chl a/chl b ratio approaching the larvae, suggesting that the larvae
gather their algae from the lower portion of the colony. Unfortunately, it was not
possible to observe larvae within the colony previous to release.
Substrate choice experiments
In the substrate choice experiment (Table III), the larvae almost unanimously
chose the dark substrata, indicating that they are capable of differentiation between
dark and light surfaces.
Larvae settled in a somewhat distinct band around the outer edge of the un-
dersides of settling panels. Figure 6 shows the frequency distributions of larvae settled
on panels at 2 and 4 m depths. Table IV gives the mean values for the edge distances.
Although there was considerable variation, there was a clear tendency for larvae to
settle closer to the edge at the deeper site. Comparison of observed edge distances
with the expected distribution based on random settlement and area alone, shows
that larvae settled predominantly near the edge (Fig. 7).
Light intensities were measured beneath the settling plates at the mean edge
distances (Table IV) of settled larvae. The light intensity at the mean edge distance
measured at the shallow and deep sites was 100 and 1 10 fiE irT2s ', respectively
(Fig. 8). Thus larvae appear to seek a light intensity of approximately 100 ;uE m
This means that at deeper sites, where light intensity on the top of surfaces is less
than 100 nE irT2s ', larvae should settle on the upper surfaces of substrata. At
Lizard Island, this light intensity occurs around 1 5 m depth. At this depth, juveniles
were found living in unshaded sites.
There are rare instances when newly settled larvae are found in unshaded habitats
in shallow water, but these are certainly the exception. As a part of another study
of D. molle recruitment at Lizard Island, settlement of larvae on settling panels at
2 m depth was recorded over five days every two weeks. Of over 3000 settled larvae,
only three settled on the topside of the settling panels. The rest settled on the
undersides or occasionally on the legs of the panels. Juvenile colonies were never
observed living in full sunlight. They are generally found on the undersides of coral
plates. Figure 9 shows the size distribution of D. molle colonies on the topside and
underside of a coral plate collected from 2 m depth. The 0. 1 gm size class of the
underside population was composed primarily of newly settled juveniles, still green
in color. None of the topside colonies were green.
TABLE III
Substrate choice experiment *
Roof
Trial N Black White Bottom Side Not Attached
1
35
65
0
6
3
26
2
33
40
3
30
24
3
3
71
79
0
15
6
0
* Roof of chamber was painted with black and white squares. Data are percent of all larvae in
treatment (N).
ASCIDIAN-ALGAL LARVAL ECOLOGY
231
25 H
NUMBER
0 F
LARVAE
SETTLED
0
25
0 12345
012345
DISTANCE FROM EDGE (cm)
FIGURE 6. Distributions of edge distances of larvae that settled on the undersides of settling panels
at 2 and 4 m depth during one week in December, 1981, and January, 1982.
Survivorship experiment
Four days of full sunlight was lethal to the newly settled larvae (Table V). Healthy
juveniles are colored bright green by their symbiotic algae. They have few spicules
and thus the algae can be seen through the tunic. Juveniles exposed to full sunlight
changed from bright green to light green to grayish-brown, then withered and died.
Colonies in the shade appeared healthy and of normal size and color, as did the
control colonies. Colonies beneath the clear plexiglas roof did not die as rapidly as
those fully exposed, probably because they were shaded by a small amount of sed-
iment which accumulated each day on the plexiglas roof. They did, however, show
the same evidence of deterioration and their mortality was also high. This is inter-
esting since plexiglas is an effective filter of ultraviolet radiation (Jagger, 1977).
Larval swimming time experiments
D. molle larvae are capable of swimming for more than an hour if they do not
find a suitable site for settlement (Fig. 10). In the clear treatment, all nonsettled
TABLE IV
Mean edge distances (X) of larvae settled on the bottoms of settling panels at 2 and 4 m depths*
Depth (m)
December sample
January sample
X
s.d.
X
s.d.
1.75
1.09
0.80
0.70
1.95
1.55
0.78
0.78
Measurements are cm from the edge of the panel, s.d. = standard deviation.
232
R. R. OLSON
50
NUMBER
OF
LARVAE
SETTLED
2 M
DEC
2 M
JAN
50
4 M
DEC
4 M
JAN
12345678 9
DISTANCE
I 23456789
FROM
EDGE
cm)
FIGURE 7. Comparison of observed (solid line) and expected random (dashed line) distributions
of edge distances of settled larvae. Expected distributions were calculated by multiplying the total number
of larvae by the amount of area in the distance interval.
larvae were lying on the bottom of the chamber after one hour, only occasionally
swimming up off the bottom. By 1.5 hours, all larvae had ceased swimming. This
is in contrast to the dark experiment in the lab (Table VI) in which almost half of
the larvae were still active after two hours. Thus it appears that larvae can swim
longer in lower light environments. This would be important to larvae swept into
deeper water.
What happens if larvae do not find a suitable substratum for settlement? It has
already been shown that larvae prefer dark substrata over light (Table III). This
result is seen again in the larval swimming experiment (Fig. 10). In the roof and
shade treatments the majority of larvae settled within the first twenty minutes. The
300
LIGHT
INTENSITY
(AiE/mVs)
200
100
10 20
DISTANCE FROM EDGE
OF PLATE (mm)
30
FIGURE 8. Light intensities measured on the undersides of settling panels. Points labeled "m'
correspond to mean "edge distances" of settled larvae at respective depths.
ASCIDIAN-ALGAL LARVAL ECOLOGY
233
10
UJ
TOP
fLTL
o
o
CC
UJ
QQ
2
ID
Z
15-
10-
5-
BOTTOM
0.5
1.0
WET WEIGHT (gm)
FIGURE 9. Size distributions of colonies on the top and bottom of a dead coral plate collected from
2 m depth. Note that the largest size class of colonies on the underside of the plate is the 0. 1 g size group.
larvae settled primarily on the top of the chamber in the roof and shade treatments
(Table VII).
In the clear treatment, where larvae were given no dark substrata or shade, the
larvae swam continuously upward. After about 45 minutes, most of the larvae lay
on the bottom, still swimming, but seldom raising above the bottom. Eventually
most of these larvae attached themselves to the bottom where they metamorphosed.
The survivorship experiment (Table V) showed that an unshaded settlement site is
lethal. Thus the denial of a suitable site (shade) eventually results in the larvae
settling in a much less suitable or certain-death habitat.
234
R. R. OLSON
TABLE V
Juvenile survivorship experiment*
Total
Mean %
number
survivorship
Standard
Level of
Treatment
of larvae
after 4 days
deviation
significance
Shade
34
77.3
28.0
Control
54
82.2
12.6
N.S.
Clear plexiglas
35
38.3
20.5
P < 0.01
Full sunlight
28
2.7
4.6
P < 0.001
* See text for explanation of treatments. Data tested for significant difference from shade treatment
using single factor analysis of variance (Fs = 12.23, P < 0.01) with Student-Newman-Keuls multiple
comparisons test. Data were arcsine transformed (Zar, 1974). Each treatment was replicated 3 times.
The ability of larvae to delay their settlement is important in an habitat like the
Lizard Island lagoon. Patch reefs provide plenty of suitable habitats for the larvae,
but between them lie bright white sand flats with little or no shaded substrata. By
postponing settlement, larvae can drift over the sand flats until they encounter
another patch reef, thus achieving inter-reef recruitment.
DISCUSSION
The larval stage ofDidemnum molle is not substantially different from the typical
colonial ascidian larval phase as described by Millar (1971). Upon release, larvae
are positively phototactic, swimming towards bright light. They gradually change
to negative phototaxis and negative geotaxis, swimming upwards and settling on the
undersides of dark surfaces. What appears to be distinguishing about D. molle (and
perhaps all ascidian-algal associations, see Table II) is that the larvae are released
only in the middle of the day with a peak shortly after meridian passage of the sun.
This phenomenon held true for D. molle during the three seasons (spring, summer,
winter) in which it was studied. Duyl el al. (1981) reported a similar midday timing
of larval release for Trididemnum solidum, a Caribbean ascidian-algal symbiosis.
Larval release by colonial ascidians has previously been reported to occur pri-
marily at dawn or first light after a long period of darkness (Table I). However, as
Kott (pers. comm.) notes, most of the ascidians studied have been temperate species.
In all of the papers cited in Table I, there is little speculation as to the functional
PERCENT
STILL
SWIMMING
BLACK SURFACE
SHADE
10 20 30 40
TIME (minutes)
FIGURE 10. Larval swimming time experiment. See text for explanation of treatments. Bars rep-
resent 95% confidence limits. Dark treatment was examined only at 60 minutes. Each treatment was
replicated four times, except the dark treatment (N = 3). Each replicate contained ten larvae.
ASCIDIAN-ALGAL LARVAL ECOLOGY
235
TABLE VI
Percent larvae swimming (sw), metamorphosed and attached (m/a). and metamorphosed but not
attached (m/na) after two hours in total darkness
Tnal
N
sw
m/a
m/na
1
25
48
40
12
2
16
44
44
12
3
12
33
33
33
significance of the timing of larval release. Watanabe and Lambert (1973) noted
that the larvae of Distaplia occidentalis are released only during daylight and pri-
marily in the morning. Their behavior and settlement is closely attuned to light
conditions, with the larvae settling in dark habitats. This presumably enables them
to find cracks and crevices which provide refuge from predators and physical stress
such as strong currents. But no experiments were performed to test whether sur-
vivorship is greater in cracks and crevices.
Experiments with the larvae of D. molle suggest a clear purpose for the midday
timing of larval release. The light intensity of the juvenile habitat appears to be a
very important (if not the most important) factor determining the suitability of the
settlement site. Too much light is lethal to the juvenile (Table V), too little light
reduces the growth and photosynthetic rate of the algae which probably has a direct
effect on the growth of the ascidian. By releasing larvae at midday, when light
intensity is greatest, adult colonies enable their larvae to search for settlement sites
TABLE VII
Settlement sites oj larvae in swimming experiment conducted on reef*
Larval settlement site
Treatment
Sw
T
B
Si
M
Multiple comparison
Clear
X%
47.5
10.0
30.0
12.5
0.0
Sw B Si T M
x%a
47.5
5.0
24.0
12.23
0.0
Sa
1.0
7.7
11.8
0.5
0.0
Shade
X%
5.0
45.0
15.0
35.0
0.0
T Si B Sw M
x%a
2.6
43.6
2.0
28.8
0.0
Sa
3.4
14.2
11.4
19.1
0.0
Black Surface
X%
12.5
50.0
27.5
7.5
2.5
T B Sw Si M
x%a
9.1
49.6
25.4
3.8
0.6
Sa
5.5
5.1
5.7
5.4
2.5
Dark
x%
3.3
6.7
20.0
56.7
13.3
Sw B M T Si
x%a
0.3
1.4
11.7
56.8
5.3
sa
1.0
4.0
8.8
1.4
4.0
* See text for explanation of treatments. Results of treatments underlined at right were not signif-
icantly different from each other using Student-Newman-Keuls multiple comparisons test (P < 0.05).
N = 4 trials for each treatment, except for dark treatment (N = 3). Each trial included ten larvae. Sw —
larvae still swimming; T — settlement on top of chamber; B — settlement on bottom of chamber; Si-
settlement on side of chamber; M — larvae that metamorphosed but did not attach; X% — mean percent
settlement; X%a — mean percent settlement using arcsin transform on data, then back transforming; sa —
standard deviation of transformed data.
236 R. R. OLSON
under the most extreme conditions, minimizing the chance of settling in a location
that is too bright.
Given that the algae within the parent colony are photoadapted to different light
levels according to their depth in the colony (Fig. 5), why should the larvae have
evolved to collect algae from the more shade adapted portion of the parent colony?
The larvae and juveniles lack the photoadaptations of the adult colonies and thus
must settle in a low light habitat. Adult colonies contain calcareous stellate spicules
[40-80 nm in diameter (Kott, 1980)] and a dark brown animal pigment (D. Parry,
pers. comm.) in the outer test of the colony. These materials shield the algae from
much visible and ultraviolet radiation. The test of the larvae and young juveniles
is transparent, lacking both the spicules and brown pigment. Juveniles are the color
of their symbiotic algae due to this transparency. After two days, juveniles begin to
produce spicules which originate around each zooid (Fig. 2b). Although calcareous
spicules are found in many colonial ascidians without symbiotic algae (Van Name,
1945), in the algal-ascidian symbioses they appear to have been modified into a
photobiological role. The Caribbean species Trididemnum solidum produces a sig-
nificantly higher proportion of spicule versus tissue in higher light intensity habitats
(Olson, 1980).
At about two weeks of age, brown pigmentation begins to appear in the test of
juvenile D. molle colonies. Experiments by Jokiel (1980) have demonstrated a lethal
effect of ultraviolet radiation on invertebrates that normally live in the shade. D.
molle has obviously evolved a means of protecting itself in the adult stage from the
damaging effects of ultraviolet radiation. It is probably the spicules and brown pig-
ment that achieve this. By gathering shade-adapted algae prior to release from the
parent colony, the larva is prepared to settle in a low light environment and thus
does not require spicules and brown pigment for a shield. The small size of the
larvae probably prohibits them from already possessing these photoadaptations upon
release.
Examination of dead coral plates and rubble on the shallow reefs around Lizard
Island shows that the undersides are the nurseries for D. molle (Fig. 9). Juvenile
colonies (green in color) are found only on the undersides of such substrata. It
appears that juveniles grow on the undersides until they have acquired the proper
photoadaptations (spicules and pigment), then migrate around the edge of the coral
plates into the full intensity of sunlight (up to 2600 nE irT2s') (Fig. 1 1). Colonial
ascidians have long been known to move, through the extension of stolonic test
vesicles (Carlisle, 1961). Birkeland et al. (1981) documented whole colony move-
ment by didemnid ascidians in general, and Cowan (1981) reported movement by
D. molle. I attempted to follow juveniles through this stage in the field by placing
markers beside them, but this proved impossible since it required daily monitoring
for more than a month. When occasional rough weather interrupted the monitoring,
the colonies were lost. Nevertheless, juveniles are found on the undersides and
adults are found mostly on the topsides (Fig. 9). The migration probably takes
several weeks.
Larval ecology of algal-invertebrate symbioses
Relatively little research has been conducted on the larval ecology of algal-in-
vertebrate symbioses. The findings of this study of D. molle are relevant to other
invertebrates with lecithotrophic larvae containing algae. These include corals such
as Pocillopora damicornis (Edmundson, 1929; Kawaguti, 1941; Harrigan, 1972),
Seriotopora hystrix (Atoda, 1951), and Favia fragum (Lewis, 1974).
ASCIDIAN-ALGAL LARVAL ECOLOGY
237
FIGURE 1 1. The life cycle of D. molle. Stages A-D correspond to photos in Figure 1. Larvae are
released and settle at midday. Tail resorption by the larva is completed approximately 20 minutes after
attachment. Complete metamorphosis takes approximately 3 hours. By late afternoon most colonies have
three functioning zooids (B). At six days all zooids divide to form a total of 6 zooids. Another synchronous
division usually occurs at 12 days, after which division is asynchronous. As a colony produces more
spicules and acquires brown pigmentation (D) it presumably migrates into full sunlight (E).
Lewis (1974) studied settlement of the hermatypic coral Favia fragum in the
laboratory. The larvae showed a preference for dark surfaces and settled primarily
on the undersides of substrata in dishes. When the substrata were inverted shortly
after settlement, the larvae detached themselves and moved to the underside again.
Lewis, surprised that a photosynthetic organism would settle in the shade, conjec-
tured that it was probably a predator avoidance phenomenon. No consideration was
given to the differences in photoadaptations between the adults and larvae. It is
possible that the juveniles of F. fragum, similar to D. molle, cannot survive in full
sunlight. Harrigan (1972) found that the larvae of P. damicornis also prefer to settle
on dark surfaces.
Goreau et al. (1981) examined settling patterns and mortality of planulae larvae
from the coral Porites porites. Larvae which settled on the sides of aquaria had a
much higher mortality rate than those which settled on the bottom. They suggested
that this might be due to reduced food availability. However, no consideration was
given to light as a factor. Juveniles on the side were illuminated from all sides,
whereas those on the bottom were dark on their undersides as well as being deeper
in the water. Survivorship of larvae in shade was not investigated. Mortality was
greatly reduced once the juveniles produced a skeleton. This may be analogous to
spicule production in D. molle juveniles which results in a tolerance for greater light
intensities.
238 R. R. OLSON
Birkeland (1977) found that Caribbean corals at 9 m depth recruited more
frequently on the sides and undersides of cement blocks than upper surfaces. At
deeper sites the recruitment shifted towards the upper surface. Birkeland et al. ( 198 1 )
report the same result for Pacific corals around Guam. His explanation is that
macroalgae and sediment on the upper surface inhibit coral larval settlement. No
mention is made of possible differences in photoadaptations between larvae and
adults. The shift towards the upper surfaces at greater depths is similar to larval
settlement edge distances for D. molle (Figs. 6, 7). Loya (1976) also found that the
coral Stylophorapistilata settled and survived primarily on the undersides of surfaces
in shallow- water habitats at Eilat, Israel.
Competition for space among sessile invertebrates is a popular explanation of
community structure on coral reefs (Lang, 1973; Jackson and Buss, 1975; Connell,
1976; Bak et al., 1977; Sheppard, 1979; Benahayu and Loya, 1981). It has been
assumed that adult sessile invertebrates usurp potential larval settlement area (Ma-
guire and Porter, 1977; Benahayu and Loya, 1981). If coral planulae are unable to
withstand full sunlight, then adults may generate more suitable settlement space
(their shade and undersides) than they consume. For the colonial ascidian Didem-
num molle, there is little overlap between the habitats of the adults and newly
metamorphosed larvae. Many other algal-invertebrate symbioses, upon close in-
spection, may follow the same pattern.
ACKNOWLEDGMENTS
The Lizard Island Research Station and the Australian Institute of Marine Sci-
ences generously provided research facilities. C. Cavanaugh, C. Crockroft, P. Filmer-
Sankey, B. and L. Goldman, V. Harriot, M. Harriss, T. Givnish, R. Grosberg, R.
Pardy, M. Patterson, H. Sweatman, K. Sebens, and L. Vail contributed helpful
discussion. Figure 2a was kindly provided by P. Parks. P. Mather donated collecting
materials. M. Harriss, H. Olson, and M. Patterson gave endless moral support for
which I am grateful. Lastly, I thank R. T. Paine and T. Suchanek for their initial
encouragement. This research was supported by the Franklin B. Knox Fellowship
of Harvard University and NSF grant OCE 81-14299.
LITERATURE CITED
ABBOTT, D. P. 1955. Larval structure and activity in the ascidian Metandrocarpa taylori. J. Morphol.
97: 569-594.
ATODA, K. 1951. The larval and postlarval development of some reef building corals. V. Seriotopora
hystrix Dana. Sci. Rep. Tohoku Univ. Fourth Ser. 19: 33-39.
BAK, R. P. M., J. J. W. M. BROUNS, AND F. M. L. HAYS. 1977. Regeneration and aspects of spatial
competition in the scleractinian corals Agaricia agaricites and Montastrea annularis. Proc. 3rd
Int. Symp. Coral Reefs 1: 143-148.
BENAYAHU, Y., AND Y. LOYA. 1981. Competition for space among coral-reef sessile organisms at Eilat,
Red Sea. Bull. Mar. Sci. 31(3): 514-522.
BIRKELAND, C. 1977. The importance of rate of biomass accumulation in early successional stages of
benthic communities to the survival of coral recruits. Proc. 3rd Int. Symp. Coral Reefs 1:15-
21.
BIRKELAND, C., L. CHENG, AND R. LEWIN, 1981. Motility of didemnid ascidian colonies. Bull. Mar.
Sci. 31: 170-173.
BIRKELAND, C., D. ROWLEY, AND R. H. RANDALL. 1981. Coral recruitment patterns at Guam. Proc.
4th Int. Coral Reef Symp. 2: 339-344.
BOARDMAN, R. K. 1977. Comparative photosynthesis of sun and shade plants. Ann. Rev. Plant Physiol
28: 355.
CARLISLE, D. B. 1961. Locomotory powers of adult ascidians. Proc. Zool. Soc. Lond. 136: 142-146.
ASCIDIAN-ALGAL LARVAL ECOLOGY 239
CONNELL, J. H. 1976. Competitive interactions and the species diversity of corals. Pp. 51-58 in Coel-
enterate Ecology and Behavior, G. O. Mackie, ed. Plenum Press, N. Y.
COSTELLO, D. P., AND C. HENLEY. 197 1 . Methods for obtaining and handling marine eggs and embryos,
2nd ed. Mar. Biol. Lab., Woods Hole, MA.
COWAN, M. E. 1981. Field observations of colony movement and division of the ascidian Didemnum
molle. Mar. Ecol. Prog. Ser. 6: 335-337.
CRISP, D. J., AND A. F. A. A. GHOBASHY. 1971. Responses of the larvae of Diplosoma listerianum to
light and gravity. Pp. 443-465 in 4th European Mar. Biol. Symp., D. J. Crisp, ed.
DUYL, F. C. VAN, R. P. M. BAK, AND J. SYSBESMA. 1981. The ecology of the tropical compound ascidian
Trididemnum solidum. I. Reproductive strategy and larval behaviour. Mar. Ecol. Prog. Ser. 6:
35-42.
EDMONDSON, C. H. 1929. Growth of Hawaiian corals. Bull. Bernice P. Bishop Mus. 58: 1-38.
FALKOWSKI, P. G. 1980. Light-shade adaptation in marine phytoplankton. Pp. 531-549 in Primary
Production in the Sea, P. G. Falkowski, ed. Plenum Press, New York.
GOREAU, N. I., T. J. GOREAU, AND R. L. HAYES. 1981. Settling, survivorship and spatial aggregation
in planulae and juveniles of the coral Porites porites (Pallas). Bull. Mar. Sci. 31(2): 424-435.
GRAVE, C. A. 1936. Metamorphosis of ascidian larvae. Carnegie Inst. Wash. Pap. Tortugas Lab. 29:
209-291.
GRAVE, C. A. 1937. Notes on the culture of eight species of ascidians. Pp. 560-564 in Culture Methods
for Invertebrate Animals, P. S. Galtsoff, F. E. Lutz, P. S Welch, and J. G. Needham, eds.
Comstock Press, Ithaca, NY.
GRAVE, C. A., AND G. McCosH. 1923. Perophoris viridis Verrill: The activities and structure of the free
swimming larvae. Wash. Univ. Stud. Sci. Ser. 11: 89-116.
GRAVE, C. A., AND H. WOODBRIDGE. 1924. Botryllus schlosseri (Pallas): The behavior and morphology
of the free swimming larva. J Morphol. 39: 207-247.
GROSBERG, R. K. 1981. Competitive ability influences habitat choice in marine invertebrates. Nature
290: 700-702.
GROSBERG, R. K. 1982. Intertidal zonation of barnacles: the influence of planktonic zonation of larvae
on vertical distribution of adults. Ecology 63(4): 894-899.
HARRIGAN, J. F. 1972. Behavior of the planula larva of the scleractinian coral Pocillopora damicornis
L. Am. Zool. 12: 723.
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., AND L. W. Buss. 1975. Allelopathy and spatial competition among coral reef inver-
tebrates. Proc. Nat. Acad. Sci. U.S.A. 72(12): 5160-5163.
JAGGER, J. 1977. Phototechnology and biological experimentation. Pp. 1-26 in The Science of Photo-
biology. Plenum Press, New York.
JEFFREY, S. W., AND G. F. HUMPHREY. 1975. New spectrophotometric equations for determining chlo-
rophylls a, b, c 1 , and c2 in higher plants, algae and natural phytoplankton. Biochem. Physiol.
Pflanz. 167: 191-194.
JOKJEL, P. 1980. Solar ultraviolet radiation and coral reef epifauna. Science 207: 1069-1071.
KAWAGUTI, S. 1941. On the physiology of reef corals V. Tropisms of coral planulae considered as a
factor of distribution. Palao Trap. Biol. Sta. Stud. 2: 319-328.
KOTT, P. 1980. Algal-bearing didemnid ascidians in the Indo-West Pacific. Mem. Queensl. Mus. 20(1):
1-47.
KEOUGH, M. J., AND B. J. DOWNES 1982. Recruitment of marine invertebrates: the role of active larval
choices and early mortality. Oecologia 54: 348-352.
LAFARGUE, F., AND G. DUCLAUX. 1979. Premier example en Atlantique tropical d'une association
symbiotique entre une ascidie didemnidae et une cyanophycee chroococale: Trididemnum cy-
anophorum nov. sp. et Synechocystis trididemni nov. sp. Ann. Inst. Oceanogr. Paris 55: 163-
184.
LAMBERT, G. 1979. Early post-metamorphic growth, budding and spicule formation in the compound
ascidian Cystodytes lobatus. Biol. Bull. 157: 464-477.
LANG, J. C. 1973. Interspecific aggression by scleractinian corals. 2. Why the race is not only to the swift.
Bull. Mar. Sci. 23: 260-279.
LEWIN, R. A. 1976. Prochlorophyta as a proposed new Division of algae. Nature 261: 697-698.
LEWIN, R. A., AND N. W. WITHERS. 1975. Extraordinary pigment composition of a prokaryotic alga.
Nature 256: 735-737.
LEWIS, J. B. 1974. The settlement behavior of planulae larvae of the hermatypic coral Favia fragum
(Esper). J. Exp. Mar. Biol. Ecol. 15: 165-172.
LOYA, Y. 1976. Settlement, mortality and recruitment of a Red Sea scleractinian coral population. Pp.
89-100 in Coelenterate Ecology and Behavior, G. O. Mackie, ed. Plenum Press, New York.
240 R. R. OLSON
MAGUIRE, L., AND J. W. PORTER. 1977. A spatial model of growth and competition strategies in coral
communities. Ecol. Model. 3: 249-271.
MAST, S. O. 1921. Reactions to light in the larvae of the ascidians, Amaroucium constellatum and
Amaroucium pellucidum with special reference to photic orientation. J. Exp. Zool. 34: 149-
187.
MEADOWS, P. S., AND J. I. CAMPBELL. 1972. Habitat selection by aquatic invertebrates. Adv. Mar. Biol.
10: 271-382.
MILLAR, R. H. 1971. The biology of ascidians. Adv. Mar. Biol. 9: 1-100.
MORGAN, T. O. 1977. Growth rate, age at sexual maturity, longevity, and seasonality in three West
Indian colonial ascidians. Masters thesis, Univ. Puerto Rico.
MORIARTY, D. J. W. 1979. Muramic acid in the cell walls of Prochloron. Arch. Microbiol. 120: 191-193.
NEWCOMB, E. H., AND T. D. PUGH. 1975. Blue-green algae associated with ascidians of the Great Barrier
Reef. Nature 253: 533-534.
OLSON, R. R. 1980. Sun-shade adaptations of a colonial ascidian with a procaryotic symbiont. Am. Zool.
20: 778.
PALMER, A. R., AND R. R. STRATHMANN. 1981. Scale of dispersal in varying environments and its
implications for life histories of marine invertebrates. Oecologia 48: 308-318.
RYLAND, J. S. 1974. Behavior, settlement, and metamorphosis of bryozoan larvae. Thalassia Jugosl. 10:
239-262.
SEBENS, K. P. 1981. Recruitment in a sea anemone population: Juvenile substrate becomes adult prey.
Science 213: 785-787.
SEEWALDT, E., AND E. STACKEBRANDT. 1982. Partial sequence of 16S ribosomal RNA and the phylogeny
of Prochloron. Nature 295: 618-620.
SCOTT, S. F. M. 1954. Metamorphic differentiation in Amaroucium constellatum treated with nitrogen
mustard. J. Exp. Zool. 127: 331-365.
SHEPPARD, C. R. C. 1979. Interspecific aggression between reef corals with reference to their distribution.
Mar. Ecol. Prog. Ser. 1: 237-247.
SOKAL, R., AND F. ROHLF. 1969. Biometry. W. H. Freeman Co., San Francisco, CA. 776 pp.
VAN NAME, W. G. 1945. The North and South American ascidians. Bull. Am. Mus. Nat. Hist. 84: 1-
45.
WATANABE, H., AND C. C. LAMBERT. 1973. Larva release in response to light by the compound ascidians
Distaplia occidentalis and Metandrocarpa taylori. Biol. Bull. 144: 556-566.
WHATLEY, J. M. 1977. The fine structures of Prochloron. New Phytol. 79: 309-313.
YAMAGUCHI, M. 1975. Growth and reproduction cycles of the marine fouling ascidians dona intestinalis,
Styela plicata. Botrylloides violaceus, and Leptoclinum mitsukurii at Aburatsubo-Moroiso Inlet
(Central Japan). Mar. Biol. 29: 253-259.
ZAR, J. H. 1974. Biostatistical Analysis. Prentice-Hall, Inc. Englewood Cliffs, NJ. 620 pp.
Reference: Biol. Bull. 165: 241-264. (August, 1983)
A NEW STRAIN OF PARATETRAMITUS JUGOSUS FROM
LACUNA FIGUEROA, BAJA CALIFORNIA, MEXICO
LAURIE K. READ1, LYNN MARGULIS1, JOHN STOLZ1, ROBERT OBAR2,
AND THOMAS K. SAWYER3
Department of Biology and2 Department of Chemistry. Boston University, Boston, MA 02215, and
^National Marine Fisheries Service, U. S. Department of Commerce, Oxford, MD 21654
ABSTRACT
A euryhalic, moderately temperature tolerant, fast growing strain of the amoe-
bomastigote1 Paratetramitus jugosus was isolated from the North Pond flat lami-
nated microbial mat at Laguna Figueroa, Baja California del Norte, Mexico. The
morphology was studied with phase contrast, differential interference contrast, scan-
ning, and transmission electron microscopy. On the basis of its life cycle charac-
teristics, growth rate, salt and heat tolerance, fluorescence excitation and emission
spectra, and isozymes, the organism was determined to be a new strain, P. jugosus
baja californiensis. This new strain, unlike the type specimen (ATCC 30703), grows
vigorously on half strength sea water and slowly at 0.5 1 M (nearly 3 per cent) sodium
chloride. It tolerates the hypersaline conditions of the evaporite flat that prevail
when the North Pond mats are dominated by Microcoleus and other bacteria, grow-
ing well during periods of influx of fresh water. Its cysts survive complete dryness
of the sediment for at least three years.
The microbial mats in which this Paratetramitus jugosus has been found are
thought to have Archean analogues over 3 billion years old. The discovery of resistant
abundant small eukaryotes within a setting dominated by bacteria may be important
for the interpretation of the Proterozoic microbial fossil record.
INTRODUCTION
We report here the isolation and identification of an extremely fast growing,
hardy, desiccation resistant new strain of the amoebomastigote (=amoeboflagellate)'
Paratetramitus jugosus from the North Pond of Laguna Figueroa, Baja California
del Norte, Mexico (Fig. 1).
The living microbial mats of Laguna Figueroa, Baja California, have been com-
pared with the 3400 million year old carbon-rich Fig Tree cherts of the Swaziland
System of rocks from South Africa (Margulis, et al, 1980). These laminated rocks,
as well as others also deposited over 3 billion years ago from western Australia
(Lowe, 1980), show a textural and paleoecological resemblance to the flat laminated
bacterial mats of Baja California. Fossils of bacteria have been found in some of the
most ancient cherts (Knoll and Barghoorn, 1977; Awramik et al, 1983) and there
is a continuous record of such microfossils from over three billion years ago to the
present. However, the time of appearance of the first eukaryotic microorganisms
Received 8 November 1982; accepted 18 May 1983.
' The terms "flagella" and "flagellate" are ambiguous since they refer to both flagellin-containing
bacterial structures and tubulin (9 + 2) eukaryotic structures, and the organisms which bear them,
respectively (Margulis, 1980). In this paper we restrict "flagella" and "flagellate" to the bacteria we
describe and use "undulipodia" for the eukaryotic structures and "mastigote" for the organism that bears
them. We replace "amoeboflagellate" with "amoebomastigote."
241
242
READ ET AL.
INDICE DE HOJAS
RANCHO RENDON
CUADRANTE SUOESTE
Sccretariq da la D*Un>a
Nocional, EXL de Menco
Baia California Zono 2,
San Quintin
(Hoio I de U <
Noro«ile, Son Qu.nlin)
ESCAIA I iO.OOO
I bO.OOO
I16°a0
FIGURE 1. Map of Baja California del Norte. A. Location of Laguna Figueroa and field sites. B.
The organism has been isolated from North Pond (site 1) and South Salinas (site 2).
in the fossil record is not known with certainty (Francis el al, 1978). The first protists
are thought to have appeared before 1400 million years ago (Knoll, 1982). We
initiated this study on the protistan composition of the bacterial mats to identify
the major eukaryotes in ecosystems overwhelmingly dominated by bacteria in the
hope of providing a better interpretation of the fossil record of laminated mats and
microorganisms preserved in cherts.
During the spring of 1979 an unusually severe flood occurred at Laguna Figu-
eroa, submerging the mats under one meter of fresh water until late August. The
flood water, which contained terrigenous sediment from the neighboring alluvial
plains, subsided by the late summer of 1979 but the rains of winter 1979-80 were
even more severe. From December 1979 until late summer 1981 the mats were
PARATETIL4MITUS 243
continuously flooded with fresh water. Never during the entire summer of 1980 did
the Microcoleus community emerge and grow. These episodes drastically altered the
composition of the mats from their relatively stable former state (described by Hor-
odyski el al, 1975, 1977). When the fresh water finally subsided, the productive
cyanobacterial community was replaced by a different community of heterotrophs
and purple photosynthetic sulfur bacteria, mainly by thiocapsoids (Margulis et al.,
1983; Stolz, 1983a). From both submerged samples of the Microcoleus mat and
from reemerged samples, Paratetramitus jugosus amoebae were recovered in im-
pressive numbers from every mat sample transferred onto permissive plates.
Several features of this amoebomastigote including its morphology, fluorescence,
as well as salt, heat, and desiccation tolerance are described here. The Baja California
isolate is compared with the original American Type Culture Collection strain
(ATCC 30703).
MATERIALS AND METHODS
Growth and isolation
The two kinds of media used in this study (modified K and manganese acetate.
Table I) were taken to the field study site at North Pond, Laguna Figueroa, Baja
TABLE I
Media
Modified K medium
MnSO4-4H2O 0.1 g
Bacto-Peptone l.Og
Yeast extract 0.25 g
Agar 7.5 g
500 ml ASW (autoclaved together)
Manganese acetate medium
Mn(C,H3O,), • 4H,O 0.002 per cent
(w/v) in ASW
(unless other concentration specified, i.e., from 2 X 10~5 to 10~3)
Artificial sea water (ASW)
CaCl2-2H2O 1.45g
MgSO4-7H,O 12.35 g
KC1 0.75 g
NaCl 17.55 g
Tris buffer (1.0 A/, pH 7.5) 50 ml
Distilled water 950 ml
Tris buffer
HC1 (cone) 33.3 ml
Trizma Base 60.55 g
bring to 500 ml with distilled water
Sawyer medium
Malt extract 0. 1 g
Yeast extract 0. 1 g
Difco agar 10 g
Distilled water 1000 ml
244 READ ET AL.
California (Margulis et al, 1980). About 1 mm2 samples of the flat laminated mi-
crobial mat (Margulis et al., 1980; Stolz, 1983a) were placed directly on sterile plates.
After 48 hours plates were then covered with about 1 ml of sterile distilled water
to resuspend the organisms and initiate a new growth cycle. After vigorous growth
occurred, cysts and amoebae were repeatedly subcultured on fresh medium by
streaking with a sterile platinum loop. In the final step of the isolation the organisms
were inoculated onto plates of modified K or manganese acetate medium with 2.4
per cent sodium chloride and checked for uniformity of cyst morphology. They were
then transferred from this medium to different conditions for study.
The organisms were routinely grown on modified K medium or manganese
acetate (McAc) medium (both of which contain half-concentrated sea water), or on
nonnutrient fresh-water agar with or without Klebsiella as food (Sawyer medium,
Table I). Growing cultures were kept at room temperature or in an incubator at
30°C. The major food source for P. jugosus was a gram positive, flagellated, fac-
ultatively aerobic rod which grew readily on modified K and MnAc media, and was
called the B bacillus.
Mastigotes were obtained by adding distilled water to agar plates of young cul-
tures 24-48 h old. Samples taken on the following day revealed that approximately
one third of the organisms had transformed into mastigotes, generally with more
than 2 undulipodia each.
American Type Culture Collection (ATCC) Paratetramitus jugosus strain no.
30703 was obtained in axenic medium no. 1034 (ATCC catalog, 1982). MnAc plates
containing 0. 1 M NaCl were inoculated with the food inoculum dominated by the
B bacillus and the ATCC P. jugosus. The ATCC P. jugosus grew better on this
medium with the food bacillus than on medium no. 711, the one routinely used
(ATCC catalog, 1982).
Storage of live material
The isolate was most easily preserved by storage of desiccated agar plates at 4°C.
Over the past 3 years cultures have been resuscitated within two or three days by
replating on modified K or MnAc medium. Healthy cultures have also been rees-
tablished from desiccated field samples or desiccated plates. A portion of the dry
sample was placed on fresh medium and flooded with about 1 ml of sterile distilled
water for at least 10 minutes.
P. jugosus also survived freezing. About 2 ml of sterile distilled water was placed
on each of several plates, 48-72 h old, containing healthy cultures of rounded forms
and cysts. The organisms were pipetted into centrifuge tubes, spun in a desk top
centrifuge at medium or high speed for about 10 min, and then resuspended in
Page's salt solution (Table I, ATCC catalogue, 1982, p. 633). The cyst concentration
was from 106 to 107 organisms per ml as determined by a counting chamber. The
amoeba suspension was then diluted by half with Tris buffer (Table I) to which 15
per cent dimethylsulfoxide (DMSO) had been added to yield a final DMSO con-
centration of 7.5 per cent (v/v). The DMSO-buffer- P. jugosus suspension was divided
into 1 ml plastic capped vials and frozen at -70°C. For resuscitation of the culture
contents of the vials were poured onto plates containing fresh medium.
Light microscopy
Living amoebae and cysts were observed using wet mounts with bright field
Nomarski, phase contrast and fluorescence optics (Nikon Optiphot and Fluorophot).
Agar coated slides were prepared to observe growing cultures. Alcohol-cleaned slides
PARATETR.4MITUS 245
were dipped into hot 1.5 per cent agar, and the undersides were wiped clean with
sterile cheesecloth. The cooled slides were inoculated down the center by streaking
the slide with a sterile platinum needle. Sterile cover slips, held up with bits of
plasticene, were placed on the inoculated preparations. The slides were incubated
in sterile Coplin jars or petri plates to which a few drops of sterile distilled water
were added from time to time. Growing amoebae and food bacteria could be main-
tained for at least a week under these conditions with very little contamination.
Measurements of live amoebae and cysts were made with a calibrated ocular mi-
crometer. Fifty amoebae and fifty rounded forms including mature cysts were
measured.
Nuclear division patterns were studied after staining with Kernechtrot (Dar-
byshire et ai, 1976). Prior to staining, blocks of agar containing amoebae and cysts
were transferred, upside down, into distilled water on microscope slides. These were
allowed to sit for about 45 min, in which time the amoebae swam into the water
and away from the agar. The agar was removed and the amoebae were fixed for 1 5
s in Nissenbaum's fixative (Nissenbaum, 1953) and treated with saturated iodine
alcohol. The fixed amoebae were then stained for 8 min in Kernechtrot (0.1 g in
a 5 per cent aqueous solution of A12(SO4)3). The preparations were dehydrated in
ethanol (70, 95, and 100 per cent) and xylene. for Protargol staining, the methods
of Zagon (1969) were used with modifications as suggested by Eugene Small, Uni-
versity of Maryland (pers. comm.).
Electron microscopy
For transmission electron microscopy, amoebae and cysts were fixed, embedded
and observed according to the methods described in Margulis et al., 1983.
For scanning electron microscopy organisms were suspended in distilled water
to produce mastigotes. The distilled water from suspensions harvested from several
petri plates was collected and the organisms were washed 12 times in 0.5 artificial
sea water (ASW, Table I) using a desk top centrifuge. The resuspended organisms
were fixed for 5 min in Parducz's fixative (6 parts 2 per cent osmium tetroxide in
0.5 ASW to 1 part saturated HgCl2 in distilled water) and washed 10 times in distilled
water. Amoebae were affixed to broken pieces of coverslip with 1 per cent polylysine
in distilled water. They were then dehydrated in a series of alcohols, dried in a
critical point dryer (Denton DC31 ), evaporated with a vacuum evaporator (Denton
DV502) and observed using SEM (AMR 1000) at 10 Kv at the University of Mas-
sachusetts at Boston.
Salt tolerance
Growth of P. jugosus as a function of salt concentration was measured between
0.0 and 0.60 M NaCl. An inoculation of 0.1 ml of the suspended culture in 5 ml
of distilled water was plated on each test plate. From the 3rd until the 26th day
plates were scored every 2 to 3 days for appearance of cysts relative to their food
bacteria. Using a dissecting microscope, outlines of areas covered by bacteria only
were compared to outlines of areas of bacterial colonies riddled with cysts. The
outlines were pencil-copied onto filter paper, cut out and weighed. The weight of
each cyst-covered outline was divided by the weight of the bacteria-covered outline
to yield relative amounts of amoeba growth. Since P. jugosus growth is limited to
the very surface of the plate and the results were consistent from experiment to
experiment, we felt this procedure was adequate to estimate the relative growth as
a function of salt and temperature. Growth was defined as continued production
246 READ ET AL.
of amoebae and cysts after three transfers 9 days apart. These experiments were
repeated three times for the new Baja California isolate, and twice for the ATCC
strain.
Heat tolerance
Growth of the amoebae as a function of temperature was measured by incubating
plates made with MnAc media at temperatures from 4°C to 48 °C. Ability to survive
high temperatures was tested by suspending samples of P. jugosus and their food
bacteria in distilled water and exposing them to elevated temperatures in water baths
for 10 min. The samples were then poured onto plates containing MnAc media and
incubated at 30°C to check for growth, which was denned as in the salt experiments.
Fluorescence
Chlorophyll fluorescence was used routinely to aid in the identification of cy-
anobacteria in mixed cultures on media designed to enrich for photosynthetic mi-
crobes. On such plates the yellow-green fluorescence associated with the cysts of P.
jugosus was observed. The Nikon Fluorophot microscopic observations were doc-
umented with a 35 mm mounted camera back and supplemented by measurements
of the excitation and emission peaks using a Perkin-Elmer MPF-44A Fluorescence
Spectrophotometer. Approximately 0.2 ml of concentrated mature cysts from plates
about two weeks old was spread on alcohol cleaned microscope slides which were
secured in the spectrophotometer either with tape or with a model no. 063-0502
solid sample holder attachment.
Enzyme analysis
Starch gel electrophoretic techniques for enzyme patterns were conducted under
a contract with the American Type Culture collection, Rockville, MD (Nerad and
Daggett, 1979). Both strains were tested for three isoenzyme systems: propionyl
esterase, leucine aminopeptidase and acid phosphatase (Daggett and Nerad, 1983).
RESULTS
Field studies and isolation
Recognition of cysts. Many of the mat samples plated in the field in 1979 and
1980 showed sporadic clumps of cysts. Some field sample plates were overrun by
cysts and others apparently lacked them entirely. Unidentified cysts appeared in low
numbers on media designed to enrich for manganese oxidizing bacteria in the sum-
mer of 1980. Samples of mixed bacteria and cysts were prepared for transmission
electron microscopy. A separate ultrastructural study of mat organisms, coccoid
chlorophytes grown on photosynthetic medium containing no carbon source (ASN
III, Rippka el al., 1979), also revealed cysts. These were very similar to those pre-
viously seen in the bacterial cultures (Fig. 2). Cysts on the photosynthetic medium
were well fixed and more abundant than those on heterotrophic media (Margulis
et al., 1983). Easily overlooked by light microscopy, the cysts could be differentiated
from the coccoid algae by their fluorescence spectra. Transfer of cysts onto fresh
low nutrient heterotrophic media (Table I) resulted in a higher yield of clearly
distinguishable cysts. Characterization of the cyst ultrastructure led to the recognition
of the same cysts in situ from 1977 laminated mat dominated by the cyanobacterium
Microcoleus chthonoplastes (Stolz, 1983a, b).
PA&4TETR.4MITUS
247
FIGURE 2. TEM of cysts in an algal culture, showing different stages of cyst development. Note
pore in younger, lighter cyst, a = alga, b = bacterium, c = cyst. Bar = 2
The amoebomastigote was subsequently recognized easily within 48 h on several
types of mixed culture plates: manganese acetate, K medium, or various photosyn-
thetic media either fresh from the field or in transferred or stored samples. The
appearance of white spots in dark colonies of manganese oxidizing bacilli (color
plate I, II, III) were taken as a presumptive test for the presence of cysts. With higher
magnification (200X or greater) the numerous amoeba cysts were seen among
clumps of manganese-coated spores (color plate IV) and distended food bacteria
(color plate V). The abundance of bacilli decreased as they were fed upon by the
amoebae and the area covered by white spots (which are the cysts) increased over
time as the amoebae digested the bacteria.
The food bacillus. The B bacillus was easily recognized: it measured about 4
Mm long and 1 ^m wide (Fig. 3A). It produced subterminal spores (Fig. 3B, C) and
formed smooth colonies that were beige and became brown centered in a few days.
When first isolated from the Laguna Figueroa mats in 1980 it oxidized manganese,
coating its spores within 4-7 days of incubation (Margulis et ai, 1983). During
subculture this ability to oxidize manganese was lost. In the presence of P. jugosus
these colonies became spotted with white cysts, then riddled with cysts, and finally
replaced entirely by cysts (color plate I, II, III, IV). However at least three other
types of bacteria were also present in this "B + cyst" inoculum in far smaller numbers
(Fig. 4). The B bacillus has been isolated in pure culture on at least two occasions
by taking advantage of the spores' resistance to temperatures up to 85°C for at least
ten minutes. This treatment killed the amoebae and cysts and all but one or oc-
casionally two of the bacterial types in the inoculum. The B bacillus colonies were
then easily picked and transferred to sterile plates and maintained indefinitely. When
inocula of P. jugosus were introduced into a pure culture of B bacilli, however, they
brought with them several other types of bacteria, presumably by adherence to their
248
READ ET AL.
•r
-
COLOR PLATE
I. Dark bacterial colonies riddled with light cysts after 2 days growth. Bar = 600
II. Dark bacterial colonies riddled with light cysts after 4-5 days growth. Bar = 1.2 ^m.
III. Colonies of manganese oxidizing (dark) and other heterotrophic bacteria taken directly from
the field. The "plaques" or cleared areas represent the growth of P. jugosus within colonies of manganese
oxidizing bacteria. At the lower left an entire dark colony has been converted to cysts. Bar = 100 nm.
IV. At higher power cysts can be seen among the manganese coated bacterial spores. Bar = 10
PAR.4TETR.4MITUS
249
FIGURE 3. B bacillus colonies with cysts. A. Agar slide preparation of a young culture prior to spore
formation (less than 1 day old) Nomarski differential interference microscopy. B. Subterminal spores
phase contrast microscopy C. Same as 3B. but Nomarski optics. Bar = 5
cysts. For this reason the P. jugosus cultures contained several bacterial types but
in fewer numbers than the B bacillus.
Morphology
Amoebae. The amoeboid form was monopodial when moving forward (Fig. 5A-
C). When stationary the amoebae often exhibited bulging forms typical of vahlk-
ampnds (Fig. 6A-C). Monopodial forms range in length from 12-24 /urn, averaging
17.2 ^tm. This fell in the lower part of the size range reported by Darbyshire (et al,
1976) for other strains of P. jugosus. The average length:breadth ratio was 3.2:1.
Occasional binucleate amoebae were seen, but fewer than the 7 per cent reported
V. Spaghetti-like masses of B bacillus in young culture infected with ectoplasmic forms, phase
contrast Bar = 20 nm.
VI. Phase contrast white light micrograph of fluorescent cysts. Bar = 5 jum.
VII. Matching fluorescence micrograph. Note dense granules which may correspond to autolyso-
somal or even chromatin bodies (arrows) (see figure 8D). Bar = 5 jim.
VIII. Mature cysts, note the binucleate cyst (arrow), phase contrast. Bar = 5 Mm.
IX. Mature cyst with pore (arrow). Bar = 5
250
READ ET AL.
y
.: #
I
_
FIGURE 4. TEM of a cyst in a mixed bacterial culture. Bar = 2 ^m. Amoeba at upper right, am
amoeba, b = bacteria, c = cyst, s = spore.
g
i *
FIGURE 5. Monopodial amoebae. A. Phase contrast Bar = 5 nm. B. SEM Bar = 2 nm. C. TEM
Note mitochondria with tubular cristae and granules, arrows = mitochondria, g = granules, n = nucleus,
s = spore. Bar = 2
PARA TETRAMITUS
251
FIGURE 6. Irregular vahlkampfids. A. Feeding, phase contrast, Bar = 5 Mm. B. Contractile vacuole
(black) and nucleus phase contrast. Bar = 5 nm. C. TEM with bacteria in food vacuoles. n = nucleus,
s = spore, v = food vacuole. Bar = 2
by Page (1967, 1976) for some strains. The cytoplasm contained many granules and
conspicuous vacuoles which contained bacteria, interpreted to be food vacuoles.
Ectoplasmic and small rounded forms. In actively growing cultures rounded
forms with thin or indistinct walls and often with an outer clear ectoplasmic layer
were by far the most obvious forms on the plates (Fig. 7A-H). These forms ranged
from 3-15 nm and often had large vacuoles containing bacteria, bacterial spores
and cytoplasmic granules. Except for occasional bulging the rounded forms were
stationary. In those with a distinct ectoplasm, the inner granular cytoplasm was
observed in various positions of protrusion beyond the ectoplasmic layer (Fig. 7E).
We surmise that the bulging cytoplasm protrudes through an organized opening.
Up to three such pore-like openings per rounded form could be distinguished by
phase microscopy (color plate IX) and in electron micrographs (Fig. 4).
Small rounded forms from less than 3 to about 5 /im in diameter were extremely
conspicuous in young healthy cultures and more mature ones which had been
flooded with distilled water to stimulate more growth. The small forms even out-
numbered the larger cyst-sized forms, especially in low salt medium. We compared
252
READ ET AL.
FIGURE 7. A-H. Ectoplasmic forms and affected bacteria, all of these are common in active cultures,
Bar = 5 nm. I-O. Mature cysts. I-L. Nomarski optics. M-O. Phase contrast. Bar = 5
the ATCC P. jugosus in axenic liquid medium immediately after it was received.
That culture also was filled with small bodies and rounded forms. We suggest that
these forms are stages in the life cycle of P. jugosus and certainly not contaminants,
possibly precysts or encysting amoebomastigotes.
Mature cysts. The mature cysts which appeared on the second or third day
generally had a distinct smooth round endocyst and an irregular ectocyst (Fig. 7
I-O). At some points the ectocyst contacted the endocyst. Cysts averaged 8 ^m in
diameter and ranged from 5.5-10 fj.m. The majority of the cysts were uninucleate
but binucleate cysts were seen occasionally (color plate VIII). Mature cysts tended
to become smaller as they aged and desiccated further. They also became more and
more fluorescent as they desiccated. Fluorescent materials seen as granules in moist-
ened cysts may be transferred to the walls as maturation proceeds. Bodies tradi-
tionally referred to as "autolysosomal bodies" in electron micrographs of rounded
forms are assumed to be related to breakdown of cell material and to rapid wall
PARA TETR.4MITUS
253
formation (Page, 1981). However, these bodies, which apparently contained ribo-
somes, and material that resembled chromatin may be related to the rapid prolif-
eration off. jugosus. These bodies were very conspicuous in rewet cysts and growing
cultures, like the nucleus they stained green with acridine orange. It is our judgment,
whatever their nature and development, that these intracellular bodies seen in the
electron micrographs (Fig. 8A-C) correspond to the bright bodies seen with phase
contrast microscopy (Fig. 8D) and are the source of fluorescence observed on the
light micrographic level (color plate VI, VII).
Mastigotes. Mastigotes were never observed on routine culture plates. We were
not aware of the ability to form a mastigote stage until it was brought to our attention
by F. C. Page of the Culture Centre of Algae and Protozoa, Cambridge, England.
Page, on the basis of the morphology of live cultures sent to him, kindly identified
the organism as P. jugosus. When suspended in distilled water overnight about one
third of the organisms transformed into mastigotes overnight. The mastigote stage
persisted for 1-2 days. Mastigotes were spherical, or more frequently, elongated in
shape; they had 2, sometimes more, forward directed undulipodia (Fig. 9). They
tend to be smaller than the amoebae. Bacterial spores could be seen in food vacuoles
through the transparent mastigotes. Whether or not the mastigote form actively
feeds is unknown; undigested bacteria and spores may have been residues from
feeding immediately prior to transformation.
FIGURE 8. Intracellular inclusions in cysts. A. TEM of "autolysosomal bodies" (a) and mitochondria
(m). Note the unidentified crystals inside the mitochondria (arrow). Bar = 1 pm. B. Cysts contain food
vacuoles (v) and "autolysosomal" bodies with ribosomes. Bar = 0.5 ^m. C. Cyst with bodies that may
contain chromatin (arrow), n = nucleus. Bar = 1 nm. D. Phase contrast light micrograph. Bar = 1 /urn.
254
READ ET AL.
FIGURE 9. Mastigotes. A-G The number of undulipodia per cell vary from 1 to as many as 1 1.
Bar = 5 j/m.
Growth and reproduction
The new strain off. jugosus grew extremely rapidly. New isolates from the field
entirely covered the B bacillus colonies with cysts within 3 days at room temperature
(color plate I-I V). About sixty cysts per colony developed from plated colonies using
a loop. After about a year in culture the growth rate slowed somewhat: it took from
4 to 5 days to entirely replace the food colonies with cysts. The ATCC P. jugosus
grew more slowly, not forming visible cysts at all until after the 12th day. It never
formed populations as dense as the Baja California isolate on any media tested (for
example the maximum number of cysts per colony was about 30 in the same test
that the Baja California strain developed about 60 cysts per colony). Even after
transfer from higher salt concentration (0. 1 M NaCl) the ATCC P. jugosus grew
very slowly and to low cell densities on our routine MnAc media
(Fig. 16).
We interpret the rounded forms to be active feeding stages. In young cultures
virtually devoid of monopodial amoebae the effects of P. jugosus on bacteria were
easily seen. Motility was lost and the bacteria became severely clumped and elon-
gated. Apparently P. jugosus arrested bacterial cell division, for when infected with
P. jugosus the bacilli would grow to up to 10 times their normal length and in some
cases spaghetti-like masses of unhealthy-appearing bacteria were seen (color plate
V, Fig. 10A, compare with Fig. 3). Thread-like material in which bacteria were
embedded could be seen in scanning electron micrographs (Fig. 10B). The material,
PAR.4TETRAMITUS
255
FIGURE 10. B bacillus infected with P. jugosus A. Swollen, elongated bacteria with ectoplasmic
forms, phase contrast. Bar = 20 ^m. B. Ectoplasmic forms, bacteria and apparent exudate SEM. Bar
consistently seen as cotton-like fluff in all active amoebae cultures and absent in
pure cultures of the B bacillus, may be part of the feeding process. Typical engulfing
by pseudopods was rare as it was for at least one other vahlkampfid described by
Page (1967). It is likely that P. jugosus can digest bacillus spores, even manganese-
coated ones. As P. jugosus grow on older colonies of bacteria which have all trans-
formed to manganese-encrusted spores, cyst-ridden plaque-like holes on plates re-
place the bacteria. Electron micrographs show bacterial spores in the cytoplasm of
the amoebae (Fig. 1 1).
FIGURE 1 1 . TEM of two vahlkampfid amoebae with spores (arrow) and bacteria in food vacuoles
(V). Bar = 2
256 READ ET AL.
When divisions were seen they were promitotic, characterized by the persistent
nuclear envelope as seen in two other vahlkampfids (Fig. 12). However, in over
two years of continuous and frequent observation including close monitoring of
agar slide cultures, divisions were rarely observed. Samples taken from 1-2 day old
cultures at frequent intervals (1 to 2 hours) failed to reveal any divisions after
examination with oil immersion microscopy. Indeed, there were few amoebae. Sim-
ilarly, amoebae stained with Kernechtrot or Protargol (Fig. 13) showed only a few
figures that could be interpreted as in division. However, these same active cultures
were replete with great numbers of small round bodies. These bodies contained
yellow-green fluorescent granules (such fluorescence is absent in pure cultures of
food bacillus). We suggest that these bodies may be involved in reproduction, thus
explaining the scarcity of vegetative amoebae and of their mitotic figures. The fre-
quency of amoebae was highest on moist plates. Generally fewer than 10 percent
of the forms in growing young cultures were amoebae and sometimes none at all
were seen. Protargol staining confirmed this observation. Small bodies appeared
entirely purple whereas only the nuclei of vegetative amoebae retained the stain.
The large round bodies contained purple nuclei and cytoplasmic bodies which also
stained (Fig. 1 3, A, B). As the cysts desiccated and matured, the entire round bodies,
large and small, stained heavily (Fig. 13 C-I). Both wall material and chromatin
stain heavily. On many occasions small round bodies associated with cysts were
observed (Fig. 14). These tiny rounded amoebae-like forms were often clumped in
groups of seven or eight. Their abundance and association with cysts and large
amoebae suggest they may be the product of a rapid series of standard mitoses or
multiple fission. Some of the released bodies were fecal pellets that were seen in the
amoebae (Fig. 15A) and in the medium (Fig. 15B). Fecal bodies, which are striped
and contain partially digested bacteria (Fig. 15C), could be distinguished from the
small amoeboid-like bodies. Nothing short of a sequential, carefully timed ultra-
structural study of development will solve the question. However, the astonishingly
fast reproductive rate, paucity of dividing amoebae, and the omnipresence of spher-
ical bodies which appear to contain chromatin suggest another mode of division in
addition to promitotic binary fission of amoebae.
Apparently, only amoebae transform directly into mastigotes. Plates containing
abundant amoebae formed mastigotes whereas old plates, predominantly mature
cysts, produced very few mastigotes when flooded.
FIGURE 12. Mitosis (A-D Bar = 5 nm): A. Interphase, B. Metaphase, C. Anaphase, D. Telophase.
PAR.4 TETR.4MITUS
257
FIGURE 13. Protargol stained preparations. Bar = 5 Mm. A. 12 day old cultures. B. 65 day old
cultures. C-I. Ectoplasmic, small rounded, granulated forms all of which are typical of young cultures.
Bar = 5 ^m.
When old plates with mature cysts were moistened and carefully observed, amoe-
bae could be seen to emerge from the encysted form. On one occasion a single cyst
was seen to convert to the monopodial amoeba form in about 10 minutes.
Salt
Although the growth of this P. jugosus isolate was most rapid in media made
with distilled water, it also grew well in half strength sea water medium (1.7 per
cent NaCl). Furthermore growth occurred in NaCl concentrations up to 0.50 M
(2.92 per cent). Figure 16A shows typical data from one of three experiments in
which growth was shown to be an inverse function of NaCl concentration. When
the inoculum size was large, growth was more vigorous even on 0.4 M (2.34 per
cent) NaCl, and the cysts covered the food bacterial colonies within 7 days. There
was even some continued growth (through three transfers) at 0.55 M (transferred
from 0.05 M). The ATCC P. jugosus strain also grew optimally in media without
NaCl, but did not grow in concentrations of salt above 0.3 M (Fig. 16B).
Cyst morphology changed as a function of salt, as is common in encysting
amoebae (Fig. 17). Presumably due to shrinkage of the cell, the space between the
endo- and the ectocyst widened at higher concentrations of salt and the cysts became
more refractile (Fig. 1 7C, F).
258
READ ET AL.
\
FIGURE 14. Small bodies released? (A., D., E., F., G., Phase contrast; B., C., Nomarski optics.
Bar = 5 pm). A. Clump of newly released amoebae? B. and C. enormous cyst from which (A., D. E.,
F., G., H.) have young amoebae been released? I. SEM spherical small amoebae? Bar = 2
The small, round wall-less forms so conspicuous in actively growing cultures
were less frequent at higher salt concentrations. The correlation of these bodies with
media that support the most rapid growth rate reinforces the hypothesis that these
bodies are directly involved in reproduction.
PARA TETRAMITUS
259
FIGURE 15. Fecal pellets. A. Amoeba containing fecal pellet (Bar = 1 f/m). B. Clumps of fecal
pellets (0 (Bar = 1 Mm)- C. TEM thin section of fecal pellet with disintegrating bacteria (b). s = spore.
Bar = 0.5
Temperature
The Baja California strain off. jugosus grew well between 20 and 36°C. It did
not grow at 37°C. It survived temperatures up to 56°C for 10 minutes, but did not
survive heat treatment for 10 minutes at 59°C. The ATCC P. jugosus also survived
temperatures up to 56°C for ten minutes. However, it did not grow when incubated
at 36°C, but grew well at 30°C. No temperatures were tested between these two
points.
Fluorescence
Yellow-green fluorescence emission from vahlkampfids is unreported. Yet in
our studies of Baja California microbial communities we have seen this phenomenon
consistently not only in small amoebic cysts but in larger unidentified acantham-
oebids. From the fluorescence data in Table II it can be seen that the two strains
of P. jugosus differ from one another in their emissive properties. Little is known
about the chemical basis or possible significance of this fluorescence, but the pos-
sibility of its use as a tool in diagnostics is obvious.
Fluorescence in these amoebomastigotes was strongly correlated with life cycle
stage. Amoebae do not fluoresce, yet the small rounded bodies had faint fluorescence.
The larger round bodies usually contained 2-6 strongly fluorescing bodies that
measured from 1-3 pm (Color plate VI, VII). Mature cyst walls fluoresced most
strongly; the cysts themselves fluoresced more and more strongly as they desiccated.
Although Page (1967) did not report fluorescence data he did describe cytoplasmic
granules in Vahlkampfia ornata. Dense spherules (1.5 to 2 /urn in diameter) were
260
READ ET AL.
PER CENT
120 -r-
fa»»T[TKaitlTUS JU8OSUS BttJft CRLlfORHin
«o -
20
11 13 13 DAYS
pap/>T£TnaniTus Jueosus arcc
PER CENT
120
100
40
FIGURE 16. Growth of P. jugosus as a function of salt. A. The Baja California strain: area of colony
covered by cysts in media made up from 0.0 to 0.4 M (2.34 per cent) sodium chloride. B. The ATCC
strain: area of colony covered by cysts in media made up from 0.0 to 0.3 M sodium chloride.
present in immature V. ornata cysts whereas mature cysts showed only fine gran-
ulation. There is most likely a relationship between the fluorescent material, the
granules, and the conspicuous autolysosomal bodies seen in electron micrographs
(Fig. 8).
Isoenzymes
The electrophoretic mobility pattern for three different enzymes of P. jugosus
from Baja California was compared with that from the P. jugosus from the ATCC.
The patterns for propionyl esterase and leucine aminopeptidase were nearly iden-
tical. However, there were conspicuous differences between the two stains with re-
spect to their alkaline phosphatase (Fig. 18).
DISCUSSION
Paratetramitus jugosus was proposed by Darbyshire (et al., 1976) as a new genus
and species of amoebae isolated from a stream near Moscow, Idaho. Before the
PAIL4TETR.4MITUS
261
FIGURE 17. Cyst morphology as a function of salt concentration. A. and D. 0.5 per cent. B. and
E. 1.2 per cent. C. and F. 2.4 per cent. A., B., and C. Bar = 10 ^m. D., E., and F. Bar = 5 /urn.
mastigote stage had been seen, P. jugosus had been introduced by Page (1967) into
the literature as Vahlkampfia jugosus. These amoebae have a closed nuclear division
pattern (promitosis), eruptive monopodial pseudopods, and temporary amoebom-
astigote stages. They belong to the family Vahlkampfidae (Page, 1976).
Different geographical strains have been isolated from Scottish soil samples, fresh
water lakes and streams in England and in the United States (Darbyshire et al,
1976), and from a Czech swimming pool (Cerva, 1971). All of the strains originally
identified as Vahlkampfia jugosus also transformed into mastigotes and thus were
reclassified by Page (1976) as Paratetramitus jugosus.
We believe that the differences in growth rate, salt and heat tolerance, isoenzyme
pattern, fluorescence emission maxima, and extraordinary desiccation resistance
constitute enough difference to recognize this protist as a new strain, Paratetramitus
jugosus baja californiensis.
On the coastal evaporite flat of Laguna Figueroa the protist survives but does
not grow during normal periods of cyanobacterial organic mat deposition and ex-
Fluorescence of mature cysts
TABLE II
Strain
Maxima
Excitation (nm)
Emission (nm)
Paratetramitus jugosos BC
Paratetramitus jugosus ATCC
488
488
611
592
262
READ ET AL.
PjBC
ATCC
•o
m
PjBC
ATCC
1
1
>
•o
PjBC
ATCC
TT
T
>
•o
FIGURE 18. Isoenzymes: PE = propionyl esterase, LAP = leucine aminopeptidase, AP = acid
phosphatase. Starch gel electrophoresis, movement right to left.
tremely high evaporation rates. Once a year during winter-spring rains, however,
conditions become ideal for the rapid growth of P. jugosus; blooms become obvious.
The unusual weather conditions of 1979-1980 conspired to retain such superb
growing conditions for P. jugosus that even after severe desiccation three years later
it was the dominant organism enriched from mat material in the laboratory.
During the winter and spring of 1982 there was extremely little or no rainfall
at the field site. As the flood water evaporated, the mat condition became more
saline, recolonization by halophilic bacteria began, and the growth of P. jugosus
diminished, decreasing the frequency with which the protist was isolated in the
summer of 1982. We conclude that this amoebomastigote is highly adapted to the
transient appearance of fresh water.
As described elsewhere in detail (Margulis et al., 1983; Stolz, 1983a, b), during
the summer of 1982 the laminated microbial mat which developed from before
1965 thru 1979, although covered by several centimeters of terrigenous sediment
due to the flood, was found buried from 10-15 cm below the surface of newly
forming mat. Between the older laminated mat and the new growth we observed
a smooth organic-rich mud smelling of sulfide which contained remains of cyano-
bacterial sheaths, and heterotrophic bacteria of many kinds. It also contained the
abundant P. jugosus. If silicified, this smooth black mud layer, as it lithified to chert,
would likely preserve entrapped microbes and their remains. Because of their high
population densities, euryhalinity and environment of deposition, hardiness, and
resistance, P. jugosus is likely to have a high preservation potential. Indeed it is
possible that it has already been reported in the microbial fossil record as "acritarchs"
or other problematica (Knoll, 1982; Vidal and Knoll, 1983).
Only further study can solve the mystery of the mode of reproduction in young,
relatively dry cultures on agar plates that contain many spheres and nearly no
amoebae. Multiple fissioning of some kind that produces small spheres may occur.
In studies of similar small amoeboid forms from oysters, Hogue (1914) diagrammed
amoebal multiple fission uncannily similar to what we have seen. Even though it
is likely that Hogue's studies were plagued by mixed cultures of protists we think
PARATETRAMITUS 263
her suggestion of multiple fission deserves reconsideration. The ubiquitous small
spheres may be the active feeding and multiplying forms of P. jugosus. Either very
rapid mitosis or multiple fission to produce small spherical forms occurs.
ACKNOWLEDGMENTS
This work was inspired by the NASA program of Planetary Biology and Mi-
crobial Ecology (1980, University of Santa Clara) and supported by NASA-NGR-
004-025 and the Boston University Graduate School. We acknowledge with gratitude
the aid of Mary Smith and the University of Massachusetts Boston, with the scanning
electron microscopy as well as F. C. Page and B. Dexter-Dyer Grosovsky in both
the work and the manuscript. We are grateful to Lorraine Olendzenski, Susan Lenk,
Geronimo Sahagun, and Evelyn Ball for lab and field assistance as well as to Zachary
Margulis for help with the computer-related aspects of this project. We are grateful
to Pierre-Marc Daggett and Tom Nerad, American Type Culture Collection, Rock-
ville, MD for performing the isoenzyme analyses and to Shawn Bodammer, Natl.
Marine Fisheries Lab., Oxford MD who performed the initial staining studies, de-
voting many hours searching for the very few specimens undergoing nuclear division.
Prof. Lindsey Olive, University of North Carolina, suggested the malt extract-yeast
extract culture medium.
LITERATURE CITED
American Type Culture Collection Catalogue of Strains. 15th Ed. 1982. P.-M. Daggett, R. L. Gherna,
P. Pienta, W. Nierman, J. Shun-Chang, H. Hsu, B. Brandon, and M. T. Alexander, eds. ATCC,
Washington, DC.
AWRAMIK, S. M., J. W. SCHOPF, AND M. R. WALTER. 1983. Filamentous fossil bacteria from the Archean
of Western Australia. Precambrian Res. 20: in press.
CERVA, L., 1971. Studies of limax amoebae in a swimming pool. Hydrobiologia 38: 141-161.
DAGGETT, P.-M., AND T. A. NERAD. 1983. The biochemical identification of Vahlkampfid amoebae.
J. Protozool. 30: 126-128.
DARBYSHIRE, J. F., F. C. PAGE, AND L. P. GOODFELLOW, 1976. Paratetramitus jugosus: an amoebo-
flagellate of soils and fresh water, type-species of Paratetramitus nov. gen. Protistologica 12:
375-387.
FRANCIS, S., L. MARGULIS, AND E. S. BARGHOORN. 1978. On the experimental silicification of micro-
organisms II. On the time of appearance of eukaryotic organisms in the fossil record. Precam-
brian Res. 6: 65-100.
HOGUE, M. J. 1914. Studies in the life history of an amoeba of the Limax group. Arch. Protistenkd. 35:
154-163.
HORODYSKI, R. J. 1977. Lyngbya mats at Laguna Mormona, Baja California, Mexico: comparison with
Proterozoic stromatolites. J. Sediment. Petrol. 47: 680-696.
HORODYSKI, R. J., AND S. J. YONDER HAAR, 1975. Recent calcareous stromatolites from Laguna Mor-
mona (Baja California), Mexico, /. Sediment. Petrol. 45: 894-908.
KNOLL, A., 1982. Microorganisms from the late Precambrian Draken conglomerate, Ny Friesland, Spitz-
bergen. J. Paleontol. 56: 7755-7790.
KNOLL, A., AND E. S. BARGHOORN. 1977. Archean microfossils showing cell division from the Swaziland
System of South Africa. Science 198: 396-398.
LOWE, D. R. 1980. Stromatolites 3400 million years old from the Archean of Western Australia. Nature
284: 441-443.
MARGULIS, L. 1980. Undulipodia, flagella and cilia. Biosystems 12: 105-108.
MARGULIS, L., E. S. BARGHOORN, D. ASHENDORF, S. BANERJEE, D. CHASE, S. FRANCIS, S. GIOVANNONI,
AND J. STOLZ. 1980. The microbial community in the layered sediments at Laguna Figueroa,
Baja California, Mexico: Does it have Precambrian Analogues? Precambrian Res. 11: 93-123.
MARGULIS, L., B. D. D. GROSOVSKY, J. F. STOLZ, E. J. GONG- COLLINS, S. LENK, D. READ, AND A.
LOPEZ-CORTES, 1983. Distinctive microbial structures and the prephanerozoic fossil record.
Precambrian Res. 20: (in press).
NASA Planetary Biology and Microbial Ecology Summer Course Report. 1980. University of Santa Clara.
NASA Life Sciences Office, Washington, DC. 203 pp.
264 READ ET AL.
NERAD, T. A., AND P. -M. DAGGETT, 1979. Starch electrophoresis; an effective method for separation
of pathogenic and non-pathogenic Naeglaria strains. J. Protozool. 26: 613-615.
NISSENBAUM, G. 1953. A combined method for the rapid fixation and adhesion of ciliates and flagellates.
Science 119: 31-32.
PAGE, F. C. 1967. Taxonomic criteria for limax amoebae with descriptions of three new species of
Hartmanella and three of Vahlkampfia. J. Protozool. 14: 499-521.
PAGE, F. C. 1976. A revised classification of the Gymnamoebia (Protozoa:Sarcodina). Zool. J. Linn. Soc.
58:61-77.
PAGE, F. C. 198 1 . A light- and electron microscopical study of Protacanthamoeba caledonica n. sp., type-
species of Protacanthamoeba n. g. (Amoebida, Acanthamoebidae). J. Protozool. 28: 70-78.
RIPPKA, R., J. DERUELLES, J. B. WATERBURY, M. HERDMAN, AND R. Y. STANIER. 1979. Generic
assignments, strain histories and properties of pure cultures of cyanobacteria. J. Gen. Microbiol.
110: 1-61.
STOLZ, J. F. 1983a. Fine structure of the stratified microbial community at Laguna Figueroa, Baja
California, Mexico. I. Methods of in situ study of the laminated sediments. Precambrian Res.
20: 479-492.
STOLZ, J. F. 1983b. Fine structure of the stratified microbial community at Laguna Figueroa, Baja
California, Mexico II. Transmission electron microscopy as a diagnostic tool in studying mi-
crobial communities in situ. In The Woods Hole Microbial Mat Symposium. R. Castenholz,
H. O. Halvorson, and Y. Cohen, eds., Alan Liss, Inc., New York.
VIDAL, G., AND A. H. KNOLL. 1983. Proterozoic plankton. Geol. Soc. Am. Mem. (in press).
ZAGON, I. S. 1969. Electron microscopic observations of protargol stain deposition in Carchesium po-
lypinum. J. Protozool. 16(Supp.): 7-8.
Reference: Biol. Bull. 165: 265-275. (August, 1983)
THE INITIAL CALCIFICATION PROCESS IN SHELL-REGENERATING
TEGULA (ARCHAEOGASTROPODA)
CHARLENE REED-MILLER
Department of Geology. Florida State University, Tallahassee, FL 32306
ABSTRACT
Shell regeneration was induced in the marine archaeogastropod, Tegula, by cut-
ting a window in the first body whorl of the shell. At six hour intervals for six days
after the shell window was cut, the mantle, foot, and hepatopancreas were prepared
for transmission electron microscopy, and the shell window was prepared for scan-
ning electron microscopy. Transmission electron microscopy of the three tissues
showed an increase in rough endoplasmic reticulum, Golgi complexes, and mito-
chondria, followed by the appearance of three types of inclusions. Later, intracellular
space increased and spherites were visible. Scanning electron microscopy showed
initial crystal deposition in the shell window to be in the form of small doubly-
pointed crystallites associated with an organic membrane. These spindle-shaped
crystals were frequently aggregated into radiating clusters or rosettes which coalesed
until a thin sheet of mineralized material covered the shell window, within six days
of shell injury.
INTRODUCTION
The regeneration or repair of molluscan shell is a subject of great interest. Most
of the studies of repair of mineralized tissues in molluscs have concerned terrestrial
or freshwater species (Wagge, 1951;Tsujii, 1960; 1976; Beedham, 1965; Saleuddin,
1967; Abolins-Krogis, 1968; Saleuddin and Wilbur, 1969; Kapur and Gupta, 1970;
Meenakshi et al., 1975; Blackwelder and Watabe, 1977). Reports on shell regen-
eration in marine molluscs include work on the cephalopod, Nautilus macromphalus
(Meenakshi et al., 1974), and the bivalve, Mytilus edulis (Meenakshi et al., 1973;
Uozumi and Ohata, 1977; Uozumi and Suzuki, 1979). One impression from these
studies is that marine molluscs require more time to repair their shells than do
terrestrial or freshwater species.
Meenakshi et al. (1974) report that it takes 45 days for shell regeneration to
occur in Nautilus, and 30-32 days must elapse following shell injury before the first
evidence of mineral deposition occurs in Mytilus (Meenakshi et al., 1973). Fur-
thermore, it takes at least eight weeks before the regenerated shell takes on a normal
appearance in Mytilus (Meenakshi et al, 1973; Uozumi and Ohata, 1977; Uozumi
and Suzuki, 1979). These chronologies are impressively long compared to the time
required for substantial calcium deposition in the land snails Helix and Otala, e.g.,
two to three days (Wilbur, 1973). The freshwater snail, Heliosoma, and the fresh-
water bivalve, Anodonta, require a somewhat longer time for mineral deposition —
at least five days (Chan and Saleuddin, 1974) and about 14 days (Tsujii, 1976)
respectively. But these freshwater molluscs still repair damaged shell faster than their
marine counterparts.
Received 1 November 1982; accepted 23 May 1983.
265
266 C. REED-MILLER
Preliminary observations showed that initial mineral deposition occurred about
24 to 48 hours after the creation of a shell window in the first body whorl of the
marine snail, Tegula (Reed-Miller et al, 1980). On the average, six days were re-
quired for a thin sheet of mineralized tissue to cover the shell window (Reed-Miller,
unpub. ob.). This paper describes the events of the first six days of shell repair in
the marine archaeogastropod, Tegula. Both scanning and transmission electron
microscopy were used. Preliminary accounts of this work were presented to the
American Society of Zoologists (Reed-Miller, 1981).
MATERIALS AND METHODS
Snails, Tegula funebralis and Tegula eiseni, were obtained from the Pacific
Biomarine Laboratories, Inc., Venice, CA. They were maintained in aquaria in
filtered, aerated sea water from the Gulf of Mexico (32 ppt), at 15°C. The animals
were fed marine algae from a laboratory culture.
A 4 mm2 section of shell was carefully removed from the first body whorl of the
shell using a Dremel "Moto-tool," jeweler's saw, and a triangular file. Care was
taken not to injure the underlying tissue. The opening in the shell, or window, was
covered with a small piece of a plastic coverslip, and that in turn was covered with
warm dental wax, sealing the window from the external environment.
Small pieces of the mantle from directly underneath the shell window, foot, and
hepatopancreas were carefully dissected from the animals six hours to six days after
the window was cut. As controls, the same tissues from normal, non-regenerating
snails were always prepared with those from shell regenerating animals. The ex-
periments were repeated at least three times with at least four experimental animals
examined each time.
Transmission electron microscopy
The soft tissues were dissected out and fixed at room temperature in 1% glu-
taraldehyde in filtered sea water (pH 7.2). The tissue was then washed three times
in a 1:1 sea watenglass distilled water solution. Following the third wash, the material
was postfixed for one hour in 1% osmium tetroxide in filtered sea water, rinsed with
glass distilled water, dehydrated through a graded series of ethanol, taken through
two changes of propylene oxide, and embedded in Medcast (Ted Pella, Inc., Tustin,
CA). Silver to gold sections were cut with a diamond knife, and stained with uranyl
acetate and lead citrate. The specimens were observed in a Philips 201 transmission
electron microscope operated at 60 kV.
Scanning electron microscopy
After removal of the soft parts, the shell was preserved in 70% ethanol. Then
the shell was carefully cut around the shell window with a rotary rock saw, until
a small frame of shell (about 3 mm wide) surrounded the window on all sides. This
frame and the shell window with the regenerated material were rinsed with distilled
water and air dried. The samples were mounted on aluminum scanning electron
microscopy stubs with nail polish, coated with 100-200 A of gold palladium (60:40),
using a E5 100 Polaron Sputter Coater. The material was observed with a Cambridge
S4-10 scanning electron microscope operated at 20 kV.
TEGULA SHELL REGENERATION
RESULTS
267
The structures of the three tissues from normal, nonregenerating snails were
unexceptional, and, in fact, identical to descriptions from transmission electron
microscope studies of those tissues appearing in the literature ( Abolins-Krogis, 1 96 1 ;
1963; Tsujii, 1976; Watabe et al, 1976). The purpose of this paper is to describe
the ultrastructural changes in the tissues during shell regeneration. The sequence
of events was consistent for each snail in the experimental group. However, the time
after the shell window was cut until each ultrastructural change was seen showed
some individual variation. Consequently, the results are outlined in time frames
following the creation of the shell window.
14-48 hours of regeneration
Three ultrastructural changes took place in the soft tissues during this stage of
shell regeneration. First, the amount of rough endoplasmic reticulum increased,
typically in the form of whorls or spirals (Fig. 1 ). Second, the number of Golgi
complexes also increased (Fig. 2), and third, juxtaposed with the Golgi complexes,
were open vesicles containing condensed or fibrous material (Fig. 3).
The shell window had small, doubly pointed or spindle-shaped crystals in and
on an organic membrane (Fig. 4). This was the first appearance of mineralized
material in the injured area of the shell. Often these crystals were aggregated into
radiating clusters or small rosettes (Fig. 5).
48-72 hours of regeneration
The predominant feature in the soft tissues during this phase of regeneration
were membrane-bound clusters of vesicles or vacuoles (Fig. 6). These inclusions
took several forms, some were aggregates of very dense vesicles with some of the
FIGURE 1. Foot epithelium, 36 hours of regeneration, showing a whorl of rough endoplasmic
reticulum (*), Nu = nucleus. Bar = 500 nm.
268
C. REED-MILLER
FIGURE 2. Mantle epithelium, 24 hours of regeneration, showing two Golgi complexes (arrows).
Note the light fibrillar material near the Golgi complexes. Bar = 1 urn.
FIGURE 3. Foot epithelium, 48 hours of regeneration with several Golgi complexes (arrows) and
associated vesicles containing some condensed material. Bar = 1
FIGURE 4. Scanning electron micrograph of the shell window, 48 hours of regeneration, showing
small crystals associated with an organic matrix. Bar = 10 ^m.
FIGURE 5. Higher magnification scanning electron micrograph of the shell window, 48 hours of
regeneration, showing clusters of doubly-pointed crystallites. Bar = 10
TEGUL.4 SHELL REGENERATION
269
. .
• .•.../ •• t : ••• 1 ~-~ • •
...-. •••.^•;:/v ••• •.••*.*-*•:
FIGURE 6. Mantle epithelium, 54 hours of regeneration, showing several inclusions containing dark
vacuoles. The small dark droplets are melanin. Bar = 10 nm.
FIGURE 7. Mantle epithelium, 72 hours of regeneration, showing granular and fibrous material
associated with a Type I inclusion. Bar = 500 nm.
FIGURE 8. Foot epithelium, 48 hours of regeneration, showing Type II inclusions (arrows). Small
dark droplets are melanin. Bar = 10 nm.
FIGURE 9. Hepatopancreas, 72 hours of regeneration, with Type III inclusions. Bar = 1
individual vesicles appearing granular and connected to the other vesicles and the
delimiting membrane by a fibrous network (Fig. 7, Type I inclusions). In another
form (Type II inclusions), the entire inclusion was round, and the vacuoles were
less electron-dense than Type I inclusions (Fig. 8). The third form consisted of
aggregates of two to 15 or more dark vacuoles (Fig. 9, Type III inclusions). These
270
C. REED-MILLER
three inclusions were not correlated with any particular tissue, that is, all three
morphologies were found in all three of the tissues during this stage of regeneration.
The regenerated material in the shell window consisted of spindle-shaped crys-
tals, and was virtually identical to the description for 14-48 hours of regeneration.
72 hours-six days of regeneration
As shown in Figure 10, transmission electron microscopy of the mantle, foot,
and hepatopancreas showed widened intracellular spaces and spherules. A fibrous
network linked the cores of the spherules with the surrounding membrane (Fig. 1 1 ).
By this stage of regeneration, e.g., as early as 72 hours, but no later than six
days after the shell window was cut, a thin sheet of material formed by the coales-
cence of spindle-shaped crystals covered the shell window (Figs. 12 and 13).
DISCUSSION
The present study shows that the mantle, foot, and hepatopancreas of Tegula
undergo ultrastructural alterations during the first six days of shell repair. Each of
these tissues has been implicated in shell repair and calcification (Abolins-Krogis,
1970a, b; Burton, 1972; Watabe el al, 1976; Tsujii, 1976; Watabe and Blackwelder,
1980). However, few studies concern the involvement of all three tissues at the
same time.
Since of the three tissues studied, the mantle is the one usually associated with
molluscan shell formation (Wilbur, 1964; 1972; 1976; Crenshaw, 1980), the ultra-
FIGURE 10. Mantle, 4 days of regeneration, showing wide intracellular spaces and spherules. Ar-
rowheads indicate some mitochondria. Bar = 5
. .
FIGURE 1 1 . Mantle, 4 days of regeneration, showing at higher magnification the spherule indicated
with a * in Figure 10. Note the membrane (arrow) and the granular-fibrillar appearance of the dark
material surrounding the lucent core. Bar = 100 nm.
TEGULA SHELL REGENERATION
271
FIGURE 12. Scanning electron micrograph of the shell window, 6 days of regeneration, showing
rosettes of spindle-shaped crystals associated with an organic matrix (arrowheads). Bar = 10 nm.
FIGURE 13. Scanning electron micrograph of the shell window, 6 days of regeneration, showing
the coalescence of spindle-shaped crystals and rosettes to cover the window. Bar = 100
structural observations are discussed with regard to the role of this organ. The
function of the foot and the hepatopancreas in shell repair will be compared to
previous work of these tissues in other molluscs.
The mantle edge is the region actively involved in shell growth (Tsujii, 1976;
Crenshaw, 1980). In this study, the site of shell regeneration was the first body whorl
of the shell which lies over the central zone of the mantle — an area not usually
involved in shell formation (Tsujii, 1976; Crenshaw, 1980). Following the initiation
of shell regeneration, this zone of the mantle showed ultrastructural changes that
may indicate an increased role in shell maintenance. These include a proliferation
of rough endoplasmic reticulum, typically in whorls or spirals. Comparable changes
have been reported in the mantle edge of Heliosoma (Saleuddin, 1976), Helix (Sa-
leuddin, 1970, Fig. 11), and marine bivalves (Bubel, 1973a, b).
An increase in Golgi complexes was evident in the mantles of the shell-regen-
erating gastropods in this study. Moreover, as was the case for rough endoplasmic
reticulum, increased numbers and activity in Golgi complexes have been described
in active regions of the mantle in other molluscs. For instance, during periostracum
or shell repair in Mytilus edulis and Helix pomatia (Saleuddin, 1970; Bubel, 1973c).
Precursors of the periostracum were observed in Golgi cisternae in the gland cells
of the mantle of Littorina (Bevelander and Nakahara, 1971), and Watabe et al.
(1976) found that the formation of calcareous spherules in Pomacea paludosa was
preceded by large vacuoles near the Golgi apparatus.
272 C. REED-MILLER
There was also an increase in mitochondria. Saleuddin (1970) noted 24 hours
after shell injury to Helix pomatia, that the number of mitochondria increased. The
calcium cells of Pomatia paludosa continued numerous mitochondria (Watabe et
al, 1976). In fact, these workers suggest that their finding lends support to the notion
that the mitochondria are involved in the uptake and release of calcium and phos-
phate for calcification (Spiro and Greenspan, 1969; Lehninger, 1970; Saleuddin,
1970; Elder and Lehninger, 1973; Becker et al., 1974). Certainly this could also be
the role of the increased mitochondria in the tissues of shell-regenerating Tegula.
Three types of inclusions were noted. Their appearance was preceded by pro-
liferation of rough endoplasmic reticulum and Golgi complexes. Vacuoles in the
inclusions and fibrous, matrix-like material were found in close proximity to the
Golgi complexes. Presumably, these inclusions are derived from the reticulum-Golgi
complex system as in other calcifying systems (see Spangenberg, 1976; Watabe et
al., 1976; and Simkiss, 1980, for examples and discussions of the Golgi vesicle-
reticulum system in calcification).
Distinctions as to the possible functions of each of the three types of inclusions
are impossible to make with the current data, but some speculation on their roles
is possible. It is conceivable that one role may be to provide matrix material for the
deposition of mineral for regenerated shell. The most likely candidates for this part
are the Type I and Type II inclusions, based on the similarity in their ultrastructural
appearance to inclusions that do serve as mineral deposition sites in other systems
(See Watabe et al, 1976, and Watabe and Blackwelder, 1980, for discussions of
Golgi-derived vacuoles and vesicles in another gastropod under normal and shell-
regenerating conditions).
The Type III inclusions may represent another morphology of calcifying vesicle
or vacuole, or it may be involved in cellular detoxification. Calcification and min-
eralization involve a high degree of cellular activity, and there must be a way of
ridding the cells of the resulting waste. Mason and Simkiss (1982), Kingsley and
Watabe (1982), and Simkiss (1980) discuss similar inclusions in invertebrates and
their role in detoxification.
If the inclusions are involved in providing sites for deposition of calcium in this
system, the calcium must be mobilized to the site of shell regeneration. It is inter-
esting that the appearance of the inclusions is followed by the occurrence of lucent
cored spherules. These spherules may be similar to the naturally decalcified spherule
found in the calcium cells of Pomacea paludosa (Watabe et al, 1976; Watabe and
Blackwelder, 1980).
Since the ultrastructural picture was similar to that for the mantle, it would be
repetitive to consider in detail the changes in the foot and the hepatopancreas re-
ported here. However, there are some points to be made about the possible roles
of these tissues in shell repair. For example, when calcareous spherule development
was not evident, prominent Golgi complexes and abundant rough endoplasmic
reticulum and mitochondria occurred in the calcium cells of the foot and the al-
bumin-capsule gland complex as well as in the mantle of Pomacea paludosa (Watabe
et al., 1976). Later work showed that spherule calcium was used for shell regeneration
(Watabe and Blackwelder, 1980). The spherules described in the foot of the shell-
regenerating Tegula in this study may be contributing calcium for shell repair as
did those in Pomacea.
The role of the hepatopancreas in molluscan shell regeneration has been debated.
Burton (1972) suggested that calcium from the hepatopancreas of Helix was used
for shell repair, but did not show it to be mobilized. Work on the hepatopancreas
of shell-regenerating Helix pomatia showed calcium spherites developed in protein-
TEGUL4 SHELL REGENERATION 273
containing Golgi saccules (Abolins-Krogis, 1970a), and alterations in the ultrastruc-
ture of this organ (Abolins-Krogis, 1 970b; 1 972). Simkiss ( 1 980) found that spherules
from the mantle and the foot of Helix aspersa were soluble in saline, while those
from the hepatopancreas were not. He suggested different functions for the two types
of spherules — those from the foot and the mantle would be involved in mineral-
ization, while the more insoluble ones would be responsible for cellular detoxification
(Simkiss, 1980; Mason and Simkiss, 1982). Campbell and Boyan (1976) suggested
that the function of calcium spherules in the gastropod hepatopancreas is as a
phosphate reserve. The hepatopancreas of Tegula may provide calcium for shell
repair, as well as detoxifying the animal and providing phosphate.
Regenerated shell may be different from or similar to the normal shell ultra-
structure (Saleuddin and Wilbur, 1969; Wilbur, 1972; Wong and Saleuddin, 1972).
The small doubly-pointed crystals and the stellate shapes they formed have been
described for both normal and regenerated shell in other molluscs. Spherulitic ag-
gregates have been shown in another archaeogastropod, Cittarium pica (Wise and
Hay, 1968a, b; Erben, 1971), and in the regenerated shell of Pomacea paludosa and
Cepaea nemoralis (Blackwelder and Watabe, 1977; Watabe, 1981). The crystals
formed during early shell regeneration in Nautilis macromphalus are doubly-
pointed, associated with an organic matrix, and form stellate aggregates which grow
until a Spherulitic prismatic layer is formed (Meenakshi et ai, 1974). Therefore, the
spindle-shaped crystals described in the regenerated shell of Tegula appear to be a
common morphology of calcium carbonate in some archaeogastropods as well as
in regenerated molluscan shell.
ACKNOWLEDGMENTS
Thanks go to William I. Miller, III for expert assistance with the scanning electron
microscopy, and to Dennis Cassidy for the use of the darkroom in the Antarctic
Research Facility in the Department of Geology at FSU. I am grateful to Dr. Charles
B. Metz for a helpful review of the manuscript. Supported by N.I.H. Grant
#DE05491. This is contribution number 195 from the Tallahassee, Sopchoppy and
Gulf Coast Marine Biological Association.
LITERATURE CITED
ABOLINS-KROGIS, A. 1961. The histochemistry of the hepatopancreas of Helix • pomatia (L.) in relation
to the regeneration of the shell. Ark. Zool. 13: 59-202.
ABOLINS-KROGIS, A. 1963. The histochemistry of the mantle of Helix pomatia (L.) in relation to the
repair of the damaged shell. Ark. Zool. 15: 461-474.
ABOLINS-KROGIS, A. 1968. Shell regeneration in Helix pomatia with special reference to the elementary
calcifying particles. Symp. Zool. Soc. Land. No. 22: 75-92.
ABOLINS-KROGIS, A. 1970a. Electron microscope studies of the intracellular origin and formation of
calcifying granules and calcium spherites in the hepatopancreas of the snail. Helix pomatia L.
Z Zellforsch. 108: 501-515.
ABOLINS-KROGIS, A. 1970b. Alterations in the fine structure of cytoplasmic organelles in the hepato-
pancreatic cells of shell-regenerating snail, Helix pomatia L. Z. Zellforsch. 108: 516-529.
ABOLINS-KROGIS, A. 1972. The tubular endoplasmic reticulum in the amoebocytes of the shell-regen-
erating snail, Helix pomatia L. Z. Zellforsch. 128: 58-68.
BECKER, G. L., C-H. CHEN, J. W. GREENWALT, AND A. L. LEHNINGER. 1974. Calcium phosphate
granules in the hepatopancreas of the blue crab Callinectes sapidus. J. Cell Biol. 61: 316-326.
BEEDHAM, G. E. 1965. Repair of the shell in species ofAnodonta. Proc. Zool. Soc. Lond. 145: 107-125.
BEVELANDER, G. B., AND M. NAKAHARA. 1971. The formation and growth of the prismatic layer of
Pinctada radiata. Calc. Tiss. Res. 7: 31-45.
BLACKWELDER, P. L., AND N. WATABE. 1977. Studies on shell regeneration II. The fine structure of
normal and regenerated shell of the freshwater snail, Pomacea paludosa. Biomineralization 9:
1-10.
274 C. REED-MILLER
BUBEL, A. 1973a. An electron-microscope investigation of the cells lining the outer surface of the mantle
in some marine molluscs. Mar. Biol. 21: 245-255.
BUBEL, A. 1973b. An electron-microscope study of periostracum formation in some marine bivalves. II.
The cells lining the periostracal groove. Mar. Biol. 10: 222-234.
BUBEL, A. 1973c. An electron-microscope study of periostracum repair in Mytilus edulis. Mar. Biol. 20:
235-244.
BURTON, R. F. 1972. The storage of calcium and magnesium phosphates and of calcite in the digestive
glands of pulmonata (Gastropoda). Comp. Biochem. Physiol. 43A: 655-663.
CAMPBELL, J. W., AND B. D. BOYAN. 1976. On the acid-based balance of gastropod molluscs. Pp. 109-
133 in The Mechanisms of Mineralization in the Invertebrates and Plants, N. Watabe and
K. M. Wilbur, eds. Univ. South Carolina Press, Columbia.
CHAN, J. F. Y., AND A. S. M. SALEUDDIN. 1974. Acid phosphatase in the mantle of the shell-regenerating
snail, Helisoma duryi duryi. Calcif. Tissue Res. 15: 213-220.
CRENSHAW, M. A. 1980. Mechanisms of shell formation and dissolution. Pp. 1 15-132 in Skeletal Growth
of Aquatic Organisms, D. C. Rhoads and R. A. Lutz, eds. Plenum Press, New York.
ELDER, J. A., AND A. L. LEHNINGER. 1973. Respiration dependent transport of carbon dioxide into rat
mitiochondria. Biochemistry, N. Y. 12: 976-982.
ERBEN, H. K. 1971. Anorganische und organische schalenkompenten bei Cittarium pica (L.). (Archaeo-
gastropoda). Biomineralization 3: 51-64.
KAPUR, S. P., AND A. S. GUPTA. 1970. The role of amoebocytes in the regeneration of shell in the land
pulmonate Euplecta indica (PFIEFER). Biol. Bull. 139: 502-509.
KJNGSLEY, R. J., AND N. WATABE. 1982. Ultrastructural investigation of spicule formation in the gor-
gonian Leptogorgia virgulata (Lamarck) (Coelenterata: Gorgonacea). Cell Tissue Res. 223: 325-
334.
LEHNINGER, A. L. 1970. Mitochondria and calcium ion transport. Biochem. J. 119: 129-138.
MASON, A. Z., AND K. SIMKISS. 1982. Sites of mineral deposition in metal-accumulating cells. Exp. Cell
Res. 139:383-391.
MEENAKSHI, V. R., P. L. BLACKWELDER, AND K. M. WILBUR. 1973. An ultrastructural study of shell
regeneration in Mytilus edulis (Mollusca: Bivalvia). J. Zool. Land. 171: 474-484.
MEENAKSHI, V. R., A. W. MARTIN, AND K. M. WILBUR. 1974. Shell repair in Nautilus macromphalus.
Mar. Biol. 27: 27-35.
MEENAKSHI, V. R., P. L. BLACKWELDER, P. E. HARE, K. M. WILBUR, AND N. WATABE. 1975. Studies
on shell regeneration — I. Matrix and mineral composition on the normal and regenerated shell
of Pomacea paludosa. Comp. Biochem. Physiol. 50A: 347-35 1 .
REED-MILLER, C., S. W. WISE, JR., AND H. J. SIEGEL. 1980. Correlated techniques for identifying mineral
phases and morphologies in regenerated archaeogastropod shells: A progress report. Pp. 73-78
in The Mechanisms of Biomineralization in Animals and Plants, M. Omori and N. Watabe,
eds. Tokai University Press, Tokyo.
REED-MILLER, C. 1981. Transmission and scanning electron microscopy of gastropod shell regeneration.
Am. Zool. 21: 1037.
SALEUDDIN, A. S. M. 1967. The histochemistry of the mantle during the early stage of shell repair. Proc.
Matacol. Soc. Lond. 37: 371-380.
SALEUDDIN, A. S. M. 1970. Electron microscopic study of the mantle of normal and regenerating Helix.
Can. J. Zool. 48:409-416.
SALEUDDIN, A. S. M. 1976. Ultrastructural studies of the structure and formation of the periostracum
in Heliosoma. Pp. 309-337 in The Mechanisms of Mineralization in the Invertebrates and
Plants, N. Watabe and K. M. Wilbur, eds. Univ. South Carolina Press, Columbia.
SALEUDDIN, A. S. M., AND K. M. WILBUR. 1969. Shell regeneration in Helix pomatia. Can. J. Zool. 47:
51-53.
SIMKJSS, K. 1980. Detoxification, calcification and the intracellular storage of ions. Pp. 13-18 in The
Mechanisms of Biomineralization in Animals and Plants, M. Omori and N. Watabe, eds. Tokai
Univesity Press, Tokyo.
SPANGENBERG, D. B. 1976. Intracellular statolith synthesis in Aurelia aurita. Pp. 131-144 in The Mech-
anisms of Mineralization in the Invertebrates and Plants. N. Watabe and K. M. Wilbur, eds.
Univ. South Carolina Press, Columbia.
SPIRO, I., AND J. GREENSPAN. 1 969. Are mitochondria directly involved in biological calcification? Calcif.
Tissue Res. 3: 100-102.
TSUJII, T. 1960. Studies on the mechanisms of shell- and pearl-formation. XI. The submicroscopical
observations on the mechanism of formation of abnormal pearls and abnormal shell. Rep. Fac.
Fish. Prefect. Univ. Mie. 6: 59-66.
TSUJII, T. 1976. An electron microscopic study of mantle epithelial cells of Anodonta sp. during shell
regeneration. Pp. 339-353 in The Mechanisms of Mineralization in the Invertebrates and Plants,
N. Watabe and K. M. Wilbur, eds. Univ. South Carolina Press, Columbia.
TEGULA SHELL REGENERATION 275
UOZUMI, S., AND S. OHATA. 1977. Scanning electron microscopy and electron probe microanalysis of
initial mineralization in marine shell regeneration. J. Geol. Soc. Jpn. 83: 425-432.
UOZUMI, S., AND S. SUZUKI. 1979. "Organic membrane-shell" and initial calcification in shell regen-
eration. / Fac. Sci. Hokkaido Univ. Ser. IV. 19: 37-74.
WAGGE, L. E. 1951. The activity of amoebocytes and of alkaline phosphatase during the regeneration
of the shell of the snail Helix aspersa. Q. J. Microsc. Sci. 92: 307-32 1 .
WATABE, N. 1981. Crystal growth of calcium carbonate in the invertebrates. Pp. 99-147 in Inorganic
Biological Crystal Growth, B. R. Pamplain, ed. Pergamon Press, Oxford.
WATABE, N., V. R. MEENAKSHI, P. L. BLACKWELDER, E. M. KURTZ, AND D. G. DUNKELBERGER.
1976. Calcareous spherules in the gastropod, Pomacea plaudosa. Pp. 283-308 in The Mecha-
nisms oj Mineralization in the Invertebrates and Plants, N. Watabe and K. M. Wilbur, eds.
Univ. South Carolina Press, Columbia.
WATABE, N., AND P. L. BLACKWELDER. 1980. Ultrastructure and calcium localization in the mantle
epithelium of the freshwater gastropod, Pomacea plaudosa during shell regeneration. po. 131-
144 in The Mechanisms of Biomineralization in Animals and Plants, M. Omori and N. Watabe,
eds. Tokay University Press, Tokyo.
WILBUR, K. M. 1964. Shell formation and regeneration. Pp. 243-282 in Physiology of Mollusca, K. M.
Wilbur and C. M. Yonge, eds. Academic Press, New York.
WILBUR, K. M. 1972. Shell formation on molluscs. Pp. 103-145 in Chemical Zoology. Vol. VII. Mollusca,
M. Florkin and B. T. Scheer, eds. Academic Press, New York.
WILBUR, K. M. 1973. Mineral regeneration in echinoderms and molluscs. Ciba Found. Symp. 11 (new
sen): 7-33.
WILBUR, K. M. 1976. Recent studies of invertebrate mineralization. Pp. 79-108 in The Mechanisms of
Mineralization in the Invertebrates and Plants, N. Watabe and K. M. Wilbur, eds. Univ. South
Carolina Press, Columbia.
WISE, S. W., JR., AND W. W. HAY. 1968a. Scanning electron microscopy of molluscan shell ultrastructure.
I. Techniques for polished and etched sections. Trans. Am. Microsc. Soc. 87: 411-418.
WISE, S. W., JR., AND W. W. HAY. 1 968b. Scanning electron microscopy of molluscan shell ultrastructure.
II. Observation of growth surfaces. Trans. Am. Microsc. Soc. 87: 419-430.
WONG, V., AND A. S. M. SALEUDDIN. 1972. Fine structure of normal and regenerated shell of Helisoma
duryi durvi. Can. J. Zoo/. 50: 1563-1568.
Reference: Biol. Bull. 165: 276-285. (August, 1983)
MATING AND EGG MASS PRODUCTION IN THE AEOLID
NUDI BRANCH HERMISSENDA CRASSICORNIS
(GASTROPODA: OPISTHOBRANCHIA)
RONALD L. RUTOWSKI
Department of Zoology, Arizona State University, Tempe, AZ 85287
ABSTRACT
Interactions leading to copulation in the aeolid nudibranch, Hermissenda cras-
sicornis, have a duration of only a minute or two, and intromission lasts only a few
seconds (Longley and Longley, 1982). This paper reports additional details on the
temporal structure and variation in structure of these interactions. It is also shown
that sperm from a single copulation are sufficient to fertilize only 2 to 3 egg masses
and that the state of an animal's sperm supply affects the rate at which it produces
egg masses but not the size of the masses produced. In the discussion the results are
compared to information on other simultaneous hermaphrodites in an effort to
assess the possible adaptive advantages of the structure of the reproductive behavior
of H. crassicornis.
INTRODUCTION
Longley and Longley (1982) reported recently that copulation in the nudibranch,
Hermissenda crassicornis, is much briefer than that reported for many other op-
isthobranchs. In particular, copulation in this simultaneous hermaphrodite involves
an explosive and reciprocal eversion of the penises with intromission lasting only
a few seconds. In contrast, copulation in most other sea slugs lasts for many minutes
or hours.
The gross sequential and temporal features of copulatory interactions in H.
crassicornis are known, in part by default (Zack, 1975; Rutowski, 1982). Copulation
is most likely to occur in interactions that begin when two animals meet head-to-
head. After initial contact the participants stop forward movement and reciprocally
touch each other with their tentacles. These repeated contacts were termed "flag-
ellation" by Zack and last for about 45 s (Fig. 1 A). The animals then begin moving
slowly toward one another until they begin to pass, with the head of each animal
moving along the right side of the other (Fig. 1 B). Zack called this "sidling", and
it brings the gonopores on the right side of the body a little behind the head into
apposition. In Zack (1975) and Rutowski (1982) the cerata prevented observation
of the eversion of the penises that occurs when the gonopores come into contact
(Fig. 1C). Hence, interactions with sidling were not known to be copulatory until
Longley and Longley's (1982) report. About 5 s after the beginning of sidling both
animals erect their cerata and begin to move apart. This separation is often con-
current with lunging and biting by one or both animals directed at its partner (Fig.
1 D). Within 20 s after erection of the cerata, the animals are fully separated.
The data presented in this report are directed at answering two questions about
the intriguing mating behavior of//, crassicornis. First, what are the details of these
Received 3 February 1983; accepted 23 May 1983.
276
NUDIBRANCH REPRODUCTIVE BEHAVIOR
277
FIGURE 1. Events in a copulatory interaction in Hermissenda crassicornis. (A) Flagellation (ct, area
above gonopore from which cerata have been cleared); (B) beginning of sidling; (C) the moment of
intromission; (D) shortly after withdrawal (p, withdrawn but unretracted penis).
copulatory interactions? Second, how are egg fertility and output related to copu-
lation? Special attention will be devoted to determining how long the sperm from
a single copulation lasts relative to the duration of an individual's reproductive life.
Quantitative and detailed answers to these questions are of interest because the
Longleys' observations were largely qualitative and because the behavior of simul-
taneous hermaphrodites is generally so little known (Fischer, 1980). The discussion
examines the adaptive features of the mating behavior of//, crassicornis relative to
that of other opisthobranchs and other simultaneous hermaphrodites.
MATERIALS AND METHODS
Hermissenda crassicornis was collected between January and June in 1982 in
the intertidal Zostera marina beds in Elkhorn Slough, Monterey County, CA. All
animals were taken to the Long Marine Laboratory, Santa Cruz County, CA, where
they were housed individually in small plastic cups (about 250 ml) each with its
own supply of fresh running sea water (approximately 1 1-1 3°C) from a holding
tank on the station property (Rutowski, 1982). The animals were fed either fresh
mussel (Mytilus californianus) or, rarely, fresh squid mantle (Loligo spp.) every other
day. At each feeding cups were cleared of food remaining from previous feedings
and any detritus introduced by the sea water system.
278 R. L. RUTOWSKI
The production and fertility of egg masses were carefully monitored for all
isolated animals. If an animal produced an egg mass the animal was removed from
its container and placed in a new one. The egg mass diameter was measured and
then incubated undisturbed in unaltered running sea water for 4 to 7 days at which
time the shells of developing veligers become clearly visible (Williams, 1980). At
this time the proportion of fertilized eggs in each mass was assessed by estimating,
to the nearest 25%, the proportion of eggs that had developed into shelled veligers.
In the results, any egg mass in which less than 50% of the eggs developed into veligers
is referred to as an "infertile" egg mass. Those with 50% of more of the eggs de-
veloping into veligers are called "fertile" egg masses.
Isolated animals were, under conditions of constant observation, permitted to
contact and mate with other individuals. The staging and videotaping of these in-
teractions followed the techniques described in Rutowski (1982) except that larger
containers (12 cm diameter watchglass or dish) were typically used.
Throughout the study the area above and slightly behind the gonopore of each
animal was kept free of cerata (Fig. 1). This was done about once every week or
two by plucking away with watchmakers forceps any cerata that might impede
observations of penis eversion and intromission.
All parametric summary statistics are given as: mean ± standard deviation.
Statistical evaluations were made at the 0.05 level of significance.
RESULTS
Form ofcopulatory interactions
Over 60 interactions leading to sidling and copulatory attempts were observed
during the course of this study. Of these, 43 were videotaped and analysed for the
timing of events before and after penis eversion (Table I, Fig. 1 ).
The form of all these interactions was similar to the copulatory interactions
described by Longley and Longley (1982) and for interactions with flagellation and
sidling described by Zack (1975) and Rutowski (1982). I would add that during
flagellation there is a pronounced swelling of the gonopore region that continues
until penis eversion. If flagellation is terminated for some reason the swelling de-
creases and a small amount of semen is released from the gonopore. Also, new
temporal information to add to these descriptions includes the time from the be-
ginning of sidling (1) to penis eversion and (2) until both animals have uncoupled
(Table I). These data confirm the relatively rapid progression of events in these
matings. The average duration of intromission was only about 4 s, and in all but
a few cases differences in the time of eversion between participants were not re-
solvable with the video system used.
TABLE I
The timing oj events in a copulatory interaction from the beginning oj sidling until the animals
begin to retreat or move apart
Time of occurrence relative
Event to beginning of sidling (s) Source
Cerata movement 4.1 1 ± 0.859 s (n = 19) Rutowski, 1982
Intromission 6.53 ± 2.77 s (n = 39) This study
Withdrawal 10.5 ±2.88s(n = 31) This study
Begin retreat 13.1 ± 6.2 s (n = 19) Rutowski, 1982
NUDIBRANCH REPRODUCTIVE BEHAVIOR 279
Sidling does not always lead to intromission. First, on a few rare occasions one
or both animals did not evert the penis. When neither animal attempted intromis-
sion, they simply moved apart after a brief apposition of the gonopores. Second, as
pointed out by the Longleys, not all penis eversion resulted in intromission. Typ-
ically, during the process of eversion in such an interaction, the penis appeared to
ricochet off the body wall of the partner next to its gonopore and ejaculated the
semen at the moment of full extension. Of the 4 1 videotape records of interactions
with sidling and penis eversion, 37 were of adequate quality to determine whether
or not penis eversion with intromission occurred. In 38% of these both animals
attained intromission. However, in another 49%, although both animals tried, one
animal missed the other's gonopore. In the remaining interactions neither animal
attained intromission either because both missed (8%) or because one missed and
one did not attempt intromission (5%). Interactions in which neither animal at-
tempted penis eversion were not included in this count. In summary, sperm transfer
was not reciprocal in almost 50% of the interactions in which penis eversion by both
animals occurred.
After penis eversion and separation a new behavior pattern was observed which
was directed at the mass of semen that was frequently seen in the vicinity of the
gonopore particularly if intromission did not occur. An animal stopped, turned its
head so that the mouth was positioned near the gonopore, and consumed all or part
of the semen in the vicinity of the gonopore (Fig. 2). Data on the frequency of
occurrence of this behavior was not recorded. It also occurs in land snails in the
genus Partula (Lipton and Murray, 1979).
Effect of copulation on egg production
Rates of egg production: fertile versus infertile. If an individual of//, crassicornis
is isolated and denied contact with conspecifics it will typically produce a few com-
FIGURE 2. Hermissenda crassicornis immediately after copulation ingesting semen from the vicinity
of the gonopore.
280 R. L. RUTOWSKI
pletely (100%) fertile egg masses and then either stop producing eggs or begin to
produce masses with an increasing proportion of infertile eggs (Table II). In egg
masses with fertile and infertile eggs, the infertile eggs are concentrated in the outer
loops of the spiral while those eggs near the middle of the spiral are mostly fertile.
The average interval between fertile egg masses (as denned earlier) was 4.32 ± 2.22
days (Fig. 3). The time between the last fertile mass produced by an animal and the
first infertile mass was 9.35 ± 10.06 days, which is significantly different from the
interval between fertile masses (Wilcoxon rank sum test, P = 0.00007; / = 5.47,
P< 0.001). The average interval between masses of mostly infertile eggs was
7.36 ± 6.5 days which was also significantly longer than the time between two fertile
masses (Wilcoxon rank sum test, P ---- 0.00005; / : 5.19, P < 0.0001) but not
significantly different from the time between the last fertile and the first infertile egg
mass (Wilcoxon rank sum test, P = 0.17; / - 1.17, P > 0.2).
Another way of summarizing these data is to ask, if an animal produces a fertile
egg mass how does the probability that the next mass will be infertile change with
the time elapsed between the two? As Table III shows, if the interval is greater than
1 1 days the probability that the next mass is infertile is 100%. Hence for purposes
of this study an animal was regarded as sperm depleted if (1) it had produced 2 or
more infertile egg masses, or (2) if it had not produced a fertile egg mass in 20 days
or more.
Egg mass size: fertile versus infertile. Egg mass size is known to be a function
of the size of the producer in H. crassicornis (Harrigan and Alkon, 1978). To min-
imize this effect I examined the diameter of egg mass pairs that were sequentially
produced by the same animal, although no more than 1 1 days apart, and that varied
by at least 50 percent in the proportion of eggs fertilized. Of 28 pairs of egg masses
produced the more fertile mass was the larger of the two in 14 while the opposite
situation appeared in 1 3 pairs. In one pair there was no difference. I conclude that
the state of an animal's sperm supply has no consistent effect on the size of the egg
mass that it produces.
Effect of copulation on the rate of fertile egg mass production. Thirty-eight animals
that were sperm depleted (by the criteria above) were each permitted to engage in
one sidling interaction with another animal and then returned to isolation. As in-
dicated by the videotape records, 28 of these animals were successfully intromitted,
while 15 were not. For 93% of those that were successfully intromitted, the next egg
mass they produced was fertile. All produced egg masses an average of 2.64
± 1.33 days after the copulation. These included 1 1 animals that had not produced
an egg mass in over 20 days. In contrast, of the 1 5 animals that were not inseminated,
none produced fertile egg masses within 20 days of the interaction with sidling. In
TABLE II
A summary of the fertility of egg masses produced after the last completely (100%)
fertile egg mass produced
Percent of egg masses in which the %
of eggs that developed was:
Lgg mass alter last
100% fertile mass
0-24
25-49
50-74
75-100
Sample size
First
11%
48%
15%
26%
27
Second
63%
31%
6%
0%
16
Third
100%
0%
0%
0%
11
Fourth and beyond
100%
0%
0%
0%
6
NUDIBRANCH REPRODUCTIVE BEHAVIOR
281
30i
20-
10-
A
0=149
U
^-
LU
Z>
o
LU
n =
B
n
n n..
20-1
10-
0J
n = 80
c
-TUn
0
5 10 15
INTERVAL (DAYS)
20 >20
FIGURE 3. Intervals between successive egg masses. (A) Fertile to fertile; (B) fertile to infertile; (C)
infertile to infertile.
fact, 5 of these animals did not produce any egg mass in the first 20 days after the
interaction.
The likelihood that an isolated animal will produce an infertile egg mass increases
with the number of egg masses produced since the last successful copulation (Table
IV). By the fifth egg mass after copulation the probability that the mass is infertile
is 50% or greater. The average number of fertile masses produced by an isolated
animal after a single copulation and before it shows clear signs of sperm depletion
is 2.65 ± 1.66 masses (n = 20).
Egg production of isolated wild-caught individuals
Ten animals were isolated from the time of capture in the field and their sub-
sequent egg production was monitored in the laboratory for a period of 24 days.
282 R- L. RUTOWSKI
TABLE III
The relationship between the fertility of an egg mass and the time of its production relative
to the last fertile egg mass (no intervening masses)
Interval (days) % Infertile Sample Size
0-2 12.5 16
3-5 18.2 99
6-8 23.3 30
9-11 33.3 6
12 or more 100 10
During this time all produced at least one fertile egg mass; but before the end, 7
showed signs of sperm depletion. Two of these 7 produced infertile egg masses while
the other five did not produce any egg masses during the final 10 days or more of
isolation. The average number of fertile egg masses produced during the 24 day
period was 3.7 ± 2.5. This information suggests that about 80% of these animals
carried stored sperm from a recent copulation or recent copulations.
DISCUSSION
Comparison with reports on copulation in other opisthobranchs
These data confirm Longley and Longley's (1982) report that copulation in H.
crassicornis is a rapid affair. Copulatory interactions last a few minutes, but the
actual duration of intromission is only a few seconds. It appears that one conse-
quence of such a rapid attempt at intromission is the high frequency of unsuccessful
attempts.
The Longleys point out that the hesitant approach and high willingness to turn
away from contact that characterize the behavior of participants in the early phases
of all interactions in H. crassicornis may be an effort on the part of one or both
animals to avoid cannibalism, which has been observed in several studies of this
animal (Zack, 1975; Rutowski, 1982). It may also be that the speed of copulation
also reflects an adaptation that minimizes the duration of contact with potentially
cannibalistic conspecifics. Supporting this notion is the observation that cannibalism
and apparently cannibalistic attacks on conspecifics have not been reported for
species of nudibranchs in which intromission is known to last many minutes or
hours, such as Embletonia fuscata (Chambers, 1934), Coryphella stimpsoni (Morse,
1971), Precuthona peachii (Christensen, 1977), Tritonia hombergi (Thompson,
TABLE IV
A summary of the fertility of egg masses produced by initially sperm depleted animals
that were intromitted
Egg mass after copulation % Infertile Sample Size
First
7.1
28
Second
18.5
27
Third
38.5
26
Fourth
65
20
Fifth
50
10
Sixth
57
7
Seventh and Eighth
80
5
NUDIBRANCH REPRODUCTIVE BEHAVIOR 283
1961), and Chromodoris zebra (Crozier, 1918). Longley and Longley (1982) report
a very brief copulation in Aeolidia papillosa in which I have observed cannibalistic
attacks on conspecifics. However, the tectibranch, Navanax inermis, is also known
to feed readily on conspecifics but its copulation is prolonged (Paine, 1965).
The continued production of egg masses in spite of sperm depletion is known
for a number of nudibranchs (Hadfield, 1963; Harris, 1975; Rivest, 1978; Chris-
tensen, 1977) and is puzzling. Why an animal continues even at a reduced rate to
invest energy and resources in egg production when most will not develop is not
currently clear.
Comparisons with other simultaneous hermaphrodites
In the opisthobranchs, reciprocal and internal fertilization is the rule (Costello,
1938; Beeman, 1977). In most, reciprocal insemination is simultaneous. Only in
Aplysia has the assumption of single sex roles by individual animals been reported
(Lederhendler and Tobach, 1977). In Hypoplectrus nigricans, a simultaneously her-
maphroditic serranid fish, fertilization is external but reciprocal in that the animals
take turns playing male and female roles in mating interactions (Fischer, 1980). The
prevalence of reciprocity in these simultaneous hermaphrodites is in some ways
unexpected. The best interests of the male and female functions of a hermaphrodite's
reproductive tract are not the same. Charnov (1979) points out that in simultaneous
hermaphrodites copulation is probably primarily an effort to give sperm rather than
receive it. In other words, the fitness through male function can best be maximized
through repeated copulation. It follows then that in mating interactions simultaneous
hermaphrodites should be more often willing to play a male than a female role. One
might expect to see more one-way inseminations or efforts on the part of individuals
to give sperm without receiving it. However, at any given time in a population there
will be many more animals willing to give sperm than to receive it so that to
maximize the rate of giving sperm an animal must reduce its reluctance to receive
it. As Charnov states, "each should be inclined to accept sperm in order to give its
sperm away." In this respect the mating behavior of simultaneous hermaphrodites
is more like that expected in cooperative interactions than those of gonochoric
organisms. This view has been promulgated by Axelrod and Hamilton (1981) in
their analysis of the selection pressures shaping the form of cooperative interactions
in animals.
There are two ways in which this analysis might be relevant to understanding
the reproductive behavior of H. crassicornis. First, it is possible that the missed
intromissions represent efforts by animals to give sperm without receiving it by
actively deflecting the penis of the other individual. Second, it is at least plausible
that the rapidity of intromission is a result of animals attempting to give sperm
quickly without giving the other animal a chance to intromit. These ideas both seem
unlikely, however, in that most of the missed individuals in this study were animals
in a sperm-depleted state who could have greatly benefited from receiving as well
as giving sperm.
Evolution of simultaneous hermaphroditism in H. crassicornis
The data presented here and elsewhere permit a partial evaluation of the ap-
plicability of competing hypotheses for the adaptive significance of simultaneous
hermaphroditism in H. crassicornis. Currently, there are three major explanations
of this sort. The first two have been summarized by Ghiselin (1969). First, simul-
taneous hermaphroditism may increase an animal's reproductive efficiency by per-
284 R. L. RUTOWSKI
mitting self-fertilization. This idea does not apply to H. crassicornis because this
animal has never been observed to self-fertilize (Harrigan and Alkon, 1978; this
study). The second explanation is the low density model which suggests that si-
multaneous hermaphroditism will be favored in organisms with low population
densities in which contacts with conspecifics are rare or infrequent. Any conspecific
is then an appropriate mate. Although this model is often invoked to explain the
occurrence of simultaneous hermaphroditism in nudibranchs (e.g., Todd, 1978) it
does not appear to fit H. crassicornis well. If contact with conspecifics are rare, one
might expect that (1) sperm from a single copulation would last the better part of
an individual's active reproductive life and (2) individuals should readily mate in
any contact with a conspecific. Neither of these conditions is met in H. crassicornis.
Individuals of H. crassicornis live several months in the laboratory (Harrigan and
Alkon, 1978; pers. obs.) and presumably in the field as well. The sperm from a
single copulation is sufficient to fertilize most eggs in about 3 egg masses which if
produced once every 4 to 5 days will last an animal about 15 days or less. This
coupled with the surprisingly high frequency of unsuccessful copulation attempts
suggests that these animals must mate several times during their life to maintain a
maximal egg production rate. The lack of precision in intromission and in the
production of fertile eggs is also not to be expected if the low density model were
in force. Furthermore, as Zack (1975) points out, most contacts and interactions
in H. crassicornis do not end in the sidling that often leads to copulation. If anything,
the animals appear more inclined to cannibalize than to copulate with conspecifics.
In sum, the behavior of these animals does not conform to some simple expectations
about behavior derived from the low density model.
The third explanation is a form of Maynard Smith's (1978) resource allocation
model, which states that the fitness return per egg will diminish as the number of
eggs produced increases, especially in species that produce myriads of small pro-
pagules. Hence, over evolutionary time selection might favor a partitioning of re-
productive energies into both male and female modes. The only observation sup-
porting the application of this model to H. crassicornis is that they do appear readily
able to produce millions of eggs (Williams, 1980) and so simultaneous investment
in male functions might well payoff. Of the three adaptationist models, this one,
largely by elimination, appears to be the most likely explanation for the evolution
of simultaneous hermaphroditism in this nudibranch.
In closing, it cannot be discounted at this time that the occurrence of simul-
taneous hermaphroditism may be a result of phylogenetic inertia. In other words,
this reproductive mode need not be adaptive in H. crassicornis but may have retained
during evolution as an incidental effect of other advantageous life history traits. This
view is supported by the prevalence of this reproductive mode in other opistho-
branchs which suggests that the ancestors of H. crassicornis were simultaneous
hermaphrodites. Hence, this model and the resource allocation model appear to be
the major competing hypotheses for the explanation of simultaneous hermaphro-
ditism in H. crassicornis.
ACKNOWLEDGMENTS
I thank: the Director and staff of the Long Coastal Marine Laboratory for the
space and support services they provided, Patricia Rutowski for assistance in the
field, Drs. John Alcock and Richard Satterlie for their comments on an earlier draft
of the manuscript, and John Schaefer for his assistance in preparation of the manu-
script.
NUDIBRANCH REPRODUCTIVE BEHAVIOR 285
LITERATURE CITED
AXELROD, R., AND W. D. HAMILTON. 1981. The evolution of cooperation. Science 211: 1390-1396.
BEEMAN, R. D. 1977. Gastropoda: Opisthobranchia. Pp. 115-179 in Reproduction of Marine Inverte-
brates. Volume 4, A. C. Giese and J. S. Pearse, eds. Academic Press, New York.
CHAMBERS, L. A. 1934. Studies on the organs of reproduction in the nudibranchiate mollusks, with
special reference to Embletonia fuscata Gould. Bull. Amer. Mm. Nat. Hist. 66: 599-641.
CHARNOV, E. L. 1979. Simultaneous hermaphroditism and sexual selection. Proc. Nat. Acad. Sci. 76:
2480-2484.
CHRISTENSEN, H. 1977. Feeding and reproduction in Precuthona peachii (Mollusca, Nudibranchia).
Ophelia 16: 131-142.
COSTELLO, D. P. 1938. Notes on the breeding habits of the nudibranchs of Monterey Bay and vicinity.
J. Morphol. 63: 319-343.
CROZIER, W. J. 1918. Assertive mating in a nudibranch, Chromodoris zebra Heilprin. /. Exp. Zool. 27:
247-292.
FISCHER, E. A. 1980. The relationship between mating system and simultaneous hermaphroditism in
the coral reef fish, Hypoplecirus nigricans (Serranidae). Anim. Behav. 28: 620-633.
GHISELIN, M. T. 1969. The evolution of hermaphroditism among animals. Q. Rev. Biol. 44: 189-208.
HADFIELD, M. G. 1963. The biology of nudibranch larvae. Oikos 14: 85-95.
HARRIGAN, J. F., AND D. I. ALKON. 1978. Larval rearing, metamorphosis, growth and reproduction of
the aeolid nudibranch Hermissenda crassicornis (Eschscholtz, 1831) (Gastropoda: Opistho-
branchia). Biol. Bull. 154: 430-439.
HARRIS, L. G. 1975. Studies on the life history of two coral-eating nudibranchs of the genus Phestilla.
Biol. Bull. 149: 539-550.
LEDERHENDLER, I. I., AND E. TOBACH. 1977. Reproductive roles in the simultaneous hermaphrodite
Aplysia dactylomela. Nature 270: 238-239.
LIPTON, C. S., AND J. MURRAY. 1979. Courtship of land snails of the genus Partula. Malacologia 19:
129-146.
LONGLEY, R. D., AND A. J. LONGLEY. 1982. Hermissenda: agonistic behavior or mating behavior?
VeligerlA: 230-231.
MAYNARD SMITH, J. 1978. The Evolution of Sex. Cambridge University Press, Cambridge. 222 pp.
MORSE, M. P. 1971. Biology and life history of the nudibranch mollusc, Coryphella stimpsoni (Verrill
1879). Biol. Bull. 140: 84-94.
PAINE, R. T. 1965. Natural history, limiting factors and energetics of the opisthobranch Navanax inermis.
Ecology 46: 603-619.
RIVEST, B. R. 1978. Development of the eolid nudibranch Cuthona nana (Alder and Hancock, 1942)
and its relationship with a hydroid and hermit crab. Biol. Bull. 154: 157-175.
RUTOWSKJ, R. L. 1982. The temporal structure of behavioral interactions in Hermissenda crassicornis
(Opisthobranchia). Veliger 24: 227-229.
THOMPSON, T. E. 1 96 1 . The structure and mode of functioning of the reproductive organs of Tritonia
hombergi (Gastropoda Opisthobranchia). Q. J. Microsc. Sci. 102: 1-14.
TODD, C. D. 1978. Gonad development of Onchidoris muricata (Muller) in relation to size, age, and
spawning (Gastropoda: Opisthobranchia). J. Molluscan Stud. 44: 190-199.
WILLIAMS, L. G. 1980. Development and feeding of larvae of the nudibranch gastropods Hermissenda
crassicornis and Aeolidia papillosa. Malacologia 20: 99-1 16.
ZACK, S. 1975. A description and analysis of agonistic behavior patterns in an opisthobranch mollusc,
Hermissenda crassicornis. Behaviour 53: 238-267.
Reference: Biol. Bull. 165: 286-304. (August, 1983)
SETTLEMENT AND METAMORPHOSIS OF A TEMPERATE
SOFT-CORAL LARVA (ALCYONIUM SIDERIUM VERRILL):
INDUCTION BY CRUSTOSE ALGAE
KENNETH P. SEBENS
Biological Laboratories and Museum of Comparative Zoology,
Harvard University, Cambridge, MA 02138
ABSTRACT
The temperate soft-coral Alcyonium siderium Verrill has a demersal planula
larva which usually settles and metamorphoses on vertical rock surfaces near the
parent colony. Such surfaces are covered by a variety of encrusting invertebrate
species and by three common crustose algae (Lithothamnium glaciale, Phymatoli-
thon rugolosum, and Waernia mirabilis). Larvae settle and metamorphose most
frequently on these three algal species in the field (Sebens, 1983).
Contact with each of the three crustose algae induced settlement and meta-
morphosis within 1 -5 days in laboratory experiments. Rock or shell fragments, even
with naturally filmed surfaces, did not induce metamorphosis in the same time
period. A few larvae did metamorphose on the rock, shell, and glass or plastic
surfaces of the containers, taking up to 30 days to do so. Larvae were kept alive up
to 194 days but their competence to metamorphose declined significantly after ten
days. The half-life of larvae that did not metamorphose was approximately 25 days.
Larvae presented with coralline algae in darkness delayed metamorphosis by ap-
proximately 10-20 days, but most of them did metamorphose by 30 days. Neither
sea water incubated with coralline algae, nor coralline algae in close proximity (4-
5 mm) to the larvae, but without contact, induced metamorphosis. Induction of
settlement and metamorphosis is thus mediated by surface contact with the algae
and probably not by a dissolved chemical. Presence of the colonial ascidian, Apli-
dium pallidum, inhibited metamorphosis even when larvae were able to contact
coralline algae, and also caused early larval death.
INTRODUCTION
The planulae of octocorals are usually brooded by the adult colony to a swim-
ming stage (Matthews, 1917; Gohar, 1940; Hartnoll, 1975, 1977; Weinberg, 1979;
Weinberg and Weinberg, 1979) which settles and crawls on the substratum. They
may also be released as demersal crawling larvae (Hartnoll, 1977). The swimming
larvae are similar in morphology and behavior to those of certain scleractinian corals
(Abe, 1937; Atoda, 1947a, b, 195 la, b, c, 1953; Kawaguti, 1941, 1944; Harrigan,
1972a, b; Lewis 1974), hydroids (Nishihara, 1967a, b, 1968a, b; Donaldson 1974),
scyphozoans (Brewer, 1976a, b; Neumann, 1979), and sea anemones (Chia and
Spaulding, 1972; Siebert, 1973). Behavior of the demersal planulae is similar to that
described for scleractinian corals (Gerrodette, 1981; Fadlallah and Pearse, 1982;
Fadlallah, 1983), certain hydroids (Williams, 1965, 1976), and hydrocorals (Ostar-
ello, 1973, 1976). Settlement and substratum choice has been studied for few an-
thozoans [reviewed by Chia and Bickell (1978)], and for even fewer octocorals
(Theodor, 1967; Chia and Crawford, 1973; Weinberg, 1979; Weinberg and Wein-
berg, 1979).
Received 23 November 1982; accepted 1 April 1983.
286
CORAL LARVAL SETTLEMENT 287
The soft-coral Alcyoniwn siderium Verrill is common on vertical rock surfaces
at 6-17 m depth along the coast of Northern Massachusetts and further north in
the Gulf of Maine. It broods lecithotrophic demersal planulae which are released
in late July or August (Feldman, 1976). The released planulae either drift with the
current or crawl down the parent colony and onto the nearby substratum. Alcyonium
hibernicum in the British Isles has a similar demersal larva (Hartnoll, 1977) which
differs from the better-known swimming larva of Alcyonium digitatum (Hartnoll,
1975, 1977).
I observed larval settlement and metamorphosis of Alcyonium siderium planulae
in the field during August of 1978, 1979, 1980, and 1981 and quantified availability
of substratum types and frequency of larval metamorphosis on all available substrata
(Sebens, 1983). Two species of coralline algae (Lithothamnium glaciale, Phymato-
lithon rugulosum) and one fleshy red crustose alga ( Waernia mirabilis proposed, R.
Wilce manuscript) were by far the most common substrata chosen by the larvae.
Laboratory studies were then designed to find out if metamorphosis could be induced
by the presence of these algae or whether larvae were just attaching to any piece of
hard substratum near the parent colony. The following questions were addressed
experimentally: 1 . Can any of the three algal species induce settlement and meta-
morphosis?, 2. Is this induction mediated by surface contact or by substances dis-
solved in sea water?, 3. Can any of the common encrusting larval invertebrates be
used as substratum or induce metamorphosis?, and 4. How long can larvae survive,
and are they competent to settle and metamorphose if they do not receive the
appropriate stimulus within the first few days? There is good evidence that certain
bryozoan larvae can avoid settling near colonial ascidians which are known to ov-
ergrow established bryozoan colonies (Grosberg, 1981; Young and Chia, 1981).
Small Alcyonium colonies are overgrown by the ascidian Aplidium pallidum in the
field (Sebens, 1982). Therefore, additional laboratory experiments were designed to
test whether Alcyonium larval settlement would be inhibited in the present of
Aplidium.
MATERIALS AND METHODS
Fifteen large colonies of Alcvonium were collected at the Shag Rocks, Nahant,
MA (42°24'50" N; 70°54'20" W) from vertical rock surfaces at 6-9 m depth. Corals
were scraped off carefully and placed in plastic containers. Only colonies with visible
planulae in the anthocodia were taken. Collections were made during August of
each year (1980-1982) when ambient temperature ranged from 8-2 1°C for the
month. Larvae were visible in most colonies in early August (1978-1982) but were
present in very few colonies by the end of each August. Colonies were maintained
in the laboratory in aerated sea water at 1 1-1 3°C overnight.
Colonies were slit lengthwise along the lobes with a razor blade, then swished
back and forth in filtered (80 M mesh) sea water to remove the larvae. The larvae
and sea water were passed through 80 M Nitex, followed by two rapid washes in
clean filtered sea water. The mesh was then quickly inverted into clean filtered sea
water. Larvae, eggs, and some colony fragments settled to the bottom of the dish,
from which colony fragments were then removed individually. Elongate crawling
planulae (2 mm long) were removed by pipette for each experimental replicate (15
in 1980, 30 in 1981, 1982). This technique probably prevented larvae which were
still at early stages of development from being included in the experiment.
Settlement experiments were carried out either in a refrigerated chamber ( 1 1 °C,
1980, 1981) or in a cold room (13°C, 1982). Containers for the 1980 experiments
288 K. P. SEBENS
were glass vials 5 cm tall, 2 cm diameter. Those for the 1981 experiments were
plastic Petri dishes 4.5 cm diameter. All vessels had been soaked in flowing sea water
for 60 days prior to the experiments, then rinsed in fresh water to remove the organic
film. The 1980 and 1981 containers held 6 ml of filtered sea water. The 1982
experiments used wide mouth jars (3 cm tall, 4 cm bottom diameter) containing
12 ml of filtered sea water 1 cm deep. All vessels (1981, 1982) were mounted on
a rocking platform that stirred the water by tilting to 15° each 5 seconds. In the
1982 experiments 12 hours of agitation were alternated with 12 hours at rest because
the constant agitation in the 198 1 experiments caused many larvae to metamorphose
without attaching. Water was replaced every 48 hours by pipetting off the old water
and adding fresh filtered sea water. Two fluorescent bulbs (40 watt) at 30 cm from
the containers were used as the light source. Darkened treatments were kept on the
same apparatus in an aluminum foil box with spaces to allow air flow. The 1980
experiments were not continuously agitated, but instead were aerated with an air
pump and pipette twice daily. Water was changed daily.
Substrata to be used in treatments were collected from the same site as were the
corals, on rock (1982) or mussel (Modiolus modiolus) shell (1980, 1981). The rock
or shell was fractured and trimmed to produce pieces ^ 1 X 1 X 0.5 cm with
appropriate test substratum on the upper side. Controls were the same rock or shell
without algae on the surface. At least one surface of the shell or rock was the original
exposed surface but without algae or invertebrates. Lithothamnium glaciale, Phy-
matolithon rugulosiim, the red crustose alga Waernia mirabilis, the sponge (Hali-
sarca dujardini), and colonial tunicates (Aplidium pallidum) accounted for most of
the space cover on walls with Alcyonium (Sebens, 1982, 1983). Each of these or-
ganisms was also used as an experimental substratum.
Controls were prepared with only the glass or plastic container as substratum,
in both light and darkness. In the 1982 series of experiments, vigorously aerated
treatments were also included. Glass tubing from a vibrator aquarium pump was
used to bubble air through these containers. This treatment was an attempt to
determine whether the oxygen production of crustose algae alone could have induced
settlement.
If the presence of any of the experimental substrata induced metamorphosis, it
would be of interest to determine whether induction could be mediated by chemicals
released by the substratum and dissolved in sea water. In the 1980 experiments, sea
water was incubated with each substratum for 24 hours at 1 1 °C (termed 'super-
natant') before being poured off and used in the experimental treatment. This would
allow metabolic products of the algae or invertebrates to concentrate before being
introduced into the larval containers. This treatment was repeated daily with fresh
supernatant.
The 1980 experiments indicated that coralline algae could induce metamor-
phosis. An experiment was thus designed in 1981 to find out whether contact with
the alga was necessary. In this experiment, the Lithothamnium substratum was
suspended by fine monofilament line 4-5 mm above the bottom of the container
without touching the walls. This design would allow exudate from the algae to
contact the larvae but would prevent contact with the algal surface.
Abalone larvae settle on coralline algae and can be induced to settle by the
presence of algal extracts or by the chemical GABA (7-aminobutyric acid) (Morse
et al, 1979). Since coralline algae induced settlement in Alcyonium siderium (1980
experiments), it was of interest to test for possible mediation by GABA. Groups of
larvae were kept in the light with GABA in sea water ( 1 ^M/\, 50 nM/\, 1 mA//l),
CORAL LARVAL SETTLEMENT 289
changed daily, since induction of metamorphosis by coralline algae occurred much
more rapidly in the light.
Statistical analysis of data (Analysis of variance (ANOVA), Student-Newman-
Keuls multiple comparisons test (SNK test) and Chi-squared nonparametric test)
were based on methods in Sokal and Rohlf (1969). Table I summarizes the exper-
imental protocol, conditions, and results for all three years.
RESULTS
Survey of potential substrata
The first set of experiments (August 1 980-May 1981) pointed out the importance
of coralline algae as inducers of metamorphosis. Lithothamnium was the only sub-
stratum that induced settlement within the first three days, and was certainly the
only substratum which caused large numbers of larvae (27 of 45) to metamorphose.
The sea water control treatment had three larvae metamorphose between days 3
and 5 and Waernia had only one after 49 days (Table II, Fig. 1 ). Alcyonium colonies
did not induce settlement and metamorphosis ruling out larval aggregation around
adult colonies as a result of adult chemical mediation. Halisarca did not induce
settlement, but some larvae remained alive until the end of the experiment. Aplidium
did not induce settlement either, but most larvae died within the first week.
Sea water incubated for 24 hours with each of the substrata (termed 'superna-
tant') failed to induce metamorphosis. Since Lithothamnium supernatant did not
have the same effect as Lithothamnium itself, it appeared that there was no chemical
dissolved in sea water that was mediating the effect of the coralline alga. It was also
evident that settlement of the larvae in the presence of corallines did not necessarily
occur on the surface of the alga itself. In fact, more larvae metamorphosed on the
bottom of the glass vials. There was also no larval swimming or negative geotaxis
(i.e., crawling up the walls of the vial). All settlement was on the bottom. A few
larvae, however, did crawl to the top surface of the shell fragment and attached
directly to the alga or to the shell surface.
This set of larval settlement and metamorphosis experiments had several less
than optimal conditions. The temperature ranged from 8-12°C, the water was not
agitated constantly, and treatments were kept in darkness most of each day. The
temperature range was well within that observed for the August period in the field
(8-2 1°C). However, later experiments pointed out the importance of light in in-
ducing settlement and the short light period may have slowed down the rate of
settlement. Agitation of the water did not appear necessary for larval survival, which
continued for up to nine months (at 5°C for months 3-9), even without daily
aeration.
Mechanism of induction of metamorphosis by coralline algae: effects of contact
and light regime
During the August-September 1 98 1 experiments temperature was kept constant
( 1 1 ± 1 °C), treatments were continuously agitated, and were maintained under
constant low irradiance. The percentage of larvae that settled in the presence of
coralline algae, and the rapidity with which they metamorphosed, indicated that
this set of conditions was more conducive to their substratum selection process.
Constant slow agitation did prevent a fairly large percentage (10-40) of the meta-
morphosed individuals from attaching to any surface during the entire experiment.
290
K. P. SEBENS
TABLE I
Summary of experiments for induction of settlement and metamorphosis o^ Alcyonium planulae
Experiment
Date
H
Light
H
Dark
Significant
Purpose Settlement
Lithothamnium on shell
1980
3
21
Test for possible induction by
this substratum
Y
Waernia on shell
1980
3
21
Test for possible induction by
this substratum
N
Halisarca on shell
1980
3
21
Test for possible induction by
this substratum
N
Aplidium on shell
1980
3
21
Test for possible induction by
this substratum
N
Alcyonium on shell
1980
3
21
Test for possible induction by
this substratum
N
Shell substrate alone
1980
3
21
Control for effects of other
substrata
N
Lithothamnium supernatant
1980
3
21
Test for possible induction by
N
Waernia supernatant 1980 3 21
Halisarc a supernatant 1980 3 21
Aplidium supernatant 1 980 3 2 1
Alcyonium supernatant 1980 3 21
Lithothamnium on shell 1981 24 0
Lithothamnium on shell 1981 0 24
Lithothamnium on shell 1981 12 12
Lithothamnium on shell 1981 24 0
suspended
Lithothamnium on shell 1981 24 0
using old larvae
Lithothamnium on shell 1981 24 0
with Aplidium
Phymatolithon on shell 1981 24 0
soluble chemicals released
by this substratum
Test for possible induction by N
soluble chemicals released
by this substratum
Test for possible induction by N
soluble chemicals released
by this substratum
Test for possible induction by N
soluble chemicals released
by this substratum
Test for possible induction by N
soluble chemicals released
by this substratum
Test for induction by this Y
substratum in light
Test for induction by this Y
substratum in dark
Test for induction by this Y
substratum in light/dark
cycle
Test for induction by this N
substratum without direct
contact
Test for competence of larvae N
denied induction stimulus
for 10 days
Test for inhibition of N
induction by Aplidium
Test for induction by this Y
substrate in light
CORAL LARVAL SETTLEMENT
291
TABLE I (Continued}
Experiment
H H
Date Light Dark
Purpose
Significant
Settlement
Shell substratum alone
Shell substratum alone
Sea water alone
Sea water alone
GABA in sea water, 3
concentrations
Lithothamnium on rock
Lithothamnium on rock
Phymatolithon on rock
Waernia on rock
Rock substratum alone
Lithothamnium on rock
with Aplidium
Sea water alone, aerated
1981
24
0
1981
0
24
1981
24
0
1981
0
24
1981
24
0
1982
24
0
1982
1982
1982
0
24
24
1982 24
1982 24
1982 24
24
0
0
0
0
Control for effects of other
substrata in light
Control for effects of other
substrata in dark
Control for effects of shell
substratum in light
Control for effects of shell
substratum in dark
Test for induction by GABA
Test for induction by this
substratum in light (repeat
of 1981 treatment)
Test for induction by this
substratum in dark (repeat
of 1981 treatment)
Test for induction by this
substratum in light (repeat
of 1981 treatment)
Test for induction by this
substratum in light
(conditions different than
in 1980)
Control for effects of other
substrata
Test for inhibition of
settlement by Aplidium
Test for induction of
settlement by increased
oxygen tension alone,
asmight occur with
crustose algae in the light
N
N
N
N
Y
N
N
N
Lithothamnium again induced metamorphosis of the greatest numbers of larvae
(Table III, Fig. 2). Phymatolithon, the other coralline alga, also induced a large
proportion of larvae to metamorphose. Note that there was a great difference between
two subsequent sets of three replicate groups with both Lithothamnium and Phy-
matolithon (A and B in Table III). In a light/dark cycle (12h each), metamorphosis
in the presence of Lithothamnium was comparable to that with constant light (Fig.
3). In constant darkness, most larvae did not metamorphose until after 10 days (Fig.
3). However, almost all larvae did metamorphose by 30 days.
When Lithothamnium was separated from the larvae by 4-5 mm, metamor-
phosis was drastically reduced (not statistically different from the control, Fig. 3).
This agrees with the previous year's results that indicated that the induction was
not mediated by a chemical diffused through sea water, and that contact with the
292
K. P. SEBENS
TABLE II
Alcyonium larval metamorphosis experiments conducted during August 1980 to May 1981
Lithothamnium
Waernia
Sea Water Control
Days
1
0
0
0
2
1.0 ± 1.0
0
0
3
3.0 ± 1.0**
0
0
5
5.0 ± 2.6
0
1.0 ±
1.7
10
6.3 ± 2.3**
0
1.0 ±
1.7
49
8.3 ± 4.7**
0
0
194
8.3 ± 4.7**
0.3 ± 0.6
0
Treatments without metamorphosis: Halisarca (sponge), Aplidium (tunicate), Alcyonium, Litho-
thamnium supernatant, Waernia supernatant, Halisarca supernatant, Aplidium supernatant, Alcyonium
supernatant. All treatments were given 2-3 hours light per day at 8-12°C, then 5°C after day 49. Values
are mean number of larvae metamorphosed, out of an initial 1 5, ±S.D. for three replicates.
** Denotes treatments significantly different than the control (ANOVA, P < 0.05 at least).
coralline alga was necessary. Constant aeration of sea water alone (1982) did not
induce metamorphosis. Therefore, it is unlikely that the addition of oxygen to the
water by the crustose algae could, by itself, be the factor mediating induction of
metamorphosis. I considered using dead coralline algal skeletons to see if the in-
duction was mediated by surface texture rather than by contact chemoreception.
However, this would not differentiate the potential role of surface texture of the live
ALCYONIUM SIDERIUM METAMORPHOSIS
CORALLINE 3
9 10 49
9/24/80
194
4/4/81
DAYS
FIGURE 1. Number of larvae that had metamorphosed (of initial 15), during the 1980 experiments,
on each of the three coralline algae replicates (Lithothamnium glaciate), and on the crustose red alga
Waernia mirabilis. One of the 3 replicates in the control group also had some metamorphosis.
CORAL LARVAL SETTLEMENT
293
< "
2
+1
o
^^
O
o :
O
U.
1/5 >
Q
q
E
•^
™
=
2
04
,
(N
(N
(N
r-j
^
<J" UJ
_•
"-
—
—
—
£
UJ *~
rh
o
+1
+1
+1
+i
3
on
t/1 <
j
r-
r-
r-
r-
3
d
d
d
d
0
O
g
^
r-
d
d
-
C
a
H
*
o
o
+1
+1
+1
I
M
z
Q
v->
10
q
U
K
u
d
d
(N
%
Si
UJ
o
r--
r-
vO
q
<u
>
X
d
d
d
t
CT3
O
~
[—
_
o
+1
+1
+1
+1
«
J
z
o
u
11
d
d
r-
d
vO
—
K
Q
0
r-
vO
m
__
t*-
03
d
d
(N
•-C
o
H
CQ
0
+1
+1
+i ;
+1 1
•g
^.
g
d
vO
i —
1 —
E
|
C
"S
^
t—
*/")
r*
rt
fl
Q
i
Q.
<
J
+\
q
+1
r-j
+1
0
q
I
d
—
—
d
—
+i
^
73
g
I
X Q
UJ
>
o
-H
r-
d
+1
O
fN
+1
O
S
a.
^
•^
— Q
~1
r-
r*~t
rn
O
_±i
• o
UJ
"S
J
-J
d
d
—
S
73
< "^
_j
co
H
1
O
9
r-
d
d
(N
-.
E
u
ca
q
d
VI
0 ^
Z J=
< M
to
j_
^
+1
+i
+1
+1
o
l_
a.
— u
Q
"^
^
m
en
r-
q
> w
d
d
d
— '•
J3
^
4-> C3
^
o
^>
• o
~
u
0
C "°"
~
0
^
d
00
u-i
d
£
E
z
(U k.
1
X
H
<
Q
o
o
+1
+1
+1 *
q *
fN
3
C
C
U
00
IS
2|
1
'G
•s
O H
X X
o.
o
o
d
+1
vO
d
+1
vb
+1
Si
(0
(XI
u
3
: control
O ed
VI S
a, -
o
J J
1/1
d
r-
q
C3
>
•S
I! CT
^j
c
P3 ^
r-«
r-
VO
10
rr>
^
C
S.
"5 3
experimel
LITHO
Q
j
+1
q
(N
+1
O
(N
+1
q
+1
q
d
-H »
q
r-
0
+1
0
y different
C 0
0 £
O ~
ti O
E £
(-.
•^.
O OJ
_t<n
op
c
to
,
r-~
*o
CO
~ ^3 *^
O
t
(N
<N
rn
?
u
•-C
E
Ji < '|
amor
H
X
o
03
o
O
+i I
+1
q
+1 J
d
c
s/3
op
LO
C
IIJ
S $ •$
53
S
"5
o
I
-
-
r-
-
oo
1*1
c
0
o
1
treatme
C K ^
OJ ri *"•
E -c -3
S 1 £
i^
™J
+1
+1 »
+1 •
+1 *
-H *
l/J
w)
^~ ^
_2
*^
"> *
0 *
0 *
q *
c
0
£ "*i •**
6
—
d
(N
d
(N
(N
u
E
O
c
o O 5
3
c3
Q
"-1 E §
'S
IJ
u C u
o
CO
*
» J OJ
>•*
X*
_^
(N
</"»
O
^O
o
o
^£
— —
ro
US
<;
Q
^
294
K. P. SEBENS
N 20
N 20
LITHOTHAMNIUM SUSPENDED
30 -
N 20
10
5 10
15 20
30 35 40 45
FIGURE 2. Cumulative number of larvae metamorphosed (of initial N = 30 planulae) in treatments
with Lithothamnium (in light and suspended in light) and Phymatolithon (in light). Values are mean
number of metamorphosed larvae ± S.D. for three replicates of each treatment (1981 or 1982).
30
N 20
,1
N 2°
10
LITHOTHAMNIUM
LIGHT/ DARK CYCLE
FIGURE 3. Cumulative number of larvae metamorphosed (of initial N = 30 planulae) in treatments
with Lithothamnium (in light/dark cycle, and in the dark). Values are mean number of metamorphosed
larvae ± S.D. for three replicates of each treatment (1982).
CORAL LARVAL SETTLEMENT 295
alga. The surface contacted by the larva is living cell surface, not the carbonate
skeleton.
Control treatments included the same rock or shell material that the algae were
growing on, with its natural surface. This surface was probably covered with a
bacterial film, known to induce settlement in several invertebrate larvae (Crisp and
Ryland, 1960) including bryozoans (Mihm et al, 1981; Brancato and Woollacott,
1982), polychaetes (Kirchman et al., 1982), hydroids (Spindler and Miiller, 1972),
and scyphozoans (Brewer, 1976b; Neumann, 1979). Treatments with naturally
filmed rock or shell surfaces alone did not cause more larvae to metamorphose than
sea water controls with only the cleaned glass or plastic surfaces available (Table
III, Fig. 5). However, bacterial films can develop in a matter of hours and the
artificial surfaces were probably also covered with bacteria since the experiments
lasted for many days. Neither rock, shell, nor the artificial surfaces ever had the
rapid effects of the crustose algae, and it is unlikely that bacteria alone are inducing
settlement in the Alcyonium larvae, unless there are specific bacteria associated with
the algal surface that are being recognized.
Larvae that were kept for 10 days in filtered sea water (old larvae, Table III) had
very low rates of metamorphosis even with Lithothamnium present. This is sur-
prising since many of the larvae kept with Lithothamnium in the light metamor-
phosed between days 5 and 20 (B, Table III) and most of those in the dark meta-
morphosed between days 10 and 36. The results of the 1980 experiments indicated
that some larvae remained competent even after 49 days. Clearly there is some
reduction in the larvae's ability to metamorphose given increased time without a
stimulus.
There was no induction of settlement (attachment) or metamorphosis by GABA
at any of the experimental concentrations. The only visible effect of GABA at the
highest concentration (1 mM/1) was that the planulae were thin and extremely
elongate, up to twice as long as normal. Crawling was discerned at the 1 p.M/\ and
50 fj.M/1 concentrations but not at 1 mM/\. The lack of attachment or metamor-
phosis in the presence of GABA argues for a different mediation by corallines from
that suggested for abalones (Morse et al., 1979) or for chitons (Rumrill and Cameron,
1983). It is possible that introduction of GABA occurred before larvae were com-
petent. This sometimes prevents larvae from ever responding to the stimulus (e.g.,
gastropods, Hadfield, 1977). However, presence of the known inducer, Lithoth-
amnium, did induce metamorphosis in larvae from the same batch (Table III). The
attachment and initial change from elongate to rounded morphology takes many
hours and some larvae had completed this process within the first 24 hours. Larvae
were thus competent initially or became so rapidly during the first day.
Experiments with Waernia
The experiments conducted during August-September 1982 (13° ± 1°C, con-
stant low light) introduced intermittent agitation so that metamorphosing larvae
had time to become firmly attached. In fact, only 0-20 percent of metamorphosed
individuals in each treatment were unattached by the end of the experiment. As in
the previous year's experiments, Lithothamnium and Phymatolithon were strong
inducers of metamorphosis (Table IV, Fig. 2). Waernia was tested again because
many larvae in the field metamorphosed on it (Sebens, 1983). This time Waernia
was as successful in inducing settlement as were the corallines (Fig. 4). The control
had slightly more metamorphosis this year than previously (Fig. 5). The lit control
had more larvae metamorphose than did Lithothamnium in the dark, but the dif-
ferences were not significant.
296
K. P. SEBENS
TABLE IV
Alcyonium larval metamorphosis experiments conducted during August to September 1982
DAYS
PHYM
LIGHT
WAER
LIGHT
LITHO
LIGHT
LIGHT
CONTROL
LITHO
DARK
APLIDIUM AERATED
+ LITHO SEAWATER
1
0.7 ±
1.2
0.3 ± 0.6
0
0
0
0
0
3
9.0 ±
1.7*
9.0 ± 3.0*
9.7
± 2.5*
4.3 ± 2.5
1.7 ±
2.1
1.7 ± 2.0
0**
5
10.7 ±
3.2*
9.3 ± 3.1*
10.3
± 3.5*
5.3 ± 2.3
2.3 ±
3.2
1.7 ± 2.1
0**
11
10.7 ±
3.2
11.0 ± 2.7*
11.7
± 4.7*
7.3 ± 2.5
3.0 ±
3.6
1.7 ±
2.1**
0**
30
14.0 ±
3.5**
11.6 ± 2.1*
11.3
± 4.0*
7.0 ± 2.5
4.7 ±
3.8
0
**
0**
Experiments were run at 13°C with low light levels and intermittent slow stirring. Values are mean
number of larvae metamorphosed, out of an initial 30, ±S.D. of 3 replicates in each treatment.
** Denotes treatments significantly different than the light control (ANOVA, P < 0.05 at least).
* Denotes treatments different from the light control at the P < 0. 10 confidence level (ANOVA).
LITHO = Lithothammum, PHYM = Phvmatolithon. WAER = Waernia.
Effects o/"Aplidium
The 1980 experiments had no settlement in treatments with the tunicate Apli-
dium, with the sponge Halisarca or with the Alcyonium colonies present. Field
experiments (Sebens, 1983) showed that settlement did not occur on sponge or
30
20
N
10
30
20
N
10
WAERNIA 1982
t I )
APLIDIUM
---o 1981 WITH LITHOTHAMNIUM
— • 1962 WITH LITHOTHAMNIUM
AND AERATION
f-r1 fr
io
I5
20
25
30
DAYS
FIGURE 4. Cumulative number of larvae metamorphosed (of initial N = 30 planulae) in treatments
with Waernia (in light 1 982) and Aplidium plus Lilholhamnium (in light 1 98 1, in light with aeration
1 982). Values are mean numbers of metamorphosed larvae ± S.D. for three replicates of each treatment.
CORAL LARVAL SETTLEMENT
297
30
20
N
10
30
20
N
i
DARK CONTROL
— a 1982
• 1981
LIGHT CONTROL
o 1982
• 1981
FIGURE 5. Cumulative number of larvae metamorphosed (of initial N = 30 planulae) in control
treatments with only rock (1982) or Modiolus shell (1981) substratum. Values are mean number of
metamorphosed larvae ± S.D. for three replicates of each treatment.
tunicate surfaces. Since there seemed to be a negative effect of Aplidium on larvae
in 1980, 1 examined its effect in the presence of a known inducer of metamorphosis
(Lithothamnium).
When Aplidium was present with Lithothamnium, larval metamorphosis and
survivorship were again poor (Table IV, Fig. 4). This time the treatments were
constantly aerated to reduce the possibility that the Aplidium was depleting available
oxygen during the experiments. Colonies of Aplidium remained alive and apparently
healthy throughout this set of experiments. However, all such experiments were
within containers, allowing maximum concentration of released metabolites or other
chemicals.
Substratum orientation by larvae
In all three years of experiments, many larvae metamorphosed on the bottoms
of the glass or plastic containers, but never on the walls. Larvae never swam (as
suggested by Feldman, 1976) after removal from adult colonies or during natural
release. The corner where the bottom met the wall was the most common site of
attachment but there was no evidence of aggregation. When the number of meta-
morphosed larvae on the rock or shell surface was compared to that on the bottom
of the container (corrected for surface area), there was no difference in treatments
with Phymatolithon or Waernia in the light, or in the lit controls (1981, 1982
combined) (Table V). However, there was significant preference for the rock or shell
substratum in the Lithothamnium (light or dark) and dark control treatments (Table
V). When there was a preference shown, it was always for the natural substratum.
The large number of larvae metamorphosing on the glass or plastic argues against
298
K. P. SEBENS
TABLE V
Metamorphosis o/"Alcyonium larvae after 30 days, on the container bottom (plastic or glass)
and on the rock or shell material used as substratum (1981 and 1982)
ON
ON
CONTAINER
ON ROCK
CONTAINER
(CORRECTED)
OR SHELL
TREATMENT
(9.6 cm2)
(3.0 cm2)
(3.0 cm2)
x2
P
LITHOTHAMNIUM (light)
84
26
57
20.5
<0.01
PHYMATOLITHON (light)
47
15
22
1.95
>0.05
WAERNIA (light)
23
7
5
0.76
>0.05
LITHOTHAMNIUM (dark)
9
3
64
156.03
<0.01
LIGHT CONTROL
22
7
12
2.61
>0.05
DARK CONTROL
10
3
35
69.30
<0.01
Top (with crustose algae) and bottom of the rock (without) were combined for this comparison.
The number of larvae on the container bottom (9.6 cm2) was corrected to 3.0 cm2.
X2 = chi-squared statistic, P = significance level.
thigmotaxis for coralline or other algal surfaces, although surface texture recognition
by the larvae is certainly possible.
There were several treatments in which it appeared that larvae were primarily
on the bottom, or the top, of the substratum offered. When Phymatolithon covered
the upper surface of the rock or shell, significantly more larvae settled on the bottom
and sides than on the top (algal) surface in the lit treatments (Table VI). There was
also less attachment on the top surface of Waernia and Lithothamnium (light or
light/dark cycle). Only Lithothamnium is darkness had more settlement on the top
(algal surface) than expected by its area. In the lit controls, most settlement was on
the top surface. This indicates that while contact with the crustose algae induces
settlement, the algae may also be able to deter settlement directly onto their living
surfaces.
Larval survivorship
It was clear from the 1980 experiments that long-term survival of actively crawl-
ing planulae was possible (to at least nine months). Survivorship was better in
TABLE VI
Metamorphosis of Alcyonium larvae on the top or bottom of the rock or shell substratum
after 30 days (1981 and 1982)
SUBSTRATE
TREATMENT
TOP
BOTTOM
2
X"
P
LITHOTHAMNIUM (light)
16
41
0.9
>0.05
LITHOTHAMNIUM (light/dark)
8
14
2.1
>0.05
LITHOTHAMNIUM (dark)
32
30
8.9
<0.05
WAERNIA
2
5
0
>0.05
PHYMATOLITHON
9
47
8.0
<0.05
LIGHT CONTROL
11
1
8.3
<0.05
DARK CONTROL
18
15
0.3
>0.05
The top of the rock was covered with the encrusting algae in the first 5 treatments.
X2 = chi-squared statistic, P = significance level.
Area of top surface = 34%, area of bottom and sides combined = 64% of total area.
CORAL LARVAL SETTLEMENT
299
100
50
WAER
LITH
LIGHT CONTROL
10 15 50
DAYS
200 0
10 15 50
DAYS
200
FIGURE 6. Survivorship of planulae during the 1980 experiments expressed as the percentage of
all initial larvae that did not go on to metamorphose. Values are mean ± S.D. for arcsine transformed
data, backtransformed for the graph. At days 1 and 3 treatments were statistically indistinguishable. At
days 7 and 10 the Aplidium treatment was significantly different than all others (P < 0.05, ANOVA and
Student-Newman-Keuls (SNK) multiple comparisons test). At days 49 and 194 the Lithothamnium and
Waernia treatments were different from the rest (SNK test). All other combinations of treatments at each
time were indistinguishable (statistical analysis from Sokal and Rohlf 1969). Abbreviations as follows:
LITH = Lithothamnium. light, WAER = Waernia. LITH SUPER = Lithothamnium supernatant, WAER
SUPER = Waernia supernatant, APLID = Aplidium. APLID SUPER = Aplidium supernatant.
treatments with crustose algae than in treatments with other substrata or in controls.
Survivorship was worst in treatments with Aplidium (Fig. 6). Data on numbers of
live planulae were not taken regularly during 1981. During the 1982 experiments,
live planulae were again counted every other day. In this set of experiments mean
survivorship was between 1 1 and 39 percent for 30 days for larvae that never did
metamorphose (Fig. 7). 28 percent of the light control group, and 61 percent of the
100
10 20
DAYS
10 20
DAYS
30
FIGURE 7. Survivorship of planula larvae during the 1982 experiments expressed as percentage of
all initial larvae that did not go on to metamorphose. Values are mean ± S.D. for arcsine transformed
data, backtransformed for the graph. At days 1 and 3 all treatments were statistically indistinguishable.
At days 5, 7, and 1 1 the aerated treatment was different from the rest (P < 0.05, ANOVA and SNK test).
At days 1 7 and 22 the Aplidium and aerated treatments were indistinguishable but different from all but
the dark control group (on day 17) and the Lithothamnium and Waernia treatments (on day 22); all
other treatments were indistinguishable from each other. At day 30 the 5 treatments that still had living
larvae (all but aerated and Aplidium treatments) were indistinguishable (statistical treatment based on
methods in Sokal and Rohlf (1969). Abbreviations as follows: LITH = Lithothamnium. PHYM = Phy-
matolithon. WAER = Waernia. APLD = Aplidium. LITH DARK = Lithothamnium in darkness, LIGHT
CON = control, in light.
300 K. P. SEBENS
dark control group were still alive after 30 days. All larvae in the aerated treatments
and in the treatments with Aplidium died within the 30 day period. Aeration may
have increased the larvae's metabolism causing them to lose their energy reserves
rapidly. On the other hand, the agitation itself may have caused the larvae to damage
themselves by hitting the walls of the container.
DISCUSSION
The crustose coralline algae, Lithothamnium glaciale and Phymatolithon ru-
gulosum, as well as the fleshy red crustose alga Waernia mirabilis, induced settlement
of Alcyonium siderium planulae in laboratory experiments. Rock surfaces around
Alcyoniwn colonies in the field are covered with colonial invertebrates (tunicates,
sponges, hydroids) and the three crustose algae used in this experiment (Sebens,
1982, 1983). Field studies of larval settlement (Sebens, 1983) showed significant
metamorphosis only on these algae and on adjacent bare rock, although settlement
on Lithothamnium was less than expected by its percent cover and settlement on
Waernia was greater.
Any of the three algae, but not the common encrusting invertebrates Aplidium
pallidum, Halisarca dujardini, or the mussel shell (Modiolus modiolus), can induce
metamorphosis in laboratory experiments. Once the inducing substratum has been
contacted metamorphosis can then occur on nearby rock surfaces, but not neces-
sarily on the algal surface itself. Even so, there was no field settlement of planulae
on any of the encrusting invertebrates adjacent to algal crusts (Sebens, 1983). In a
few vertical rock wall community samples collected by scraping rock surfaces, I have
noted single polyps of Alcyonium attached to erect bryozoans, small red algal fronds,
or to the sides of Aplidium colonies that were encrusted with detritus (unpublished
observations). In the field studies, some larvae settled in the mat of small polychaete
tubes, amphipod tubes, and bound detritus that sometimes covers the encrusting
algae (Sebens, 1983). These individuals were probably attached directly to the algal
surface beneath.
There was distinct inhibition of metamorphosis in darkness, even with Lithoth-
amnium present. It is possible that Lithothamnium does not produce or release the
stimulus in the dark. It is more likely that the larvae are inhibited from receiving,
or responding to, the stimulus in darkness. This mechanism would allow them to
discriminate between deep crevices, underhangs, and open vertical rock surfaces,
especially since they often crawl for several days before metamorphosis. Inhibition
of settlement in darkness may keep them out of microhabitats that are likely to be
far from the greatest water flow thereby allowing the best chance of capturing zoo-
plankton prey. Weinberg (1979) found a positive photokinesis in a Mediterranean
gorgonian coral planula (Eunicella singularis), and a total lack of light-related re-
sponse in that of a second species (Corallium rubrum). It is not clear that Alcyonium
shows either a phototaxis or photokinesis, but instead simply fails to attach and
metamorphose in the dark. Although Alcyonium siderium has a similar habitat
distribution (vertical walls) to Corallium rubrum (Weinberg, 1979), it does not ap-
pear to share a negative geotaxis that would lead the planula up walls or to the
undersides of rock ledges. Release of larvae directly onto the substratum surrounding
the parent colony may alleviate any need for this behavior.
Larvae did not settle significantly, nor survive well, in the presence of the com-
pound ascidian Aplidium pallidum, even when treatments were aerated intermit-
tently (1980) or continuously (1982). Field studies (Sebens, 1982) indicate that
Aplidium overgrows, and probably kills, small colonies of Alcyonium. Larvae will,
CORAL LARVAL SETTLEMENT 301
however, settle near Aplidium in the field (Sebens, 1983). Grosberg (1981) dem-
onstrated that swimming bryozoan larvae avoid settling on experimental plates with
the compound ascidians Botryllus schlosseri and Botrylloides leachi. Both ascidians
overgrew established bryozoan colonies. Young and Chia (1981) found a similar
result in laboratory studies of bryozoan larvae in the presence of other compound
ascidians. In both the present study and that of Young and Chia ( 198 1 ), larvae were
confined with the ascidians in relatively small volumes of water. In Grosberg's study,
settling plates were suspended in the relatively still water of the Eel Pond, Woods
Hole, MA. In all such cases ascidian metabolites or other exuded chemicals could
concentrate at levels that would not be found in more turbulent conditions such
as the field sites where Alcyonium has been studied (Sebens, 1982, 1983). Bryozoan
larvae can swim away if they contact the ascidians; the Alcyonium planulae can only
crawl. Thus, Alcyonium is probably not absolutely restricted from settling near
Aplidium in the field, thereby avoiding overgrowth. If there is a chemical recognition
of the ascidian by the larva, it probably keeps the planula from crawling onto the
ascidian rather than preventing nearby settlement.
The vertical rock wall community is in constant spatial flux. Invertebrates are
often observed overgrowing coralline algae, Waernia, and sometimes small Alcyon-
ium colonies. The presence of uncovered algal crusts indicates either that a grazer
(e.g., the sea urchin Strongylocentrotus droebachiensis) has recently cleared off the
tunicates, sponges, or hydroids, or that those encrusting organisms have receded on
their own (after reproduction or starvation). On vertical walls, such algae are ideal
settlement sites for the soft-coral in that they are hard, stable surfaces that will persist
for long periods of time. Horizontal surfaces adjacent to the vertical walls are com-
pletely covered by Lithothamnium, Phymatolithon, and other corallines but are con-
stantly grazed by sea urchins. Nothing that settles on these algae survives such
grazing very long. On vertical surfaces, grazers are much less common and Alcyon-
ium can probably grow to a size sufficient to be avoided before the area is grazed.
Planulae would probably be induced to metamorphose if they were to drift onto
horizontal surfaces with corallines, but they would not survive.
Coralline algae induce settlement in mollusks which later graze the algal surface
(chitons, Barnes and Gonor, 1973; Rumrill and Cameron, 1983; abalone Morse et
ai, 1979). Harrigan (1972a, b) found that Pocillopora damicornis planulae would
settle on coral rubble with coralline algae on its surface. Breitburg (1983), however,
found that settlement of a variety of invertebrates and algae in the field was less
successful on the surface of corallines than on scraped rock areas. She notes that
corallines are easily overgrown by invertebrate colonies expanding laterally onto
them rather than by direct settlement onto their living surface. Alcyonium will
certainly settle on coralline surfaces under both field and laboratory conditions.
However, there is some evidence that it prefers to settle on the rock, shell, or glass
adjacent to the coralline algae rather than on the algal surfaces after having contacted
the algae in the laboratory. This agrees with field evidence that bare rock is preferred
to corallines (Sebens, 1983).
Alcyonium larvae leave the parent colony and crawl across the substratum for
periods up to several days (Sebens, 1983). However, it appears that most larvae settle
within a few centimeters of the adult colonies. They probably do not have a chance
to leave the local habitat unless they are washed off the colony by wave surge as
they emerge. Similar local dispersal by crawling demersal planulae has been shown
for the temperate Pacific coral Balanophyllia elegans (Gerrodette, 1981; Fadlallah,
1983). Substratum choice is not a matter of settlement, testing and then swimming
away as in barnacle cyprids (Crisp, 1965, 1974), polychaete larvae (Wilson, 1948,
302 K. P. SEBENS
1952, 1954, 1968), hydroid planulae (Nishihara, 1967a, b; 1968a, b; Spindler and
Miiller, 1972, Miiller, 1973), and many other invertebrate larvae (reviewed by Mil-
eikowsky, 1971; Meadows and Campbell, 1972). Crawling larvae are in constant
substratum contact and must respond by either settlement, continued crawling, or
active avoidance of each substratum type. Substrata may be either suitable surfaces
for metamorphosis, or less suitable attachment sites but still inducers of metamor-
phosis. Larvae will settle on non-inducing substrata (rock, shell, glass) after having
contacted Lithothamnium, Phymatolithon, or Waernia. These algae serve as indi-
cators of suitable habitat for the larva rather than as necessary attachment sites.
ACKNOWLEDGMENTS
I thank the following for their field and laboratory assistance: M. Ashenfelter,
D. Denninger, D. Levitan, S. Norton, R. Olson, M. Patterson, J. Sigda, D. Smith,
W. Stotz, and T. VanWey. Edward Elbers and JoAnn Resing helped set up and
monitor substantial portions of the laboratory experiments. I also thank R. Olson
for reading and commenting on the manuscript and R. Steneck and R. Wilce for
algal identification.
The Marine Sciences and Maritime Studies Center of Northeastern University,
the Museum of Comparative Zoology and the Biological Laboratories of Harvard
University provided laboratory space and equipment. The research was supported
by NSF grants OCE 78 08482 and OCE 80 07923 and the Milton Fund of Harvard
University. This is M.S.M.S.C. Contribution no. 117.
LITERATURE CITED
ABE, N. 1937. Post-larval development of the coral Fungia acliniformis var. palawensis. Palao Trap.
Biol. Stn. Stud 1: 73-93.
ATODA, K. 1947a. The larva and postlarval development of some reef-building corals. I. Pocillopora
damicornis cespitosa (Dana). Sci. Rep. Tohoku Univ. (4, biol.) 18: 24-47.
ATODA, K. 1947b. The larva and postlarval development of some reef-building corals. II. Stylophora
pistillata (Esper). Sci. Rep. Tohoku Univ. (4, biol.) 18: 48.
ATODA, K. 195 la. The larva and postlarval development of the reef-building corals. III. Acropora briig-
gemanni (Brook). J. Morphol. 89: 1-13.
ATODA, K. 195 Ib. The larva and postlarval development of the reef-building corals. IV. Galaxea asperea
Quelch. J. Morphol. 89: 17-30.
ATODA, K. 1951c. The larva and postlarval development of some reef-building corals. V. Seriatopora
hystrix Dana. Sci. Rep. Tohoku Univ. (4, biol.) 19: 33-39.
ATODA, K. 1953. The larva and postlarval development of the reef-building coral. Sci. Rep. Tohoku
Univ. (4, biol.) 20: 105-121.
BARNES, J. R., AND J. J. GONOR. 1973. The larval settling response of the lined chiton Tonicella lineata.
Mar. Biol. 20: 259-264.
BRANCATO, M. S., AND R. M. WOOLLACOTT. 1982. The effect of microbial films on larval settlement
in three species of Bugula (Bryozoa). Mar. Biol. 71: 51-56.
BREITBURG, D. L. 1983. Residual effects of grazing: inhibition of competitor recruitment by Lilhoth-
amnion. Ecology in press.
BREWER, R. H. 1976a. Larval settling in Cvanea capillata (Cnidaria: Scyphozoa). Biol. Bull. 150: 183-
189.
BREWER, R. H. 1976b. Some microenvironmental influences on attachment behavior of the planula of
Cvanea capillata (Cnidaria: Scyphozoa). Pp. 347-354 in Coelenterate Ecology and Behavior,
G. O. Mackie, ed. Plenum, New York.
CHIA, F. S., AND B. J. CRAWFORD. 1973. Some observations on gametogenesis, larval development and
substratum selection of the sea pen Ptilosarcus guerneyi. Mar. Biol. 23: 73-82.
CHIA, F. S., AND J. R. SPAULDING. 1972. Development and juvenile growth of the sea anemone Tealia
crassicornis. Biol. Bull. 142: 206-218.
CHIA, F. S., AND L. R. BICK.ELL. 1978. Mechanisms of larval attachment and the induction of settlement
and metamorphosis in coelenterates: a review. In Settlement and Metamorphosis of Marine
Invertebrate Larvae. F. S. Chia and M. Rice, eds. Elsevier North-Holland, New York.
CORAL LARVAL SETTLEMENT 303
CRISP, D. J. 1965. Surface chemistry, a factor in the settlement of marine invertebrate larvae. Bot.
Golhoburg 3: 51-65.
CRISP, D. J. 1974. Factors influencing the settlement of marine invertebrate larvae. Pp. 177-265 in
Chemoreception in Marine Organisms. P. T. Grant and A. M. Mackie, eds. Academic Press,
New York.
CRISP, D. J., AND J. S. RYLAND. 1960. Influence of filming and of surface texture on the settlement of
marine organisms. Nature 185: 1 19.
DONALDSON, S. 1974. Larval settlement of a symbiotic hydroid: specificity and nematocyst responses
in planulae of Proboscidactyla flavicirrata. Biol. Bull. 147: 573-585.
FADLALLAH, Y. H. 1983. Population dynamics and demography of the solitary coral Balanophyllia
elegans. Oecologia in press.
FADLALLAH, Y. H., AND J. S. PEARSE. 1982. Sexual reproduction in solitary corals: overlapping oogenic
and brooding cycles, and benthic planulas in Balanophyllia elegans. Mar. Biol. 71: 223-232.
FELDMAN, S. L. 1976. Validation of the species of soft coral Alcyonium siderium: description and re-
productive biology. M. S. Thesis, Northeastern University. 54 pp.
GERRODETTE, T. 198 1 . Dispersal of the solitary coral Balanophyllia elegans by demersal planular larvae.
Ecology 62: 611-619.
GOHAR, H. A. F. 1940. The development of some Xeniidae (Alcyonaria) (with some ecological aspects).
Publ. Mar. Biol. Stn. At Ghardaqa 3: 27-79.
GRIGG, R. W. 1979. Reproductive ecology of two species of gorgonian corals: relations to vertical and
geographical distribution. Pp. 49-59 in Reproductive Ecology of Marine Invertebrates, S. E.
Stancyk, ed. University of South Carolina Press, Columbia, SC.
GROSBERG, R. K. 1981. Competitive ability influences habitat choice in marine invertebrates. Nature
290: 700-702.
HADFIELD, M. G. 1977. Chemical interactions in larval settling of a marine gastropod. Pp. 403-413 in
Marine Natural Products Chemistry, D. J. Faulkner and W. H. Fenical, eds. Plenum, New
York.
HARRIGAN, J. F. 1972a. The planula larva of Pocillopora damicornis: lunar periodicity of swarming and
substratum selection behavior. Ph.D. Thesis, University of Hawaii.
HARRIGAN, J. F. 1972b. Behavior of the planula larva of Pocillopora damicornis (L.). Am. Zool. 12: 723.
HARTNOLL, R. G. 1975. The annual cycle of Alcvonium digitatum. Estuarine Coastal Mar. Sci. 3: 71-
78.
HARTNOLL, R. G. 1977. Reproductive strategy in two British species of Alcyonium. Pp. 321-328 in
Biology oj Benthic Organisms, B. F. Keegan, P. O. Ceidigh, and P. J. S. Boaden, eds. Pergamon
Press, New York.
KAWAGUTI, S. 1941. On the physiology of reef corals, 5. Tropisms of coral planulae, considered as a
factor of distribution on the reefs. Palao Trop. Biol. Stn. Stud. 2: 319-328.
KAWAGUTI, S. 1944. Zooxanthellae as a factor of positive phototropism in those animals containing
them. Palao Trop. Biol. Stn. Stud. 2: 681-682.
KIRCHMAN, D., S. GRAHAM, D. REISH, AND R. MITCHELL. 1982. Bacteria induce settlements and
metamorphosis ofJamia (Dexiospira) brasiliensisGrube (Polychaeta: Spirorbidae). J. Exp. Mar.
Biol. Ecol. 56: 153-164.
LEWIS, J. B. 1974. The settlement behavior of planulae larvae of the hermatypic coral Favia fragum
(Esper). J Exp. Mar. Biol. Ecol. 15: 165-172.
MATTHEWS, A. 1917. The development of Alcyonium digitatum with some notes on early colony for-
mation. Q. J. Microsc. Sci. 62: 43-94.
MEADOWS, P. S., AND J. I. CAMPBELL. 1972. Habitat selection by aquatic invertebrates. Adv. Mar. Biol.
10: 271-382.
MIHM, J. W., W. C. BANTA, AND G. I. LOEB. 1981. Effects of adsorbed organic and primary fouling
films on bryozoan settlement. J. Exp. Mar. Biol. Ecol. 54: 167.
MILEIK.OWSKY, S. A. 1971. Types of larval development in marine bottom invertebrates, their distribution
and ecological significance: a reevaluation. Mar. Biol. 10: 193-213.
MORSE, D. E., N. HOOKER, H. DUNCAN, AND L. JENSEN. 1979. -y-aminobutyric acid, a neurotransmitter,
induces planktonic abalone larvae to settle and begin metamorphosis. Science 204: 407-410.
MULLER, W. A. 1973. Metamorphose-induction bei planulalarven I. Wilhelm Roux' Arch. 173: 107-121.
NEUMANN, R. 1979. Bacterial induction of settlement and metamorphosis in the planula larvae of
Cassiopea andromeda (Cnidaria: Scyphozoa, Rhozostomeae). Mar. Ecol. Prog. Ser. 1: 21-28.
NISHIHARA, M. 1967a. Observations on the selection of algal substrata by hydrozoan larvae, Sertularella
miurensis in nature. Bull. Mar. Biol. Sta. Asamushi 13: 35-48.
NISHIHARA, M. 1967b. Dispersal of the larvae of a hydroid, Sertularella miurensis. Bull. Mar. Biol. Sta.
Asamushi 13: 49-56.
304 K. P. SEBENS
NISHIHARA, M. 1968a. Experiments on the algal selection by the larvae of Coryne uchidai Stechow
(Hydrozoa). Bull. Mar. Biol. Sta. Asamushi 13: 83-89.
NISHIHARA, M. 1968b. Brief experiments on the effect of algal extracts in promoting settlement of the
larva of Coryne uchidai Stechow (Hydrozoa). Bull. Mar. Biol. Sta. Asamushi 13: 91-101.
OSTARELLO, G. L. 1973. Natural history of the hydrocoral Allopora californica Verrill (1866). Biol. Bull.
145: 548-564.
OSTARELLO, G. L. 1976. Larval dispersal in the subtidal hydrocoral Allopora californica Verrill. Pp. 33 1-
337 in Coelenterate Ecology and Behavior, G. O. Mackie, ed. Plenum, New York.
RUMRILL, S. S., AND R. A. CAMERON. 1983. Effects of gamma-aminobutyric acid on the settlement of
larvae of the black chiton Kalharina tunicata. Mar. Biol. 72: 243-247.
SEBENS, K. P. 1982. Competition for space: growth rate, reproductive output, and escape in size. Am.
Nat. 120: 189-197.
SEBENS, K. P. 1983. The larval and juvenile ecology of the temperate octocoral Alcyonium siderium
Verrill. I. Substratum selection. J. Exp. Mar. Biol. Ecol. 70: 1-17.
SIEBERT, A. E. 1973. A description of the sea anemone Stomphia didemon sp. nov. and its development.
Pac. Sci. 27: 363-376.
SOKAL, R. R., AND F. J. ROHLF. 1969. Biometry. W. H. Freeman, San Francisco.
SPINDLER, K. D., AND W. A. MULLER. 1972. Induction of metamorphosis by bacteria and by a lithium
pulse in the larva of Hydractinia echinata (Hydrozoa). Wilhelm Roux' Arch. 169: 271-280.
THEODOR, J. 1967. Contribution a F etude des gorgones (VII): ecologie et comportement de la planula.
Vie Milieu 18: 291-301.
WEINBERG, S. 1 979. The light-dependent behavior of planula larvae ofEunicella singularis and Corallium
riibrum and its implication for octocorallian ecology. Bijdr. Dierkd. 49: 16-30.
WEINBERG, S., ANDF. WEINBERG. 1979. The life cycle of agorgonian: Eunicella singularis (Esper, 1791).
Bijdr. Dierkd. 49: 1-15.
WILLIAMS, G. B. 1965. Observations on the behavior of the planulae larvae of Clava squamata. J. Mar.
Biol. Assoc. U. K. 45: 257-273.
WILLIAMS, G. B. 1976. Aggregation during settlement as a factor in the establishment of coelenterate
colonies. Ophelia 15: 57-64.
WILSON, D. P. 1948. The relation of the substratum to the metamorphosis of Ophelia larvae. J. Mar.
Biol. Assoc. U. K. 31: 413-438.
WILSON, D. P. 1952. The influence of the nature of the substratum on metamorphosis of the larvae of
marine animals, especially the larvae of Ophelia bicornis Savigny. Ann. Inst. Oceanog. Monaco
27:49-156.
WILSON, D. P. 1954. The attractive factor in the settlement of Ophelia bicornis Savigny. J. Mar. Biol.
Assoc. U. K. 33: 361-380.
WILSON, D. P. 1968. The settlement behavior of the larvae of Sabellaria alveolata (L.). J. Mar. Biol.
Assoc. U. K. 48: 387-435.
YOUNG, C. M., AND F. S. CHIA. 1981. Laboratory evidence for delay of larval settlement in response
to a dominant competitor. Int. J. Invertebr. Reprod. 3: 221-226.
Reference: Biol. Bull. 165: 305-320. (August, 1983)
ON THE EVOLUTIONARY CONSTRAINT SURFACE OF HYDRA
L. B. SLOBODKJN AND KENNETH DUNN
Department of Ecology and Evolution, State University of New York. Stony Brook, L.I., NY 1 1794
ABSTRACT
Food consumption, body size, and budding rate were measured simultaneously
in isolated individual hydra of six strains. For each individual hydra the three mea-
surements define a point in the three dimensional space with axes: food consump-
tion, budding rate, and body size. These points lie on a single surface, regardless of
species. Floating rate and incidence of sexuality map onto this surface. We suggest
that this surface is an example of a general class of evolutionary constraint surfaces
derived from the conjunction of evolutionary theory and the theory of ecological
resource budgets. These constraint surfaces correspond to microevolutionary
domains.
INTRODUCTION
While there may be many conceivable solutions to the ecological and evolu-
tionary problems faced by organisms, not all of these solutions are equally practicable
from the standpoint of the organisms themselves (Wright, 1932). An ideally designed
organism, able to meet all contingencies, need neither evolve nor reproduce. How-
ever, organisms are constrained in their structure and capacities as if, as noted by
Bateson (1963), there were an "economics" of somatic response and evolution. As
a rule, while the existence of these constraints is accepted, they cannot be explicitly
and completely described for any group of organisms, due primarily to gaps in our
knowledge of natural history and development. As a rule, properties to be studied
are selected for either interest or convenience and there is no attempt at explicitly
describing any organism's complete evolutionary strategy (in the sense of Slobodkin
and Rapoport, 1974, and Plotkin and Odling-Smee, 1981). This is due, in part, to
the inherent complexity of most organisms.
Hydra seemed simple enough in anatomy and sufficiently restricted in their
behavior to facilitate an attempt at a reasonably complete explicit description and
quantitative analysis of evolutionary restrictions. We present part of this description
here. Further descriptive experiments are underway and a mathematical analysis,
suggested by the descriptive work to date, is being developed by Gatto, Matessi, and
Slobodkin (in prep.).
Hydra are generally similar in shape. Species differ in body size, budding rate,
and the presence or absence of symbiotic algae. The spectrum of physiological and
behavioral responses does not differ markedly among hydra species, but they do
differ in the circumstances which elicit these responses. It was hypothesized by
Slobodkin (1979) that perhaps all individual hydra, regardless of species, could be
considered to show the same basic patterns of growth and development, differing
only in the way that a given amount of food energy is partitioned between the
maintenance of the adult's body and reproduction.
This hypothesis was presented in geometric form as a curved surface in a three
Received 14 February 1983; accepted 25 May 1983.
305
306
L. B. SLOBODKIN AND K. DUNN
dimensional space, with the axes steady-state body size, steady-state budding rate,
and food consumption. In Slobodkin (1979) this was referred to as an "Adaptive
Response Surface." Since then the word "Adapted", and its etymological relatives,
have become embroiled in almost polemical discussions. We would therefore prefer
to use the term "Constraint" Surface. The term "steady-state" restricts predictions
to hydra individuals that have had a relatively constant food supply for long enough
that neither body size nor budding rate are changing. It is implicitly assumed that
senescence does not occur in hydra. That is, any hydra, regardless of species was
assumed to lie on a two-dimensional surface in the space defined by the three
dimensions body size, food income, and budding rate. This hypothetical surface is
shown in Figure 1.
The hypothesis also asserts that either a clone of hydra in which a series of
individuals are each equilibrated to a different food level or a single individual with
a very slowly changing food supply, will trace a line on the surface. The animals
are assumed to have already completed their transition from bud to potentially
reproductive adult.
The shape of the surface takes account of the well known fact that budding rate
increases with food consumption of individual hydra, that larger hydra species re-
produce more slowly than smaller ones at any given food supply, and that hydra
stop budding and become smaller when starved (Slobodkin, 1964; Stiven, 1965;
Hecker and Slobodkin, 1976; Gurkewitz et ai, 1980; i.a.).
It is not tautological that a single surface should account for the variation between
hydra species. It is possible to imagine, for example, that all hydra partition energy
FIGURE 1. A surface relating body size, budding rate, and food supply for all species of hydra,
hypothesized by Slobodkin (1979). Each meridional line represents the locus of a particular genotype in
the size, budding rate, and food space. The possible states of individuals of a particular species would
be represented by a stripe on this surface, covering several such lines. It is assumed that below some food
level, A, all hydra will die of starvation. It is further assumed that there exists a food level A', such that
at food levels between A and A', even the smallest species are considered unable to reproduce.
EVOLUTIONARY CONSTRAINTS IN HYDRA 307
between growth and reproduction but that the efficiency of the growth and repro-
ductive processes themselves differ between species or with age. If this were so we
would expect a cloud of points in three dimensional space, whose upper bound
might be similar to Figure 1 . To be on a single surface requires that the organisms
be relatively constant in efficiency and that a sufficient number of dimensions has
been considered.
Several questions are immediately apparent:
1 . Is there in fact such a surface?
2. If the surface does exist, can we map significant physiological or ecological
properties on it?
3. What are the theoretical implications of positive answers to 1 and 2 with
reference to hydra and other organisms?
These questions will be considered in turn, after consideration of our methods.
MATERIALS AND METHODS
The experimental animals were taken from a variety of strains all of which are
being maintained in our laboratory. All of the strains had been in the laboratory
for at least a year prior to the start of the experiments, some as long as ten years.
Green hydra were represented by a small strain collected in the Nissequogue River
on Long Island. Studies on other properties of this strain are discussed in Bossert
and Slobodkin (1983). Hydra americana were from the laboratory of Richard D.
Campbell, as were Hydra cauliculata. Hydra fusca were from Lago Maggiore, Italy.
There was also a very large strain ("Connetquot") from the Connetquot River, Long
Island and a slightly smaller brown hydra ("5-tentacle") from the Carmans River,
Long Island. These animals are available to investigators on request. We have not
attempted rigorous identification of the wild caught strains, since our experiments
refer to the genus Hydra in its entirety. These strains have persisted in having
different sizes and slightly different coloration over many months of culture under
closely similar conditions.
M solution was used for all stocks and experimental animals (Lenhoff and Brown,
1970). The animals were maintained in controlled temperature chambers under
constant overhead illumination at seventeen degrees centigrade. The experimental
animals were fed ad lib with Anemia nauplii. The Anemia nauplii had been hatched
within twenty-four hours and washed briefly in distilled water, before being sus-
pended in M solution and offered to the hydra.
Experimental hydra were maintained as isolated individuals in the laboratory
for periods of from three weeks to two months. They were offered large numbers
of Anemia nauplii as food and after each feeding the number of nauplii actually
ingested was determined by shining light through the gastric region and counting
them in the gastric cavity. The feeding counts were made after the animals had
stopped "swallowing" but before digestion made counting too difficult.
To estimate size of the hydra, the animals were photographed. All photographs
were taken prior to feeding. The photographic procedure was constant and standard
throughout. The single lens reflex camera was on a permanent frame used for this
purpose only. Focus and enlargement were not changed. Standards were photo-
graphed at each photography session to check on the possibility of inadvertant
rearrangements of the apparatus. The length and area were measured using a bright-
ness thresholding algorithm on computer digitized video images of photographic
308
L. B. SLOBODKIN AND K. DUNN
negatives, which is part of an optical measurement computing program, SPOT,
under development by Rohlf and Person, at Stony Brook.
One source of error in this procedure is that moribund tissue at the pedal end
of a hydra need not be sloughed off immediately. A sausage-like post-peduncle may
persist for a while and then drop off quite suddenly. This occurs most often in the
larger species.
While every effort was made to standardize the state of contraction of the hydra
during the photography, there was the possibility of a major source of error being
introduced by differences in contractile state. We assumed that each hydra was a
constant volume cylinder lying on its side so that projected area would be a function
of length. The relevant equation is:
In A == '/2(ln V + In 4 - In TT + In 1)
in which 1 is the observed length, A is projected area, and V is the constant volume.
When a series of photographs of hydra individuals in different contractile states
was made it was found that the curves of area against length for individual animals
of all species conform to this simple equation. The average of the slopes of the
relation between log length and log projected area for eleven animals of three species
was .514 with standard deviation ± .0110 with an average coefficient of determi-
nation of .95 ± .0122. With the apparent verification of the above model, volume
can be computed. This measure of volume, being demonstrably independent of
contractile state, was taken as our size estimate. Mass, determined as freeze-dried
weight, was found to correlate well with calculated volume (Fig. 2).
Budding rate could be immediately determined, since animals were maintained
in isolation. Ambiguity was avoided by counting buds after they have dropped off
their mothers and using an average budding rate over the period of observation.
Other times of origin of buds, as for example, appearance of first tentacles etc., could
have been considered without changing things, since there is effectively no death
of buds. Any buds that were on animals at the initiation of the period of experimental
150
- 100
co
CO
o
LU
cr
co
50
I I T I
0 / R ' 0.83
0.5
ESTIMATED VOLUME
1.0
FIGURE 2. The relation between estimated volume, based on a single photograph for each animal,
and freeze dried mass of 28 hydra weighed individually on a Cahn Electronic Microbalance.
EVOLUTIONARY CONSTRAINTS IN HYDRA 309
observation were not included in the bud counts, but buds that were attached at
the time of termination of the experiment were included.
Floating and sexuality were noted for one subset of experimental animals.
Notice that the animals had all been taken from stock cultures, so that there
was a non-equilibrated transition period during the early portion of their history in
isolation. Also we have no guarantee that all animals equilibrated during the ob-
servation period. One set of animals was maintained under experimental conditions
for ten days and the remainder for twenty-one days prior to the first collecting of
data. Rather than arbitrarily omitting data, all of the data were used, and the non-
equilibrium may be assumed to have added to our variance.
RESULTS
We now return to the questions listed in the Introduction.
1 . Is there in fact a surface of the sort indicated?
The series of measurements for each hydra produced a single point (measured
as the triplet; mean body size, mean budding rate, and mean feeding rate). It was
found that the green hydra were discordant, having excessively high budding rate
and body sizes per unit food consumption, in comparison with the brown species.
Since it is known (Muscatine, 1961; Slobodkin, 1964;Stiven, 196 5;) that green hydra
can receive approximately three times as much energy from their algae as from
animal food, the measured food consumption of the green hydra was multiplied by
four and the product was used as our estimate of their food consumption. A similar
procedure was followed in Slobodkin, ( 1 964). This is obviously a first approximation,
and may also have introduced variance. We are now performing experiments de-
signed to estimate the fraction of energy that actually comes from algae under
different circumstances. (See also Bossert and Slobodkin, 1983.)
The data for each animal are presented in Table I, and as a three dimensional
graph in Figure 3.
The complete set of points using a total of 39 hydra of six strains was tested for
fit to a two-dimensional surface embedded in three space.
While the shape of the surface will prove of importance (cf. Gatto, Matessi and
Slobodkin, in prep.), our immediate concern is the presence or absence of a surface,
rather than its precise shape.
Consider a resource budget consisting of a set of mutually exclusive ways of
expending resources, which sum to the total resources income. In our case, bud
production and body size maintenance are the result of these expenditures. The
resources expended for bud production plus those expended in body maintenance
are assumed to equal total resource income. If different strains of hydra apportion
resources differently between these expenditures, but the efficiencies are constant
between strains (i.e., body size per unit resource expended for body maintenance
and buds per unit resource alloted to bud production), then the measurements of
individual hydra will generate a monotonic surface in the space whose dimensions
consist of an axis for resource income and an axis for each of the modes of expen-
diture. The term "monotonic surface" requires definition in the present context.
The intuitive meaning is of a surface with neither hills nor valleys. In three dimen-
sional space a monotonic surface, in our sense, is one in which the locus of the
points of intersection between the surface itself and any flat plane that intersects the
axis of resource income will be a monotonic curve passing through the origin.
If the surface in Figure 1 is a monotonic plane folded in three space, rankit
310
L. B. SLOBODKIN AND K. DUNN
^
VO VD vo ^3" OO 01 O 1 — ^ ^J" O VO ol O VD O ^ VO ?*^ ^t O *^ OO O ol O
OO I — • O] t~~~
00
q
O Ol — r«")
TT o r~-
JQ
UJ
OOOONONO-OlTf — — i/IOOOOON^t-VO OONOO — 011-OO ON —
— m — —
o
oo O oo O
00 — 04
—
PU
^
(/5
X
LLJ
OOOOOOOO — — — — OOOOOOOOOimONOl OOOOOOOO — — — — OOOO
00 — — —
01
oo
00 — — —
^^r;
2
Q
!ir
Ol fN fN (N fN fN <N fS fN fS fN fN fN fN -™ fN (N (N fN fN fN (N fN fN C^ <N
<N (N PN! CN)
04
O4 O4 O4 O4
04 04 04
OJ
LU
^
UJ
Lu
s
Q
Z
UJ
O
t-sj ~ — —
04 — — —
UJ
C/>
X
r£
's-
—
OOOO OOOO — u-ioioo
r-j — ri
r- 04 04
_ m I—
04
~
<
<
_.___-.
— —
—
J~,
0.
o
cc
C
Lt
Q
LU
C/3
_i
^
>
p
OOOO OfNO— r^ — ON —
O 0 O
01 O —
— vO OO
01
LU
%
X
CO
tA
0
Z
oo
vO r*^ 04
<r
S
i/~i ^^ \o r — • ^2 oo f^ ~~ o^ oo f^i ^ Vs! •/"> ^o (-^ ^D r^ r^ r^- ~~ oo r^i t^ to oo
vo O F- r~-
•£.
•cr
04 O t- 0
— — I/"} —
04 VO ^
>n 04 oo
O
»f—
s
fN <"*"* r*^ C4 f^-l — fN fS (N <"^ >/^ fN r*^ to — (N <"*~i fN <N ro fS — <N — r*"* r*"i
rr, (N — —
q
— 04 — 04
O O —
<
—
LU
C2S;?SS?25o ^S-S^o??SS ^£8?S2£^ -^
ON "^ r~~ ^^
04 O vO O>
~c
OO ^^ r''! ON
0 0 rq
vO
i
LU
sOrOr-r-OO^'-t^O TfO-^t^fN — — — ro^i^r-^DTj-(N<N ^m
vo <N vo >n
q
O-l — >n —
V-> •* ON
•*
2
m
^~
<^*
^i
Tj1 r^i ^ ro — t^ ^o r"~J r~^ oo ro tO ^C to fN oo ~^ ^f t*~* oo ^^ tO *^ ~~ ~~ ^D
f^t — ^^ i —
Ol
— vO ON O4
00 04 00
—
*—•
- . ',
rvj
tO -^ to */~) 0s ^0 to to oo r~~- ^c ^ ^o ^ ^o oo ~^ ^o ~^ ^^ ^-t" ^D ^^ r~~- r~~- oo
•^ — 01 (N
m
VO ON OO ON
i/-j so ^D
•/"}
^>
—
En
^- ^- Tj- ^- rO <N tO tO tO tO Tf fO — CN — •" C^l ^" ^" to "^ i/^ »/^ */^ CO f~~~ OO
oo vo ^3 |n
*n
TI- 10 r~ •*
r*^ OJ T}-
ro
(/5
LU
C
O
C
•nr-vO«->vOvOvOvO r-r~OOOOTr<NTr<-) OO^OOvOvOvOvOvO OOOO
r- vo vo vo
l~-
OO vo vo vO
^ vo vo
VO
*
X
n.
0
£
fNfNrNfSONONONON fNfNfNfNOOTt^.^ rNfNfNrMONONONON fN<N
04 ON ON ON
fN
VO
O4 ON ON ON
ON ON ON
0,
*
Q
_]
1
<
C
D
_rsrr)^.1/^vOr,oo —.rsirnTfinvor-oo — (Nr'-i-a-mvor-oo — rs
f"} ^f *O NO
,
04 m T «->
— 04 m
^
X
O
< UJ H
STRAIN
< o ^ B
p B 1 %
^^ (*™V ,**] Mj
^ % LU 5
•J r*^ *-•
<
«
ENTACLE
< S2 S O
U Z < U
E
•/i
EVOLUTIONARY CONSTRAINTS IN HYDRA
311
SIZE
BUDS
FOOD
FIGURE 3. Rankit transformed data from Table I, plotted as a three dimensional graph with axes
food consumption, body size and budding rate. Both a three dimensional and one dimensional repre-
sentation of these data can be rejected by Bartlett's test of sphericity at P > .001. (Key to symbols — box:
H. americana, hourglass: Nissequogue strain, triangle: H fusca Italian strain, cross: H. cauliculata, dia-
mond: Connetquot strain, circle: 5-tentacle strain).
transformation will project the data onto a flat plane. Principal components analysis
and associated tests of significance can then be used to test the fit of the transformed
data to a two dimensional surface. The data were therefore converted to rankits
(Rohlf and Sokal, 1969). The rankit transformation discards information about the
particular shape of the curves relating food, budding rate and size. This transforms
any monotonic curve to a plane. The use of rankit transformation in facilitating
statistical tests of energy budget data is being addressed, in detail, elsewhere (War-
tenburg, Slobodkin and Dunn, in prep.). We assume nothing about the shape of
Figure 1 other than its monotonicity.
Principal components for the rankit converted data were calculated using the
NTSYS program of Rohlf et al. (1982). The first, second, and third eigenvalues and
their power to explain variance were 1.627, 1.037, and .3354 with elimination of
54%, 35%, and 1 1%, respectively, of the data variance.
The rankit data meet the assumptions for Bartlett's Test for Sphericity (Bartlett,
1950; Green and Douglas Carroll, 1978). This test permits assignment of a prob-
ability value to the null hypotheses that the data in Figure 3 are adequately rep-
resented by a spherical cloud of points (i.e., require three dimensions), or by a cigar
shaped cloud varying around a line (i.e., require only one dimension). Both of these
hypotheses can be rejected at P < .00 1 . That is, we can assert that a three dimensional
representation is not necessary, while a one dimensional representation is inade-
quate, hence we conclude that two dimensions are an appropriate representation.
312
L. B. SLOBODKIN AND K. DUNN
Departure from three dimensions was checked by Monte Carlo simulation in
which the food income, size estimate and budding rate, expressed as rankits, for
each hydra were randomized among hydra. The distribution of the resultant triplets
was then tested. This was done one hundred times, and the actual, non-randomized
data was found to more closely approximate a plane surface than any of these one
hundred replicates. We conclude that, in fact, the surface exists.
All of the animals in our experiments were sufficiently well fed to permit budding.
We believe that we were in a relatively narrow range of the possible feeding rates.
While we intend to study more fully the actual shape of the constraint surface, the
region for which we now have data shows a significant correlation between food
consumption and budding rate, but not between food consumption and body size.
We suggest that hydra more readily adjust their budding rate than their body size
to food consumption, once they are sufficiently well fed to bud at all. Otto and
Campbell (1977) and Hecker (1978) found that body size does respond to feeding
rate at high food levels, and also reported that, at very high food intake rates, hydra
may lose the capacity to maintain a steady state in size.
2. Does position on the surface matter to the physiology of the animals?
Slobodkin (1979) suggested that the surface presented in Figure 1 would be
divisible into regions, within which hydra would have particular properties. This
hypothesis is presented graphically in Figure 4.
At low levels of food intake not only are budding rate and body size reduced
but also particular physiological responses are found (Fig. 5).
Large individuals float more readily (see Lomnicki and Slobodkin, 1966). Sex-
uality was found predominantly in intermediate sized, low food level, brown ani-
mals. The green hydra were in general smaller than the brown.
FIGURE 4. Localizations of physiological and behavioral properties on the surface of Figure 1 as
hypothesized by Slobodkin (1979).
EVOLUTIONARY CONSTRAINTS IN HYDRA 313
In short, position on the surface is related to physiological state, as predicted by
Slobodkin (1979). Obviously, the ecological relationships of a floating animal are
different than those of a settled animal in many ways. We have thus demonstrated
an affirmative answer to the question of whether position on the surface matters
both physiologically and ecologically.
DISCUSSION
The third question stated in the introduction, (i.e., the possible significance of
these results), will now be addressed. The results will be discussed in four contexts —
the idea of constraints in evolution; the relation between constraint systems and
resource budgets; the search for other, similar, constraint systems; and finally the
implications of our findings for the natural history of hydra.
Evolutionary constraint systems
Clutton-Brock and Harvey (1979), in their review of constraint systems, distin-
guish between "generic constraints" and "evolutionary constraints". Generic con-
straints are those sets of properties which are found to be correlated with physio-
logical or ecological categorizations of organisms, without being, necessarily,
confined to single taxonomic categories. For example, herbivory may imply the co-
occurrence of one set of properties, while carnivory implies another. All homeo-
therms may share certain characteristics, all poikilotherms another. Evolutionary
constraints, in contrast, are inferred from comparisons between members of different
subcategories within a larger taxonomic category. We consider that we have dem-
onstrated an evolutionary constraint system in hydra. Note, however, that both
Clutton-Brock and Harvey (1979) and Gatto, Mattessi and Slobodkin (in prep.)
discuss the fact that an apparent surface may actually consist of a series of separate
surfaces, each perhaps representing a genotype or species, that resemble a single
surface on the generic level in much the same way that the individual slats of a
"Venetian blind" are seen as one surface from across the room. Our data are in-
determinate on this issue.
Individual hydra can equilibrate at various locations on the surface as a con-
sequence of environmental factors. The fact that, at least within the statistical limits
of our data, different species share the same surface, leads us to believe that mi-
croevolutionary changes in hydra would tend to move them about on the surface
rather than orthogonal to it.
Gould (1980) has presented the metaphor of objects resting on a surface to help
explain what is meant by an evolutionary constraint. In this metaphoric context,
denial of the existence of constraints on evolutionary direction is taken as imagining
a ball rolling on a flat plane. This is taken by Gould and Lewontin (1979) as the
image underlying what they refer to as the "Adaptationist Programme." How far
the ball rolls depends only on the force with which it is pushed, not on the direction.
Gould goes on to suggest that evolutionary changes for any particular kind of or-
ganism may be more restricted in their direction, resembling a polygonal solid,
whose motion will depend on both force and direction of the propulsive forces, as
well as on which of its faces it is resting. An actual polygonal solid cannot roll, but
can be more readily tipped over in certain directions. In a sense we have explored
this metaphor. We believe that on experimental and theoretical grounds we have
demonstrated explicitly a set of ecological and physiological constraints on the genus
Hydra. On the basis of this demonstration we suggest adding to Gould's metaphor
314
L. B. SLOBODKIN AND K. DUNN
IO
M
CD
DAYS FLOATING OUT OF 12 DAYS OBSERVED
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
..
x Cauliculata
• Nissequogue
A Americana
i Connetquot
• Fusca
o Rye Tentacle
KID
02)
0(7)
nnn
0(2)
X(0)
X(0)
• (0)
0(5) X(0)
• (0)
.(0)
1 234567
MEAN FEEDING RATE (ARTEMIA/DAY)
8
FIGURE 5a. The relation between fraction of days during which animals were floating, body size
estimated photographically, and mean number of Anemia nauplii consumed. The P value associated
with this distribution arising at random was determined by the Spearman Rank Correlation Coefficient
of the order of the points when projected onto a line with a negative forty five degree slope and their
order in floating rate. P was less than .001.
the image of a non-spherical solid with rounded edges, or perhaps no clear edges
at all, which is capable of rolling easily in only certain directions, and must be
toppled over if it is to roll in other ways. The mental image is that of the conical
egg of the murre, which rolls in tight curves, thereby avoiding falling off ledges
(Heinroth and Heinroth, 1958).
Constraint systems as consequences of resource budgets
There is an obvious connection between analyses of budgets and constraints and
discussions of ecological and evolutionary "strategies." Various theories of evolu-
tionary strategy build on the assumption that organisms are constrained so that their
capacity to do a particular thing or have a particular property carries a "cost" which
interferes to some degree with their capacity to do another thing or have another
property. This approach is recently summarized by Townsend and Calow (1981)
and McCleery (1978).
The analyses of energy, material, and time budgets for individuals and for pop-
ulations demonstrate that there are restrictions on the present activities of organisms.
Energy used for running can not be used for growth. Material used for seeds can
EVOLUTIONARY CONSTRAINTS IN HYDRA
DAYS SEXUAL OUT OF 12 DAYS OBSERVED
315
0.9
-
X Couliculota
0.8
B(0) • Nissequogue
A Americana
i Connetquot
0.7
• Fusca
o Five Tentacle
xfico
t*-^.
0.6
(2)«A(9) I"°'
o
E
(7)
E
(10)3^
~~~^
0.5
• (1) A""
UJ
0(8)
M
O)
0.4
X(0)
odi) A(n>
UJ
O(12)
X(0)
0.3
X(0)
• <U
0(6) X(0)
• (2)
0.2
-
• (0)
0.1
-
n
1 1 1 1 1 1 1 1
234567
MEAN FEEDING RATE (ARTEMIA/DAY)
8
FIGURE 5b. The relation between the occurrence of gonads, body size and food consumption. The
values are not randomly distributed (.05 > P > .01) nor are they monotonically distributed along the
axes of food consumption and body size. Sexuality occurs most often at intermediate levels of food and
body size.
not make leaves. Time spent preening can not be used for feeding, etc. A sufficient
number of inverse correlations have been demonstrated between properties of or-
ganisms that excellent summaries now appear in elementary texts [for example,
Begon and Mortimer's chapter on "Life history strategies" (1981)].
The observed constraint surface in hydra, which would be included in the cat-
egory of "evolutionary constraints" of Clutton-Brock and Harvey (1979), may be
considered an elementary consequence of energy budget considerations.
If any two activities or properties of a single organism require sufficiently large
allotments or utilization rates of energy or some other resource, development of one
of these properties will tend to inhibit or reduce the development of the other. If
we consider several organisms, all similar in developmental and physiological po-
tentialities, but differing somewhat as a result of environmental differences, then
those individuals that have enhanced one of our hypothetical properties will to some
degree have reduced the other. It is possible for organisms to have the same, or very
similar anatomy and developmental properties, but to differ genetically in how they
partition available resources between the two properties. In particular, this applies
to organisms which are of different, but closely related species. In fact, the term
"closely" in the previous sentence may be denned by the existence of such simi-
316 L. B. SLOBODKIN AND K. DUNN
larities. Such pairs of properties meet Clutton-Brock and Harvey's criterion for being
under evolutionary constraint. Only if the development of the pair of properties use
different resources, or use resources very sparingly, can both properties be increased
in the same organisms and under the same circumstances.
We are not confined to considering only pairs of properties. As larger sets of
properties are taken into consideration the development of all the properties of the
set is more and more likely to constitute a significant fraction of the organisms'
resources, and constraints are more likely to become apparent over the set as a
whole, whether they are in evidence for any pair of properties. Notice that any
property that is found to always be enhanced as some other property is increased
is likely to be artifactual, in the sense of Gould and Lewontin's (1979) discussion
of the primate chin.
Assume that the degree of development of the properties in question can be
measured in the same units as the resource income. If the income of energy, or any
other resource, to an individual organism or population of organisms is known,
then for any set of properties which are under physiological constraint, it is possible
to construct a budget by assuming that the total supply of the resource in question
allotted to the set of behaviors is equal to the measured income of that resource.
The usual technique would be by multi-dimensional regression analysis. Examples
and discussions of this procedure in this context may be found in Slobodkin (1980)
and McFarland (1976).
The two properties, body size and budding rate in hydra both require the pro-
duction and maintenance of tissue. The tissues of a bud are not noticeably different
from those of its mother, and our data indicate that evolutionary constraint exists
on this pair of properties. No hydra can simultaneously increase both its body size
and budding rate above the constraint surface unless it can make a fundamental
improvement in the efficiency of its biochemical processes. This apparently has not
been possible. In this sense, the area above the constraint surface is free of hydra
due to thermodynamic limitations.
Notice that green hydra have energy resources that are unavailable to brown
ones. We estimated the amount of energy supplied by the algae, and this permitted
us to consider green and brown hydra to be on the same surface. If we think in
terms of a constraint set by animal food income, then the green hydra must be
thought of as being above the constraint surface. We expect that there exists a
constraint surface for all species of green hydra. In this sense, evolutionary loss or
gain of the capacity to maintain symbiosis with algae would constitute a macro-
evolutionary step for hydra.
The area beneath the surface is kept free of hydra by evolutionary considerations.
Conceivably some hydra with a low capacity to maintain tissue and at the same
time a low budding capacity could perhaps have some kind of selective edge. For
example, hydra are unable to eat certain kinds of cladocera. One of these, Anchis-
tropus, actually feeds on hydra (Hyman, 1926; Borg, 1935; Griffing, 1965; Personal
Observation, L.S.). If Anchistropus were to become extremely common, we might
expect that a strain of hydra that was immune to its attack, or even capable of
feeding on it, might have a selective advantage. Under these circumstances we might
expect that efficiency of growth and reproduction would be evolutionarily unim-
portant.
We have some evidence that aposymbiotic Hydra viridis might fall below the
observed surface (Stiven, 1965; Pardy and Dieckmann, 1975; but see Cantor and
Rahat, 1982). There is no evidence that aposymbiotic Hydra viridis occur in nature.
EVOLUTIONARY CONSTRAINTS IN HYDRA 317
The search for other constraint systems
Raup and Stanley (1971) studying snail shell evolution, Hutchinson (1968) for
Bdelloid rotifers, and Porter (1976) for some of the Scleractinian corals, among
others, all have evidence for restrictions on evolutionary possibilities. Raup and
Stanley present their data in the space denned by the mathematical representation
for a coiling shell, which contains three parameters. Both Hutchinson and Porter
present their data as clouds in two dimensions. On purely formal grounds it is
understood that often data which appear as points on a surface of a given dimen-
sionality will, when projected onto a space of lower dimension, appear as a cloud.
Conversely, we believe that many of the taxonomically restricted scatter diagrams
published in ecological literature will appear as surfaces if third or higher dimensions
are added, and that some of these surfaces will permit mapping of particular phys-
iological or behavioral properties. We expect that, while which, and how many,
measurement axes will define a surface for a particular group of organisms is not
obvious, all such sets of axes will share certain properties. We believe that they all
will be related to resource budgets. One axis will consist of some resource and the
others will be different ways in which that resource is expended. This will guarantee
suitable concavity and monotonicity of the surfaces.
Principal components analyses test dimensionality. Our hypothesis, presented
in Figure 1, assumes monotonic curves. Fortunately the rankit transformation maps
monotonic curves onto flat planes, permitting our use of the Bartlett's test for spher-
icity. For reasons presented above, we expect that most evolutionary constraint
surfaces will also project as monotonic curves in a space of sufficient dimensionality.
In general, sufficient dimensionality will have been achieved in a constraint surface
when rankit transformed data can be significantly explained by a number of com-
ponents one less than the total number of measured variables. Principle components
analysis, combined with either special tests, of the sort we used, or Monte Carlo
simulations, may provide probability estimates for measuring the quality of the
surfaces.
The natural history of hydra
Note that Figure 1 is drawn as if the entire surface were available for hydra. We
believe that the edges will tend not to be occupied by actual organisms. This is due
to the fact that the particular environmental problems which arise for hydra at
various points on the surface are likely to differ.
Excessively large hydra have very high maintenance costs, so that budding can
only occur if the food supply is very abundant. The capacity to float may permit
these larger hydra to survive in an unpredictable environment. They are capable of
surviving for an extended period without food. During this period floating animals
may encounter richer concentrations of prey. Being excessively small probably nar-
rows the range of possible food items and also narrows the time available for a hydra
to starve between meals and still be large enough to capture prey. Floating until
new feeding grounds are encountered does not seem as useful for small hydra, since
not only is their ability to survive starvation while floating limited, but their range
of acceptable animal foods is restricted. Symbiotic algae may serve small hydra in
essentially the same way that floating serves large ones, since the symbionts extend
the period that these animals can survive between feedings. Bossert and Slobodkin
(1983), Thorington and Margulis (1980), and others (cf. Hyman, 1940; Kaenev,
318 L. B. SLOBODKIN AND K. DUNN
1969) have shown that at least the largest of the green hydra may, under some
circumstances, suffer damage from their algal symbionts. That is, at particular re-
gions on this surface of constraints special ecological problems arise. Particular
mechanisms for solving these problems have evolved. These include symbiotic algae
supplementing the food supply and the capacity to float to richer food areas.
We believe that environmental changes may distort or rotate the constraint
surface. We know that those species of hydra so far examined have a lower budding
rate and larger body size at lower temperatures (Hecker, 1976) and that floating rate
is sensitive to temperature (Slobodkin, 1 979).
In hydra the empirical evidence suggests that a reasonably complete and explicit
description of the constraints of both physiological and evolutionary responses con-
sists of a surface embedded in a three dimensional space, on which physiological
and behavioral properties may be mapped. We suggest the possibility that similar
descriptions, consisting of a mapped surface in a minimum of three dimensions
may exist in other groups of closely related species. It seems likely that physiological,
developmental, or evolutionary alterations which result in movement on such a
surface occur more frequently than alterations which successfully permit changes
which are orthogonal to the surface. This may relate to the problems of the contrast
between micro- and macro-evolution.
Constraint surfaces of this type may be viewed as consequences of resource
budget considerations in groups of organisms that share most of their developmental
and anatomical properties, but differ in their "Policy" (in the sense of Gatto, et al.,
in prep) of apportioning resources to different uses. There is an intimate connection
between evolutionary constraint surfaces, optimality theory, and resource budgets.
ACKNOWLEDGMENTS
These studies were supported by grants from the Biological Sciences Division
of the U. S. National Aeronautics and Space Agency, the Mobil Oil Foundation,
and the Italian National Research Council. Profs. R. Armstrong, L. Ginzberg, J.
Rohlf, and H. Lyman of Stony Brook provided criticism, encouragement, and ad-
vice. Scott Person, Dan Wartenburg, and Patricia Bossert, graduate students in the
Ecology and Evolution Department each provided their own kind of help. Several
undergraduate students were invaluable. These include Jay Fader, who did most of
the photography, John Le Guyader and Heidi Chapnick who participated in the
experimental work and who, along with Louise Holbrook helped in the tedious
work of maintaining stocks. The animals from Central Park were provided by Mike
Lewandowsky. R. De Bernardi of Pallanza, Italy, facilitated the collection of Hydra
fusca. The 1st. di Genetica, Biochemica, ed Evolutionistica, C.N.R. at Pavia provided
hospitality and an opportunity for discussion. This is contribution number 452 from
the Ecology and Evolution Program, SUNY, Stony Brook.
LITERATURE CITED
BARTLETT, M. S. 1950. Tests of significance of factor analysis. Brit. J. Psvchol. (Statistical Section) 3:
77-85.
BATESON, G. 1963. The role of somatic change in evolution. Evolution 7: 529-539.
BEGON, M., AND M. MORTIMER. 1981. Population Ecology: A Unified Study of Plants and Animals.
Sinauer Associates. Sunderland, MA. 200 pp.
BORG, F. 1935. Zur kentniss der Cladoceran-gattung Anchistropus. Zoo/. Bidr. Upps. 15: 289-330.
EVOLUTIONARY CONSTRAINTS IN HYDRA 319
BOSSERT, P., AND L. B. SLOBODKIN. 1983. Thee effect of fast and regeneration in light vs. dark on
regulation in hydra-algal symbiosis. Biol. Bull. 164: 396-405.
CANTOR, M. H., AND M. RAHAT. 1982. Regulation of respiration and photosynthesis in Hydra viridis
and in its separate cosymbionts: Effect of nutrients. Physiol. Zoo/. 55: 281-288.
CLUTTON-BROCK, T. H., and P. H. HARVEY. 1979. Comparison and adaptation. Proc. R. Sac. Land.
Ser. B 205: 547-565.
GATTO, M., C. MATESSI, AND L. B. SLOBODKIN. (In prep.) Evolution within ecological and physiological
constraints: Explicit predictions for hydra.
GOULD, S. J. 1980. The evolutionary biology of constraint. Daedalus, Proc. Am. Acad. Arts Sci. 109:
39-52.
GOULD, S. J., AND R. C. LEWONTIN. 1979. The spandrels of San Marco and the Panglossian paradigm:
A critique of the adaptationist programme. Proc. Roy. Soc.. London. Ser. B 205: 581-598.
GREEN, P. E., AND J. DOUGLAS CARROL. 1978. Pp. 361-364 in Analyzing Multivariate Data. Dryden
Press. Hinsdale, IL.
GRIPPING, T. 1965. Dynamics and Energetics of Brown Hydra. Ph.D. Thesis in the Department of
Zoology, University of Michigan, Ann Arbor.
GURKEWITZ, S., M. CHOW, AND R. D. CAMPBELL. 1980. Hydra size and budding rate: influence of
feeding. Int. J. Invert. Reprod. 2: 199-201.
HECKER, B. 1978. Adaptive Responses in Hydra oligactis. Ph.D. Thesis in the Department of Ecology
and Evolution, S.U.N.Y., Stony Brook.
HECKER, B., AND L. B. SLOBODKIN, 1976. Responses of Hydra oligactis to temperature and feeding rate.
Pp. 175-183 in Coelenterate Ecology and Behaviour, G. O. Mackie, ed. Plenum Press, New
York.
HEINROTH, O., AND K. HEINROTH. 1958. The Birds. University of Michigan Press, Ann Arbor, MI.
HUTCHINSON, G. E. 1968. When are species necessary? Pp. 177-186 in Population Biology and Evolution,
R. C. Lewontin, ed. Syracuse University Press.
H YMAN, L. 1 926. Note on the destruction of hydra by a Chydorid Cladoceran, Anchistropus minor Birge.
Trans. Am. Microsc. Soc. 45: 298-301.
HYMAN, L. 1940. The Invertebrates: Protozoa through Ctenophora. McGraw-Hill Book Company, NY
726 pp.
KANAEV, I. I. 1969. Hydra: Essays on the Biology of Freshwater Polyps. Translated by E. T. Burrows
and H. M. Lenhoff; H. M. Lenhoff, editor and publisher. Miami. 452 pp.
LENHOFF, H. M., AND R. D. BROWN. 1970. Mass culture of hydra: an improved method and its appli-
cation to other aquatic invertebrates. Lab. Anim. 4: 139-154.
LOMNICKI, A., AND L. B. SLOBODKIN. 1966. Floating in Hydra littoralis. Ecology. 47: 881-889.
McFARLAND, D. J. 1976. Form and function in the temporal organization of behaviour. Pp. 55-93 in
Growing Points in Ethology, P. G. Bateson and R. A. Hinde, eds. Cambridge University Press.
McCLEERY, R. H. 1978. Optimal behaviour sequences and decision making. Pp. 377-410 in Behavioural
Ecology, an Evolutionary Approach, J. R. Krebs and N. B. Davis, eds. Sinauer Associates.
Sunderland, MA.
MUSCATINE, L. 1961. Symbiosis in marine and freshwater coelenterates. Pp. 255-264 in The Biology
of Hydra and Some Other Coelenterates. H. M. Lenhoff and W. F. Loomis, eds. University of
Miami Press, Coral Gables.
OTTO, J. J., AND R. D. CAMPBELL. 1977. Tissue economics of hydra: regulation of cell cycle, animal
size, and development by controlled feeding rates. J. Cell Sci. 28: 1 17-132.
PARDY, R. L., AND C. DIECKMANN. 1975. Oxygen consumption in the symbiotic hydra Hydra viridis.
J Exp. Zoo/. 194: 373-378.
PLOTKIN, H. C., AND F. J. ODLING-SMEE. 1981. A multiple level model of evolution and its implications
for sociobiology. The Behavioral and Brain Sciences 4: 225-268.
PORTER, J. W., 1976. Autotrophy, heterotrophy, and resource partitioning in Caribbean reef-building
corals. Am. Nat. 110: 731-742.
RALIP, D. M., AND S. M. STANLEY. 1971. Principles of Paleontology W. H. Freeman and Co., San
Francisco. 388 pp.
ROHLF, F. J., AND R. R. SOKAL. 1969. Statistical Tables. W. H. Freeman and Co., San Francisco. 253
pp.
ROHLF, F. J., J. KISHPAUGH, AND D. KIRK. 1 982. NT-SYS, numerical taxonomy of multivariate statistical
programs. Technical Report, State University of New York, Stony Brook.
SLOBODKIN, L. 1980. Growth and Regulation of Animal Populations. Second Enlarged Ed. Dover Pub-
lications Inc., New York. 234 pp.
320 L. B. SLOBODKIN AND K. DUNN
SLOBODKJN, L. 1964. Experimental populations of Hydrida. In British Ecological Society Jubilee Sym-
posium, A. Macfadyen and P. J. Newbould, eds. /. Anim. Ecol. 33(Suppl.): 131-148.
SLOBODKJN, L. B. 1979. Problems of ecological description: The adaptive response surface of hydra. In
Proc. Svmp. "Biological and Mathematical Aspects of Population Dynamics, " R. de Bernardi,
ed. Mem. 1st. Idrobiol. Suppl. 37: 75-93.
SLOBODKIN, L. B., AND A. RAPOPORT. 1974. An optimal strategy of evolution. Q. Rev. Biol. 49: 181-
200.
STIVEN, A. E. 1965. The relationship between size, budding rate, and growth efficiency in three species
of hydra. Res. Popul. Ecol. 1: 1-15.
THORINGTON, G., AND L. MARGULIS. 1981. Hydra viridis; Transfer of metabolites between hydra and
symbiotic algae. Biol. Bull. 160: 175-188.
TOWNSEND, C. R., AND P. CALOW, eds. 1981. Physiological Ecology: An Evolutionary Approach to
Resource Use. Sinauer Associates, 393 pp.
WRIGHT, S. 1932. The roles of mutation, inbreeding, crossbreeding and selection in evolution. Proc.
Sixth Intern. Cong. Genetics. 1: 356-366.
Reference: Biol. Bull. 165: 321-329. (August, 1983)
ECDYSTEROID TITERS DURING THE MOLT CYCLE OF THE
BLUE CRAB RESEMBLE THOSE OF OTHER CRUSTACEA
CYNTHIA SOUMOFF AND DOROTHY M. SKINNER
University of Tennessee-Oak Ridge Graduate School of Biomedical Sciences and Biology Division,
Oak Ridge National Laboratory, Oak Ridge, Tennessee 37830
ABSTRACT
Callinectes sapidus is the only true crab (brachyuran) whose pattern of ecdyste-
roid tilers has been described as departing from the pattern seen in other decapods.
While ecdysteroids in other crabs reach a peak just prior to ecdysis, those of C.
sapidus were claimed to reach their maxima after ecdysis. The data reported here
challenge these findings. We have measured ecdysteroids in hemolymph, ovaries,
and whole animal extracts of blue crabs using a radioimmunoassay. In hemolymph
and whole animals, ecdysteroid levels rose during premolt to a maximum at stage
D3. Ecdysteroids declined rapidly from late premolt stage D4 through postmolt stage
A2, increased slightly at postmolt stage B, and returned to low levels where they
remained during intermolt stage C. Ecdysteroid levels in males and immature fe-
males were not significantly different but mature females, having reached a terminal
anecdysis, had significantly lower ecdysteroid levels. Ovaries of mature females ac-
cumulated ecdysteroids during vitellogenesis while the concentration of ecdysteroids
in hemolymph was low.
INTRODUCTION
Ecdysteroids in crustaceans, measured in whole animals or hemolymph, rise
during proecdysis, reach peak levels shortly before ecdysis, and decline to basal levels
before or soon after ecdysis (Spindler et al, 1980; Skinner, in press). This pattern
is consistent with the role of 20-hydroxyecdysone (20HE) in initiating premolt.
When ecdysteroids were examined in female blue crabs Callinectes sapidus, 20HE,
inokosterone, and makisterone A were identified and, surprisingly, the ecdysteroid
peak, consisting principally of 20HE, occurred after ecdysis (Faux et al., 1969). It
was suggested that the hormone peak during postmolt was involved with hardening
of the exoskeleton (Faux et al., 1969). Because of the decline in hormone titers
following ecdysis in the crayfish Orconectes limosus, Willig and Keller (1973) con-
cluded that calcification of exoskeleton was independent of hormonal control.
Until the experiments described here, there has been no investigation of cir-
culating ecdysteroid titers nor of ecdysteroids in individual tissues of C. sapidus.
These are important data since many arthropods regulate ovarian maturation and
embryonic development by sequestering ecdysteroids in the ovaries during the re-
productive stage; regulation of the molt cycle is distinguished by changes in circu-
lating ecdysteroids. Several insects accumulate ecdysteroids in the ovary (Garen et
Received 12 January 1983; accepted 16 May 1983.
By acceptance of this article, the publisher or recipient acknowledges the U. S. Government's right
to retain a nonexclusive, royalty-free license in and to any copyright covering the article.
Research supported by the Office of Health and Environmental Research, U. S. Department of
Energy, under contract W-7405-eng-26 with the Union Carbide Corp. C.S. is a Postdoctoral Investigator
supported by subcontract 3322 from the Biology Division of ORNL to the University of Tennessee.
321
322 C. SOUMOFF AND D. M. SKINNER
al, 1977; Lagueux et ai, 1977; Hoffman et ai, 1980) as do the crabs Carcinus
maenas (Lachaise and Hoffman, 1977) and Acanthonyx lunulatus (Chaix and De
Reggi, 1982). Although Carcinus continues to molt after its reproductive phase,
Acanthonyx and other oxyrhynchans enter a terminal anecdysis (stage C4T; Carlisle,
1957) at the puberty molt. Similarly, Callinectes, a brachyrhyncan, enters terminal
anecdysis after reaching the puberty molt (Churchill, 1919). It was therefore im-
portant to determine ecdysteroid concentrations in both hemolymph and ovaries
of crabs in this terminal anecdysis. To that end we examined the ecdysteroid titers
in hemolymph, ovaries, and whole animals at different stages of the molt cycle using
a radioimmunoassay (RIA; Soumoff et ai, 1981). We compared males and females
to determine whether there were any hormonal differences between sexes and com-
pared sexually immature females which still undergo ecdyses with sexually mature
females that are in a terminal anecdysis.
MATERIALS AND METHODS
Animals
Crabs were collected off the Virginia coast during June and July of the molting
season. They ranged in size from 6.3 cm to 11.4 cm carapace width. Animals
collected in various phases of the molt cycle were staged by the coloration on the
distal segments of the swimming legs (Churchill, 1919) and by the extent of skeletal
resorption at the epimeral suture (Warner, 1977; Passano, 1960). Initially, four stages
were examined: intermolt (C4), early premolt (D, or green crabs), late premolt (D3
or peeler crabs) and postmolt (A,-B2 or soft crabs). A second series of experiments
examined crabs divided into several substages from A, through D4 (see Passano,
1960; Skinner, 1962; Warner, 1977 for descriptions of stages). Mature females,
immature females, and males were distinguished by the characteristic shapes of the
abdomen.
Treatment of biological material
Hemolymph was withdrawn by syringe puncture through the pericardial space,
the arthrodial membrane at the base of a limb, or the mid-joint of a claw. Clotted
hemolymph was disrupted and centrifuged to obtain serum. Aliquots were taken
for radioimmunoassay (RIA) and the remaining serum was pooled by stage and sex.
Ovaries and bursa copulatrix were excised from mature females, blotted dry, and
weighed prior to exhaustive hemolymph removal or hemolymph, bursa, and ovary
removal. Individual tissues or whole animals were homogenized in 75% MeOH and
centrifuged. Pellets were reextracted in 75% methanol and supernatants were evap-
orated under reduced pressure and resuspended in a small volume of 75% methanol.
Samples were examined by RIA.
Radioimmunoassay
Antiserum was that of Soumoff et al. ( 198 1 ) produced from 20-hydroxyecdysone
2-hemisuccinate conjugated to thyroglobulin. [3H]ecdysone (S.A. 50 Ci/mmol or
80 Ci/mmol) was the tracer ligand. 20HE (Simes, Italy) was used as a standard to
estimate ecdysteroid levels. All titers are given as 20HE equivalents, although the
antiserum has different reactivities toward closely related ecdysteroids (Soumoff et
al., 1981). The RIA protocol has been described elsewhere (Chang and
O'Connor, 1979).
ECDYSTEROID TITERS OF BLUE CRABS
323
RESULTS
An initial survey revealed that serum ecdysteroids were at basal levels in inter-
molt crabs, began rising in early premolt crabs, and reached peak liters in late
premolt crabs (Fig. 1A). By postmolt serum titers dropped, but not as low as in-
termolt levels. Males and females showed no statistically significant differences at
any given stage. Variance was greater among males than females and was not related
to size or limb loss. Blue crabs readily autotomize limbs as a result of handling;
most of the animals lost from 1 to 4 limbs while two crabs lost six limbs. Regenerating
limb buds from previously autotomized limbs were small on intermolt crabs but
60-
50-
40-
30-
cn
c
20-
10-
CC
UJ
00
CO
a
Male
Female
(5
)
-
[J
)
15)
(5)
(6)
(5)
T(5)
TW)
a
(4)
Post Inter Eorly Lote Inter Early Lote Post Inter
Pre Pre Pre Pre
MOLT CYCLE STAGE
B
UJ 60~
Male
Female
1 —
CO
85°-
LJ
•
~
40-
(12)
(2)
(10)
30-
Li
-
20-
(2)
T
-
T(4)
10-
ft'' T,fl6>
1 *'[
(12)
r1,
1
r>
-
(9) |(6)
(ID (3) T (3)
°2 C4 D, D? D3
C4 D
MOLT CYCLE STAGE
FIGURE 1 . Serum ecdysteroid levels during the molt cycle in male and female blue crabs collected
in (A) June, 1981 and (B) June, 1982. Values are the means ± standard deviations. Number of animals
assayed are given in parentheses. Hatched bars represent mature females. Ecdysteroids were calculated
as 20HE equivalents.
324 C. SOUMOFF AND D. M. SKINNER
were large on premolt crabs. It has been shown that ecdysteroid liters are elevated
in crabs in advanced stages of limb regeneration (Soumoff and Skinner, 1980).
Multiple autotomy acts as a stimulus to molt (Skinner and Graham, 1970, 1972;
Holland and Skinner, 1976; Mykles and Skinner, 1981) and limb regeneration is
a sign that a crab is in the premolt stage (Emmel, 1906, 1907; Bliss, 1956).
Since the puberty molt is the final molt for females of this species, mature females
are found only in the postmolt and subsequent C4T stages. Although immature
females should be available in all stages of the molt cycle, we were unable to obtain
postmolt immature females during this initial survey. Mature C4T females had lower
serum ecdysteroids than immature intermolt females. The difference was significant
(P < .05) and is probably related to changes in hormone production and metabolism
causing the terminal anecdysis of mature females. In one case an immature female
was assayed in late premolt, completed the molt to maturity overnight, and was
reassayed in postmolt. The premolt ecdysteroid level, 43.4 ng/ml, decreased to 6.7
ng/ml overnight.
A second examination of serum ecdysteroid levels was undertaken during the
next annual molting season (Fig. IB) and the molt cycle stages were defined more
precisely. The observed hormone levels confirmed the data obtained previously (Fig.
1A). Ecdysteroid concentrations rose during the initial stages of premolt, declined
in stage D4 and continued to decline through stage A2. There was a slight rise in
ecdysteroid concentration in stage B, . The apparent rise in stage B2 males was caused
by one exceptionally high value that may have been an artifact. There were no
significant differences between males and females throughout premolt. Mature fe-
males had significantly lower ecdysteroid levels than immature females at stages A2
and C (P < .05) and males at stages A, and C (P < .02). Among thirteen mature
C4T females examined, twelve showed no detectable ecdysteroids and one had a
level of 5 ng/ml. Intermolt juvenile females averaged 7. 1 ng/ml and intermolt males
averaged 1.3 ng/ml.
Some crabs that survived several premolt and postmolt stages in captivity were
sampled in consecutive stages. Figure 2A shows that serum ecdysteroids rose in
individual specimens as they proceeded from stage D, to stage D3. Crabs that were
collected at later premolt stages had rapidly declining serum ecdysteroids (Fig. 2B).
These data illustrate that although there may be wide variations between crabs, a
pattern is maintained within individuals of rising ecdysteroids through stage D3 and
declining ecdysteroids from stage D4 through A2 .
In several species of insects (Luu et al, 1976; Lagueux et al., 1977; Ohnishi et
al, 1977; Bollenbacher et al., 1978) and in the crab C. maenas (Lachaise and
Hoffmann, 1977) reproductively active ovaries contain ecdysteroids which regulate
vitellogenesis (Hagedorn et al., 1975; Handler and Postlethwait, 1978) and embry-
onic development (Hoffmann et al., 1980). We examined the ecdysteroid concen-
tration in ovaries of mature female blue crabs to determine whether they stored
significant amounts of ecdysteroids. As a control tissue we examined the bursa
copulatrix, the storage sacs for sperm introduced during copulation.
The reproductive stages were determined according to criteria which distinguish
changes in the gross appearance of the ovaries (Hard, 1942). Stage I describes crabs
immediately following the puberty molt when ovaries are small. Stage II describes
the period during which the ovary enlarges and becomes orange as vitellogenesis
progresses. Stage III describes the mature ovary which is very large and bright orange.
The ecdysteroid content of ovaries of C. sapidus increased as vitellogenesis pro-
gressed (Table I) although ecdysteroid concentration per unit weight declined 2.5-
ECDYSTEROID TITERS OF BLUE CRABS
325
DAYS
234
B DAYS BEFORE/AFTER ECOrSIS
-2 -1 E +1 +2
D3 D4 A, A2 B, B2
MOLT CYCLE STAGE
FIGURE 2. Serum ecdysteroid levels in individual crabs at consecutive stages of the molt cycle. Each
symbol represents a single crab whose serum was examined at the intervals shown. At each interval, the
stage of the cycle was determined by the condition of the exoskeleton and coloration of an appendage.
(A) Crabs in stages D, through Dj. The upper axis shows the number of days between measurements.
(B) Crabs in stages D3 through B2. All animals reached ecdysis. The upper axis shows the number of
days between measurements in relation to the time of ecdysis.
fold during yolk deposition as the weight of the ovary increased almost thirty-fold.
In contrast, ecdysteroids in the closely associated bursa copulatrix decreased from
stage I to stage III. Ecdysteroid accumulation in the ovaries of C4T females occurred
at a time when ecdysteroids were low in both serum (Fig. 1) and whole animals
(Table II). Although ovaries accumulated ecdysteroids during vitellogenesis, their
content of ecdysteroids did not contribute significantly to the whole animal liter.
Total ecdysteroid content in both males and females rose to maximum levels
during late premolt and declined precipitously by postmolt (Fig. 3). The pattern of
ecdysteroid liters measured throughout the molt cycle is similar to the pattern for
serum or carcass alone. These results are contrary to those of Faux et al. (1969)
who observed maximal ecdysteroids during postmolt in whole animal extracts of
females.
TABLE I
Ecdysteroid levels in female reproductive tissue
Ecdysteroid Cone.
Weight
Tissue
Stage
N
(mg/organ pr)
(ng/organ pr)
(ng/g)
Ovary
I
5
130 ± 20
0.35 ± 0.12
2.86 ± 1.19
II
3
660 ± 80
1.39 ± 0.12
2.14 ± 0.44
III
3
3240 ± 40
3.56 ± 1.09
1.10 ± 0.35
Bursa
I
5
710 ± 190
3.35 ± 1.38
4.70 ± 1.57
Copulatrix
II
3
1120± 620
1.58 ± 1.01
1.45 ± 0.36
III
3
180 ± 60
0.54 ± 0.32
3.22 ± 1.27
326
C. SOUMOFF AND D. M. SKINNER
TABLE II
Mature female whole animal ecdy steroids
Stage
N
Weight (g)
Ecdysteroid (ng/g)
A,
C4T
4
6
94.08 ± 11.06
117.73 ± 19.99
6.34 ± 2.25
2.48 ± 1.19
DISCUSSION
Contrary to previous results in which ecdysteroids reached a peak after ecdysis
(Faux et ai, 1 969) the results described here indicate that ecdysteroid concentrations
in Callinectes sapidus are at basal levels during intermolt, increase an average of
seven-fold by late premolt, and decline in postmolt. Whole animal ecdysteroid titers
for both sexes average 10.4 ng/g fresh weight, 74.8 ng/g fr. wt. and 15.8 ng/g fr. wt.
respectively at these stages. The antiserum we used has varying sensitivity toward
different ecdysteroids. It is three-fold more sensitive to ecdysone than to 20HE while
its sensitivity toward all other ecdysteroids tested is less than that to 20HE (Soumoff
et ai, 1981). This will have some effect on measurements of complex mixtures of
ecdysteroids. The concentrations we observed, however, are consistent with ec-
dysteroid levels in other crustaceans. Titers measured in the crab Carcinus maenas
(Adelung, 1969) range from 5 ng/g at intermolt to 1 10 ng/g during premolt. In the
amphipod Orchestia gammarella, the range is from 12 ng/g at intermolt to 63
ng/g at late premolt (Blanchet et ai, 1976). Ecdysteroids in the crayfish Orconectes
limosus range from 0.3 ng/g during intermolt to 60 ng/g during premolt (Willig and
Keller, 1973). In adult female lobsters (Homarus americanus) ecdysteroids are 6 ng/
g at postmolt (Gagosian et ai, 1974). Quantitation of the values for Orchestia was
by RI A, for Carcinus and Orconectes by bioassay, and for Homarus by high pressure
liquid chromatography and gas chromatography. Although the method of quanti-
co
00
cr
lOO-i
80-
if
60-
9 en 40-
O ^
rv O"
cr x
co
>-
Q
O
LJ
20-
Male
-i Female
n
Post Inter Early Late Inter Early Late Post Inter
Pre Pre Pre Pre
MOLT CYCLE STAGE
FIGURE 3. Whole animal ecdysteroid levels in male and female blue crabs at different stages of the
molt cycle. Hatched bars represent mature females. Ecdysteroids were calculated as 20HE equivalents.
Three or four animals from each stage were pooled and assayed. Hemolymph from both sexes and ovary
and bursa from mature females at each stage were assayed separately from remaining carcass and the
values were added to calculate the titers in whole animals.
ECDYSTEROID TITERS OF BLUE CRABS 327
tation determines, to some extent, the titer of hormone measured, these examples,
utilizing several different techniques, are consistent with each other.
Ecdysteroids measured by Faux el al. ( 1 969) for female blue crabs are incon-
sistent with the values reported here. In that analysis, the peak of ecdysteroids was
observed after ecdysis (280 ng/g 20HE and 24 ng/g makisterone A) and was twelve-
fold greater than the concentration at late premolt (20 ng/g inokosterone and 4 ng/
g 20HE). The method of quantitation of ecdysteroids was not specified and may
account for the discrepancy. One other example of a major peak of hormone titer
during postmolt was reported for O. gammarella (Blanchet et al., 1976). The hor-
mone titer reached a maximum in late premolt, declined by stage A, but showed
some indication of a second peak during stage B; a large standard deviation at this
stage made interpretation of the data difficult.
Measurements of circulating ecdysteroids are more variable between species than
are whole animal liters. However, all species exhibit a trend of increasing ecdysteroid
levels during premolt to a maximum prior to ecdysis, followed by a decline to basal
intermolt levels. The range of ecdysteroids in Callinectes serum, 5 ng/ml at intermolt
to 44 ng/ml in late premolt, is comparable to hemolymph tilers of the crayfish
Orconectes sanborni ranging from 4 ng/ml to 30 ng/ml (Stevenson et al., 1979).
Ecdysleroids in hemolymph of Ihe crab Pachygrapsus crassipes vary from near zero
jusl after ecdysis lo 120 ng/ml in premoll (Chang and O'Connor, 1978). The crab
Gecarcinus lateralis has a minimal liler of 10 ng/ml al inlermoll and a maximum
of 150 ng/ml al D3 when induced lo moll by mulliple limb aulolomy (Soumoff and
Skinner, 1982). Serum levels are in lhat same range in the fiddler crab Ucapugilator
(Hopkins, In press) during a natural molt cycle. Lachaise et al. (1976) reported
circulaling ecdysleroid lilers ranging from 62-470 ng/ml for Ihe crab C. maenas,
while lilers of 30-15,000 ng/ml hemolymph for Ihis species have also been reported
(Andrieux et al., 1976). Juvenile lobslers, Homarus americamis, exhibited basal
levels of ecdysteroids of less than 35 ng/ml and peak lilers of 350 ng/ml (Chang
and Bruce, 1980). These values were all quanlilaled by RIA.
Whole animal and serum ecdysleroid lilers in malure Callinectes females during
poslmoll were significanlly higher lhan Ihose in malure females al Ihe subsequenl
inlermoll slage. Despite Ihis, inlermoll ovaries conlained higher levels of ecdysteroids
lhan poslmoll ovaries; Ihe former were vilellogenic while Ihe laller were nol. Sim-
ilarly, Ihe ecdysleroid concenlralion in ovaries increased al vilellogenesis while Ihe
ecdysteroids in hemolymph remained low in C. maenas (Lachaise and Hoffmann,
1977) as well as in Ihe spider crab Acanthonyx lunulatus (Chaix and de Reggi, 1982).
Females of Ihe oxyrhynchan species Maja squinado and A. lunulatus reach re-
productive malurily al Iheir lasl moll, when Ihey enter terminal anecdysis. Their
Y-organs become inactive and degenerate (Carlisle, 1957; Chaix et al., 1976) and
hemolymph ecdysteroids decline (Chaix and de Reggi, 1982). Similarly for male
isopods (Sphaeroma serratum), Ihe Y-organs degenerate following Ihe puberty moll,
a terminal anecdysis is reached, and ecdysteroids gradually disappear from Ihe he-
molymph (Charmanlier, 1980). The very low hemolymph ecdysteroids in malure
C4T females of C. sapidus is consislenl wilh Ihese observations and, similarly, may
resull from degenerative changes in Ihe Y-organs.
ACKNOWLEDGMENTS
We are grateful lo Dr. C. P. Mangum for Ihe use of Ihe facilities al Ihe Virginia
Inslilule of Marine Sciences, Easl and for help in Ihe early slages of Ihis work. We
lhank Dr. M. Caslagna and Ihe slaffal VIMS for Iheir generous assislance and Dr.
328 C. SOUMOFF AND D. M. SKINNER
P. Hopkins for critical reading of the manuscript. The ecdysteroid antiserum was
kindly supplied by Dr. J. D. O'Connor (Univ. of California, Los Angeles), and the
[3H]ecdysone by Dr. D. S. King.
LITERATURE CITED
ADELUNG, D. 1969. Die Ausschiittung und Funktion von Hautungshormon wahrend eines Zwischen-
hautungs-Intervalls bei der Strandkrabbe Carcinus maenas L. Z. Naturforsch. 24B: 1447-1455.
ANDRIEUX, N., P. PORCHERON, J. BERREUR-BONNENFANT, AND F. DRAY. 1976. Determination du
taux d'ecdysone au cours du cycle d'intermue chez le Crabe Carcinus maenas: comparaison
entre individus sains et parasites par Saccitlina carcini. C. R. Acad. Sci. Paris 283: 1429-1432.
BLANCHET, M.-F., P. PORCHERON, AND F. DRAY. 1976. Etude des variations du taux des ecdysones au
cours du cycle d'intermue chez le male d'Orchestia gammarella Pallas (Crustace Amphipode)
par dosage radioimmunologique. C. R. Acad. Sci. Paris 283: 651-654.
BLISS, D. E. 1956. Neurosecretion and the control of growth in a decapod crustacean. Pp. 56-75. in
Bertil Hanstrom. Zoological Papers in Honour of his 65th Birthday, K. G. Wingstrand, ed. Zool.
Inst., Lund.
BOLLENBACHER, W. E., H. ZVENKO, A. K. KUMARAN, AND L. I. GILBERT. 1978. Changes in ecdysone
content during postembryonic development of the wax moth, Galleria mellonella: the role of
the ovary. Gen. Comp. Endocrinol. 34: 169-179.
CARLISLE, D. B. 1957. On the hormonal inhibition of moulting in decapod Crustacea. II. The terminal
anecdysis in crabs. J. Mar. Biol. Assoc. U. K. 36: 291-307.
CHAIX, J. C., ANDM. DE REGGI. 1982. Ecdysteroid levels during ovarian development and embryogenesis
in the spider crab Acanthonyx lunulatus. Gen. Comp. Endocrinol. 47: 7-14.
CHAIX, J. C., J. P. TRILLES, AND G. VERNET. 1976. Degenerescence de 1'organe Y chez les males puberes
d'Acanthonvx lunulatus (Risso) (Crustacea Decapoda Oxyrhyncha). C. R. Acad. Sci. Paris 283:
523-525.
CHANG, E. S., AND J. D. O'CONNOR. 1978. In vitro secretion and hydroxylation of a-ecdysone as a
function of the crustacean molt cycle. Gen. Comp. Endocrinol. 36: 151-160.
CHANG, E. S., AND J. D. O'CONNOR. 1979. Arthropod molting hormones. Pp. 797-814 in Methods of
Hormone Radioimmunoassay. 2nd ed., B. M. Jaffe and H. R. Behrman, eds. Academic Press,
New York.
CHANG, E. S., AND M. J. BRUCE. 1980. Ecdysteroid titers of juvenile lobsters following molt induction.
J. Exp. Zool. 214: 157-160.
CHARMANTIER, G. 1980. Controle endocrine et neuroendocrine de la mue de puberte chez les males de
Sphaeroma serratum (Fabricius, 1787) (Crustacea, Isopoda, Flabellifera). Gen. Comp. Endo-
crinol. 41: 349-364.
CHURCHILL, E. P., JR. 1919. Life history of the blue crab. Bulletin, U. S. Bureau of Fisheries for 1917-
191836: 93-136.
EMMEL, V. E. 1906. The relation of regeneration to the molting process in the lobster. Ann. Rep. Comm.
Inland Fish.. Rhode Island, 36: 258-313.
EMMEL, V. E. 1907. Relations between regeneration, the degree of injury, and moulting of young lobsters.
Science 15: 785.
FAUX, A., D. H. S. HORN, E. J. MIDDLETON, H. M. FALES, AND M. E. LOWE. 1969. Moulting hormones
of a crab during ecdysis. Chem. Commun. 175-176.
GAGOSIAN, R. B., R. A. BOURBONNIERE, W. B. SMITH, E. F. COUCH, C. BLANTON, AND W. NOVAK.
1974. Lobster molting hormones: isolation and biosynthesis of ecdysterone. Experientia 30:
723-724.
GAREN, A., L. KAUVAR, AND J.-A. LEPESANT. 1977. Roles of ecdysone in Drosophila development.
Proc. Nat. Acad. Sci. USA 74: 5099-5103.
HAGEDORN, H. H., J. D. O'CONNOR, M. S. FUCHS, B. SAGE, D. A. SCHLAEGER, AND M. K. BOHM.
1975. The ovary as a source of a-ecdysone in an adult mosquito. Proc. Nat. Acad. Sci. USA
72: 3255-3259.
HANDLER, A. M., AND J. H. POSTLETHWAIT. 1978. Regulation of vitellogenin synthesis in Drosophila
by ecdysterone and juvenile hormone. J. Exp. Zool. 206: 247-254.
HARD, W. L. 1942. Ovarian growth and ovulation in the mature blue crab, Callinectes sapidus Rathbun.
Chesapeake Biol. Lab. Bull. 46: 1-17.
HOFFMANN, J. A., M. LAGUEUX, C. HETRU, M. CHARLET, AND F. GOLTZENE. 1980. Ecdysone in
reproductively competent female adults and in embryos of insects. Pp. 431-465 in Progress in
Ecdysone Research. J. A. Hoffmann, ed. Elsevier/North-Holland Biomedical Press, Amsterdam.
ECDYSTEROID TITERS OF BLUE CRABS 329
HOLLAND, C. A., AND D. M. SKINNER. 1976. Interactions between molting and regeneration in the land
crab. Biol. Bull. 150: 222-240.
HOPKINS, P. M. 1983. Patterns of serum ecdysteroids during induced and uninduced proecdysis in the
fiddler crab, Uca pugilator. Gen. Comp. Endocrinol. In press.
LACHAISE, F., M. LAGUEUX, R. FEYEREISEN, AND J. A. HOFFMAN. 1976. Metabolisme de 1'ecdysone
au cours du developpement de Carcinus maenas (Brachyura, Decapoda). C. R. Acad. Sci. Paris
283: 943-946.
LACHAISE, F., AND J. A. HOFFMANN. 1977. Ecdysone et developpement ovarien chez un Decapode,
Carcinus maenas. C. R. Acad. Sci. Paris 285: 701-704.
LAGUEUX, M., M. HIRN, AND J. A. HOFFMANN. 1977. Ecdysone during ovarian development in Locusta
migratoria. J. Insect Physiol. 23: 109-119.
Luu, B., M. LAGUEUX, M. HIRN, AND J. A. HOFFMANN. 1976. Identification par chromatographie en
phase vapeur et spectrometrie de masse de 1'ecdysone synthetisee dans les ovaries de femelles
adultes de Locusta migratoria (Insecte Orthoptere). C. R. Acad. Sci. Paris 283: 1081-1084.
MYKLES, D. L., AND D. M. SKINNER. 1981. Preferential loss of thin filaments during molt-induced
atrophy in crab claw muscle. J Ultrastruct. Res. 75: 314-325.
OHNISHI, E.. T. MIZUNO, F. CHATANI, N. IKEKAWA, AND S. SAKURAI. 1977. 2-Deoxy-a-ecdysone from
ovaries and eggs of the silkworm, Bombyx mori. Science 197: 66-67.
PASSANO, L. M. 1960. Molting and its control. Pp. 473-536 in The Physiology of Crustacea, Vol. 1,
T. H. Waterman, ed. Academic Press, New York.
SKINNER, D. M. 1962. The structure and metabolism of a crustacean integumentary tissue during a molt
cycle. Biol. Bull. 123: 635-647.
SKINNER, D. M. In press. Interactions between molting and regeneration. In The Biology of Crustacea,
D. E. Bliss, and L. H. Mantel, eds. Academic Press, New York.
SKINNER, D. M., AND D. E. GRAHAM. 1970. Molting in land crabs: stimulation by leg removal. Science
169: 383-385.
SKINNER, D. M., AND D. E. GRAHAM. 1972. Loss of limbs as a stimulus to ecdysis in Brachyura (true
crabs). Biol. Bull. 143: 222-233.
SOUMOFF, C. AND D. M. SKINNER. 1980. Effect of Y-organectomy and autotomy on circulating ecdyste-
roid levels and regeneration in the crab Gecarcinus. Arner. Zoo/. 20: 933.
SOUMOFF, C., AND D. M. SKINNER. 1982. In vitro assay of hydroxylase activity in the crab Gecarcinus
lateralis. Amer. Zoo/. 22: 938.
SOUMOFF, C., D. H. S. HORN, AND J. D. O'CONNOR. 1981. Production of a new antiserum to arthropod
molting hormone and comparison with two other antisera. /. Steroid Biochem. 14: 429-435.
SPINDLER, K..-D., R. KELLER, AND J. D. O'CONNOR. 1980. The role of ecdysteroids in the crustacean
molting cycle. Pp. 247-280 in Progress in Ecdysone Research, J. A. Hoffmann, ed. Elsevier/
North-Holland Biomedical Press, Amsterdam.
STEVENSON, J. R., P. W. ARMSTRONG, E. S. CHANG, AND J. D. O'CONNOR. 1979. Ecdysone liters during
the molt cycle of the crayfish Orconectes sanborni. Gen. Comp. Endocrinol. 39: 20-25.
WARNER, G. F. 1977. In The Biology of Crabs. Van Nostrand Reinhold Co., New York.
WILLIG, A., AND R. KELLER. 1973. Molting hormone content, cuticle growth and gastrolith growth in
the molt cycle of the crayfish Orconectes limosus. J. Comp. Physiol. 86: 377-388.
Reference: Biol. Bull. 165: 330-341. (August, 1983)
GEOGRAPHIC LIMITS AND LOCAL ZONATION: THE BARNACLES
SEMIBALANUS (BALANUS) AND CHTHAMALUS
IN NEW ENGLAND
DAVID S. WETHEY
Department of Biology and Marine Science Program,
University of South Carolina, Columbia, SC 29208
ABSTRACT
The interactions between the intertidal barnacles Semibalanus (Balanus) bal-
anoides and Chthamalus fragilis were examined in order to determine whether the
factors which influence local zonation in the intertidal also contribute to the estab-
lishment of geographic limits.
Both physical and biotic factors influence intertidal zonation at the northern
limit of Chthamalus in New England. On sloping surfaces Semibalanus died at all
shore levels higher than mid tide level, apparently as a result of desiccation associated
with high summer temperatures. Chthamalus settlement occurred at all shore levels
above mean tide level, and postsettlement mortality apparently restricts Chthamalus
to high shore locations where Semibalanus growth and survival is inhibited. North
of the northern limit of Chthamalus, Semibalanus does not suffer summer heat
death, so it occupies the zone where Chthamalus would have a refuge from com-
petition further south.
The northern limit of Chthamalus is set not by factors directly related to cold
acting on Chthamalus. Rather the northern limit appears to be set by cold which
allows the dominant competitor to exclude Chthamalus from its refuge zone. South
of the northern limit the competitor, Semibalanus, is excluded from the high shore
by high summer temperatures.
INTRODUCTION
One of the goals of ecology is to determine the mechanisms responsible for the
patterns of distribution and abundance of organisms. The rocky intertidal zone has
been used very successfully to make experimental tests of a wide variety of hy-
potheses about the organization of communities and the dynamics of assemblages
of species. Much of this work has been designed to elucidate patterns of local dis-
tribution and abundance, rather than large scale geographic patterns. Here I examine
whether the same mechanisms that control local zonation are responsible for large
scale geographic patterns, those of geographic limits of species.
The strong physical gradient in the intertidal zone was long considered to be
fully responsible for the zonation patterns observed. Upper and lower limits of
distribution were thought to be set by physiological tolerances (Colman, 1933; He-
watt, 1937; Doty, 1946). Upper limits on the shore are now known to be generally
determined by physical factors. Foster (1969) and Hatton (1938) demonstrated that
barnacles die if transplanted above their usual shore zone, and that both heat and
moisture influence the rate of death. There is little field evidence that intolerance
of submersion sets the lower limit of marine species in the intertidal zone. The
Received 6 January 1983; accepted 23 May 1983.
330
NORTHERN LIMIT OF CHTHAMALUS 331
majority of the evidence is consistent with the hypothesis (Connell, 196 la, b) that
local lower limits are set by interactions with predators or competitors (e.g., reviews
by Connell, 1972; Paine, 1974; Menge, 1976; Lewis, 1977; Lubchenco and Menge,
1978; Schonbeck and Norton, 1978; Lubchenco, 1980).
Geographic limits have been correlated with physical conditions in much the
same way as have local zonation patterns. Hutchins (1947) hypothesized that the
most likely factors limiting geographic distribution were lethal temperatures for
adults and what he termed critical temperatures within which reproduction is suc-
cessful. In some cases the lethal physiological limits of species as determined in
laboratory studies correspond to geographic limits (e.g., Vernbergand Tashian, 1959;
Vernberg and Vernberg, 1967), but in other cases, geographic ranges are narrower
than predicted from studies of lethal limits (e.g., Barnes, 1958; Southward, 1958).
Since local zonation is not entirely controlled by lethal physiological limits, and
biotic interactions are often locally dominant, it is likely that biotic interactions also
play an important role in determining geographic limits.
In this paper I discuss biotic and physical factors which appear to strongly in-
fluence the northern geographic limit of the intertidal barnacle Chthamalus fragilis
on the Atlantic coast of North America. Chthamalus fragilis ranges from the Car-
ibbean to Cape Cod (Dando and Southward, 1980). At the northern end of its
distribution it is restricted to a narrow zone at the upper levels of the intertidal.
Below this zone lives an arctic barnacle species Semibalanus balanoides. This type
of zonation is also found near the northern limit of Chthamalus in Scotland, where
Connell (196 la) demonstrated that the upper shore limit of Chthamalus was set by
desiccation and the lower shore limit was set by competition with Semibalanus.
Semibalanus was renamed by Newman and Ross (1976); it is referred to as Balanus
balanoides in all previous literature.
MATERIALS AND METHODS
This study was carried out 100 km north and 150 km south of the recorded
northern limit of Chthamalus on Cape Cod, Massachusetts. The northern site was
East Point, Nahant, Massachusetts (42 25 N, 70 54 W), near the Northeastern
University Marine Science Institute. At this location only Semibalalanus is present.
Here, the tidal range is approximately 3.5 meters. The southern sites were the Yale
University Peabody Museum Field Station at Guilford, Connecticut (41 16 N, 72
44 W), and nearby Horse Island in the Long Island Sound (41 15 N, 72 45 W). At
these sites, both Semibalanus and Chthamalus coexist. The tidal range at these sites
is approximately 1.9 meters. Semibalanus settles at all sites between March and
May, and Chthamalus settles in Connecticut in August and September.
The zonation patterns of Semibalanus and Chthamalus were quantified by tran-
sects of contiguous 0.5 m X 0.5 m quadrats, which were photographed with a 70
mm camera held perpendicular to the shore with a focal framer. Permanent quadrats
were marked with stainless steel screws set in the corners. Percent cover of live and
dead organisms was estimated by placing a transparent plastic sheet with 49 uni-
formly plotted dots on its surface over enlargements of the photographs. Percent
cover was then estimated by counting the number of dots overlying each species
(e.g., Menge, 1976). Transects were established in a variety of locations in order to
determine the influence of shore orientation and aspect. Heights of the marker
screws relative to mean low water were estimated by the tables in the Tide Tables
(NOAA, 1982). Percent cover data are based on samples taken in August and
October.
332 D. S. WETHEY
Settlement of Chthamalus in the absence of Semibalanus was estimated by
removing Semibalanus with a paint scraper in a checkerboard pattern in perma-
nently marked quadrats. In this way Semibalanus removals were performed at all
shore levels. Removals were performed in August 1982, at the beginning of the
Chthamalus settlement season. Settlement was measured in mid October, 1982, by
counting newly settled spat in 4 cm X 4 cm quadrats in the field.
RESULTS
In Connecticut, Chthamalus occupies a narrow zone near mean high water of
neap tides (Figs. 1-4). The zonation is strongly influenced by slope and aspect. On
north-facing vertical surfaces, Chthamalus occupies a very narrow zone on the high
shore (Fig. 1). Maximum percent cover is 50% near mean high water of neap tides
(Fig. 1). Below this level, Semibalanus occupies 100% of the space (Fig. 1). On
south-facing vertical surfaces, Chthamalus occupies a wider zone. Its upper shore
limit is similar to that on north facing surfaces, but its lower limit is 25 cm lower
(Fig. 1). Its maximum percent cover is almost 100% on west facing vertical surfaces
(Fig. 1). Semibalanus occupies 100% of the space below Chthamalus, down to mid-
tide level. Below this zone, predation by the gastropod Urosalpinx apparently reduces
the percent cover of Semibalanus. Urosalpinx densities are as high as 200 per square
meter at mean low water of neap tides. On horizontal surfaces, Chthamalus occupies
a wider zone, and Semibalanus reaches its abundance peak very low on the shore
(Fig. 1 ). In the region below the Chthamalus zone, there was evidence of widespread
death of small Semibalanus (3 mm to 5 mm basal diameter) on sloping and hori-
zontal surfaces. Settlement of Semibalanus occurred throughout the intertidal zone
in March, and the newly settled individuals died in mid-summer on much of the
shore above mid-tide level. The dead individuals were tightly crowded, indicating
that the Semibalanus settlement had occupied almost 100% of the space below the
Chthamalus zone. The most likely cause of death of small individuals on horizontal
surfaces is desiccation related to summer high temperatures. There was no evidence
of Chthamalus death from desiccation.
More evidence of Semibalanus death resulting from high temperatures may be
seen in the zonation on a surface with a 70° slope which has a slow drip from a
deep crevice in the rock surface. Two transects were enumerated within the per-
manent 0.5 m X 0.5 m quadrats. The transects were 0.25 m apart. One ran through
the area with the water drip, and the parallel transect was dry. In the area with the
water drip, the upper shore limit of Semibalanus was 25 cm higher than in the
adjacent dry transect (Fig. 2). In the dry transect there were tightly crowded small
dead Semibalanus at the same shore level where individuals survived in the damp
location.
The influence of shade is clearly seen in a series of three parallel transects set
up close to the laboratory dock. In the partial shade of the dock the Chthamalus
zone is very narrow. Semibalanus occupies most of the space in the mid shore, and
Fucus occupies all space at mid tide level and below (Fig. 3A). In a parallel transect
0.5 meters away from the dock, there is less shade, the Chthamalus zone is wider,
and Semibalanus occupies a narrower zone, with Fucus at mid tide level (Fig. 3B).
In a third parallel transect 0.5 meters farther still from the dock, there is little shade,
the Chthamalus zone is even wider, and the upper shore limit of Semibalanus is
0.5 meters lower than it was in full shade (Fig. 3C).
The summer heat death documented here was common on all sloping shores
near the Yale Field station, and on the island shores visible en route to Horse Island
NORTHERN LIMIT OF CHTHAMALUS
333
100 -
50 -
V - N
o:
UJ
>
o
o
100 -
50
UJ
o
o:
UJ
0.
V - W
100
H - S
50 -
MHWN
MLWN
TIDAL HEIGHT
FIGURE 1 . Percent cover of Chthamalus (open circles) and Semibalanus (solid squares) as a function
of tidal height on transects at Horse Island, Connecticut. Tide levels are mean high water of neap tides
(MHWN) and mean low water of neap tides (MLWN). Top panel is a vertical north facing surface ( V-
N), center panel is a vertical west facing surface (V-W), bottom panel is a 10 degree slope facing south
(H-S).
in Long Island Sound. The total length of shoreline observed exceeded 5 kilometers.
This appeared to be a widespread mortality event on the high shore.
In order to test the hypothesis that the lower limit of Chthamalus was set by
postsettlement mortality associated with the presence of Semibalanus, a series of
Semibalanus removals were set up at all shore levels. Smothered individuals of
Chthamalus were encountered several times during the process of scraping Semi-
balanus off the rock during establishment of the Semibalanus removals. Chthamalus
subsequently settled in the Semibalanus removal areas. The heaviest settlement of
334
D. S. WETHEY
rr
UJ
>
O
o
00 i
50 •
DRIP
z
UJ
O
or
100 -i
50 -
DRY
MHWN
MLWN
TIDAL HEIGHT
FIGURE 2. Percent cover of Chthamalus (open circles) and Semibalanus (solid squares) as a function
of tidal height on transects at the Yale Field Station in Guilford, Connecticut. Tide levels marked as in
Figure 1. The two panels are from parallel transects 25 cm apart. Upper panel transect has water seepage
from a deep crevice in the rock surface located at the position of the arrow. Lower panel transect is dry.
Note the upward displacement of Semibalanus in the damp area below the water seepage.
Chthamalus on vertical surfaces was near mid tide level, in the zone where indi-
viduals usually die as a result of overgrowth by Semibalanus (Fig. 4). There was
very little settlement in the Chthamalus zone itself (Fig. 4). On horizontal surfaces,
settlement was most intense near mid tide level, in the zone where Semibalanus
died as a result of summer heat (Fig. 4).
Semibalanus removals were established in August, and settlement of Chthamalus
occurred prior to the October samples (Fig. 4). These sites were surveyed at the end
of April, at the height of the Semibalanus settlement season. Semibalanus had settled
at densities in excess of 50 per square centimeter. In the vertical sites (Figs. 1,4), at
all but the highest shore levels, Chthamalus, in the Semibalanus removal quadrats,
were overgrown by newly settled 2-week-old Semibalanus. When the nearly 100%
cover of newly settled Semibalanus was removed with a toothbrush, live Chthamalus
were found beneath it. Presumably these totally smothered Chthamalus, although
tolerant of desiccation (Foster, 197 la), would die within a few weeks with no direct
access to food, water, or oxygen.
Approximately 5% of the live Chthamalus (approximately 5 mm diameter) were
being undercut (sensu Connell, 196 la), lifted from the substratum, and expelled
from the growing surface of Semibalanus spat (approximately 1 mm diameter). No
crushing of Chthamalus by newly settled Semibalanus was observed. Semibalanus
spat were completely occluding the opercular valves of the majority of Chthamalus
in the zone where the adults of the two species co-occur, yet there was no settlement
100 -
50 -
or
> 100
o
o
UJ
o
(T
UJ
CL
50 •
100 !
B
50
MHWN
MTL
TIDAL HEIGHT
FIGURE 3. Percent cover of Chthamalus (open circles) and Semibalanus (solid squares) and Fucus
(open triangles) as a function of tidal height on transects at the Yale Field Station in Guilford, Connecticut.
Tide levels are mean high water of neap tides (MHWN) and mean tide level (MTL). The panels are from
three parallel transects separated from one another by 0.5 m. Panel A is adjacent to the laboratory dock
and is shaded for most of the day. Panel B is 0.5 meters farther from the dock and has more exposure
to sun. Panel C is 0.5 meters still farther from the dock and is exposed to sun for more than half of the
day. Shore has a 45 degree slope and faces east. Note the downward displacement of the Semibalanus
upper shore limit as the shore receives more sun.
335
336
D. S. WETHEY
1.0
.5
C\J
I
V-N
E
o
CO
z
ui
*
UJ
2.0
1.0
v-w
UJ
CO
1.0
.5
H- S
MHWN
MLWN
TIDAL HEIGHT
FIGURE 4. Settlement density of Chthamalus in numbers per cm2 in Semibalanus removals. Means
and standard deviations from 5 to 10 replicate counts of 4 cm X 4 cm quadrats are reported as a function
of tidal height. Symbols and locations are as in Figure 1 .
on the opercular valves of adjacent Semibalanus individuals. In sites where Semi-
balanus settlement was less intense (shore sites used for Figs. 2 and 3), Semibalanus
had not yet overgrown Chthamalus but were likely to do so by June or July.
These results indicate that Chthamalus is capable of settlement and survival for
at least 8 months (August through April) at mid-tide level in the absence of Semiba-
lanus. Postsettlement mortality as a result of competition with Semibalanus is the
most likely mechanism causing the restriction of Chthamalus to the high shore.
Although Chthamalus settles most heavily in the mid-shore, it survives only in its
refuge from competition high on the shore, where Semibalanus is restricted by
desiccation. Postsettlement mortality of Chthamalus is likely to be very intense in
spring when Semibalanus settlement occurs, thereby smothering Chthamalus.
NORTHERN LIMIT OF CHTHAMALUS
337
In the northern site, beyond the northern limit of Chthamalus, zonation varies
as a function of slope and aspect, but there was no evidence of the widespread heat
death that characterized sloping shores in Connecticut. On vertical surfaces, the
upper shore limit of Semibalanus is higher on north-facing localities than in south-
facing shores (Fig. 5). On sloping surfaces Semibalanus survives from mean high
water of neap tides down to mid-tide level, where it is excluded by competition with
the mussel Mytilus (Fig. 5). In the seven summers for which I have data on the
distribution and abundance of barnacles (1976-1982), Semibalanus populations in
northern Massachusetts have never suffered summer heat death of the kind docu-
mented from the Connecticut shore in 1982 (Wethey, 1979; personal observation).
100
50 -
V - N
cc
UJ
>
o
o
100 -,
50 i
UJ
O
<r
UJ
o.
V - S
100 n
H - N
50 -
MHWN
MLWN
TIDAL HEIGHT
FIGURE 5. Percent cover of Semibalanus (solid squares) and Mytilus (open circles) as a function
of tidal height on transects at Nahant, Massachusetts. Tide levels are as in Figure 1 . Top panel is a vertical
north facing surface (V-N), center panel is a vertical south facing surface (V-S), bottom panel is a 30
degree slope facing northwest (H-N).
338 D. S. WETHEY
DISCUSSION
This study was set up to determine whether the factors which influence local
zonation in the intertidal might also contribute to the establishment of geographic
limits. Both physical and biotic factors appear to influence zonation at the northern
limit of Chthamalus in New England. The upper shore limit of Semibalanus is
apparently set by desiccation associated with high summer temperatures. In damp
or shaded locations, Semibalanus occupies the shore up to mean high water of neap
tides (Figs. 1-3). In sunny locations the upper shore limit of Semibalanus is lower
than in shaded locations (Figs. 1-3). On sloping surfaces Semibalanus died appar-
ently as a result of desiccation at all shore levels higher than mean tide level (Figs.
1, 2). Chthamalus survives in locations where Semibalanus fails to persist (Figs. 1-
3). Settlement of Chthamalus occurs at all shore levels down to mean tide level (Fig.
4), and apparently post-settlement-mortality subsequently limits Chthamalus to lo-
cations where Semibalanus growth is restricted. Warmer, drier sites have wider
Chthamalus zones because these locations are apparently too hot or dry for Semiba-
lanus (Figs. 1-3). These same factors may also be important in setting the northern
limit of Chthamalus. North of the northern limit of Chthamalus, Semibalanus does
not suffer summer heat death, so it occupies the zone where Chthamalus would
have a refuge from competition further south (Fig. 5). In the absence of a refuge,
any Chthamalus larvae that settle on the high shore are likely to be crushed or
overgrown by Semibalanus. This in turn reduces the pool of adult Chthamalus
which contribute larvae to the plankton. The reduced number of larvae available
for settlement and the reduced settlement success as a result of competition pre-
sumably combine to restrict Chthamalus from more northern locations. Thus the
northern limit of Chthamalus is not set by factors which are directly related to cold
acting on Chthamalus. Rather, the northern limit appears to be set by cold which
allows the dominant competitor to exclude Chthamalus from its refuge zone. South
of the northern limit the competitor, Semibalanus, is excluded from the high shore
by high summer temperatures. The northern limit of Chthamalus is likely to be
more strongly influenced by competition between Semibalanus and Chthamalus
than by direct physiological limitation of Chthamalus itself.
These results are consistent with those of Connell (196 la), who documented the
importance of competition in setting local limits of zonation in Semibalanus bal-
anoides and Chthamalus near the northern limit of Chthamalus in Scotland.
Chthamalus was successful in the zone where Semibalanus suffered mortality from
desiccation. Chthamalus settled at shore levels below the zone where adults survived.
Post-settlement mortality as a result of competition with Semibalanus limited
Chthamalus to the high shore (Connell, 196 la). Barnes (1956) maintained Chtham-
alus (on stones from Connell's 196 la experiments) under conditions of total sub-
mersion on a raft for two years. He found that the growth rate under these conditions
was equivalent to that of individuals in the intertidal zone. He reported that post-
settlement mortality of Chthamalus as a result of space competition with Semi-
balanus restricted it from the low shore: on the raft "a 6-month-old Chthamalus
settlement (2 mm long) was obliterated in a few weeks by a moderate spat fall of
\Semi\Balanus and full grown Chthamalus (9-15 mm) were completely overgrown
in 2 months."
All of these results are consistent with the hypothesis that competition with
Semibalanus is a major determinant of local distribution and abundance of Chtham-
alus. The restriction of Semibalanus to shaded habitats in more southern locations
has been reported by Barnes (1958) for Woods Hole, where summer heat apparently
NORTHERN LIMIT OF CHTHAMALUS 339
killed off 95% of 5 mm basal diameter individuals on south-facing and horizontal
surfaces in 1956. On north-facing surfaces mortality was only 50% in the same
period (Barnes, 1958). These individuals were about the same size as those found
dead in the present study (3 mm basal diameter). Several authors (Southward and
Crisp, 1956; Lewis, 1957, 1964; Crisp and Southward, 1958; Bowman, 1983) have
reported effects of shore slope and aspect similar to those described here. Near its
northern limit in Scotland, Chthamalus is more common on south-facing vertical
surfaces which dry out at low tide, while Semibalanus dominates at the same tide
levels in more horizontal locations which remain wet. The most favorable location
for Semibalanus in southwest England is under rocks and overhangs (Southward
and Crisp, 1954). Semibalanus in southwest England is almost completely absent
from the south facing coast, is rare on the west-facing portion, and is abundant on
the north-facing section (Crisp and Southward, 1958). On the north Cornwall coast,
Semibalanus becomes rare along the eastern section where more of the coast faces
west (Crisp and Southward, 1958). Summer heat death of Semibalanus in 1976 in
northern Scotland resulted in a lowering of the lower limit of Chthamalus on those
shores (Bowman, 1983). These distribution patterns are consistent with the hy-
pothesis that Semibalanus is limited by desiccation and high temperatures on the
high shore and in the more southern localities. Direct tests of the temperature
tolerances of Semibalanus and Chthamalus indicate that the latter species is far
more tolerant of desiccation and high temperatures, and that the larval stages and
newly metamorphosed spat are more susceptible than are adults (Southward, 1958;
Crisp and Ritz, 1967; Foster, 1969, 197 la, b).
Southward and Crisp (1956) hypothesized that year to year fluctuations in tem-
perature influenced the relative abundance of Semibalanus and Chthamalus by
changing the intensity of competition between the species. Many details of the
geographic distribution of Semibalanus and Chthamalus were recorded in the 1930's
(Moore, 1936; Moore and Kitching, 1939), and these distributions have been studied
at the same localities by Southward and Crisp. After a number of warm years
Chthamalus increased in abundance, and after a number of cold years Semibalanus
increased (Southward and Crisp, 1956; Southward, 1967; Crisp et al, 1981). They
argued that the mechanism might be related to competition for food (Southward
and Crisp, 1956, p. 220). Lewis (1964, pp. 251-252) hypothesized that the principal
effect of temperature was mediated through competition for space with Semibalanus.
The evidence for cold limitation of Chthamalus is far less strong than that of
heat limitation of Semibalanus. Crisp (1950) transplanted Chthamalus beyond its
northern limit to Whitley Bay in Northumberland on the North Sea coast of En-
gland. The individuals survived two winters and produced viable larvae. In the
extremely cold winter of 1 962- 1 963, there was no increased mortality of Chthamalus
in North Wales or in the south or southwest coasts of England (Crisp, 1964). Mor-
tality was higher than usual in south Wales in Mumbles Pier, where there was 25%
mortality on horizontal surfaces (Crisp, 1964). During this particular winter a num-
ber of species including the commercial oyster and the New Zealand barnacle El-
minius modestus, suffered extremely high mortality as a result of cold (Crisp, 1964).
Southward (1967) stated that the decreases in Chthamalus during 1963 were more
strongly influenced by the previous cool summer than by the exceptionally cold
winter. He suggested that very cold winters were not a major factor controlling the
distribution of Semibalanus and Chthamalus (Southward, 1967). At the northern
limit of its geographic distribution, Chthamalus is found at the highest shore levels,
where the effect of cold air temperatures would be the most severe in winter. If it
were not tolerant of cold, Chthamalus ought to die in winter at its northern limit,
340 D. S. WETHEY
but it does not appear to do so (e.g., Crisp and Southward, 1958; Lewis, 1964, pp.
251-252). All of these data indicate that direct limitation of the geographic distri-
bution of Chthamalus by cold is unlikely.
The northern limit of Chthamalus in New England appears to be influenced by
temperature as mediated through competition with Semibalanus. Post-settlement
mortality of Chthamalus within the Semibalanus zone apparently excludes it from
living low on the shore. Chthamalus survives in southern New England in the zone
where Semibalanus dies from desication and/or heat stress (Figs. 1-3). The northern
limit of Chthamalus occurs where Semibalanus no longer dies from desiccation on
the high shore (Fig. 5). The absence of adult Chthamalus in northern New England
is also likely to contribute to a reduced pool of larvae available for settlement,
because reproductive populations exist only south of Cape Cod. Failure of larval
or juvenile stages has been suggested as setting the northern limit of Chthamalus
in Scotland (Lewis et ai, 1982). In northern Scotland, recruitment declines regularly
towards the geographic limit of Chthamalus (Lewis et ai, 1982; Bowman, 1983).
It is likely that this comes about partly because of the progressive restriction of
Chthamalus by competition with Semibalanus to narrower and narrower zones at
the highest levels on the shore.
ACKNOWLEDGMENTS
This work was supported by grants from the National Science Foundation (OCE
8208176, OCE 7726503) and a grant from the Research and Productive Scholarship
Fund of the University of South Carolina. M. P. Morse, N. W. Riser, R. B. Shepard,
and H. Werntz allowed access to the Massachusetts site at the Northeastern Uni-
versity Marine Science Institute at Nahant, and provided laboratory space. M. P.
Morse provided housing and transportation. L. W. Buss allowed access to the Con-
necticut sites at the Yale University Peabody Museum Field Station and Horse
Island and provided lab space, housing, and transportation. L. Haas took the pho-
tographs at Nahant and M. W. Reed took the photographs at Guilford Connecticut.
S. A. Woodin, J. P. Sutherland, and S. Ortega provided field assistance during
establishment of the permanent sites. J. R. Lewis sparked my interest in geographic
limits while I was at his laboratory as a National Needs Postdoctoral Fellow (NSF
Grant SPI-7914910). S. A. Woodin and an anonymous reviewer provided many
helpful comments.
LITERATURE CITED
BARNES, H. 1956. The growth rate of Chthamalus stellatus (Poli) J. Mar. Biol. Assoc. U.K. 35: 355-361.
BARNES, H. 1958. Regarding the southern limits of Balanus balanoides (L.). Oikos 9: 139-157.
BOWMAN, R. S. 1983. The role of stochastic events in Balanus/Chthamalus interactions on Scottish
shores, ms.
COLMAN, J. S. 1933. The nature of intertidal zonation of plants and animals. J. Mar. Biol. Assoc. U.K.
18: 435-476.
CONNELL, J. H. 196 la. The influence of interspecific competition and other factors on the distribution
of the barnacle Chthamalus stellatus. Ecology 42: 710-723.
CONNELL, J. H. 1961b. The effects of competition, predation by Thais lapillus. and other factors on
natural populations of the barnacle Balanus balanoides. Ecol. Monogr. 31: 61-104.
CONNELL, J. H. 1972. Community interactions on marine rocky intertidal shores. Ann. Rev. Ecol. Svst.
3: 169-192.
CRISP, D. J. 1950. Breeding and distribution of Chthamalus stellatus. Nature 166: 311-312.
CRISP, D. J., ed. 1964. The effects of the severe winter of 1962-63 on marine life in Britain. J. Anim.
Ecol. 33: 165-210.
CRISP, D. J., A. J. SOUTHWARD, AND E. C. SOUTHWARD. 1981. On the distribution of the intertidal
barnacles Chthamalus stellatus, Chthamalus montaqui and Euraphia depressa. J. Mar. Biol.
Assoc. U. K. 61: 359-380.
NORTHERN LIMIT OF CHTHAMALUS 341
CRISP, D. J., AND A. J. SOUTHWARD 1958. The distribution of intertidal organisms along the coasts of
the English Channel. J. Mar. Biol. Assoc. U. K. 37: 157-208.
CRISP, D. J., AND D. A. RITZ 1967. Changes in the temperature tolerance of Balanus balanoides during
its life cycle. Helgol. Wiss. Meeresunters. 15: 98-1 15.
DANDO, P. R., AND A. J. SOUTHWARD. 1980. A new species of Chthamalus (Crustacea Cirripedia)
characterized by enzyme electrophoresis and shell morphology: with a revision of the other
species of Chthamalus from the western shores of the Atlantic Ocean. J. Mar. Biol. Assoc.
V. K. 60: 787-831.
DOTY, M. S. 1946. Critical tide factors that are correlated with the vertical distribution of marine algae
and other organisms along the Pacific coast. Ecology 27: 315-328.
FOSTER, B. A. 1969. Tolerance of high temperature by some intertidal barnacles. Mar. Biol. 4: 326-332.
FOSTER, B. A. 197 la. Desiccation as a factor in the intertidal zonation of barnacles. Mar. Biol. 8: 12-
29.
FOSTER, B. A. 197 Ib. On the determinants of the upper limit of intertidal distribution of barnacles
(Crustacea: Cirripedia). J Anitn. Ecol. 40: 33-48.
HATTON, H. 1938. Essais de bionomie explicative sur quelques especes intercotidales d'algues et
d'animaux. Ann. Insl. Oceanogr. 17: 241-348.
HEWATT, W. B. 1937. Ecological studies on selected marine intertidal communities of Monterey Bay,
California. Am. Midi Nat. 18: 161-206.
HUTCHINS, L. W. 1947. The bases for temperature zonation in geographical distribution. Ecol. Monogr.
17: 325-335.
LEWIS, J. R. 1957. Intertidal communities of the northern and western coasts of Scotland. Trans. R. Soc.
Edmb. 63: 185-220.
LEWIS, J. R. 1964. The Ecology of Rocky Shores. Hodder & Stoughton, London.
LEWIS, J. R. 1977. The role of physical and biological factors in the distribution and stability of rocky
shore communities. Pp. 417-424 in Biology of Benthic Organisms, B. F. Keegan, P. O'Ceidigh,
and P. J. S. Boaden, eds. Pergamon, Oxford.
LEWIS, J. R., R. S. BOWMAN, M. A. KENDALL, AND P. WILLIAMSON, 1982. Latitudinal variations in
population dynamics: possibilities and realities in some littoral species. Neth. J. Sea Res. 16:
18-28.
LUBCHENCO, J. 1980. Algal zonation in the New England rocky intertidal community: an experimental
analysis. Ecology 61: 333-344.
LUBCHENCO, J., AND B. A. MENGE. 1978. Community development and persistence in a low rocky
intertidal zone. Ecol. Monogr. 59: 67-94.
MENGE, B. A. 1976. Organization of the New England rocky intertidal community: role of predation,
competition and environmental heterogeneity. Ecol. Monogr. 46: 355-393.
MOORE, H. B. 1936. The biology of Balanus balanoides. V. Distribution in the Plymouth area. / Mar.
Biol. Assoc. U. K. 20: 701-716.
MOORE, H. B., AND J. A. KJTCHING. 1939. The biology of Chthamalus stellatus (Poli). J. Mar. Biol.
Assoc. U. K. 23: 521-541.
NATIONAL OCEANIC AND ATMOSPHERIC AGENCY. 1982. Tide Tables, East Coast of North and South
America including Greenland. U. S. Government Printing Office, Washington.
NEWMAN, W. A., AND A. Ross. 1976. Revision of the Balanomorph Barnacles; including a catalog of
the species. San Diego Soc. Nat. Hist. Mem. 9: 1-108.
PAINE, R. T. 1974. Intertidal community structure: experimental studies on the relationship between a
dominant competitor and its principal predator. Oecologia 15: 93-120.
SCHONBECK, M., AND T. A. NORTON, 1978. Factors controlling the upper limits of Fucoid algae on the
shore. J. Exp. Mar. Biol. Ecol. 31: 303-313.
SOUTHWARD, A. J. 1958. Note on the temperature tolerance of some intertidal animals in relation to
environmental temperature and geographic distribution. J. Mar. Biol. Assoc. U. K. 37: 49-66.
SOUTHWARD, A. J. 1967. Recent changes in the abundance of intertidal barnacles in south west England:
a possible effect of climatic deterioration. J. Afar. Biol. Assoc. if. K. 47: 81-95.
SOUTHWARD, A. J., AND D. J. CRISP, 1954. Recent changes in the distribution of the intertidal barnacles
Chthamalus stellatus Poli and Balanus balanoides L. in the British Isles. J. Anim. Ecol. 23: 163-
177.
SOUTHWARD, A. J., AND D. J. CRISP 1956. Fluctuations in the distribution and abundance of intertidal
barnacles. J. Mar. Biol. Assoc. U. K. 35: 21 1-229.
VERNBERG, F. J., AND R. E. TASHIAN. 1959. Studies on the physiological variation between tropical and
temperate zone fiddler crabs of the genus Uca. I. Thermal death limits. Ecology 40: 589-593.
VERNBERG, F. J., AND W. B. VERNBERG. 1967. Thermal limits of southern hemisphere Uca crabs.
Studies on the physiological variation between tropical and temperate zone fiddler crabs of the
genus Uca. IX. Oikos 18: 118-123.
WETHEY, D. S. 1979. Demographic variation in intertidal barnacles. Ph.D. Dissertation, University of
Michigan (University Microfilms No. 80-07857).
Continued from Cover Two
secutive Arabic numbers, on a separate sheet at the end of the paper. Footnotes should be limited to
authors' current addresses, acknowledgments or contribution numbers, and explanation of unusual ab-
breviations. All such footnotes should appear on the title page. Footnotes are not normally permitted
in the body of the text.
4. A condensed title or running head of no more than 35 letters and spaces should appear at the top
of the title page.
• /"f .tS/V7 \ *~:^<
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 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, con-
forming 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 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 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 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 abbre-
viations invented by the author. For example, use Rit Visindajjelags Islendinga without abbreviation.
F. All single word 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 preferred forms rather than WORLD LIST or
USASI usage (e.g. Nature, Science, Evolution NOT Nature, Lond., Science, N. Y.; Evolution, Lancaster,
Pa.)
6. Reprints, charges. The Biological Bulletin has no page charges. However, authors will be re-
quested to help pay printing charges of manuscripts that are unusually costly due to length or numbers
of tables, figures, or formulae. Reprints may be ordered at time of publication and normally will be
delivered about two to three months after the issue date. Authors (or delegates or 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).
CONTENTS
ANNUAL REPORT OF THE MARINE BIOLOGICAL LABORATORY 1
AGUDELO, MARIA I., KENNETH KUSTIN, GUY C. MCLEOD, WILLIAM E.
ROBINSON, AND ROBERT T. WANG
Iron accumulation in tunicate blood cells. I. Distribution and oxidation
state of iron in the blood of Boltenia ovifera, Styela clava, and Molgula
manhattensis 100
ANDERSON, WINSTON A., AND WILLIAM R. ECKBERG
A cytological analysis of fertilization in Chaetopterus pergamentaceus 1 10
BlCKELL, LOUISE R., AND STEPHEN C. KEMPF
Larval and metamorphic morphogenesis in the nudibranch Melibe
leonina (Mollusca: Opisthobranchia) 119
CRONIN, THOMAS W., AND RICHARD B. FORWARD, JR.
Vertical migration rhythms of newly hatched larvae of the estuarine
crab, Rhithropanopeus harrisii 139 '
FORWARD, RICHARD B., JR., AND KENNETH J. LOHMANN
Control of egg hatching in the crab Rhithropanopeus harrisii (Gould) 154 *
HAND, STEVEN C., AND GEORGE N. SOMERO
Energy metabolism pathways of hydrothermal vent animals: adapta-
tions to a food-rich and sulfide-rich deep-sea environment 167
HILLER-ADAMS, PAGE, AND JAMES J. CHILDRESS
Effects of feeding, feeding history, and food deprivation on respiration
and excretion rates of the bathypelagic mysid Gnathophausia ingens 182
INCZE, LEWIS S., AND A. J. PAUL
Grazing and predation as related to energy needs of stage I zoeae of
the tanner crab Chionoecetes bairdi (Brachyura, Majidae) 197 *
MACKIE, G. O., AND C. L. SINGLA
Coordination of compound ascidians by epithelial conduction in the
colonial blood vessels 209
OLSON, RICHARD RANDOLPH
Ascidian-/V0c/r/0r0/i symbiosis: the role of larval photoadaptations in
midday larval release and settlement 221
READ, LAURIE K., LYNN MARGULIS, JOHN STOLZ, ROBERT OBAR, AND
THOMAS K. SAWYER
A new strain of Paratetramitus jugosus from Laguna Figueroa, Baja
California, Mexico 241
REED-MILLER, CHARLENE
The initial calcification process in shell-regenerating Tegula (Archaeo-
gastropoda) 265
RUTOWSKI, RONALD L.
Mating and egg mass production in the aeolid nudibranch Hermissenda
crassicornis (Gastropoda: Opisthobranchia) 276
SEBENS, KENNETH P.
Settlement and metamorphosis of a temperate soft-coral larva (Al-
cyonium siderium Verrill): induction by crustose algae , 286 "
SLOBODKIN, L. B., AND KENNETH DUNN
On the evolutionary constraint surface of hydra 305
SOUMOFF, CYNTHIA, AND DOROTHY M. SKINNER
Ecdysteroid titers during the molt cycle of the blue crab resemble those
of other Crustacea 321 •
WETHEY, DAVID S.
Geographic limits and local zonation: the barnacles Semibalanus (Bal-
anus) and Chthamalus in New England 330
Volume 165 Number 2
u
T-TTTj!
n™'"
BIOLOGICAL
- • ;| - •
-— \-~ . U'/7 <-r V* v w*
PUBLISHED BY
THE MARINE BIOLOGICAL LABORATORY
Editorial Board
•o;
DANIEL L. ALKON, National Irsututes of Health and MICHAEL G. O'RAND, Laboratories for Cell Bioic^y,
Marine Biological Laboratory University of North Carolina at Chapel Hill
ROBERT B. BARLOW, JR., Syracuse University RALPH s- QUATRANO, Oregon State University at
Corvallis
WALL.S H. CLARK, JR., University of California jrt LlQNEL L R£BHUN> University of virginia
JOEL L. ROSENBAUM, Yale University
DAVID H. EVANS, University of Florida
DOROTHY M. SKINNER, Oak Ridge National
HARLYN O. HALVORSON, Brandeis University Laboratory
W Y-'7
RONALD R. HOY, Cornell University JOHN D. STRANDBERG, Johns Hopkins University
SAMUEL S. KCMDE, The Population Council, JOHN M' TEAL' Woods Hole Oceanographic
Rockefeller University Institution
J. RICHARD WHITTAKER, Boston University
FRANK J. LONGO, University of Iowa Marine Program and Marine Biological Laboratory
CHARLOTTE P. MANGUM, The College of GEORGE M. WOODWELL, Ecosystems Center, Marine
William and Mary Biological Laboratory
Editor: CHARLES B. METZ, University of Miami
OCTOBER, 1983
Printed and Issued by
LANCASTER PRESS, Inc.
PRINCE &. LEMON STS.
LANCASTER, PA.
THE BIOLOGICAL BULLETIN
THE BIOLOGICAL BULLETIN is published six times a year by the Marine Biological Laboratory, MBL
Street, Woods Hole, Massachusetts 02543.
Subscriptions and similar matter should be addressed to THE BIOLOGICAL BULLETIN, Marine Bio-
logical Laboratory, Woods Hole, Massachusetts. Single numbers, $10.00. Subscription per volume (three
issues), $27.00 (this is $54.00 per year for six issues).
Communications relative to manuscripts should be sent to Dr. Charles B. Metz, Editor, or Pamela
Clapp, Assistant Editor, at the Marine Biological Laboratory, Woods Hole, Massachusetts 02543 between
May 1 and October 1, and at the Institute For Molecular and Cellular Evolution, University of Miami,
521 Anastasia, Coral Gables, Florida 33134 during the remainder of the year.
Copyright © 1983, by the Marine Biological Laboratory
Second-class postage paid at Woods Hole, Mass., and additional mailing offices.
ISSN 0006-3185
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). Very short
, apers (less than 10 manuscript pages includipg tables, figures, and bibliography) will be published in
a separate section entitled "Short Reports." A limited number of solicited review papers may be accepted
after formal review. A paper will usually appear within four months after its acceptance.
The Editorial Board requests that manuscripts conform to the requirements set below; those manu-
scripts which do not conform will be returned to authors for correction before review.
1. Manuscripts. Manuscripts, including figures, should be submitted in triplicate. (Xerox copies
of photographs are not acceptable for review purposes.) The original manuscript must be typed in double
spacing (including figure legends, footnotes, bibliography, etc.) on one side of 16- or 20-lb. bond paper,
8'/2 by 1 1 inches. 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). Manu-
scripts should conform to the Council of Biology Editors Style Manual, 4th Edition (Council of Biology
Editors, 1978) 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. Figures. Figures should be no larger than 8'/2 by 1 1 inches. The dimensions of the printed page,
5 by 73/» inches, should be kept in mind in preparing figures for publication. We recommend that figures
be about \lh 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 illus-
tration 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 repro-
duced 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. The author's name and an arrow indicating orientation should appear on
the reverse side of all figures.
3. 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 con-
Continued on Cover Three
THE
BIOLOGICAL BULLETIN
PUBLISHED BY
THE MARINE BIOLOGICAL LABORATORY
Editorial Board
DANIEL L. ALKON, National Institutes of Health and MICHAEL G. O'RAND, Laboratories for Ce.i Biology,
Marine Biological Laboratory University of North Carolina at Chapel Hill
ROBERT B. BARLOW, JR., Syracuse University
RALPH S. QUATRANO, Oregon State University at
Corvallis
WALLIS H. CLARK, JR., University of California at LIONEL I. REBHUN, University of Virginia
Davis
JOEL L. ROSENBAUM, Yale University
DAVID H. EVANS, University of Florida
HARLYN O. HALVORSON, Brandeis University
RONALD R. HOY, Cornell University
SAMUEL S. KOIDE, The Population Council,
Rockefeller University
FRANK J. LONGO, University of Iowa
CHARLOTTE P. MANGUM, The College of
William and Mary
DOROTHY M. SKINNER, Oak Ridge National
Laboratory
JOHN D. STRANDBERG, Johns Hopkins University
JOHN M. TEAL, Woods Hole Oceanographic
Institution
J. RICHARD WHITTAKER, Boston University
Marine Program and Marine Biological Laboratory
GEORGE M. WOODWELL, Ecosystems Center, Marine
Biological Laboratory
Editor: CHARLES B. METZ, University of Miami
OCTOBER, 1983
Printed and Issued by
LANCASTER PRESS, Inc.
PRINCE & LEMON STS.
LANCASTER, PA.
The BIOLOGICAL BULLETIN is issued six times a year at the
Lancaster Press, Inc., Prince and Lemon Streets, Lancaster, Penn-
sylvania.
Subscriptions and similar matter should be addressed to The
Biological Bulletin, Marine Biological Laboratory, Woods Hole,
Massachusetts. Single numbers, $10.00. Subscription per volume
(three issues), $27.00 ($54.00 per year for six issues).
Communications relative to manuscripts should be sent to Dr.
Charles B. Metz, Marine Biological Laboratory, Woods Hole, Mas-
sachusetts 02543 between May 1 and October 1, and to Dr. Charles
B. Metz, Institute For Molecular and Cellular Evolution, University
of Miami, 521 Anastasia, Coral Gables, Florida 33134 during the
remainder of the year.
THE BIOLOGICAL BULLETIN (ISSN 0006-3185)
Second-class-postage paid at Woods Hole, Mass., and additional mailing offices.
LANCASTER PRESS, INC., LANCASTER, PA
11
ERRATA
The Biological Bulletin, Volume 165, Number 1, Page 2.
The following correction should be made in the Marine Biological Laboratory's
Eighty-fifth Report, for the Year 1982, Ninety-fifth Year:
Joel P. Davis, Seapuit, Inc. should be added to the Class of 1984 — Trustees.
The Biological Bulletin, Volume 165, Number 1, Page 203.
The following corrections should be made in the paper by L. S. Incze and A. J.
Paul entitled, Grazing and predation as related to energy needs of stage I zoeae of
the tanner crab, Chionoecetes bairdi (1983, Biol. Bull., 165: 197-208):
The exponent —2 was deleted from two equations in Table III. The table should
appear as follows:
TABLE III
Ingestion rate (I) of 24 hour old zoeae grazing on phytoplankton (Gonyaulax grindleyi, Coscinodiscus
spp., Thalassiosira spp.) at various cell concentrations ((€}) at 5°C, and percent contribution to
respiratory requirement (%R)
Cell type
Carbon <C>
(Mgceir1) (cells T1)
I (zoea '
d")
%R'
No. cells
MgC
G. grindleyi
2.9 X 1(T3 7.3 X 103
4.3
1.2 X 10'2
1.4
3.1 X 104
19.0
5.0 X 10~2
5.8
6.5 X 104
33.6
9.7 X 10~2
11.3
1.2 X 105
39.9
1.2 X 10~'
14.0
Coscinodiscus spp.
2.66 X 1(T2 7.8 X 102
2.6
6.8 X JO'2
7.9
8.40 X 102
2.9
7.6 X 1Q-2
8.9
8.80 X 102
2.7
7.0 X 10'2
8.2
9.40 X 102
2.4
6.3 X 10~2
7.4
9.83 X 102
2.3
6.0 X 10~2
7.0
9.83 X 102
2.3
6.0 X JO'2
7.0
1.68 X 103
2.2
5.7 X 10~2
6.6
1.68 X 103
2.5
6.6 X 10 2
7.7
1.68 X 103
4.2
1.1 X 10'1
12.8
Thalassiosira spp.
2.30 X 10'2 6.26 X 102
1.5
3.6 X JO'2
4.2
1.66 x 103
2.7
6.4 X 10'2
7.5
1 Calculation is based on a mean respiratory requirement of 0.6 ^g C zoea ' d ' (from Table II), an
RQ of 0.9 and an assimilation efficiency of 0.70.
Reference: Biol. Bull. 165: 343-352. (October, 1983)
GROWTH AND REGENERATION RATES IN THINLY ENCRUSTING
DEMOSPONGIAE FROM TEMPERATE WATERS
AVRIL L. AYLING'
Marine Laboratory, University of Auckland, Leigh, New Zealand
ABSTRACT
Thinly encrusting species of subtidal sponge grow at slow but measurable rates
over natural surfaces by lateral spreading. Of the eleven species studied here, Aplysilla
rosea had the highest undisturbed rate of growth and Microciona sp. the lowest with
an overall negative change in size. Using the mean growth rate it can be estimated
that the largest sponge patches observed in the field may be over seventy years old.
Growth rates of individual patches were varied but this variation was not synchro-
nous within a species nor did it show any regular temporal pattern. Similarly,
no relation between the normal thickness of the species, the wet weight, or true
organic content of the species with undisturbed rates of growth could be found.
However, the mean patch size of the species was correlated with the undisturbed
growth rates. If the tissues of the sponges were damaged, rapid regeneration was
initiated at rates many times greater than the undisturbed growth rate of the species.
It was also found that even very small sponge patches could recover after almost
all living tissue was scraped from the rock.
INTRODUCTION
Almost no data exists on the rates of growth and regeneration, or estimates of
the age of, thinly encrusting species of marine sponge. Similarly, little information
is available for other invertebrate groups with a sheet-like growth form such as
compound ascidians, crustose bryozoans and corals (Jackson, 1979). It is thought
that growth in these types of sessile organisms is indeterminant, the colony increasing
exponentially in size with time (Jackson, 1977). From studies of sponge explant
outgrowths it has been shown that the tissue initially spread out is undifferentiated
and only slowly thickens and develops functional units (Simpson, 1963). The rate
of growth in subtidal thinly encrusting sponges is apparently slow. Bryan (1973)
studying a tropical species of Terpios over several weeks found that it could grow
over unoccupied space at a rate of 0-0.02 mnr/cm border/day but this rate increased
to 0.08-0.10 mnr/cm border/day when the sponge grew over living coral. In a
temperate water community A. M. Ayling ( 198 1 ) found over a month's study period
that Stylopus sp. grew at an undisturbed mean rate of increase of 0.02 mm2/cm
border/day.
The growth of intertidal sponges (usually of thickly encrusting habit) has been
studied in a little more detail. Although these species are not directly relatable to
the subtidal thinly encrusting forms because of their seasonal life history modifi-
cations (see Fell, 1976), some of the features of their actual growth are pertinant.
Fell and Lewandrowski (1981) found that the smallest and largest sponge patches
of Halichondria sp. grew the most slowly. Patches of these sponges could also merge
Received 20 December 1982; accepted 1 July 1983.
1 Present address: P.M.B. 1, Daintree, Queensland 4873, Australia.
343
344 A. L. AYLING
and disintegrate (Elvin, 1976; Johnson, 1978; Fell and Lewandrowski, 1981), a
feature also observed in corals (Hughes and Jackson, 1980). The rates of such fission
and fusion processes may have an important bearing on estimates of age in these
organisms.
A. M. Ayling (1981) found that when the thinly encrusting sponge Stylopus sp.
was damaged in an experiment simulating the effects of urchin grazing, the sponge
tissues regained lost space at a rate 200 times the normal growth rate of the species.
This regeneration rate refers to the formation of a thin layer of tissue over the lost
space, not to the production of the normal thickness of the species. The large dif-
ference between growth and regeneration rates may help explain the abundance of
thinly encrusting forms of sponges in areas subject to grazing and other disturbances
(Jackson and Palumbi, 1978; A. M. Ayling, 1981). The high 'growth' rates attributed
to sponges in recolonization experiments (Kay and Keough, 1981) may also involve
this regeneration process. Other examples of regeneration rates reported by A. M.
Ayling (1981) range from 1.6 mm2/cm border/day for Tedania sp. (orange) to 4.0
mnr/cm border/day for Anchinoe sp. (yellow).
This paper provides growth rates for eleven species of thinly encrusting subtidal
sponge taken from two years monitoring of sponge patches in the natural habitat.
From these data estimates of longevity were derived. Data were also collected on
wet and dry weights and true organic content and related to the growth rate. Sim-
ilarly, the effect of seasonal and reproductive state of the sponges on growth rates
is considered. The regeneration rates of the sponges were experimentally determined
and the ability of small sponges to survive damage also investigated.
MATERIALS AND METHODS
Description of study area
With the exception of one species, all the thinly encrusting sponges were located
on the walls of a narrow canyon, 12 m in depth, on the exposed north-easterly side
of Goat Island, a small island near the Leigh Marine Laboratory off the north coast
of New Zealand (38° 16'S: 174° 48'E). The other species, Eurypon sp., was found
only in the Sponge Garden at a depth of 18 m north-west of Goat Island. This
species was abundant beneath a layer of sand between 2 and 5 cm in depth. The
physical characteristics of the Goat Island area are summarized in Leum and Choat
(1980) and A. M. Ayling (1981). The abundance of the sponges is given in A. L.
Ayling (1978).
Wet weight, dry weight, and composition of living sponges
Five or more pieces of each species were collected still attached to the rock
substratum and transferred to the laboratory where the area of the sponge was traced
onto acetate sheet and thickness measured. The tissues were then carefully removed
with a scalpel and paint brush, placed on filter paper and weighed. Sponges were
placed in a drying oven at 90°C until constant weight was obtained. A wet weight/
dry weight ratio was calculated and the dried residue of the sponge further examined
for ash (assumed to be all SiO2 for siliceous sponges), water of hydration and organic
fractions. At normal drying temperatures (80-100°C) the water of hydration is only
partly removed from the siliceous skeleton. As ash values can thus be underestimated
(Vinogradov, 1953; Paine, 1964) corrections were made by measuring the weight
loss of spicule samples after incineration. Spicule samples were collected from two
SPONGE GROWTH AND REGENERATION 345
species. Tissue samples from the two species were digested in Sodium hydroxide
and then repeatedly washed in distilled water. Cleared spicules were dried at 90°C,
weighed and incinerated at 500°C for four hours. All species were ashed at 500°C
for four hours.
Growth rates of sponges over natural habitat
Ten or more patches of varying sizes of each encrusting species were selected
and marked with labeled masonry nails driven into the rock adjacent to the sponge
patch. At the end of the study only those patches which had not suffered visible
damage from grazing or other sources of disturbance were chosen for estimating
growth rates. It is possible that some of these 'undamaged' sponges may have suffered
minor injuries and regenerated between monitoring intervals. Preliminary moni-
toring of growth at weekly then monthly intervals showed no measurable changes
in size in most of the species and hence monitoring was continued at three monthly
intervals over a two year period (June, 1976-June, 1978). Sponge patches were
photographed and the color negatives projected at actual size onto graph paper and
the outlines of the sponge traced. The area cover of each sponge was recorded with
an estimated error of ±0.5%.
Damage simulation experiments
A ten centimeter square was outlined on the surface of the sponge and then
scraped almost clean of tissue to simulate the grazing activities of the abundant
urchin Evechinus chloroticus. Five sponge patches of each species were then cleared
and black and white photographs were taken of the damaged areas. Cleared areas
were rephotographed a month later and percentage regeneration measured.
Can small sponges survive damage?
The recovery capability of small sponges was investigated by scraping patches of
between 0.1-42.0 cm2 area of the species Microciona sp. and Stylopus sp. almost
completely off the rock. After two weeks the percentage recovery of the original area
was recorded.
RESULTS
Wet weight, dry weight and composition of living sponges
Wet weight and dry weight per unit area of the sponge is shown in Table I. The
species with the highest wet weight per centimeter square tissue were Tedania sp.
(orange) and Hymedesmia sp. (orange). The high wet weight of Chondropsis sp. is
due to the inclusion of sand in its skeleton.
A wet/dry weight ratio was calculated and the ash, water of hydration and the
organic fractions of each species obtained (Table II). Results from this analysis
indicate that the species with the least proportion of organic matter in their body
include species where spongin forms a major part of the skeleton (Chelonaplysilla
sp.), or sediments (Chondropsis sp.) or the sponge produced large quantities of mucus
(Tedania sp. (orange)). In general these temperate water encrusting sponges had a
greater proportion of organic matter, but less water content than the species from
Antarctica analyzed by Dayton et al. (1974).
346
A. L. AYLING
TABLE I
Thickness, mean patch size, wet weight, and dry weight of thinly encrusting sponges
Wet weight
Dry weight
Mean patch
•K
g/cnr
tissue
g/cm2
tissue
No.
Thickness
size (cm )
Species
samples
(mm)
area
X
Sx
X
Sx
Sty/opus sp.
6
3-10
58.4
0.16
0.06
0.05
0.02
Hymedesmia sp.
(orange)
5
5
8.2
0.19
0.08
0.08
0.04
Hymedesmia sp.
(red)
8
2
13.6
0.14
0.02
0.09
0.03
Microciona sp.
14
3
22.2
0.07
0.01
0.03
0.004
Anchinoe sp.
5
2-15
21.9
0.06
0.01
0.03
0.01
Stylopus sp. (pink)
10
2-5
30.7
0.13
0.08
0.02
0.003
Tedania sp.
(orange)
7
5-15
14.5
0.19
0.04
0.04
0.01
Chondropsis sp.
16
5-20
45.8
0.37
0.03
0.16
0.01
Aplysilla rosea
7
2-6
151.8
0.09
0.02
0.04
0.01
Chelonaplysilla sp.
7
3-5
83.8
0.09
0.02
0.05
0.01
Eurypon sp.
10
1-2
7.8
0.03
0.003
0.02
0.01
Natural growth rates
The thinly encrusting sponges grew in slow but measurable increments over the
two year study period. In the majority of cases this growth was not a steady un-
interrupted process; during a year a single sponge patch could stop growing or retract
from areas it had occupied. Whether this retraction was spontaneous or due to
TABLE II
Composition of living sponges*
A
B
Proportion
C
Proportion
of dry wt.
D
Proportion
E
Proportion
true organic
Proportion
dry
that is false
true ash B
matter
Species
N,
H2O ± SE
(1.000 - A)
N2
ash ± SE
X C/0.91
(B-D)
Stylopus sp.
6
.695 ± .026
.305
5
.645 ± .047
.196
.109
Hymedesmia sp. (red)
8
.502 ± .030
.498
5
.327 ± .066
.115
.383
Hymedesmia sp.
(orange)
5
.664 ± .058
.335
5
.569 ± .017
.209
.126
Stylopus sp. (pink)
10
.560 ± .174
.440
5
.454 ± .017
.219
.221
Tedania sp. (orange)
7
.787 ± .014
.213
5
.710 ± .061
.166
.047
Microciona sp.
14
.544 ± .033
.456
5
.581 ± .011
.201
.165
Anchinoe sp.
5
.457 ± .180
.643
5
.409 ± .02 1
.289
.354
Chondropsis sp.
16
.564 ± .010
.436
5
.692 ± .160
.414**
.022
Chelonaplysilla sp.***
7
.425 ± .070
.550
5
.938 ± .043
.516
.034
Aplysilla rosea***
7
.542 ± .075
.458
5
.418 ± .023
.191
.267
Eurypon sp.
5
.716 ± .041
.783
—
—
—
—
* N, = number specimens used for determination of proportion H2O (A); N2 = number of specimens used for
determination of proportion of false ash (C). True ash is false ash/0.91 - water held by spicules. The composition of
sponges is given by (A) = (D) + (E).
** True ash is false ash/0.729 - sand and spicules.
*** Sponges without spicules, true ash (B) (C).
SPONGE GROWTH AND REGENERATION 347
undetected disturbance could not be determined in this study. When individual
changes in patch size were graphed no correspondence in fluctuations were apparent
or referable to seasonal or reproductive cycles (see A. L. Ayling, 1980 for the
reproductive cycles of four of the species studied here).
A mean growth rate was calculated for each species of sponge, the large standard
errors reflecting the above mentioned fluctuations in size. Growth rates are presented
as millimeter square area change in size per centimeter border per day in Table III.
Patches of Aplysilla rosea, Stylopus sp. (pink) and Chondropsis sp. grew relatively
rapidly at 0.28, 0.23, and 0. 1 3 mnr/cm border/day respectively. It is estimated that
a Stylopus sp. (pink) of one centimeter diameter growing undisturbed could reach
a size of 1 5 cm diameter in ten years and the larger patches of this species observed
on the walls of the canyon which were one meter in diameter may be 78 years old
(based on the mean growth rate shown in Table III). If grazing was more frequent
than detected then these estimates of longevity should be considered minimum age
estimates. Eurypon sp. grew the most slowly of all the sponges, and patches of this
species were easily recognized even after six and a half years as the outlines of the
sponges changed very little (Fig. 1 ).
No significant relationship was found using the Spearman Rank Correlation
coefficient rs between wet weight and growth rates (rs = 0.52), thickness and growth
rates (rs = 0.508) or true organic content and growth rates (rs = 0.167). However,
a significant correlation was found between the mean patch size of a species and
growth rates (rs = 0.64: 0.5 > P > 0.01). Thus, in general, large species such as
Aplysilla rosea and Chelonaplysilla sp. grew more rapidly than the smaller species
e.g., Hymedesmia sp. (orange) and Eurypon sp.
The smaller sponges were more likely to fluctuate in size than the large indi-
viduals. This is shown for six of the species in initial size-increment graphs in
Figure 2.
Effect of grazing on sponges (regeneration rates)
The regeneration rates of the sponges are shown in Table IV. Sponges could
regenerate into disturbed areas at rates 22 to 2,900 times the natural growth rate.
TABLE III
Growth rates of thinly encrusting sponges over natural habitat
Postulated diameter
mm2/cm
border/day
of a 10 yr
old
Number
sponge using
mean
Species
patches
X
Sx
rate of increase (cm)
Aplysilla rosea
3
0.28
0.19
20.03
Stylopus sp. (pink)
11
0.23
0.09
15.39
Chondropsis sp.
16
0.13
0.09
9.31
Tedania sp. (orange)
22
0.08
0.05
5.70
Stylopus sp.
12
0.08
0.06
5.71
Chelonaplysilla sp.
13
0.06
0.05
5.00
Hymedesmia sp. (red)
10
0.05
0.03
4.31
Hymedesmia sp. (orange)
5
0.02
0.03
2.32
Anchinoe sp.
9
0.01
0.06
1.66
Microciona sp.
25
-0.01
0.003
—
Eurypon sp.
9
0.0003
0.031
1.02
Growth rates are presented as mean growth over a two year period.
348
A. L. AYLING
FIGURE 1 . Changes in outlines of patches of Eurypon sp. taken from color photographs over the
period June, 1975 ( ) to February, 1982 ( ) in a 25 cm2 area of the Sponge Garden.
The tissue covering these disturbed areas is initially thinner than the normal thick-
ness of the species. The species that most rapidly recovered space after damage were
Stylopus sp. (pink), Aplysilla rosea, Chondropsis sp, and Stylopus sp. However the
greatest magnitude of difference between growth and regeneration rates occurred
in the slow growing species Eurypon sp. and Anchinoe sp. Using the Spearman Rank
Correlation coefficient some relationship was found between regeneration rates and
undisturbed growth rates (rs = 0.91; P < 0.01) and regeneration rates and the mean
patch size of the species (rs = 0.64; 0.5 > P > 0.1).
Can small sponges survive damage?
All patches of the rapidly growing species, Stylopus sp. reoccupied some of the
lost space, the smallest patches recovering all of their former space in less than two
weeks. In some cases however, the slower growing species, Microciona sp. did not
recover any space nor the entire area even over a month (Fig. 3).
DISCUSSION
Growth over natural surfaces in thinly encrusting sponges from temperate subtidal
waters is very slow. The most rapid mean rate of growth recorded in this study was
that of a thin fleshy sponge, Aplysilla rosea, at 0.28 mnr/cm border/day. A settled
larvae of this species growing undisturbed could reach a size of 20 cm diameter in
ten years based on this mean rate of growth. Some of the patches of this species
growing on the walls of the canyon reached a meter diameter and these could be a
minimum of 50 years old. The slowest growing species was Eurypon sp., growing at
SPONGE GROWTH AND REGENERATION
349
\
\
»0-3-
* \
1-0-
* \
\
\
* 0-2-
•
\
•
0-5-
\
* 0-1-
» •
t
-0-1 -
100 200 300
v • 200 42°__r_
•_
0-5-
"
-0-2-
~- "
>s-0-3-
0
Stylopus sp.
1-0-
Stylopus sp. (pink)
t_
•o
o
.Q
\
• \
£ 0-3-
o
• \
• \
06-
\
\
~X 0-2-
\
0-i-
\
\
E
^ \
\
E 0-1-
• *
0-2-
V
*\'
* • "~ "~--
• ^^
• ^
L* 1 ' * I • 1 " '
N
50 _ __ •
L* » ' •' '
«" 0-1-
—
0-2-
^
c
•*•
» •'*
s
Q, 02-
0-4-
/
cn
Tedania sp ( orange )
Microciona sp
c
0 0-3-
0-6-
_c
o
\
\
1-0-
\
0-3-
x
•
•~-
.
•— .
• \
0-2-
•^
0-5-
\
*
• ~ —
0-1 -
'
'." 50 100_ 15Q_'
0-1-
, *. . 50
0-5-
s
• ^
0-2-
s
Chondropsis sp
Chelonaplysilla sp
1-0
0-3-
. area
Cm2 ) ^
FIGURE 2. Initial size-increment graphs of thinly encrusting sponges showing how small patches
generally fluctuated more in size than larger patches. Dashed lines outline the areas where there are no
points.
a mean rate of 0.0003 mrrr/cm border/day. In ten years the settled larvae of this
species would only grow to a size of one centimeter diameter. This species is very
thin and in the natural habitat forms small patches up to 10 cm in diameter, the
350
A. L. AYLING
TABLE IV
Regeneration rates of thinly encrusting sponges'
mm2/cm border/
day
Species
X
Sx
Times magnitude greater
than the natural
growth rate
Aplysilla rosea
6.18
0.98
22.07
Stylopus sp. (pink)
6.98
0.78
30.35
Chondropsis sp.
5.70
0.83
43.85
Tedania sp. (orange)
4.18
1.34
52.25
Stylopus sp.
4.60
0.70
65.70
Chelonaplysilla sp.
4.08
1.20
68.00
Hymedesmia sp. (orange)
0.53
0.43
26.50
Anchinoe sp.
3.65
0.89
365.00
Microciona sp.
0.63
0.23
**
Eurypon sp.
0.88
0.44 .
2,900
* Regeneration rates were obtained by stimulating damage to the sponge, five replicate simulations
per species. Hymedesmia sp. (red) is not included in the table as it was too small and divaricate to use in
the experiment.
** Undisturbed growth in this sponge was negative over the period of study.
outlines of which change very little over long periods of time. Microciona sp. had an
overall negative growth rate although the positively growing individuals of this species
achieved a growth rate of 0.02 mm2/cm border/day.
Every species had some individual patches which regressed over the two year
study period. In some cases the patch could increase over several months then
decrease in size. As fluctuations in size did not occur contemporaneously between
individuals no relationship could be found between changes in size and seasonal
and reproductive cycles. Changes in size did not occur over the entire border line
of the sponges but were restricted to certain sections of the border. Thus while
sections of the border could remain unchanged during the study other sections could
>>
l_
o;
o
o
100-
3fcf
80 -
60 -
X
40 -
X •
X *'
20 -
m«
X
0
X
**-
5 10
20
initial cm^
30
area
FIGURE 3. Can small sponges survive damage? Sponges were scraped almost entirely off the rock
and recovery of space was recorded after two weeks time. X = Microciona sp.; • = Stylopus sp.
SPONGE GROWTH AND REGENERATION 351
expand outwards or contract inwards. Neighboring sponges may help maintain static
border outlines and explain some tissue retractions (A. L. Ayling, in press) but
whether the removal of surrounding invertebrates may stimulate growth is uncertain
(A. M. Ayling, 1981).
The longevity of these thinly encrusting sponges may not be estimated correctly
if only the mean rate of increase is considered. Like corals (Hughes and Jackson,
1980) and intertidal sponges (Elvin, 1976; Fell and Lewandrowski, 1981), these
subtidal sponges could be broken into several fragments some of which may later
join. Thus a single patch may be the result of several fissions and fusions over time,
and the size of the sponge may not be indicative of the age of the patch. In general
these thinly encrusting sponges are likely to occupy space in the community for
long periods of time and consequently would be expected to play an important part
in the structuring of these encrusting communities where they are abundant.
The sponge species' tissue thickness did not affect the rate at which the sponge
grew over the substratum. For example, the thinnest sponge, Eurypon sp. grew the
slowest, while the thickest species Chondropsis sp. grew relatively rapidly. Nor did
the undisturbed growth rate of the different species relate to the wet weight or true
organic content. However, it was found that the larger species grew more rapidly
than the smaller species.
When thinly encrusting sponges are damaged a rapid regeneration mechanism
is activated and the sponge spreads out a thin layer of tissue over the disturbed area,
regaining the lost space. This thin tissue may be similar to the explant tissue ex-
amined by Simpson (1963) which was undifferentiated and contained only a few
cell types. The highest rate of regeneration recorded in the present study was that
ofStylopus sp. (pink) at 6.98 mnr/cm border/day, a magnitude of 30 times greater
than the undisturbed growth rate of the species. Even the slowest growing species,
Eurypon sp., rapidly regenerated tissue at a rate of 0.88 mm2/cm border/day, a
magnitude of 2,900 times the undisturbed growth rate of the species. This rapid rate
of encroachment after damage has obvious advantages in communities where grazers
are abundant. The survival chances of newly recruited sponges would also be en-
hanced by this regeneration mechanism.
ACKNOWLEDGMENTS
I would like to thank Dr. Tony Ayling, Dr. Howard Choat, Dr. David Schiel,
and Dr. P. R. Bergquist for their helpful discussion of the project. This research was
supported by a grant from the Roche Pharmaceutical company of Australia.
LITERATURE CITED
AYLING, A. L. 1978. The relation of food availability and food preferences to the field diet of an echinoid
Evechinus chloroticus (Valenciennes). /. Exp. Mar. Biol. Ecol. 33: 223-235.
AYLING, A. L. 1980. Patterns of sexuality, asexual reproduction and recruitment in some subtidal marine
Demospongiae. Biol. Bull. 158: 271-282.
AYLING, A. L. In press. Factors affecting the spatial distributions of thinly encrusting subtidal sponges
from temperate waters. Oecologia.
AYLING, A. M. 1981. The role of biological disturbance in temperate subtidal encrusting communities.
Ecology 62: 830-847.
BRYAN, P. G. 1973. Growth rate, toxicity and distribution of the encrusting sponge Terpios sp. (Had-
romerida: Subertidae) in Guam, Marianas Islands. Micronesica 9: 237-242.
DAYTON, P. K., G. A. ROBILLIARD, R. T. PAINE, AND L. B. PAINE. 1974. Biological accommodation
in the benthic community at McMurdo Sound, Antarctica. Ecol. Mon. 44: 105-128.
ELVIN, D. W. 1976. Seasonal growth and reproduction of an intertidal sponge, Haliclona permollis
(Bowerbank). Biol. Bull. 151: 108-125.
352 A. L. AYLING
FELL, P. E. 1976. Analysis of reproduction in sponge populations: an overview with specific information
on the reproduction of Haliclona loosanqffi. Pp. 51-67 in Aspects of Sponge Biology, F. W.
Harrison and R. R. Cowden, eds. Academic Press, New York.
FELL, P. E., AND K. B. LEWANDROWSKJ. 1981. Population dynamics of the estuarine sponge, Hali-
chondria sp., within a New England eelgrass community. J. Exp. Mar. Biol. Ecol 55: 49-63.
HUGHES, T. P., AND J. B. C. JACKSON, 1980. Do corals lie about their age? Some demographic conse-
quences of partial mortality, fission, and fusion. Science 209: 713-715.
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. 1979. Morphological strategies of sessile animals. Pp. 499-555 in Biology and Sys-
tematics of Colonial Organisms, G. Larwood and B. R. Rosen, eds. Academic Press, London.
JACKSON, J. B. C., AND S. R. PALUMBI. 1978. Regeneration and partial predation in cryptic coral reef
environments: Preliminary experiments on sponges and ectoprocts. Pp. 303-308 in Biologie des
Spongiaires, C. Levi and N. Boury-Esnault, eds. C. N. R. S., France.
JOHNSON, M. F. 1978. Recruitment, growth, mortality and seasonal variations in the calcareous sponges
Clathrina coriacea (Montagu) and C. blanca (Miklucho-Maclay) from Santa Catalina island,
California. Pp. 271-282 in Biologie des Spongiaires, C. Levi and N. Boury-Esnault, eds. C. N.
R. S., France.
KAY, A. M., AND M. J. KEOUGH. 1981. Occupation of patches in the epifaunal communities on pier
pilings and the bivalve Pinna bicolor at Edithburgh, South Australia. Oecologia 48: 123-130.
LEUM, L. L., AND J. H. CHOAT. 1980. Density and distribution patterns of the temperate water marine
fish Cheilodactvlus spectabilis (Cheilodactylidae) in a reef environment. Mar. Biol. 57: 327-
337.
PAINE, R. T. 1964. Ash and calorie determination of sponge and Opistobranch tissues. Ecology 48: 384-
387.
SIMPSON, T. L. 1963. The biology of the marine sponge Microciona prolifera (Ellis and Solander). A
study of cellular function and differentiation. J. Exp. Zool. 154: 135-152.
VINOGRADOV, A. P. 1953. The elementary chemical composition of marine organisms. Sears Found.
Mem. 2: 1-647.
Reference: Biol. Bull. 165: 353-369. (October, 1983)
SURFACE BROODING IN THE RED SEA SOFT CORAL
PARERYTHROPODIUM FULVUM FULVUM (FORSKAL, 1775)
Y. BENAYAHU* AND Y. LOYA
Department of Zoology, The George S. Wise Center for Life Sciences,
Tel Aviv University, Tel Aviv 69978 Israel
ABSTRACT
Parerythropodium fulvum fulvum (Forskal, 1775) is an encrusting soft coral com-
monly found between 3 and 40 m, at the coral reefs of the Gulf of Eilat. The annual
gonadal development and sexual reproduction of this species were studied both in
shallow water (3-5 m) and in the deep reef zone (27-30 m). P. f. fulvum is a dioecious
species. Sex ratio of the shallow population favors higher abundance of females, while
on the deep reef a 1 : 1 sex ratio was recorded. These differences are probably due to
local aggregations of colonies of the same sex caused by asexual reproduction. Oocytes
and sperm sacs are found even in very young colonies (1-3 years). The frequency of
sexually mature males is higher than mature females among small corals.
Young oocytes appear annually in August and within 10-11 months reach their
maximal diameter. Sperm sacs start to develop later and mature after 7-9 months.
A marked synchronization in the development of the oocytes and the testes exists
among different polyps within each colony. Spawning occurs at dusk, and is fully
synchronized by lunar periodicity (a few days after the new moon and a few days
preceeding its last quarter). Fertilization takes place inside the polyp cavities. The
shallow water population breeds prior to the deeper one with the whole reproductive
period lasting approximately two months (end of June, beginning of August).
Among anthozoans, P. f. fulvum represents a unique mode of sexual reproduc-
tion and planulae development. This species is oviparous, yet eggs cleave on the
surface of the female colonies while entangled in a mucoid suspension. We term
this mode of planula development "surface brooding". Within 6 days after fertil-
ization the planulae complete their development, detach from the surface of the
colony, and sink to the bottom.
The encrusting growth form of P. f. fulvum is characterized by a thin coenen-
chyme and short polyp cavities, yet the eggs exhibit a large diameter (500-700 ;um).
Egg production of P. f. fulvum is rather low (18-24 eggs per polyp), but it is com-
pensated for by surface brooding, which protects the offspring during embryogenesis.
It is suggested that surface brooding is an adaptation to the encrusting shape of the
colony and it maximizes fecundity.
INTRODUCTION
The soft corals (order Alcyonacea) are a large and diverse group of species among
the Octocorallia. Several studies deal with alcyonacean distribution emphasizing
their importance as space utilizers (Cary, 1931; Maragos, 1974; Veron et al, 1974;
Schuhmacher, 1975; Pearson, 1981). Other investigations discuss their ecological
Received 14 March 1983; accepted 25 July 1983.
* Present address: Florida International University College of Arts and Science, Department of
Biological Sciences, Tamiami Campus, Miami, Florida 33199.
353
354
Y. BENAYAHU AND Y. LOYA
importance in the Red Sea coral reefs (Fishelson, 1970, 1973; Benayahu and Loya,
1 977, 1981). Despite their abundance on many Indo-Pacific coral reefs (Bayer, 1 973),
little information exists on their life history and reproductive tactics.
Most of our knowledge on the reproduction of alcyonacean corals is based on
early literature dealing with the widespread boreal species Alcyonium digitatum
(Linnaeus, 1758) (Lacaze-Duthiers, 1865; Hickson, 1895; Hill and Oxon, 1905;
Matthews, 1917). More recently, this species has been investigated by Hartnoll ( 1 975,
1977). Extensive studies have been carried out on the Red Sea soft corals of the
family Xeniidae (Gohar, 1940a, b; Gohar and Roushdy, 1961). These studies are
mainly concerned with the biology and reproduction of Heteroxenia fuscescens
(Ehrenberg, 1834). Recently, Yamazato and Sato (1981) have studied the repro-
ductive biology of Lobophytum crassum Von Marenzeller, 1886.
Approximately 200 alcyonacean species have been recorded from the Red Sea
(Benayahu and Loya, in prep.), but little is known about their life histories. The
present work summarizes the results of a four-year quantitative study on the ecology
and the reproductive pattern of Parerythropodium fulvum fulvum (Forskal, 1775)
(family Alcyoniidae). This species was originally described from the Red Sea, but
its present zoogeographical distribution extends to the reefs of Madagascar and east
to Indonesia (Verseveldt, 1969).
Colonies of P. f. fulvum have an encrusting membranaceous growth form (Fig.
1 ), and is among the most abundant soft corals on the coral reefs of the Gulf of
Eilat (Benayahu and Loya, 1977). This paper is concerned with the distribution and
reproductive strategy of P. f. fulvum in shallow water (3-5 m) and in deeper reef
zones (27-30 m). We have studied the annual development of gonads, sex ratio,
colony size at first reproduction, and the mode and duration of sexual reproduction.
In addition, we examined the chronology of planulae embryogenesis, as well as the
post-larval development and morphogenesis. This study describes surface brooding,
a unique mode of external planulae development among the alcyonacean corals.
FIGURE 1 . A living colony of Parerythropodium fulvum fulvum.
SURFACE BROODING IN A SOFT CORAL 355
MATERIALS AND METHODS
The present study was carried out at two reef localities. One site was Muqebla',
12 km south of Eilat, where the shallow water population at 3-5 m depth was
studied, the deep water population (27-30 m) was studied near the Marine Biological
Laboratory of Eilat. Distributional studies and the correlation between spawning
periodicity and depth were also carried out at this location. Sampling, underwater
measurements, and observations were carried out by SCUBA diving. The living
coverage and abundance of P. f. fulvum were studied by a series of line transects
(10 m each) following the method described by Loya and Slobodkin (1971).
In order to determine the relationship between colony size and the onset of
sexual maturity, small colonies were collected prior to the breeding season. These
colonies were carefully removed from the substrate by forceps and were preserved
in 4% buffered formalin. In the laboratory, each colony was numbered, its boundaries
outlined on paper and then the drawings cut out by scissors. Each piece of paper
was separately weighed using an analytical balance with a precision of 10"4 g. The
weight of the paper pieces increased linearly with the colonies surface, and they
represented the size of the corals.
The populations at the two reef sites were studied during approximately 4 years,
from November 1977 to July 1981. Almost every month, fragments of 10-20 large
colonies were randomly sampled in Muqebla' (3-5 m) and in the Marine Biological
Laboratory (M.B.L.) reef (27-30 m). Ten large colonies were numbered with plastic
tags in shallow water and on the deep reef. Fragments of these colonies were sampled
every month during 3 years, to study the annual sequence of gonadal development
within the same colony.
The polyp cavities of the formalin-fixed material were examined with a binocular
stereoscope for genital development and sex determination. Additionally, wet
mounts of septa with gonads from 25 polyps of each colony were examined micro-
scopically; the diameter of the oocytes and sperm sacs was measured. Paraffin sec-
tions (10 ^m) were employed to study gonadal structure. Sections were stained in
hematoxylin (Delafield) and eosin after decalcification in formic acid-citrate (Rink-
evich and Loya, 1979a).
Preliminary observations during the summers of 1978 and 1979 revealed that
spawned eggs of P. f. fulvum remained on the surface of the colonies. During the
summers of 1980 and 1981, prior to the breeding season, female colonies were
collected and maintained in aquaria with running sea water. Determination of the
exact timing of egg expulsion was done by continuous observations in the laboratory,
and in the field along a depth gradient to 30 m. Fertilized eggs were reared in aerated
sea water containers. Cleavage stages were compared to field material collected
successively every 12 h. Synchronization of egg cleavage was studied by examining
hundreds of embryos.
Material for scanning electron microscopy was fixed in 2% glutaraldehyde. After
dehydration in a series of graded ethyl alcohols, the samples were dried from liquid
CO2 by the critical point method. The dried preparations were coated with gold and
examined with a Jeol-S35 scanning electron microscope at 25 kV.
RESULTS
Abundance and depth distribution
The abundance of P. f. fulvum in shallow water is extremely variable. Previous
results indicated that its coverage varies from 1.1% to 44.0% on different reef flats
356
Y. BENAYAHU AND Y. LOYA
and from 7.0% to 45.6% on different fore-reef zones (Benayahu, 1975). The present
study across the M.B.L. reef indicates a lower living coverage (5.1 ± 2.3%) per 10
m transect at 18-40 m depth. The colonies tend to aggregate: young colonies are
almost always found growing near larger ones. The smaller individuals are often
found in poorly illuminated environments such as crevices or the undersides of dead
stony corals.
Colonies of P. f. fulvum exists in two color morphs: yellow-brown and gray, but
there is no taxonomic difference between them (Verseveldt, 1969). Figure 2 exhibits
the depth distribution of the two morphs from shallow water to a depth of 30 m.
Coral abundance is expressed as number of colonies per 10 m transect. The yellow
brown colonies are the most common, while the gray corals are less abundant.
Whereas the yellow-brown morph is found along the whole depth range studied,
the gray morph is common only below 20 m. This pattern of distribution was
qualitatively observed in many other reef localities along the coral reefs of the Gulf
of Eilat.
Gonadal development
P. f. fulvum is a dioecious species. In both sexes the gonads develop on the four
lateral and two ventral mesenteries of the polyp. Each polyp produces 18-24 genital
products. The oocytes and the testes are located on the middle part of the mesentery
and directed towards the center of the polyp cavity. Occasionally, few colonies of
P. f. fulvum contain parts with thick coenenchyme. In such polyps the mesenteries
may exceed a length of 6-12 mm, whereas in the most common ones they are only
a few mm long. In the thick coenenchyme polyp-type, where much more space is
Yellow-brown colonies
0
Gray colonies
12 15 18 21 24
Depth (meters)
27 30
FIGURE 2. Depth distribution of the two color morphs of Parerythropodium fulvum fulvum. The
abundance in terms of number of colonies per 10 m transect.
SURFACE BROODING IN A SOFT CORAL 357
available, up to 100 eggs or sperm sacs may develop. Measurements of the diameter
of the oocytes and sperm sacs indicate a marked synchronization in the reproductive
state among different polyps within each colony (see below). No sex changes were
detected during the study within the 20 tagged colonies.
Oocytes of living colonies of the abundant morph are characterized by a lemon-
yellow color, while sperm sacs are transparent yellow. After preservation in formalin
or alcohol their color becomes paler. The oocytes of the gray colonies are opaque-
gray, while the testes are very transparent.
Size at sexual maturity and sex ratio
A few weeks before the spawning period (early June), 216 small (young) colonies
were randomly collected in order to determine the minimum size at sexual maturity.
We define a sexually mature specimen as one having either ripe spermatozoa or ripe
oocytes (see below). The surface area of the sampled colonies ranged from less than
1 cm2 to a maximum size of 5-7 cm2. Table I represents the breeding state of these
colonies in all size groups. Oocytes and sperm sacs are found even in the smallest
colonies, but the frequency of mature males is higher than that of mature females.
In addition, the percent of colonies with gonads increases with colony size.
Information on the population sex ratio was derived from samples collected
during May-June, throughout the entire study. In shallow water 28 1 large colonies
were examined, of which 60% were females. A X2 test, at 0.05 level, indicates a
significant deviation from a 1:1 sex ratio. A total of 220 colonies collected at 30 m
depth resulted 54% males, indicating a 1:1 sex ratio (P > 0.050).
Annual cycle of gonadal development
Figure 3 demonstrates the relative percentage of colonies with oocytes or testes
in each monthly sample. Figure 3a represents the results obtained from shallow
water and Figure 3b represents the results obtained from the deep-reef. The per-
centage of colonies in the population without gonads fluctuates during an annual
cycle, due to the timing of their development. Yet, Figure 3 shows that only a low
percentage of colonies does not contain gonads prior to the spawning season.
TABLE I
Relationship between the size of young colonies and onset of reproduction of
Parerythropodium fulvum fulvum
No. of
Percent
Group size
No. of
No. of
No. of
immature
colonies
(weight)*
colonies
males
females
colonies
with gonads
1-10
60
4
1
55
8.3
11-20
77
14
0
63
18.2
21-30
26
11
2
13
50.0
31-40
21
10
2
9
57.1
41-50
16
6
5
5
68.8
51-60
6
4
2
0
100.0
>60
10
3
7
0
100.0
Total
216
52(24.1%) 19(8.8%)
145 (67.1%)
* Weight
of paper image!
i (in g 10~4)
determined the size gr<
)up (see Materials
and Methods for
further explanation).
358
Y. BENAYAHU AND Y. LOYA
90
60-
c
o
—
030
ifi
"1 °
_0
o
c
Hi
0
Females
D
Males
(
a
r>
— i
v,.
-:
7!
f'
••'
—
—
pr
";
J1.
~~ •
'-;
—
rrr
—
-T
>>
>::
^
-»-
••
•>
'•.
,'j
-.*,
•
;?
-
J5
:-'•
£
|j
;-V
'.;'
3
.-:•
;;
{r
.•.^
;<•:
f.;
:,;_
r?
*
;:•.
^
^ I
^
j t
'.1
'-••
fi
1
1
'; I
V
:::
if.
J:
-r
•i;
£
[.
N
j
M
M
I
S
N
J
M
M
J
S
N
J
M
M
j
S
N
J
M
M
J
60 ;;' n
30
FIGURE 3. Abundance of female and male colonies of Parerythropodiumfulvumfulvum with gonads
in each monthly sample. Figure 3a represents results obtained from 3 m depth and Figure 3b represents
the results from 30 m depth. The blank spaces in some of the months indicate that no sampling was
done that period.
Figure 4 represents the annual changes in the mean maximal diameters of oocytes
and sperm sacs in shallow water (Fig. 4a) and in deep water (Fig. 4b). The first
young oocytes appear in August. They grow rapidly and within 10-11 months reach
their maximal size. The diameter of the largest oocytes was 700 ^m, however, the
majority of the ripe oocytes ranged in size from 400 to 600 ^m. Figure 4 also
demonstrates that the annual development of the sperm sacs starts a few months
after oocyte initiation. The first young spermaries are found every year during Oc-
tober, although their appearance can be delayed in part of the population until
December. The development of the sperm sacs generally takes 7-9 months. The
largest reach 480 /um, although the common diameter at maturity is about 400 ^m.
Spawning occurs mainly during June-July. The annual development of female and
male gonads exhibited the same pattern throughout the research period (Fig. 4).
This pattern is markedly synchronized within the population as indicated by the
low standard deviations around the mean maximal diameters of the oocytes and
sperm sacs.
SURFACE BROODING IN A SOFT CORAL
359
400-
in
D
£200
0
D\ f A J A O D
1978
1979
1980
F A J
1981
FIGURE 4. Mean maximal diameters of oocytes and sperm sacs of Parerythropodium fulvum fulvum
at 3 m depth (Fig. 4a) and 30 m depth (Fig. 4b).
Ultrastructure of the gonadal surface
The oocytes and the sperm sacs of P. f. fulvum are surrounded by a ciliated
follicular layer (Fig. 5a, b). These cells are derived from the endodermal epithelium
of the septa. Each oocyte or testis is attached to the mesentery by a pedicle of
approximately 100 nm (Fig. 5a). The cells of the polyp cavity are covered by cilia
of about 20 nm in length. The flagella of the gonadal surface and that of some other
endodermal cells are located in small pits. Each flagellum is surrounded at its base
by 8 elevated folds of cell surface, in a palisade formation (Fig. 5c), similar to the
arrangement described by Mariscal and Bigger (1976) in other octocorals. SEM
Y. BENAYAHU AND Y. LOYA
FIGURE 5. Ultrastructure of the gonadal surface of Parerythropodium fulvum fulvum. a: an oocyte
attached with a pedicle to the mesentery. Bar = 100 ^m. b: ciliary follicular endoderm of an oocyte. Bar
= 10 fj.m. c: endodermal cilium surrounded by 8 elevated folds, cilium base (C). Bar = 10 ^m. d: outer
surface of a sperm sac. Bar = 10 //m. e: sperm cells on the testis. Bar = 1 ^m. f: magnified sperm cell
flagellum (F), sperm cell (SP). Bar = 1
examination reveals that the outer surface of the sperm sacs is elevated into hillocks
and fold-like crests (Fig. 5d). In addition, microvilli and cilia are located among
them (Fig. 5d, e). Immature sperm cells are found attached to the surface of the
testes. They probably burst the spermary wall during fixation (Fig. 6e, f)- The di-
ameter of their rounded head is 2 /tin, while their tail exceeds a length of 12 /mi.
Spawning, fertilization, and embryogenesis
After spawning, all the eggs of P. f. fulvum remain on the surface of the female
colonies, where they develop into planula larvae (Fig. 6a). The lemon-yellow color
of the eggs make them very apparent even from a distance of several meters. The
eggs are suspended in transparent, gelatinous material secreted by the corals. This
SURFACE BROODING IN A SOFT CORAL
361
FIGURE 6. Spawning ofParerythropodiumfiilvumJulvum a: colony covered by spawned eggs embed-
ded in mucus, b: eggs and sclerites entangled in mucus (X10).
mucus cover also contains many sclerites which are torn from the polyps during
egg expulsion (Fig. 6b). Various organic and inorganic particles adhere to the mucus.
The mucus flocks remain on the surface of the colonies for a week, and during this
period cleavage takes place within the mucus.
Successive observations suggest that shortly before spawning the eggs of P. f.
fulviim are fertilized within the polyp cavities. Thus, some female colonies that were
kept in aquaria during the breeding season failed to spawn. Nevertheless, cleavage
occurred inside their polyps. Additional evidence supporting internal fertilization
was detected by SEM observations. Fixation of female colonies a few hours after
egg expulsion revealed clusters of mature spermatozoa along the mesenterial fila-
ments.
The eggs of P. f. fulvum are of the telolecithal type. Normally, cleavage occurs
on the surface of the female colonies. The fertilized eggs lack a follicular layer, which
is most probably detached before fertilization (Fig. 7a). Cleavage of the eggs begins
within 3-5 h after fertilization. The first two divisions are meridional and equatorial
(Fig. 7b). Throughout cleavage highly irregular, lobed structures are formed (Fig.
7c). The holoblastic, unequal cleavage produces a morula with large cells at the
vegetal pole and smaller cells at the animal pole (Fig. 7d). Further divisions 24 h
after fertilization lead to the formation of a round blastula (Fig. 7e). Histological
sections indicate that this is a steroblastula, lacking a blastocoel. The thin external
cell layer forms a cortex, while the inner cells are filled with yolk platelets.
The surface of the blastula (Fig. 8a) is characterized by folds and microvilli 1-
2 yum long. Numerous microvilli are located between the neighboring cells (Fig. 8b).
Ciliated ectodermal cells are recognized at a later stage on the young developing
planula (Fig. 8c). During the third day after fertilization the diameter of the embryo
is 350 ^m (Fig. 70- After four days a gastrula develops with a length of 600 ^m
(Fig. 7g). A young planula bearing an oral opening is found one day later (Fig. 7h).
The young planula is rounded and gradually changes to an egg-like and then a pear-
like shape (Fig. 7i). At this stage the young larvae are motionless, still embedded
in the mucus. By the 6th day the planulae elongate; their aboral end is tapered while
the oral side is rounded.
362
Y. BENAYAHU AND Y. LOYA
FIGURE 7. Embryogenesis of the planula larva of Parerythropodium fulvum fulvum. Bar = 100
. a: an egg without follicular layer, b: first two divisions of the egg. c: young embryos, d: irregular
embryos, e: 24 h blastula. f: 48-72 h blastula. g: gastrula, 4 days after fertilization, h: young planula,
arrow points to mouth opening, i: mature planula.
Planulae structure and behavior
Seven days after fertilization the mucus with the mature planulae in it starts to
detach from the surface of the colonies and sink near the "mother colony" (Fig.
8d). The mucoid substance starts to degrade, and the larvae begin to move with
their cilia. Figure 8e presents a fractured mature planula, where dense ciliary ec-
SURFACE BROODING IN A SOFT CORAL
363
FIGURE 8. Planula structure and post larval development of Parerythropodium fulvum fulvum. a:
blastula cells. Bar = 10 ^m. b: microvilli (MV) on the surface of 24 h blastula cells, bar = 1 //m. c: ciliated
blastula cells. Bar = 10 ^m. d: mature planula (XI 8). e: fracture of mature planula, cilia (C), ectoderm
(EC), endoderm (EN) Bar = 10 nm. f: fractured mature planula, ectoderm (EC), mesoglea (ME), en-
dodermic vacuole (EV). Bar = 10 nm. g: 12-16 day old polyp (X8). h: young colony, arrows indicate
buds of young polyps.
toderm and endodermal cells can be seen. The mesoglea of the planula is very thin,
bounded by vacuolated endodermal cells (Fig. 80, which probably serve for yolk
storage. The cilia are uniformly scattered on the ectodermis, however due to the
larval contractions, they might be hidden among the body folds. The planula larvae
are elongated, barrel-shaped and recognized by their typical lemon-yellow color.
When fully extended their maximal length reaches 2.4-3.2 mm. During the first
days after maturation the planulae tend to change their shape by body contractions,
from elongated to rounded and vice versa. Most of the time the larvae are attached
to the substrate on their oral side by mucus secretion. Occasionally swimming is
observed, typified by a corkscrew rotation along the oral-aboral axis. The larvae also
tend to crawl over the substrate for short distances of a few cm.
Post- larval development and formation of a young colony
Laboratory and underwater experiments dealing with substrate selection by the
planulae (Benayahu and Loya, in prep.) have enabled us to follow the morphological
changes occurring during planulae metamorphosis. Development within the plan-
ulae population is not synchronized; differences in the developmental stages in the
same age group may vary by as much as 3 to 5 days. During the first 3-7 days after
planulae maturation they attach to the substrate and develop into young cone-
shaped polyps, surrounded by 8 tentacular buds. During days 7-10 the tentacles
elongate, and 8 septa are observed inside the polyp cavity. The development of the
first pair of tentacular pinnules occurs during days 11-12. In days 12-16 an addi-
tional 4-7 pairs of pinnules develop on each tentacle (Fig. 8g). Within the next
364 Y. BENAYAHU AND Y. LOYA
month 2-3 secondary polyps develop in the young colony, and sclerites are seen
within the polyp body (Fig. 8h).
Rhythmicity of spawning
Table II presents the timing of egg expulsion in the population of P. f. fulvum.
The dates in the table represent the first day of each spawning (which may last 2-
3 days). Successive underwater observations indicate that spawning starts around
the middle of June and lasts for approximately two months. The process begins at
dusk, and corresponds to a lunar periodicity, lasting from a few days after the new
moon to a few days preceeding its last quarter.
Although egg expulsion is synchronized, it does not occur simultaneously within
the population. A sample of 1 30 colonies was examined underwater at the beginning
of the breeding season in June 1978, a few days after first spawning was observed.
The majority of the colonies from both sexes had not yet spawned, and only a minor
number had shed part of their gametes.
Figure 9 represents the reproductive state of the shallow water population of P.
f. fulvum sampled during summer 1980 at Muqebla'. The colonies are divided into
4 groups: males with sperm sacs, females with oocytes in the polyp cavities, females
with eggs on their surface (brooding females), and colonies without any genital
products. The first two dates represent the population reproductive structure before
the breeding season. The majority of the colonies still contain gametes in their polyp
cavities. The histograms from 21 and 22 June (Fig. 9) indicated the reproductive
state a few days after gamete expulsion, which took place on 18 June (Tabel II).
These results show a decrease in the percentage of male colonies with testes, hence,
an increase in the number of colonies without any genital products. During these
days, only a minor proportion of the population brood their larvae. Seven to ten
days after spawning, in 26 and 28 June, no brooding females could be found. Similar
reproductive structure was found at the two following dates. After the 1 5 July spawn-
ing (Table II), the population consisted of brooding females and colonies without
gonads (17 July, Fig. 9). Underwater observations over large areas at various reef
localities indicated that only a negligible percentage of corals spawned on 2 July
1980.
Figure 1 0 presents the reproductive structure of the population along a depth
gradient at the M.B.L. reef during the breeding season of 1980. The upper part of
the figure illustrates the results obtained on 1 8-20 July, and the lower part that of
2-4 August. The massive spawning of July (Table II) occurred along all the depth
range studied. A few days after spawning, brooding females were observed, especially
at a depth of 5-20 m. Consequently, a marked decrease of females with oocytes
was noted. At reef zones deeper that 5 m, the percent of male corals with sperm
TABLE II
Timing of egg expulsion in the population o/Parerythropodium fulvum fulvum
Date Moon phase Depth m
25 June 1978 Full moon — Last quarter 1-4
27 June 1979 New moon — First quarter 1-6
18 June 1980 New moon — First quarter 1-3
2 July 1980 Full moon— Last quarter 1-18
15 July 1980 New moon— First quarter 15-25
30 July 1980 Full moon— Last quarter 30-35
4 July 1981 Full moon — Last quarter 1-5
SURFACE BROODING IN A SOFT CORAL
365
80r
n
160
o
o
S20
rl
10
OL.
6 6
_ 66
176
l_J Males with testes B Brooding females
Q Females with oocytes O No genital products
21 6
22.6
- 50
- 55
- 57
- 52
-
•: '•: \
~T
-
^;
-
:'A
-."•
;;.
£A
•V- \
:•• \
ITTi
I
\
s
\
-.;
j\
r',
J
\
-
1
•'••
s
\
\
266 286 127 137
Date
17 7
FIGURE 9. The reproductive state of shallow water population of Parerythropodium fulvum fulvum
at Muqebla' during the breeding season of 1980. The numbers within each sampling date represent the
sample size.
18-20 July 1980
80
60
40
^20
_o
o
u
H- 0
c
0
^20
01
Q-
40
80
100 L
5m.
45
37
1Om
49
40
15m
50
2Om
59
Males with
testes
63
2 -4 August 1980
Females with
oocytes
25m.
51
48
3Om
47
Brooding females
No genital
products
FIGURE 10. The reproductive structure of Parerythropodium fulvum fulvum along a depth gradient
during the breeding season of 1980.
366 Y. BENAYAHU AND Y. LOYA
sacs still remained high. The spawning of 30 July (Table II) was recorded below 5
m depth. However, brooding colonies were observed only at 20-30 m depth. The
lower part of Figure 1 0 indicates that after this spawning almost the whole population
remained without genital products, except for a small number of males at a depth
below 15 m. Figures 9 and 10 point out that the shallow water population breeds
before the deeper one, and the whole reproductive period takes place during ap-
proximately two months.
DISCUSSION
During the last several years much interest has been focused on the life history
of scleractinian corals (Harrigan 1972; Stimson, 1978; Rinkevich and Loya, 1979a,
b;Szmant-Froelich<tftf/., 1980; Kojis and Quinn, 1981, Fadlallah and Pearse, 1982a,
b). Although the significance of alcyonacean corals within the coral reef environment
is well recognized, only scant surveys were conducted on their life history. The
present study elucidates for the first time various aspects of the reproductive dy-
namics of the common Red Sea soft coral P. f. fulvum.
The general morphological features of the gonads of P. f. fulvum resemble those
of Alcyonium digitatum (Hickson, 1895; Hill and Oxon, 1905) and Heteroxenia
fuscescens (Gohar and Roushdy, 1961). Field experiments dealing with the colo-
nization capacity of P. f. fulvum (Benayahu, 1982) indicate that all colonies above
the age of 3-4 years old develop gonads. Small sized colonies mostly contain male
gonads, while females become sexually mature at an older age. These results fit well
with the common pattern found in other corals (Harrigan, 1972; Hartnoll, 1977;
Grigg, 1977; Rinkevich and Loya, 1979b).
Sex ratios of P. f. fulvum differed between the shallow water and the deep reef
populations. This may be due to local aggregations of the species (Benayahu, 1975).
Such uneven distribution of individuals can cause local clumps of one sex. Addi-
tionally, asexual reproduction of P. f. fulvum formed by fragmentation (Benayahu,
1982), may cause deviation from a 1:1 sex ratio.
Fecundity of gorgonian octocorals has been determined as the number of plan-
ulae produced per polyp (Grigg, 1977). Thus, the alcyonaceans Heteroxenia fus-
cescens (Gohar, 1940a) and Alcyonium digitatum (Hartnoll, 1975) with long polyp
cavities are characterized by high egg production. However, in P. f. fulvum, which
has an encrusting growth form and short polyp cavities, fecundity is low ( 1 8-24 eggs
per polyp).
Several studies reported lunar periodicity in the reproduction of stony corals
(Harrigan, 1972; Stimson, 1978; Rinkevich and Loya, 1979b). This study documents
a distinct lunar rhythmicity in the breeding of an alcyonacean coral. Lobel (1978)
suggests that such spawning may act as a cue synchronizing simultaneous repro-
ductive readiness within a species. We further speculate that this mechanism is
critically important within colonies like P. f. fulvum, which breed only a few days
per year. It should be noted that the time lag in spawning at greater depths (Fig. 10)
is probably due to differences in time of the peak water temperature along depth
gradient, as suggested by Grigg (1977) in his study on gorgonians.
Among the anthozoans, P. f. fulvum exhibits a unique mode of sexual repro-
duction and planulae development. This coral is oviparous, yet cleavage of the eggs
takes place on the surface of the female colonies within a mucoid suspension. We
term this peculiar mode of planula development as surface brooding. Brooding in
marine invertebrates was defined by Dunn (1975) as "the retention of offspring by
SURFACE BROODING IN A SOFT CORAL 367
parent through embryonic stages usually passed in the plankton," hence, P.f.fulvum
is an external brooder. External brooding in anthozoans is uncommon. The group
which is best known are actinians of the genus Epiactis (Chia, 1976), especially E.
prolifera which broods its young on its lower column (Dunn, 1975). In this species
the embryos are enveloped by the parent, and the ectoderm of the two are closely
apposed. The intimate connnection between the offspring and the parent is oblig-
atory and essential for their development. Dunn further suggests that this might
serve a nutritional function.
External brooding has also been recorded in the octocoral species from the order
Stolonifera: Clavularia crassa (Kowalewsky and Marion, 1883), Cornularia komaii,
and C. saganiensis (Suzuki, 1971). In these species the fertilized eggs developed into
planulae in an external brooding cavity formed by the tentacles. The eggs of the
scleractian coral Goniastea australensis are expelled as masses embedded in mucus
(Kojis and Quinn, 1981). They remain on the colony and after spawning is ter-
minated, the eggs sink down to the bottom where planular development takes place.
The results of the present work indicate that the brooding behavior of P. f. fulvum
differs from that of other anthozoans with external brooding. Although no cellular
connection exists between the embryos and the colonies, cleavage occurs on the
external surface of the females. Thus, the embryos are protected from mechanical
damage such as the erosive activity of sediment or wave action.
Membanaceous growth form is rare among the octocorals. The encrusting col-
onies of P. f. fulvum are characterized by a thin coenenchyme and short polyp
cavities. Most eggs of soft corals are large in diameter (500-700 /mi: Benayahu,
1982). It is therefore presumed, that if embryogenesis had been internal, the number
of eggs per polyp would have been reduced even below the number of 18-24 oocytes
due to small polyp size. We suggest that surface brooding maximizes fecundity and
is an adaptation to the encrusting growth form. Egg production in P. f. fulvum is
rather low, but this is compensated for by surface brooding which protects the
offspring through embryogenesis. It is interesting to note that the three aforemen-
tioned external brooding Stolonifera species are also encrusting corals. Hence, the
same reproductive strategy has been adopted by two different octocoral groups.
ACKNOWLEDGMENTS
We are indebted to the late Prof. C. Lewinsohn (Tel Aviv University) for his
advice during this study. We are grateful to Dr. W. M. Goldberg (Florida Inter-
national University) for critical comments on the manuscript. We thank the M.B.L.
staff at Eilat for their hospitality and facilities. We would like to thank L. Maman
and A. Shoob for taking the photographs. Y.B. is indebted to D. Benayahu for her
endless help.
LITERATURE CITED
BAYER, F. M. 1973. Colonial organization in octocorals. Pp. 69-93 in Animal Colonies Development
and Function Through Time, R. S. Boardman, A. H. Cheethman, and W. A. Oliver Jr., eds.
Dowden, Hutchinson and Ross. Inc. Stroudsburg, Pennsylvania.
BENAYAHU, Y. 1975. Quantitative characteristics of community structure of stony corals, soft corals and
algae in the northern Gulf of Eilat (Red Sea). M.Sc. Thesis, Tel Aviv Univ. 132 pp. (In Hebrew,
with English summary).
BENAYAHU, Y. 1982. Population dynamics of soft corals (Octocorallia, Alcyonacea) at the coral reefs
of the Gulf of Eilat. Ph.D. Thesis, Tel Aviv Univ. 214 pp. (in Hebrew, with English summary).
368 Y. BENAYAHU AND Y. LOYA
BENAYAHU, Y., AND Y. LOYA. 1977. Space partitioning by stony corals, soft corals and benthic algae
on the coral reefs of the northern Gulf of Eilat (Red Sea). Helgol. Wiss. Meeresunters. 30: 362-
382.
BENAYAHU, Y., AND Y. LOYA. 1981. Competition for space among coral-reef sessile organisms at Eilat,
Red Sea. Bull. Mar. Sci. 31: 514-522.
GARY, L. R. 1931. Studies on the coral reefs of Tutila, American Samoa, with especial reference to the
Alcyonaria. Pap. Tortugas Lab. Carnegie Inst. Wash. 27: 53-98.
CHIA, F. S. 1976. Sea anemone reproduction: Patterns and adaptive radiation. Pp. 261-270 in Coelen-
terate Ecology and Behavior, G. O. Mackie, ed. Plenum Publishing Corporation, New York.
DUNN, D. F. 1975. Reproduction of the external brooding sea anemone Epiactis prolifera Verrill, 1869.
Bio/. Bull. 148: 199-218.
FADLALLAH, Y. H., AND J. S. PEARSE. 1982a. Sexual reproduction in solitary corals: Overlapping oogenic
and brooding cycles, and benthic planulas in Balanophyllia elegans. Mar. Biol. 71: 223-231.
FADLALLAH, Y. H., AND J. S. PEARSE. 1982b. Sexual reproduction in solitary corals: Synchronous
gametogenesis and broadcast spawning in Paracyathus stearnsii. Mar. Biol. 71: 233-239.
FISHELSON, L. 1970. Littoral fauna of the Red Sea: The population of non scleractinian anthozoans of
shallow waters of the Red Sea (Eilat) Mar. Biol. 6: 106-1 16.
FISHELSON, L. 1973. Ecological and biological phenomena influencing coral-species composition on the
reef tables at Eilat. Mar. Biol. 19: 183-196.
GOHAR, H. A. F. 1940a. Studies on the Xeniidae of the Red Sea. Publ. Mar. Biol. Stn. Al Ghardaqa 2:
27-118.
GOHAR, H. A. F. 1940b. The development of some Xeniidae (Alcyonaria). Publ. Mar. Biol. Stn. Al
Ghardaqa 3: 27-76.
GOHAR, H. A. F., AND H. M. ROUSHDY. 1961. On the embryology of the Xeniidae (Alcyonaria) (with
notes on the extrusion of the larvae). Pub. Mar. Biol. Stn. Al Ghardaqa 11: 45-70.
GRIGG, R. W. 1977. Population dynamics of two gorgonian corals. Ecology 58: 278-290.
HARRIGAN, J. 1972. The planula and larva of Polcillopora damicornis: Lunar periodicity of swarming
and substratum selection behavior. Ph.D. Thesis Univ. of Hawaii. 213 pp.
HARTNOLL, R. G. 1975. The annual cycle of Alcvonium dighatum. Estuarine Coastal Mar. Sci. 3: 71-
78.
HARTNOLL, R. G. 1977. Reproductive strategy in two British species of Alcyonium. Pp. 321-328 in
Biologv of Benthic Organisms, P. O. Ceidigh and P. J. S. Boaden, eds. Pergamon Press, New
York.
HICKSON, S. J. 1895. The anatomy of Alcyonium digitatum. Q. J. Microsc. Sci. 37: 343-388.
HILL, M. D., AND M. A. OXON. 1905. Notes on the maturation of the ovum of Alcvonium digitatum.
Q. J. Microsc. Sci. 49: 493-505.
KOJIS, B. L., AND N. J. QUINN. 1981. Aspects of sexual reproduction and larval development in the
shallow water hermatypic coral Goniastrea australensis (Edwards and Haime, 1857) Bull. Mar.
Sci. 31: 558-573.
KOWALEWSKY, A. O., AND A. F. MARION. 1883. Documents pour 1'histoire embryogenique des Al-
cyonairs. Ann. Mus. Hist. Nat. Marseille 1: 6-50.
LACAZE-DUTHIERS, H. 1865. Histoire naturelle du corail. Organisation, reproduction, peche ne Algerie,
industrie et commerce. J. B. Bailliere et fils, Paris, Pp. 371.
LOBEL, P. S. 1978. Diel, lunar and seasonal periodicity in the reproductive behavior of the pomacanthid
fish, Centropyge potteri and some other reef fishes in Hawaii. Pac. Sci. 32: 193-207.
LOYA, Y., AND L. B. SLOBODKIN. 1971. The coral reefs of Eilat (Gulf of Eilat Red Sea). Symp. Zool.
Soc. Land. 28: 117-139.
MARAGOS, J. E. 1974. Coral communities on a seaward reef slope. Fanning Island. Pac. Sci. 28: 257-
278.
MARISCAL, R., AND C. H. BIGGER. 1976. A comparison of putative sensory receptors associated with
nematocysts in an anthozoan and a scyphozoan. Pp. 559-568 in Coelenterate Ecology and
Behavior, G. O. Mackie, ed. Plenum Publishing Corporation, New York.
MATTHEWS, A. 1917. The development of Alcyonium digitatum with some notes on the early colony
formation. Q. J. Microsc. Sci. 62: 43-94.
PEARSON, R. G. 1981. Recovery and recolonization of coral reefs. Mar. Ecol. Prog. Ser. 4: 105-122.
RINKEVICH, B., AND Y. LOYA. 1979a. The reproduction of the Red Sea coral Stylophora pistillata. I.
Gonads and planulae. Mar. Ecol. Prog. Ser. 1: 133-144.
RINKEVICH, B., AND Y. LOYA. 1979b. The reproduction of the Red Sea coral Stylophora pistillata. II.
Synchronization in breeding and seasonality of planulae shedding. Mar. Ecol. Prog. Ser. 1: 145-
152.
SCHUHMACHER, H. 1975. Die Rolle der weichkorallen (Alcyonacea, Octocorallia) innerhalb der riffbio-
SURFACE BROODING IN A SOFT CORAL 369
zonosen des roten meeres und des australischen grossen barrieriffs. Verh. Zool. Ges. 67: 380-
384.
STIMSON, J. S. 1978. Mode and timing of reproduction of colonies in some common hermatypic corals
of Hawaii and Enewetak. Mar. Biol. 48: 173-184.
SZMANT-FROELICH, A, P. YEVICH, AND M. E. Q. PILSON. 1980. Gametogenesis and early development
of the temperate coral: Astrangia danae (Anthozoa:Scleractinia). Biol. Bull. 158: 257-269.
SUZUKJ, H. 1971. Notes on Cornularia (Stolonifera, Alcyonacea) found in the vicinity of Manazuru
marine biological laboratory. Sci. Rep. Yokohama Nat. Univ. Sec. II. 18: 1-6.
VERON, J. E. N., R. A. How, T. J. DONE, L. D. ZELL, M. J. DODKJN, AND A. F. O. FARRELL. 1974.
Corals of the Solitary Islands, Central New South Wales. Aust. J. Mar. Freshwater Res. 25: 193-
208.
VERSEVELDT, J. 1969. Octocorallia from North western Madagascar (Part I) Zool. Verh. (Leiden) 106:
1-38.
YAMAZATO, K., AND M. SATO. 1981. Reproductive biology of an alcyonarian coral, Lobophytum crassum
Marenzeller. Pp. 74-75, abstract in Proc. 4th Inter. Coral Reef Symp. Manilla.
Reference: Biol. Bull. 165: 370-378. (October, 1983)
SEXUAL DIMORPHISM AND REPRODUCTIVE BEHAVIOR IN
ALMYRACUMA PROXIMOCULI (CRUSTACEA:
CUMACEA): THE EFFECT OF HABITAT
THOMAS K. DUNCAN*
Boston University Marine Program, Marine Biological Laboratory, Woods Hole, Massachusetts 02543
ABSTRACT
Individuals of Almyracuma proximoculi are the least sexually dimorphic cu-
maceans known, because the males are progenetic, i.e., they are precociously sexually
mature at a morphologically immature state. This species lives in dense aggregations
in the upper intertidal zone and has eliminated the morphologically complex, ap-
parently pheromone-sensitive, and highly motile terminal male stage found in other
cumacean species. The sexually dimorphic characters that are present are predom-
inantly ones that facilitate the rapid removal of the female's exuviae by the male
during her fertilization molt. The removal rate is critical, because the partially de-
tached exuviae blocks access to the female's ventrum. With the exception of the
rudimentary penes found in two genera, male cumaceans do not possess an in-
tromittent organ and apparently must deposit one or more spermatophores on the
female's ventrum before the developing oostegites completely enclose this area.
INTRODUCTION
Cumaceans belong to the superorder Peracarida, which also includes amphipods,
isopods, tanaidaceans, and mysidaceans, among others. The Cumacea are infaunal
peracarid crustaceans that are primarily marine and are found world-wide from the
intertidal zone to abyssal depths (Jones, 1976). Sexually immature males and females
have very similar external morphologies and ornamentation, and most of the sex-
ually dimorphic characters are acquired in the last few molts (Zimmer, 1941).
Like most Peracarida, cumaceans brood their young in a ventral marsupium,
and the most striking change in female morphology is the rapid and complete
development of the oostegites in only two molts. The external development of the
male is typically a more gradual process and involves the sexually dimorphic de-
velopment of a variety of body parts (Forsman, 1938; Granger et al, 1979; Bishop,
1982). Commonly this differential development of the male includes, but is not
limited to, the following: an increased number and greater development of natatory
thoracic exopodites; the presence of up to five pairs of natatory pleopods which,
with the exception of one species, are never present in females; a less spinose carapace
that generally has a lower profile than that of the conspecific female; and the flat-
tening and broadening of various appendages and projections such as the epimeral
plates of the thoracic and abdominal somites. No one species possesses all of these
adaptations in their most developed forms, but typically a male cumacean will
exhibit a combination of several of them, as in Diastylis cornuta (Fig. 1 ).
In addition to the above changes, the greatest differential development occurs
in the male's second antennae. The second antennae of mature male cumaceans
Received 28 March 1983; accepted 18 July 1983.
* Present address: Department of Environmental Science, Nichols College, Dudley, Massachusetts
01570.
370
CUMACEAN DIMORPHISM AND BEHAVIOR
371
pleopods
male
female
thoracic
exopodites
FIGURE 1. Copulatory male and marsupial female of Diastylis cornula (after Sars, 1900).
are always well developed, with the exception of one species, while those of females
are always rudimentary (Jones, 1963). In many species they equal or exceed the
male's total body length (Sars, 1900). The development of these enormous antennae
only in sexually mature male instars suggests that they are probably chemosensory,
serving as the receptors for pheromones released by females before their fertilization
molts, as has been demonstrated in the Amphipoda (Dahl et al, 1970; Lyes, 1979).
These two modes of differential development produce a motile, chemically sensitive
male which is able to swim up into the water column and seek out potential mates.
The cumacean Almyracuma proximoculi Jones and Burbanck, 1959, is a small
crustacean, with sexually mature individuals ranging from about 2.3 to 4.3 mm in
length (Duncan, 1981). It has been collected in low numbers in estuarine areas from
Currituck Sound, North Carolina, to Cape Cod, Massachusetts (Jones and Burbanck,
1959; Sanders et al., 1965; Boesch and Diaz, 1974; Crandall, 1977; Ristich et al.,
1977; Menzie, 1980; T. E. Bowman, Smithsonian Institution, pers. comm.), but its
optimal habitat appears to be thermally moderated areas in the immediate vicinity
of freshwater springs in the upper intertidal zone of Long Island, New York, and
southern New England. It inhabits these areas year-round, typical densities within
a few meters of these groundwater discharges range from 3000-4500 m"2, and
extrapolated densities as high as 31,000 irT2 have been recorded in these areas
(Duncan, 1981). This species is essentially restricted to these disjunct, intertidal
aggregations with high within-habitat densities and proportionately large distances
between aggregations.
MATERIALS AND METHODS
Random samples of ten preparatory females and ten copulatory males from an
intertidal freshwater spring in West Falmouth Harbor, Massachusetts, were mea-
sured with an ocular micrometer (±0.0196 mm). All dimensions are from the left
372 T. K. DUNCAN
sides of individuals, with the exception of the cross-sectional area of the fifth ab-
dominal somite.
Laboratory observations were made on over 600 clasping pairs of individuals
(copulatory male and preparatory or marsupial female). These individuals were
collected from intertidal freshwater springs at the following localities on Cape Cod,
Massachusetts: West Falmouth Harbor, Waquoit Bay, and Pocasset. Most of the
individuals were maintained as isolated pairs in multicompartmented, transparent,
plastic trays for up to four months. In addition to the animals, each compartment
contained 20 ml of water and a small amount of sand from a collection site. Ad-
ditional observations were made on groups of individuals maintained in glass finger
bowls with varying amounts of water and substrate. Specimens for scanning electron
microscopy were fixed in 5% glutaraldehyde for 10 min at room temperature, trans-
ferred to 95% ethanol, and air-dryed on double-sided adhesive tape.
RESULTS
Morphology
The external morphology of the copulatory male of A. proximoculi is very simple
(Fig. 2) and provides a sharp contrast to typical copulatory male cumaceans (Fig. 1).
Neither sex has pleopods. Both sexes have a moderate and equivalent development
of the thoracic exopodites, show similar profiles and smoothness of the carapace, and
lack pronounced flattening and broadening of appendages or body parts. The male's
second antenna is rudimentary and comparable to that of the female (Jones and
Burbanck, 1959). With the exception of the developing oostegites of the female and
the consequent greater width of her thorax, there are few other obvious morphological
differences between the sexes.
The limited sexual dimorphism that is present in A. proximoculi is expressed
mainly in the disproportionate development of the copulatory male's third maxil-
lipeds and first pereiopods (Fig. 2) and of most of the post-thoracic region of his
body (Fig. 2, Table I). On average, the abdomens of copulatory males are 31% longer
and have a 55% greater cross-sectional area when compared to those of preparatory
females of similar carapace lengths. Additionally, the uropodal peduncles of these
males are 65% longer and 25% wider than those of the females (Table I; Fig. 3A,
B). In contrast, the male uropodal endopods are only 7% longer than the female
ones, equivalent to the average difference in carapace lengths between the two
groups.
Although there are no other major differences between the sexes in the general
shape, sculpturing, or ornamentation of the integument, the medial surfaces of the
male's uropodal peduncles and endopods are armed with two distinct types of spines
which are arranged in single rows. Those found on the endopods are simple, cone-
shaped projections which are more numerous on the male than on the female
(usually six versus two, Fig. 3A, B, D). The second type is a complex, pinnate form
(Fig. 3C) which is absent on female or less mature male stages. There are usually
six to ten of these on each uropodal peduncle of a copulatory male. The same
margin of female and earlier male instars carries only a few simple setae (Fig. 3A).
The other margins of the uropodal appendages of both sexes are either bare or carry
simple setae only (Fig. 3A, B).
Behavior
In late winter, throughout the spring, and during summer mature males will
clasp preparatory females. During precopula the female is clasped and manipulated
CUMACEAN DIMORPHISM AND BEHAVIOR
373
FIGURE 2. Scanning electron micrograph of a precopulatory clasping pair of Almyracuma proxi-
moculi. The preparatory female is being held by the male's third maxillipeds and first pereiopods.
with the male's oversized third maxillipeds and first pereiopods (Fig. 2). She is
usually carried in the same posterior-anterior alignment as the male with her dorsum
adjacent to the male's ventral surface (295 1 of 2962 observations). Unless disturbed,
clasping pairs generally lie on their sides on the bottom of the observation dish, or
if enough sediment is present, they remain buried. When disturbed they often swim
up into the water, using the thoracic exopodites of the male and occasionally those
of the female for locomotion. Males were never observed feeding while clasping
females, but clasped females continue to feed normally by grasping sand grains and
rotating them against their mouthparts. It is unknown how long pairs will remain
in a clasped position in the field, but in laboratory conditions males have clasped
374
T. K. DUNCAN
TABLE I
Mean dimensions and their standard errors of Almyracuma proximoculi
Cross-sectional
Carapace
length
Abdomen
length
area of fifth
abdominal somite
(X103)
Uropodal
peduncle
length
Uropodal
peduncle
width
Uropodal
endopod
length
preparatory
female 851 + 12.3 1494 ± 24.0 37.9 ± 1.01
copulatory
male 894 ± 15.8 1952 ± 30.4 58.9 ± 1.89
increase in
male 5.05% 30.7% 55.4%
282 ± 9.8 98 ± 1.4 255 ± 5.1
465 ± 8.8 123 ± 7.3 272 ± 3.5
64.9% 25.5% 6.67%
All dimensions are in micrometers, except for cross-sectional areas in square micrometers, and are
from random samples of ten individuals of each sex.
females for as long as four months when the fertilization molt was experimentally
delayed by lowering ambient temperatures.
Normally, the male's abdomen is straight or slightly flexed, but occasionally he
flexes it enough to grasp the female's abdomen immediately behind the thorax with
his uropods. The male then straightens his abdomen rapidly, raking the medial
surfaces of his uropods along her abdomen. If copulatory males are present that are
not already clasping females, they will approach a clasping pair and attempt to
dislodge the male. During these events and while trying to hold onto the female in
any way possible, either male uses his oversized abdomen and uropods in two ways.
He attempts to force his uropods between the other male and the female and pry
them apart and/or he grabs the other male's abdomen with his uropods and attempts
to pull him off the female.
The fertilization molt is initiated by the splitting of the female's exuviae on the
dorsal midline of the five exposed thoracic somites. Immediately after this the male
moves his first pereiopods under the loose thoracic segments of the exuviae and
forces his third maxillipeds under the posterior margin of the exuvial carapace. The
carapace then comes off in one piece. The five exuvial thoracic segments remain
attached ventrally to each other and to the exuvial carapace and abdomen. Con-
sequently, the detached portions of the exuviae hang beneath the female. The male
than arches his abdomen, grasps the female's abdomen immediately behind the
thorax with his uropods, and rakes the entire length of her abdomen with their inner
surfaces (Fig. 4). This vigorous raking, involving considerable effort by the male,
continues until the exuviae is pulled completely free from her abdomen. The male
immediately turns the female over, reverses her anterior-posterior position, and
briefly clasps her with their ventral surfaces opposed. Shortly thereafter ova can be
seen within the marsupium. Females with fully developed marsupia very seldom
elicit a response from males; but as soon as the young are released, and females
molt back into a preparatory instar ("interbrood" stage, sensu Duncan, in prep.),
mature males will clasp them.
DISCUSSION
The rudimentary state of the copulatory male's second antennae in A. proxi-
moculi is unique among the approximately 1000 known species of Cumacea (Jones
and Burbanck, 1959). The copulatory males of this species are progenetic (Duncan,
CUMACEAN DIMORPHISM AND BEHAVIOR
375
FIGURE 3. Scanning electron micrograph of the uropods of mature Almyracuma proximoculi: A)
dorsal view of preparatory female; B) same view of copulatory male; C) dorsal view of two most distal
spines on left peduncle in (B); D) dorsal view of middle spines on left endopod in (B).
1981), i.e., they are precociously sexually mature at a morphologically immature
state. I suggest that the typical distribution of this species in disjunct, dense, intertidal
aggregations has eliminated the need for a pheromone-sensitive, highly motile, copu-
latory male. This distribution has apparently permitted this species to eliminate a
morphologically complex instar that would normally be the final male stage and
possibly reduces intraspecific competition for food resources that would otherwise
be needed for the elaboration of body parts seen in the males of other species. In
Pseudocuma longicornis, another cumacean species, "young males" clasp females,
and "fully adult" males, although present, have never been observed in mating pairs
(Foxon, 1936; Corey, 1969). This species is most common in low intertidal and
376
T. K. DUNCAN
FIGURE 4. Precopulatory clasping pair of Almyracuma proximoculi, consisting of a copulatory male
(top), a mature female (middle), and a partially detached exuviae (dotted outline at bottom).
shallow intertidal zones (Corey, 1970) and appears to be another example of pro-
genetic development of copulatory males in a shallow water cumacean species.
The comparatively greater size and spination of the uropods of male cumaceans
has been known for many years (Sars, 1900; Zimmer, 1941), and it has been sug-
gested that these are adaptations for cleaning adhering material from the mouthparts
and other appendages (Dixon, 1944). This function alone can not explain the striking
sexual and ontogenetic differences seen in the uropods and abdomen of A. proxi-
moculi and other species, since both sexes and the various instars of a particular
species generally occur in the same substrate and can be expected to have the same
cleaning requirements. Additionally, there is a distinct shift of morphological em-
phasis in the males of A. proximoculi from the enhancement of natatory functions
to improving the males' ability to clasp and manipulate females.
The precopulatory clasping posture utilized by A. proximoculi (female dorsum
clasped to male ventrum with both individuals in the same anterior-posterior align-
ment) is the same as has been noted in other Cumacea (Zimmer, 1941), with the
exception ofMancocuma stellifera (Gnewuch and Croker, 1973) and Spilocuma sal-
omani (Saloman, 1981). Saloman, citing Jones and Burbanck (1959), stated that male
A. proximoculi grasp female abdomens with their second antennae. Apparently he
misread the latter paper. The rudimentary development of these antennae (Jones and
Burbanck, 1959; personal observation) makes such behavior impossible. Due to the
position of the female's body and the use of the male's appendages for clasping, this
posture probably precludes feeding by the males of most species during this period.
CUMACEAN DIMORPHISM AND BEHAVIOR 377
However, this may be unimportant, since the copulatory stage is usually a terminal
one for male cumaceans, and most males die soon after mating. A similar nonfeeding
pattern occurs in the copulatory males of several species of Tanaidacea, where the
mouthparts are reduced and the anus is fused shut (Gardiner, 1975). Conversely, the
elaborate natatory and sensory appendages seen in typical copulatory male cumaceans
are maladaptive for the infaunal, burrowing lifestyle of young males and do not
develop fully until the terminal instar.
Preparatory females and other developmental stages, including all of the im-
mature male instars, molt successfully without aid in the laboratory. Therefore, it
appears that the male's differential development and behavior serve only to accel-
erate the female's fertilization ecdysis. With the exception of the rudimentary penes
found in two genera, Archeocuma (Bacescu, 1972) and Campylaspenis (Bacescu and
Muradian, 1974), intromittent organs are unknown in the Cumacea, and sperm are
extruded from two pores on the ventrum of the fifth thoracic somite. The partially
detached exuviae blocks access to the female's thoracic ventrum, and shortly after
molting the fully developed oostegites overlap each other considerably, completely
enclosing this area. Thus the removal rate is critical, if the male is to gain access
to this area and deposit a spermatophore successfully.
The unusual habitat of A. proximoculi has influenced both the morphology and
the behavior of this species. High levels of chemosensitivity and swimming ability
may not be particularly advantageous in a species, such as this one, that has a
distributional pattern of high local densities in an intertidal area and relatively large
distances between aggregations. Instead, the ability to rapidly remove a female's
exuviae once molting has started and to deposit a spermatophore before the ventrum
is enclosed by the marsupium or interruption and/or displacement by a competing
male occurs appears to have influenced the morphology of the male of this species.
Almyracuma proximoculi represents one end of the spectrum of morphological com-
plexity and swimming ability found in male cumaceans that may be controlled,
ultimately, by the densities of potential mates and competing males.
ACKNOWLEDGMENTS
I am very grateful to Masahiro Dojiri, Arthur Humes, Howard Sanders, and two
anonymous reviewers, all of whom provided constructive criticism of earlier drafts
of this paper. A substantial part of this paper has been extracted from a thesis
submitted to Boston University in partial fulfillment of the requirements for the
Ph.D. degree, and I am grateful for the support provided by this university.
LITERATURE CITED
BACESCU, M. 1972. Archeocuma and Schizocuma, new genera of Cumacea from the American tropical
waters. Rev. Roum. Biol. Ser. Zool. 17: 241-250.
BACESCU, M., AND Z. MURADIAN. 1974. Campylaspenis, Styloptocuma, Atlantacuma. new genera of
Cumacea from the deep waters of the Atlantic. Rev. Roum. Biol. Ser. Zool. 19: 71-78.
BISHOP, J. D. D. 1982. The growth, development and reproduction of a deep sea cumacean (Crustaea:
Peracarida). Zool. J. Linn. Soc. 74: 359-280.
BOESCH, D., AND R. DIAZ. 1974. New records of peracarid crustaceans from oligohaline waters of the
Chesapeake Bay. Chesapeake Sci. 15: 56-59.
COREY, S. 1969. The comparative life histories of three Cumacea (Crustacea): Cumopsis goodsiri (van
Beneden), Iphinoe trispinosa (Goodsir), and Pseudocuma longicornis (Bate). Can. J. Zool. 47:
695-704.
COREY, S. 1970. The quantitative distribution of three Cumacea. Can. J. Zool. 48: 925-930.
CRANDALL, M. E. 1977. Epibenthic invertebrates of Croton Bay in the Hudson River. N. Y. Fish Game
J. 24: 178-186.
DAHL, E., H. EMANUELSSON, AND C. VON MECKLENBURG. 1970. Pheromone reception in the males
of the amphipod Gammarus duebeni Lilljeborg. Oikos 21: 42-47.
378 T. K. DUNCAN
DIXON, A. Y. 1944. Notes on certain aspects of the biology of Cumopsis goodsiri (van Beneden) and
some other cumaceans in relation to their environment. J. Mar. Biol. Assoc. U. K. 26: 61-71.
DUNCAN, T. K. 1981. Life history and ecology ofAlmyracuma proximoculi Jones and Burbanck, 1959
(Crustacea: Cumacea). Ph.D. Thesis, Boston University.
DUNCAN, T. K. (in prep.). Life history of Almyracuma proximoculi Jones and Burbanck, 1959, (Crustacea:
Cumacea) from intertidal freshwater springs on Cape Cod, Massachusetts. Submitted to J. Crust.
Biol.
FORSMAN, B. 1938. Untersuchungen iiber die Cumaceen des Skageraks. Zool. Bidr. Upps. 18: 1-162.
FOXON, G. E. H. 1936. Notes on the natural history of certain sand-dwelling Cumacea. Ann. Mag. Nat.
Hist. (Ser. 10) 17: 377-393.
GARDINER, L. F. 1975. The systematics, postmarsupial development, and ecology of the deep-sea family
Neotanaide (Crustacea: Tanaidacea). Smithson. Contrib. Zool. 170: 1-265.
GNEWUCH, W. T., AND R. A. CROKER. 1973. Macroinfauna of northern New England marine sand. I.
The biology of Mancocuma stellifera Zimmer, 1943. (Crustacea: Cumacea). Can. J. Zool. 51:
1011-1020.
GRANGER, D., P. BRUNEL, AND D. MESSIER. 1979. Cycle de developpement de Leucon nasica (Crustacea,
Cumacea) dans la nappe glaciale circalittorale de la baie des Chaleurs, golfe du Saint-Laurent,
en 1968 et 1969. Can. J. Zool. 57: 96-106.
JONES, N. S. 1963. The marine fauna of New Zealand: crustaceans of the order Cumacea. Bull. N. Z.
Dep. Scient. Ind. Res. l52(Mem. No. 23, N. Z. Oceanogr. Inst.): 9-78.
JONES, N. S. 1976. British Cumaceans. Academic Press, New York.
JONES, N. S., AND W. D. BURBANCK. 1959. Almyracuma proximoculi, gen. et spec. nov. (Crustacea,
Cumacea) from brackish water of Cape Cod, Massachusetts. Biol. Bull. 116: 1 15-124.
LYES, M. C. 1979. The reproductive behavior of Gammarus duebeni (Lilljeborg), and the inhibitory
effect of a surface active agent. Mar. Behav. Physiol. 6: 47-55.
MENZIE, C. A. 1980. The chironomid (Insecta: Diptera) and other fauna of a Myriophyllum spicatum.
L. plant bed in the lower Hudson River. Estuaries 3: 38-54.
RISTICH, S. S., M. CRANDALL, AND J. FORTIER. 1977. Benthic and epibenthic macroinvertebrates of the
Hudson River. I. Distribution, natural history, and community structure. Estuar. Coast. Mar.
Sci. 5: 255-266.
SALOMAN, C. H. 1981. Aspects of the biology ofSpilocuma Salomani (Cumacea: Boditriidae) [sic]. Northeast
Gulf Sci. 4: 99-104.
SANDERS, H. L., P. C. MANGELSDORF, JR., AND G. R. HAMPSON. 1965. Salinity and fauna! distribution
in the Pocasset River, Massachusetts. Limnol Oceanogr.lQ(supp\.): R216-R229.
SARS, G. O. 1900. An Account of the Crustacea of Norway with Short Descriptions and Figures of All of
the Species. 3. Cumacea. Bergen Museum, Bergen.
ZIMMER, C. 1941. Cumacea. In Bronn's Kl. Ordn. Tierreichs 5(1): 1-222.
Reference: Biol. Bull. 165: 379-393. (October, 1983)
ULTRASTRUCTURAL DIFFERENCES IN THE EGGS AND OVARIAN
FOLLICLE CELLS OF CAPITELLA (POLYCHAETA) SIBLING SPECIES
KEVIN J. ECKELBARGER1 AND JUDITH P. GRASSLE2
1 Functional Reproductive Biology Program, Harbor Branch Foundation, R.R. 1, Box 196, Fort Pierce,
Florida 33450, and 'Marine Biological Laboratory, Woods Hole, Massachusetts 02543
ABSTRACT
Ultrastructural studies of ovarian follicle cells and mature eggs in four sibling
species in the polychaete genus Capitella have revealed distinct and consistent mor-
phological differences that parallel in some respects the differences between the
species in egg size, and embryonic and larval development. Capitella spp. I and II
are extremely similar in all respects: the follicle cells lack lipid and contain a modest
amount of glycogen; the mature eggs are rich in lipid and glycogen and contain very
similar proteid yolk granules. In both species mature eggs have a characteristic
electron-dense band and a zone of mitochondria in the cortical ooplasm. These
sympatric species have eggs that are similar in size and lecithotrophic larvae that
are planktonic for only a short time. Capitella sp. HI (Capitella jonesi) has ovarian
follicle cells containing a small amount of lipid and no glycogen, while the mature
eggs have a small amount of lipid, abundant glycogen, and large proteid yolk gran-
ules. These small eggs show no evidence of an electron-dense band or any concen-
tration of mitochondria in the cortical ooplasm. This species has planktotrophic
larvae that remain in the plankton for many weeks. Capitella sp. Ilia has ovarian
follicle cells rich in both lipid and glycogen. The large mature eggs are rich in lipid,
have relatively little glycogen, and have abundant proteid yolk granules. The cortical
ooplasm contains electron-dense material similar to that observed in the eggs of
species I and II but it is distributed in a discontinuous band. This species has direct
development, and juvenile worms emerge from the parental brood tube after meta-
morphosis. The egg envelopes and microvilli of the eggs of all four sibling species
undergo substantial morphological changes following release from the ovary into
the coelom.
The significance of these morphological and biochemical differences between
the species is not known, but the lack of intraspecific variation in these characters
suggests that their presence or absence reflects specific differences in the processes
of yolk formation and utilization.
INTRODUCTION
Comparative studies of metazoan sperm structure have demonstrated consid-
erable interspecific variation unprecedented in other cell types. Since one of the
events in speciation is the creation of barriers to crosses between new species and
the parental forms, it is generally thought that the modifications in sperm mor-
phology and their properties may contribute to the establishment of such a barrier.
Baccetti and Afzelius ( 1 976) point out that species specificity not only resides in the
genetic material bound in the nucleus of the spermatozoan but is also imprinted
in the morphology of the cell itself. Thus in some nereid polychaetes, for example.
Received 31 May 1983; accepted 25 July 1983.
379
380 K. J. ECKELBARGER AND J. P. GRASSLE
we observe markedly different sperm types in morphologically similar species
(Hauenschild, 1951; Durchon, 1955). Comparative studies of egg morphology are
rare however, because at the light microscope level at least, female germ cells show
far less structural variation. Aside from differences in volume, color, general shape,
or perhaps features of the egg envelope, there are fewer morphological parameters
available for cytological comparisons than in sperm. However, comparative light
microscope observations on egg morphology in closely related polychaete species
have been reported in orbiniids (Anderson, 1961) spionids, (Blake, 1969) and cir-
ratulids (Gibbs, 1971).
It seems reasonable to assume that in some cases, barriers to cross fertilization
between incipient species might be reflected by morphological changes in the eggs
as they are in sperm. Recent comparative ultrastructural studies of oogenesis in four
species of the sibling complex of Capitella have revealed distinct and consistent
morphological differences in the ovarian follicle cells and mature eggs among mem-
bers of this group. The differences include variation in the relative quantities of
nutritive materials stored in the mature egg which in turn may reflect differences
in the energetic requirements of the larvae. These findings are the first to our knowl-
edge, to describe ultrastructural differences in the female germ cells of closely related
invertebrate species.
Capitella capitata (Fabricius), formerly regarded as an opportunistic, cosmo-
politan polychaete species characteristically present in dense populations in highly
disturbed environments (Grassle and Grassle, 1974; Pearson and Rosenberg, 1978),
recently has been shown to be a complex of more than ten sibling species (Grassle
and Grassle, 1976; Grassle, 1980). Although the morphologies of the adults are very
similar, the species show striking differences in life history features including repro-
ductive mode, breeding season, egg size, and dispersal capability of the larvae.
Marked differences are also observed in the electrophoretic mobilities of allozymes
at selected enzyme loci, indicating that genetic distances between species are great.
In addition, individuals of the various sibling species do not hybridize in the lab-
oratory or in the field (Grassle and Grassle, 1976; Grassle, 1980). The Capitella
species complex is particularly interesting because it represents a wide range of
reproductive variation from species I, which has large eggs (260 ^m), small broods
(30-400 eggs), and a lecithotrophic larval dispersal phase of only a few hours to
species III (Capitella jonesi) which has small eggs (50 nm), large broods (200-1000
eggs), and a planktotrophic larval phase of five weeks or more. The length of oo-
genesis also varies from 5-7 days in species I to 40-50 days in species Ilia. Breeding
seasons range from a short period in winter or early spring (species la and III) to
those which breed throughout the year (species I and II).
MATERIALS AND METHODS
Animals used in this study belong to four genetically distinct sympatric Capitella
species collected from the field in the vicinity of Woods Hole, Massachusetts. The
material from Capitella spp. I, II, and Ilia was obtained from laboratory strains.
Capitella jonesi (Capitella sp. Ill, Grassle and Grassle, 1976) individuals were col-
lected in the field and maintained in the laboratory. Worms were kept in filtered,
standing sea water at 1 5°C and were provided with azoic mud as food and substrate.
Food and water were changed at bi-weekly intervals. For electron microscopy, genital
segments from females and hermaphrodite individuals at various stages of sexual
maturity were cut into small pieces. Tissue fixation and preparation were according
to procedures previously outlined in Eckelbarger (1979). Sections of embedded tissue
CAPITELLA SPP. EGGS AND FOLLICLE CELLS 381
were cut on a Porter-Blum MT-2B ultramicrotome with a diamond knife, stained
with aqueous saturated uranyl acetate followed by lead citrate, and examined with
a Zeiss EM-9S2 electron microscope.
RESULTS
The ovaries of all members of the Capitella sibling species complex examined
are paired, sac-like organs, suspended by mesenteries in the ventral coelomic cavity
throughout the mid-body segments. Each ovary consists of a sac (or follicle) formed
by somatic follicle cells in which the oocytes complete vitellogenesis. The follicle
cells are modified coelomic peritoneal cells which become hypertrophic prior to
vitellogenesis and undergo marked cytological changes including the development
of extensive arrays of rough endoplasmic reticulum (RER) and numerous Golgi
complexes (Fig. 1 ). In the medial region of the ovary, developing oocytes remain
in intimate contact with the layer of follicle cells but gradually lose the association
as they reach their maximum size and expand into the lateral region of the ovary
where they cease growth and await ovulation. When release from the ovary occurs,
possibly resulting from the active migration of follicle cells from the surface of the
eggs (Eckelbarger and Grassle, 1982), the eggs enter the coelom where they float
freely for a variable period before being spawned by the female. Laboratory obser-
vations indicate that the period of coelomic egg storage in the female is minimal
when a sexually mature male is present in the culture. Most ultrastructural features
of the eggs in the lateral region of the ovary are indistinguishable from those floating
freely in the coelom, although the egg envelopes of all four sibling species and the
cortical ooplasm in the egg of species Ilia undergo additional differentiation follow-
ing ovulation. All ovulated eggs have a prominent germinal vesicle and there is no
indication that further maturation occurs before spawning. Numerous ovarian fol-
licle cells, ovarian eggs, and ovulated eggs from many individuals were carefully
examined ultrastructurally in all stages of vitellogenesis in the four sibling species
of Capitella. No intraspecific variation in follicle cell and mature egg morphology
was apparent.
Follicle cells
The ovarian follicle cells of these four members of the Capitella sibling species
complex have many similar ultrastructural features. These include the presence of
large nuclei each with a prominent nucleolus, extensive RER, Golgi complexes, a
variety of membrane-bound, heteromorphic electron-dense bodies resembling ly-
sosomes, bundles of fibrils measuring 5-7 nm, mitochondria, and often a pair of
centrioles (Fig. 2). However, there are consistent differences in the relative number
of glycogen granules and lipid droplets found in these cells throughout the life history
of each species (Table I). Species I and II follicle cells are similar in not possessing
lipid droplets at any stage of oogenesis (Figs. 2, 4) whereas species Ilia cells have
an abundant quantity (Fig. 3). The follicle cells of species III have a small number
of lipid droplets. Except for species III, the follicle cells of each of the species contain
glycogen (Figs. 3, 4). These differences are readily apparent after observing semi-
serial sections from numerous ovaries in many individuals in different stages of
sexual maturity. Since quantitative methods of comparison between the follicle cells
of various siblings would be difficult, we have made qualitative, ultrastructural com-
parisons based on the absence of lipid or glycogen or its presence in small, moderate,
or abundant quantities (Table I).
382
K. J. ECKELBARGER AND J. P. GRASSLE
•
.".'ER,',
f L
>^' * • *r
^ '., ,f., ... ::-^;
^»«^*v;:;;^: /'^^.^gV.V
.*>vi!4£j«>s. >v^--— j .iifessajSssS.STyt •-/ v*V>iP5 ..-*.*: f*K/«^^»*<r' 3 - .»-'.-
FIGURE 1. Stratified layer of follicle cells composing the wall of the ovary in Capitella species III.
N, nucleus; ER, rough endoplasmic reticulum; OC, vitellogenic oocyte. Bar = 5 nm.
FIGURE 2. Follicle cell from ovary of Capitella species I showing centrioles (C), Golgi complex
(G), fibrils (arrowhead) and rough ER. Bar = 0.6 ^m.
FIGURE 3. Lipid (L) droplets in follicle cell of Capitella Ilia ovary. N, nucleus; GL, glycogen. Bar
= 1.5 /im.
FIGURE 4. Rough ER and electron-dense glycogen granules (GL) in the follicle cells of Capitella
species II. Bar = 2 ^m.
FIGURE 5. Large membrane-bound proteid yolk (Y), small lipid droplets (L) and glycogen granules
(GL) in the mature egg of Capitella species III. Bar = 1.5
CAPITELL.4 SPP. EGGS AND FOLLICLE CELLS
383
T3
fli
aj
13
Cortical ooplasm
(coelomic eggs)
Electron-dense band
Mitochondria concentrati
in band
Electron-dense band
Mitochondria concentrati
in band
No electron-dense materi
or mitochondrial band
Random distribution of
organelles
Intermittent band of
electron-dense material
Random distribution of
organelles.
• "^
o .*;
o
T3 («
.y •"?
&SS
IS ^
*^ i«
— — u
IS
*-• -M-
c ^8
C J2 ,C8
•cog.
«-> u- U.
Egg envelo
(coelomic ej
Envelope 1.2 ^n
Slender microvi]
smooth surfac
Envelope 0.8 jun
Smooth microvi
Envelope 0.75 n
thick
Microvilli granu
on lateral suri
a. 6 c5
S|j
13 tf -^
UJ £
u t> .
^J ~~ 4-t *
S3 3 O O
V)
o c c £
• — •
•— <
I*-*
r~-_
>" ^ c3 3.
u
u
+
+
+
00
c
c
C
+ c
UJ
* 1
."2 8
+ IU
(U
+ 00
"2 o
^ 1
a5
C. •"**
'o. >»
Q. >>
O. ^
•r-
J O
JO
JO
35
*j
g,
=0
60
^:
J
+
i
+
^d
U
c
c
c
c
^
^^ V)
u
U
U
—
11
1st
i!
+ o1
.- o
+ , 0
13
+•*
U.
a >>
o, >->
D. ^
'5.
35
JO
JO
JO
ed
U
^2
.£ c
•n
%i
o
— i U?
>< ^
-Si
5 "c
2 §
2 |
O <L)
g
.<J
c5 "*
> -C
> T3
o!
o
^
J a
oo
*
<
Z
sc
a
o
bo
\-
oo
1
88|?
[jj C- 3_
X)
O
Os
0
o
(N
.2 ^~^
O
|
T3
"g
<
1
00
UJ
on
W3
UJ
a
U
U
U
U
».
U
UJ
PJ
UJ
§
Cu
D.
CM
CU
384 K. J. ECKELBARGER AND J. P. GRASSLE
Eggs
The mature eggs of all four sibling species are creamy-white to pale yellow in
color. As many as three types of nutritive material or yolk are recognizable in the
eggs including large, membrane-bound, spherical, proteid yolk bodies, small un-
bound lipid droplets, and electron-dense glycogen granules (Fig. 5). The formation
of these yolk materials has been described in a previous publication (Eckelbarger
and Grassle, 1982). The proteid yolk granules are usually spherical in shape and
vary in size within the same egg. This variation probably results, in part, from a
sectioning artifact in which only a portion of some granules are visible in any one
section. The maximum diameter of yolk granules does show considerable interspe-
cific variation (based on measurement of 100 of the largest yolk granules per egg).
The smallest (averaging 3. 1 ^m) is found in species I and II, and the largest (averaging
4.75 nm) in species III. Those of species Ilia are intermediate in size, averaging 3.7
Mm. Qualitative observations suggest that the number of these granules per unit area
is approximately the same in the ovulated egg of all four sibling species (Figs. 6-9)
with the exception of species Ilia which appears to have many more (Fig. 9).
In addition to differences in species-specific egg diameter and yolk granule di-
ameter, there are also differences in the arrangement and location of cortical or-
ganelles in the eggs. These differences are first observed in the early to middle stages
of vitellogenesis and persist in some species even after release from the ovary. The
cortical regions of the eggs of species I and II are free of all organelles except for
a distinct band of mitochondria (Figs. 10, 11). Apart from their concentration in
a cortical monolayer, these mitochondria are indistinguishable from those present
in the remainder of the ooplasm. There is also a thin (100-120 nm) layer of amor-
phous electron-dense material parallel to the oolemma (Figs. 10, 12, 15, 16, 19).
This circumferentially arrayed band sometimes appears, in favorable sections, to
consist of densely packed but randomly oriented filaments which extend into the
adjacent microvilli in some instances. This band appears during early vitellogenesis
(Fig. 12) while the mitochondrial band appears during the middle stages of yolk
formation. The eggs of species III and Ilia lack the monolayer of mitochondria. The
electron-dense band is absent from the cortical ooplasm of species III eggs but is
present as a discontinuous band in the eggs of species Ilia. The cortical region of
the egg of species III contains the same random mixture of yolk granules and mi-
tochondria as the remainder of the egg (Fig. 13) at all stages of development while
that of species Ilia possesses a unique organelle-free cortical zone up to 2.6 /xm wide
(Fig. 14) which persists until ovulation. Following release from the ovary, the or-
ganelle-free zone disappears and the ooplasmic components become evenly distrib-
uted (Fig. 18). The post-ovulatory arrangement of ooplasmic organelles remains
unchanged in the eggs of species I, II, and III (Figs. 15-17).
Prior to release of the eggs into the coelom, the egg envelopes of all four species
are similar in thickness and in having short, branching microvilli and a simple electron-
dense layer extending from the oolemma to near the tips of the microvilli (Figs. 19-
22). Following ovulation, however, substantial changes are observed in the egg envelope
and the morphology of the egg microvilli (Figs. 23-26). The envelope varies in thickness
from 1 .2 ^m in species I to 0.6 pm in species Ilia. The microvilli covering the eggs
of species I, II, and III have flattened, swollen, or branching tips (Figs. 23-25), while
those of species Ilia are short with constricted tips bearing small granules (Fig. 26).
The lateral surfaces of the microvilli in species I, II, and Ilia are relatively smooth
while those of species III have a granulated appearance (Fig. 25). Table I summarizes
the ultrastructural differences between the coelomic eggs and ovarian follicle cells of
the four sibling species.
CAPITELLA SPP. EGGS AND FOLLICLE CELLS
DISCUSSION
385
The occurrence of sibling species in polychaetes in which members of natural
populations are morphologically similar or identical yet reproductively isolated has
been revealed through the analysis of morphological data, reproductive processes,
,
FIGURES 6-9. Yolk bodies from mature (coelomic) eggs of Capitella sibling species. Figure 6, species
I; Figure 7, species II; Figure 8, species III; Figure 9, species Ilia. Bars = 3.0
386
K. J. ECKELBARGER AND J. P. GRASSLE
.*>. ^'JraSSfiis1* v- «-
fr>r^s .
•
I •":
.•. •
:•*
**^ W. »"- ' . % i
-'iTSFim
U M» " l -v.y-r-^i t. "' " .
*'* •
14
FIGURES 10, 11, 13, 14. Cortical ooplasm of eggs in lateral region of ovary. Eggs have completed
growth and vitellogenesis.
FIGURE 10. Cortical ooplasm of Capitella species I showing band of mitochondria (M). Bar = 2
urn.
FIGURE 1 1. Cortical ooplasm of Capitella species II showing band of mitochondria (M). The thin
layer of amorphous electron-dense material parallel to the oolemma is seen to the right of the mitochondria
in this section.
FIGURE 1 2. Band of amorphous material (arrowheads) adjacent to newly forming microvilli in cortical
of early vitellogenic oocyte of Capitella species I. Bar = 0.53 jmi.
FIGURE 13. Cortical ooplasm of Capitella species III egg. Bar = 1.3 j/m.
FIGURE 14. Cortical ooplasm of Capitella species Ilia showing organelle-free zone. Bar = 1.8
CAPITELL.4 SPP. EGGS AND FOLLICLE CELLS
387
"
.
FIGURES 15-18. Cortical ooplasm of coelomic eggs of Capitella sibling species. Figure 15, species
I; Figure 16, species II; Figure 17, species III; Figure 18, species Ilia. Bars = 3.0
physiological responses, and electrophoretic patterns of related enzymes (see review
by Rice and Simon, 1 980). The present paper is the first to our knowledge to describe
interspecific differences in the eggs and follicle cells of sibling species in any inver-
tebrate. These findings are especially interesting in that not only is interspecific
388
K. J. ECKELBARGER AND J. P. GRASSLE
• <f ?li~ — -" "••«.' -' i KWW •_ * W"5 *'fti~'i'
v^'-X- »*jr2£8T* JP*aSS
•1- 2
rfTnW?
FIGURES 19-22. Egg envelop)es of ovarian eggs of Capitella sibling species. Figure 19, species I;
Figure 20, species II; Figure 21, species III; Figure 22, species Ilia. Note the amorphous material (arrows)
parallel to the oolemma in Figure 19. Bars = 0.63
variation on the ultrastructural level demonstrated but also that the variation occurs
in the female gamete which generally shows little gross morphological variation.
The significance of differences in cortical organelle distribution or type of nutrient
material in the eggs of Capitella sibling species is not readily apparent but it does
not appear to bear any obvious relationship to egg size, cleavage pattern, or type
of larval development.
CAPITELLA SPP. EGGS AND FOLLICLE CELLS
389
o
23 *
HP*
.
' l£l
FIGURES 23-26. Egg envelopes of coelomic eggs of Capitella sibling species. Figure 23, species I;
Figure 24, species II; Figure 25, species III; Figure 26, species Ilia. Note the band of amorphous material
(arrowheads) adjacent to the oolemma in Figures 23 and 24. Note also the granules attached to the lateral
surfaces of the microvilli in Figure 25 and to the microvillar tips in Figure 26. Bars = 0.6
The use of ultrastructural characters in phylogeny and systematics is gradually
gaining support (see review by Tyler, 1979). With regard to Capitella, some of the
ultrastructural differences observed in the eggs of the four sibling species are further
390 K. J. ECKELBARGER AND J. P. GRASSLE
evidence of morphological divergence in this taxonomically difficult species group
and may have systematic applications. Some features of the eggs such as the cortical
mitochondria in species I and II are not strictly ultrastructural characters since they
are discernible with careful light microscopy. However, the cortical band of amor-
phous material observed in the eggs of species I, II, and Ilia, is only visible with
electron microscopy. These additional morphological features may be of systematic
importance when combined with the abundant information already available on
adult morphology, genetics, and reproductive and life history characteristics (Grassle
and Grassle, 1976; Grassle, 1980). Interspecific differences in the relative size, num-
ber, and morphological features of the large proteid yolk granules also appear to
exist, although it is difficult to establish homology, between them. Although they
appear to have similar origins (Eckelbarger and Grassle, 1982), it is likely they have
very different chemical composition despite their morphological similarity. The use
of various morphological features of yolk granules as systematic characters has been
proposed in some invertebrate oocytes (Gremigni, 1979) although this approach has
been strongly criticized (Tyler, 1981).
The concentration of mitochondria in the cortical ooplasm in the eggs of Cap-
itella species I and II is unusual for a polychaete but not uncommon for other
animal oocytes (Raven, 1961; Arnold, 1971; Boyer 1972). Localization or stratifi-
cation of ooplasmic components was termed "ooplasmic segregation" by Costello
(1948) and quite often is restricted to the animal pole or polar lobe of the egg (Allen,
1961; Raven, 1961; Anderson and Huebner, 1968; Huebner and Anderson, 1976).
In Diopatra cuprea, for example, Anderson and Huebner (1968) found yolk granules
to be vegetally located and lipid and mitochondria were found in the animal pole.
This localized stratification was even retained during early cleavage. Costello (1945,
1948) reported ooplasmic segregation in Nereis limbata only following fertilization.
Recently, Eckberg (1981), using electron microscopy to study the eggs of Chaetop-
tems pergamentaceus, reported that cytoplasmic components are localized in dif-
ferent regions of the egg and that this localization is maintained as the embryo
undergoes cleavage and differentiation. Hess ( 197 1 ) noted that ooplasmic organelles
such as yolk bodies, mitochondria, and endoplasmic reticulum, as well as cellular
products such as various types of RNA and metabolites, are unevenly distributed
during ooplasmic segregation but are later evenly distributed to the blastomeres
during the process of cleavage. The presence of a cortical, organelle-free zone in the
egg of species Ilia prior to ovulation, and its disappearance following release from
the ovary, is a developmental event previously unreported in an annelid egg. Its
significance is unknown.
The functional importance of mitochondrial segregation in Capitella eggs is
unknown. It is clearly tempting to try to relate ooplasmic segregation to mosaic egg
development. However, it has been demonstrated that displacement of cell organelles
by reverse centrifugation of some mosaic eggs does nothing to alter development
(Clement, 1968). Huebner and Anderson (1976) suggested that a similar distribution
of cortical mitochondria in the egg of the hemipteran Rhodnius prolixus might
reflect the need for an energy source for pinocytosis by the oolemma. Although this
is possible in some eggs, it seems unlikely for Capitella eggs since only the eggs of
species I and II have this mitochondrial layer even though the level of endocytotic
activity appears to be the same in the eggs of all the sibling species examined.
The significance of the amorphous electron-dense band in the cortical ooplasm
of the eggs of species I, II, and Ilia is obscure. Some micrographs suggest that this
layer is composed of fine filaments although this is uncommon in oocytes. Anderson
( 1 969) described a prominent layer of filaments parallel to the oolemma in the
CAPITELLA SPP. EGGS AND FOLLICLE CELLS 391
developing oocytes of the amphineurans Mopalia mucosa and Chaetopleura api-
culata but did not speculate as to their possible significance. The amorphous sub-
stance observed in the eggs of Capitella might represent a less organized, non-fila-
mentous form of microfilament similar to that described in the sperm duct epithe-
lium of the ascidian dona intestinalis by Woollacott and Porter (1977). If the
material in Capitella eggs indeed represents a microfilament reserve, the precise role
of the putative organelles is problematical. They could serve a structural function,
or be involved in morphogenetic movements, the fertilization reaction, or perhaps
in the movement of mitochondria into the cortical ooplasm.
Wide variation in egg envelope morphology has been reported in different genera
of polychaetes within the same family (Eckelbarger, in press) but never among
closely related species of the same genus. This variation may be related to differences
in the types of yolk precursors and metabolites being absorbed by the eggs during
vitellogenesis or to the development of cross fertilization barriers. The morphological
changes observed in the egg microvilli before and after ovulation in Capitella have
not been previously described in polychaetes. This demonstrates that additional
differentiation of the egg can occur following separation from the investing follicle
cells which appear to be crucial to yolk synthesis.
Follicle cells are often associated with developing oocytes in polychaetes (Eck-
elbarger, in press) but extensive deposits of lipid and glycogen, as reported here
in some Capitella ovaries, are uncommon. Eckelbarger (1979) reported some lipid
and extensive deposits of glycogen in the follicle cells associated with the ovary in
Phragmatopoma lapidosa. These deposits were believed to be utilized by the de-
veloping oocytes during vitellogenesis. Heacox and Schroeder (1981) observed lipid
in the follicle cells associated with the gonial cell clusters in Typosyllis pulchra which
they suggested might serve as nutrient material for the oocytes during development.
In many polychaetes, the coelomic peritoneum often stores lipid and glycogen which
are believed to serve a nutritive function during vitellogenesis, particularly in species
undergoing extraovarian oogenesis (Eckelbarger, 1983). The ovarian follicle cells of
Capitella are derived from the peritoneum, and the lipid and glycogen stores are
believed to be destined for the developing oocytes (Eckelbarger and Grassle, 1982).
The differences reported here in the relative quantities of lipid and glycogen in
the ovarian follicle cells of Capitella sibling species, presumably reflect the ultimate
differences in types and quantities of yolk materials stored in the eggs. This in turn
probably has a significant effect on embryogenesis and larval development partic-
ularly when egg size and subsequent developmental pattern (i.e., planktotrophy
versus lecithotrophy) vary so widely between the Capitella species under discussion.
There are apparent differences in the general types of nutritive materials stored in
the eggs of Capitella but unfortunately nothing is known of their chemical nature.
It is tempting to use egg size as a unit of adult energy expenditure since it has been
widely used in theoretical considerations of life history patterns (see Steams, 1976)
but it can be a misleading parameter which ignores organic composition (Turner
and Lawrence, 1979). Indeed, in a study of the eggs of several invertebrate groups
including polychaetes, Strathmann and Vedder (1977) reported that although or-
ganic matter per egg increases with egg diameter or volume, it is not directly pro-
portional to egg volume because small eggs have more concentrated organic matter
than larger eggs. It will be of interest to quantify the various organic components
stored in the mature eggs of these Capitella sibling species (i.e., proteid yolk, lipid,
glycogen), to see how these materials might be utilized during embryogenesis and
early larval development. This should help us better understand the developmental
and ecological implications of the disparate distribution of nutrient material in the
392 K. J. ECKELBARGER AND J. P. GRASSLE
follicle cells and eggs of Capitella. Laboratory studies of inbred strains of the le-
cithotrophic Capitella spp. I and II indicate that there is marked variation between
lines in the capacity of the larvae to delay settlement in the absence of suitable
substrate without suffering post-settling mortality (Grassle, unpub.)- We expect these
differences to be reflected in between-line differences in the amounts and/or types
of nutritive materials incorporated into the eggs.
ACKNOWLEDGMENTS
The authors wish to acknowledge the technical assistance of C. Gelfman, P. A.
Linley, S. W. Mills, and K. G. Panker. This paper is contribution No. 347 of the
Harbor Branch Foundation.
LITERATURE CITED
ALLEN, M. J. 1961. A cytochemical study of the developing oocytes and attached nurse cells of the
polychaetous annelid Diopatra cuprea (Bosc). Ada Embryol. Morphol. Exp. 4: 219-238.
ANDERSON, D. T. 1961. The development of the polychaete Haploscoloplos fragilis. Q. J. Microsc. 102:
257-272.
ANDERSON, E. 1969. Oocyte-follicle cell differentiation in two species of amphineurans (Mollusca), Mo-
palia mucosa and Chaetopleura apiculata. J. Morphol. 129: 89-126.
ANDERSON, E., AND E. HUEBNER. 1968. Development of the oocyte and its accessory cells of the poly-
chaete Diopatra cuprea (Bosc). J. Morphol. 126: 163-198.
ARNOLD, J. M. 1971. Cephalopods. Pp. 265-310 in Experimental Embryology of Marine and Freshwater
Invertebrates, G. Reverberi, ed. American Elsevier Publishing Co., New York.
BACCETTI, B., AND B. A. AFZELIUS. 1976. The Biology of the Sperm Cell. S. Karge Publishing, Basel.
254 pp.
BLAKE, J. A. 1969. Reproduction and larval development of Polydora from northern New England.
(Polychaeta: Spionidae). Ophelia 7: 1-63.
BOYER, B. C. 1972. Ultrastructural studies of differentiation in the oocyte of the polyclad turbellarian,
Prostheceraeus floridanus. J. Morphol. 136: 273-296.
CLEMENT, A. C. 1968. Development of the vegetal half of the Ilyanassa egg after removal of most of
the yolk by centrifugal force, compared with the development of animal halves of similar visible
composition. Dev. Biol. 17: 165-186.
COSTELLO, D. P. 1945. Segregation of ooplasmic constituents. J. Elisha Mitchell Sci. Soc. 61: 277-289.
COSTELLO, D. P. 1948. Ooplasmic segregation in relation to differentiation. Ann. N.Y. Acad. Sci. 49:
663-683.
DURCHON, M. 1955. Sur le polymorphisme presente par quelques nereidiens (annelides polychetes) au
moment de la reproduction. Bull. Soc. Hist. Nat. Afr. Nord. 46: 180-193.
ECKBERG, W. R. 1981. An ultrastructural analysis of cytoplasmic localization in Chaetopterus perga-
mentaceus. Biol. Bull. 160: 228-239.
ECKELBARGER, K. J. 1979. Ultrastructural evidence for both autosynthetic and heterosynthetic yolk
formation in the oocytes of an annelid. (Phragmatopoma lapidosa: Polychaeta). Tissue Cell 11:
425-443.
ECKELBARGER, K. J. In Press. Comparative aspects of oogenesis in polychaetes. In Polychaete Repro-
duction: Progress in Comparative Reproductive Biology, A. Fischer and H.-D. Pfannenstiel, eds.
Proc. of Symposium, Helgoland, West Germany, April 1982. Fortschritte der Zoologie, (M.
Lindauer, ed.), Wurzburg.
ECKELBARGER, K. J. 1983. Evolutionary radiation in polychaete ovaries and vitellogenic mechanisms:
Their possible role in life history patterns. Can. J. Zoo/. 61: 487-504.
GIBBS, P. E. 1971. A comparative study of reproductive cycles in four polychaete species belonging to
the family Cirratulidae. J. Mar. Biol. Assoc. U.K. 51: 745-769.
GRASSLE, J. F., AND J. P. GRASSLE. 1974. Opportunistic life histories and genetic systems in marine
benthic polychaetes. J. Mar. Res. 32: 253-284.
GRASSLE, J. P., AND J. F. GRASSLE. 1976. Sibling species in the marine pollution indicator Capitella
(Polychaeta). Science 192: 567-569.
GRASSLE, J. P. 1980. Polychaete sibling species. Pp. 25-32 in Aquatic Oligochaete Biology, R. O. Brink-
hurst and D. G. Cook, eds. Plenum Publishing Corp., New York.
GREMIGNI, V. 1979. An ultrastructural approach to planarian taxonomy. Syst. Zoo/. 28: 345-355.
CAPITELLA SPP. EGGS AND FOLLICLE CELLS 393
HAUENSCHILD, C. 1951. Nachweis der sogenannten atoken Geschlechtsform des Polychaeten Platynereis
dumerilii Aud. et. M.-Edw. als eigene Art auf Grund von Suchtversuchen. Zoo/. Jahrb. Abt.
Allg. Zoo/. Physiol. Tiere. 63: 107-128.
HEACOX, A. E., AND P. C. SCHROEDER. 1981. A light and electron microscopic investigation of ga-
metogenesis. Cell Tissue Res. 218: 641-658.
HESS, O. 1971. Fresh water gastropods. Pp. 215-247 in Experimental Embryology of Marine and Fresh-
water Invertebrates, G. Reverberi, ed. American Elsevier Publishing Co., New York.
HUEBNER, E., AND E. ANDERSON. 1976. Comparative spiralian oogenesis — structural aspects: An over-
view. Am. Zoo/. 16: 315-343.
PEARSON, T. H., AND R. ROSENBERG. 1978. Macrobenthic succession in relation to organic enrichment
and pollution of the marine environment. Oceanogr. Mar. Biol. Assoc. Rev. 16: 229-3 1 1 .
RAVEN, C. 1961. Oogenesis: The Storage of Developmental Information. Pergamon Press, New York.
247 pp.
RICE, S. A., AND J. L. SIMON. 1980. Intraspecific variation in the pollution indicator polychaete Polydora
ligni (Spionidae). Ophelia 19: 79-1 15.
STEARNS, S. C. 1976. Life history tactics: A review of the ideas. Q. Rev. Biol. 51: 3-47.
STRATHMANN, R. R., AND K. VEDDER. 1977. Size and organic content of eggs of echinoderms and other
invertebrates as related to developmental strategies and egg eating. Mar. Biol. 39: 305-309.
TURNER, R. L., AND J. M. LAWRENCE. 1979. Volume and composition ofechinoderm eggs: implications
for the use of egg size in life-history models. Pp. 25-40 in Reproductive Ecology of Marine
Invertebrates, S. E. Stancyk, ed. Univ. South Carolina Press, Columbia.
TYLER, S. 1979. Contributions of electron microscopy to systematics and phylogeny: introduction to the
symposium. Am. Zoo/. 19: 541-543.
TYLER, S. 1981. Ultrastructural characters in planarian taxonomy: another point of view. Am. Zoo/.
30(1): 71-73.
WOOLLACOTT, R. M., AND M. E. PORTER. 1977. A synchronized multicellular movement initiated by
light and mediated by micronlaments. Dev. Biol. 61: 41-57.
Reference: Biol. Bull. 165: 394-408. (October, 1983)
ULTRASTRUCTURE OF EARLY EMBRYONIC SHELL FORMATION
IN THE OPISTHOBRANCH GASTROPOD AEOLIDIA PAPILLOSA*
LINDA S. EYSTER
Marine Science and Maritime Studies Center, Northeastern University, Nahant. Massachusetts 01908
ABSTRACT
Early shell formation was examined in embryos of the opisthobranch gastropod
Aeolidia papillosa. Secretion of the first organic shell material occurs prior to closure
of the shell gland lumen, contrary to reports for other molluscan embryos. This
difference suggests that in externally shelled gastropods and bivalves initiation of
shell secretion may be coincident with narrowing of the shell gland pore rather than
with closure of the lumen. The shell growth region was examined ultrastructurally.
As no shell material is seen in the shell gland lumen, the shell gland seems not to
be actively involved in shell secretion. Initial shell material is secreted only by cells
surrounding the shell gland pore. Shell material seems to be added, not in a gap
between cells as previously described, but over the apical surface of cells at the
growing edge. The growing edge of the shell and the growing edge cells are covered
by cytoplasmic extensions arising from the neighboring cells distal to the shell gland.
No infoldings of the growing edge cell membranes are seen. The first organic shell
material is 20 nm thick, consists of two electron dense layers separated by an electron
lucent layer, and is secreted at least 33 hours (5°C) before shell mineralization, as
detected by polarizing microscopy.
INTRODUCTION
Most molluscs secrete external calcareous shells. Although this secretion begins
during early embryogenesis, most of our knowledge of shell formation is derived
from studies of post-embryonic molluscs (e.g., Wilbur and Jodrey, 1952; Bevelander
and Nakahara, 1969; Wilbur, 1972; Saleuddin, 1974; Weiner and Hood, 1975;
Young et al, 1977a, b; Wheeler et ai, 1981.). Kniprath (1981) summarized the
literature on the development, morphology, and function of the embryonic shell
gland and shell field in molluscs; little is known about how, when, and where em-
bryonic shell material is secreted. To date, the only molluscs in which embryonic
shell formation has been studied ultrastructurally are the marine bivalves Mytilus
galloprovincialis and M. edulis (Humphreys, 1969; Kniprath, 1980b), the freshwater
pulmonate Lymnaea stagnalis (Kniprath, 1977), the terrestrial pulmonate Helix
aspersa (Kniprath, 1980a), the freshwater prosobranch Marisa cornuarietis (Kni-
prath, 1979), and the chitons Lepidochitona cinera and Ischnochiton rissoa (Haas
et al, 1979; Kniprath, 1980c).
The region of ectodermal cells responsible for embryonic shell secretion is called
the shell field. Preceding embryonic shell formation in all gastropod and bivalve
molluscs, a region of the dorsal shell field invaginates to form the "shell gland"
Received 28 March 1983; accepted 20 July 1983.
a Contribution No. 1 1 3 of the Marine Science and Maritime Studies Center Northeastern University,
Nahant, MA.
394
GASTROPOD EMBRYONIC SHELL FORMATION 395
(Pelseneer, 1906; chitons lack a true shell gland, see Kniprath, 1981). It is because
this invagination always forms in externally shelled species, that it has been assumed
to have an active function in shell formation. The invaginated region has been
referred to as the shell gland since 1873 (see Kniprath, 1979) although its actual
role in shell formation has been little studied and remains unclear. After the shell
gland invaginates, its lumen "closes" (narrows to a canal open to the outside through
a pore; see Kniprath, 1979, Figs. 2a, b; 1980, Fig. le). The shell gland later evaginates
or spreads back to a non-invaginated shell field.
In 1979 Kniprath outlined three aspects of early shell formation that were in
a state of confusion and that warranted further examination: 1.) At what devel-
opmental stage of the shell gland is the first shell material secreted? 2.) Which cells
secrete the first shell material? and 3.) How does evagination of the shell gland
proceed. Although several authors have addressed the first two problems (e.g., Hum-
phreys, 1969; Kniprath, 1980a, b), conflicting results have been presented and a
clear description of the cells at the leading or growing edge of the embryonic shell
is lacking. Also absent are precise date on the shape of the shell gland at the time
of secretion of the first organic shell material.
The present work provides the first ultrastructural description of embryonic shell
formation in a marine gastropod. In this paper I describe when and where the first
organic shell material is observed in embryos of the nudibranch Aeolidia papillosa.
The fine structure of the first shell material and of the cells at the early growing
edge of the shell are also examined.
MATERIALS AND METHODS
Reproductively active specimens of Aeolidia papillosa were collected subtidally
near Nahant, MA using SCUBA and were placed in a flow-through sea water aquar-
ium. Adults and young were thus exposed to natural temperature (5°C) and salinity
(30%o) conditions. Egg masses laid on the aquarium walls soon after incarceration
of the adults were allowed to harden for a few days before they were carefully
removed and placed in wide-mesh baskets suspended in the aquarium.
Capsules containing embryos were removed mechanically from the egg masses
and examined under a compound microscope to confirm synchrony, normality, and
stage of embryonic development. Polarizing microscopy (pieces of polarizing film
set at maximum extinction) was used to ascertain initiation of shell formation since
the initial shell material is not detectable with standard light microscopy. Birefrin-
gence in the shell field observed with polarizing microscopy indicates shell miner-
alization rather than presence of the organic portion of the shell since treatment of
embryos with the calcium chelator EGTA (10 mM ethylene-glycol-bis-N,N'-tetra
acetic acid) resulted in loss of the birefringence. Since secretion of organic shell
material precedes shell mineralization, by the time birefringence was detectable I
knevv' that the first organic shell material had already been secreted. Because of the
functional relationship between the organic materials and inorganic mineralization,
timing of the various developmental stages is given in hours preceding detectable
birefringence (Fig. 19).
Once gastrulation had begun and until calcareous shell material was detectable
with polarizing microscopy, embryos within their egg capsules were removed pe-
riodically from the egg masses and fixed. Embryos were held in fixative in a refrig-
erator up to 3 days, until the last sample was fixed. A variety of primary fixatives
were tried; the best results were obtained with 3% glutaraldehyde, 1% formaldehyde
with paraformaldehyde (Karnovsky, 1965), 3% NaCl, 4.5% sucrose in 0.1 A/phos-
396 L. S. EYSTER
phate buffer, with dimethylsulfoxide added to aid penetration of the fixative (pH
7.4). Embryos were washed at room temperature in 0.1 M phosphate buffer with
8% sucrose and post-fixed in 2% OsO4 in 0.2 M phosphate buffer for 1 h. The tissue
was dehydrated in a graded series of ethanol to 100% and was infiltrated with and
embedded in Spurr low viscosity embedding medium (Spurr, 1969).
Embryos were sectioned at random orientation since they could not be oriented.
For light microscopy, from 10-72 embryos at each stage were serially sectioned
(0.5-1.0 /urn). Sections were cut using glass knives, mounted onto glass slides, and
stained with Richardson's stain (Richardson et ai, 1960). For transmission electron
microscopy (TEM), thin sections were cut on a Sorvall MT-2B ultramicrotome using
glass or diamond knives, mounted onto copper grids, and stained 15 minutes in
saturated aqueous uranyl acetate followed by lead citrate. For TEM localization of
periodate-reactive carbohydrates, thin sections were mounted onto gold grids and
exposed sequentially to periodic acid (PA), thiosem'icarbazide (TSC), and silver pro-
teinate (SP); appropriate controls were run simultaneously (Thiery, 1967; Porter and
Rivera, 1979). Thin sections were examined and photographed on an RCA EMU-
4 transmission electron microscope.
RESULTS
Initiation of shell secretion
The shell gland invagination is present by 2 days (5°) prior to first detectable
birefringence in embryos ofAeolidia papillosa (Figs. 1, 2). At this stage embryonic
ciliation is just visible with a compound microscope; the embryos move slightly
inside their capsules but do not spin actively. Based on the large size of the shell
gland pore (about 17 /im) and irregular outline of the shell gland lumen, the shell
gland seems to be still forming. No secreted shell material is evident at this stage
with transmission electron microscopy although the dense granules believed to be
involved in shell formation (see Fig. 10) are already present. At all stages of devel-
opment, the shell gland lumen is lined with scattered microvilli.
In embryos fixed three hours later, at 43 hours before birefringence is observed,
the shell gland lumen is more circular in section and up to 26 nm wide and 30
nm deep (Fig. 3). Also the shell gland pore has become smaller. The smallness of
the pore and the fact that the embryos are insufficiently differentiated to be oriented
prior to sectioning makes it difficult to obtain sections passing through both the
pore and the lumen of the shell gland at this and all later stages. No secreted shell
material is observable with transmission electron microscopy in embryos at this
stage (Fig. 4).
In embryos fixed 33 hours before the first birefringence, the shell gland has
changed to a more oval shape with a shell gland neck that is narrower than the rest
of the shell gland lumen (Figs. 5, 6). It is at this stage that the first organic shell
material is observed. The shell material covers the opening of the shell gland pore.
Additions of new organic material are made at the growing edge, away from the
shell gland (Fig. 6).
Two areas of shell growth are seen in each section. The cells directly beneath
the zone of shell growth are referred to as "GE cells" because of their proximity to
the growing edge (GE). The cells adjacent to the GE cells but distal to the shell gland
are referred to as "MV cells" because of their characteristic abundance of microvilli
(see Figs. 20, 21).
At 23 hours prior to the first detectable shell mineralization, the shell gland has
closed to a narrow canal but is still open to the outside through a small pore (Figs.
GASTROPOD EMBRYONIC SHELL FORMATION
397
^W^Jife
pore
i
SG
urn en
iff*"
*;*•*
wm .,
MV
GE
.,»- • >-
I . :-V, ~& '" w^>j
MV
- GE:
O f^ '
OvJJ
' •«§*.
*.
FIGURES 1 , 2. Micrographs of sections through Aeolidia papillosa embryos fixed 46 h prior to shell
mineralization (5°C). FIGURE 1. Light micrograph. Embryos are within egg capsule (C) and show shell
gland (SG) and archenteric (A) invaginations. Bar = 50 nm. FIGURE 2. Transmission electron micrograph
(TEM) showing pore and lumen of shell gland (SG) prior to shell secretion. Bar = 5 nm.
FIGURES 3, 4. Micrographs of sections through embryos 43 h prior to shell mineralization. The
shell gland (SG) is open to outside through a pore, not visible here or in Figure 5 due to sectioning angle.
FIGURE 3. Light micrograph. Bar = 50 ^m. FIGURE 4. TEM of region similar to box in Figure 3, showing
growing edge cells (GE) with electron dense granules (arrows), and microvilli-bearing cells (MV). Bar
= 1
398
L. S. EYSTER
FIGURES 5, 6. Micrographs of sections through embryos fixed 33 h prior to shell mineralization.
FIGURE 5. Light micrograph. Bar = 50 nm. FIGURE 6. TEM of newly secreted shell material (arrows)
lying over the pore of the shell gland (SG), the proximal cells (P), and growing edge cells (GE). Also
shown are microvilli-bearing cells (MV), distal cells (D), and one cluster of vesicles present in proximal
cells (circle). Bar = 5 nm.
FIGURE 7. Light micrograph of section through embryos fixed 23 h prior to shell mineralization.
The shell gland (SG) has "closed" to a narrow canal. Bar = 50
GASTROPOD EMBRYONIC SHELL FORMATION 399
7, 19). The lumen of the shell gland canal is still lined with scattered micro villi but
the number of microvilli seen in any section is greater after narrowing of the shell
gland lumen. This apparent increase in abundance of microvilli may reflect de-
creased distance between cells lining the lumen rather than an actual increase in
number of microvilli. Although Figures 8 and 11-18 are all from embryos fixed 23
h prior to mineralization, at this stage MV cells, GE cells, and the shell itself have
the same morphological characteristics observed in embryos fixed 10 hours earlier.
The morphology of these cells and of the secreted shell material are described below.
Morphology of the shell growth region
The zone of shell growth is near the apical surface of the GE cells. The GE cells
are columnar, have rough endoplasmic reticulum associated with sub-basal nuclei,
and have fields of periodic acid-thiosemicarbazide-silver proteinate (PA-TSC-SP)
positive material, presumed to be glycogen. These cells are readily identified by the
presence of numerous membrane-bound granules (Figs. 8, 9). In section the granules
are either circular or oblong and have a maximum length of 200 nm (Fig. 10). The
granules are frequently seen in association with Golgi apparati just apical to the
nucleus (Fig. 1 1); often near the cell apices (Fig. 8); occasionally within apical cy-
toplasmic extensions (Fig. 9); but never outside of the cell.
Electron cytochemistry is currently being utilized to determine if the granules
contain potential organic or inorganic shell components. The major organic shell
component in molluscs is protein (Wilbur, 1972), but no stains are specific for
protein (Hayat, 1970). Because polysaccharides are also present in molluscan shells
(Wilbur, 1972), the PA-TSC-SP procedure was used. Preliminary tests with the PA-
TSC-SP procedure indicate that the granules do not contain carbohydrates oxidiz-
able with periodic acid. The granules are electron lucent in glutaraldehyde-osmium
fixed sections, but are very electron dense after sequential staining with uranyl
acetate and lead citrate. No distinct substructure was observed in stained or un-
stained granules at a magnification of 500,OOOX.
At all stages prior to mineralization, the shell consists of two electron dense
layers separated by an electron lucent layer or gap (Fig. 12). In embryos fixed 33
hours prior to detectable mineralization, the shell material seen in section was up
to 10 pm long (following all contours) and 22 nm thick. Clusters of small vesicles,
most about 15-60 nm in diameter, are associated with the outer surface of the shell
(Figs. 13, 14). These clusters appear to be randomly scattered.
The growing edge of the shell either abruptly terminates (Figs. 9, 16) or consists
of small electron dense particles (Figs. 8, 18, 20). Regardless of its exact morphology,
the edge of the newly formed shell material is always located on the apices of the
GE cells and never between the lateral plasma membranes of the GE cells and the
neighboring MV cells. No secretory infoldings of the lateral plasma membranes of
the GE cells were observed.
The proximal cells, those cells adjacent to the GE cells and proximal to the shell
gland, occasionally have infolded apical plasma membranes (Fig. 17). These cells
also occasionally contain electron dense granules as described in the GE cells. One
consistent feature of these cells is their association with apical-lateral intercellular
spaces lined with microvilli. A space was consistently observed between the GE cells
and proximal cells (Figs. 16-18, 20). The shell extending over this intercellular space
is almost entirely separated from it by cytoplasmic extensions arising from the cells
lining the space (Figs. 16, 18). These cytoplasmic extensions are in intimate contact
with the inner surface of the shell (Figs. 16, 18). Groups of small uncoated vesicles
400
L. S. EYSTER
Wl
• ...
v
-.*-.
-'.••
O
MY ,
m
FIGURE 8. Transmission electron micrograph of apices of: cell distal to shell gland (D), microvilli-
bearing cell (MV), growing edge cell (GE), and proximal cell (P). The growing edge of the shell lies over
the GE cell and here consists of small electron dense particles (arrows). Bar = 1 ^m.
FIGURE 9. Apex of growing edge cell (GE), characterized by electron dense granules; from embryo
fixed 33 h prior to mineralization. A cytoplasmic extension of the microvilli-bearing cell (MV) lies over
the growing edge of the shell (arrows). Bar = 1 pm.
FIGURE 10. Electron dense granules in GE cell, from Figure 9. Bar = 200 nm.
GASTROPOD EMBRYONIC SHELL FORMATION
401
**.*
'• • '--
FIGURE 1 1. Electron dense material associated with Golgi complex. Bar = 500 nm.
FIGURE 12. Section showing two electron dense layers of shell (arrows) lying close to plasma
membrane (arrowheads) of shell field cell. Sections through microvilli are shown at top. Bar = 200 nm.
FIGURES 13, 14. Clusters of small vesicles associated with outer surface of shell (arrows). FIGURE
13. The dense layer below the shell is the plasma membrane (arrowheads). Bar = 200 nm. FIGURE 14.
Bar = 1 nm.
FIGURE 15. Small vesicles seen in proximal cells. Bar = 500 nm.
circular to pear-shaped in profile and about 100 nm in diameter are present in the
proximal cells (Figs. 6, 15). These vesicles have lucent cores but are larger and have
a much denser border than the vesicles present on the outer surface of the shell.
Only one MV cell with numerous microvilli is observed in section at each grow-
ing edge (Figs. 6, 16). In comparison, the distal cells (cells adjacent to MV cells but
distal to shell gland), never have more than a few scattered microvilli (Figs. 6, 8,
17, 21). The MV cells do not contain the electron dense granules typical of the GE
cells but both the MV cells and GE cells have numerous mitochondria apically. The
MV cells are joined apically to the GE cells by zonulae adhaerens and septate
402
L. S. EYSTER
FIGURE 16. The microvilli of the MV cell lean over the growing edge cell (GE) and over the growing
edge of the shell. Two clusters of vesicles are shown (arrowheads); rarely were clusters seen not in contact
with the outer shell surface (arrows). P = proximal cell; C = capsule. Bar = 1 ^m.
FIGURE 1 7. Infoldings (arrowheads) of the apical plasma membranes of the proximal cells (P) were
observed rarely; no infoldings of other shell fields cells were observed. The growing edge of the shell is
covered by abundant microvilli of the microvilli-bearing cell (MV). The distal cell (D) has only sparce
microvilli. Arrows = shell. Bar = 1 nm.
FIGURE 18. An intercellular space (ICS) occurs between the proximal cells (P) and growing edge
cells (GE). Some sections show long cytoplasmic extensions from the microvilli-bearing cells (MV) cov-
ering the growing edge of the shell (arrows). Bar = 1
desmosomes. The microvilli of the M V cells tend to lean over the GE cells and the
growing edge of the shell (Figs. 16, 17). Long cytoplasmic extensions that arise from
the inner edges of the MV cells also reach over the growing edge of the shell and
may completely cover the apical surfaces of the GE cells (Figs. 9, 1 8).
GASTROPOD EMBRYONIC SHELL FORMATION
403
262
312 315 325 335
96
46 43 33 23
358
FIGURE 19. Schematic diagram of changes in shell gland morphology related to time (hours) after
oviposition (top scale) and time prior to detection of shell mineralization (bottom scale), 5°C. Drawings
represent sections through embryos at gastrulation and as in Figures 1, 3, 5, and 7. Not to scale.
DISCUSSION
Timing of first shell secretion
The first shell material in embryonic molluscs is secreted sometime during the
existence of the shell gland (Gather, 1967; Demian and Yousif, 1973; Kniprath,
FIGURE 20. Schematic diagram showing arrangement of shell and early shell field cells, at about
30 h prior to detection of shell mineralization. The shell (S) consists of two electron dense layers (arrows).
At its growing edge, the shell consists of small electron dense particles lying on the apical surface of the
growing edge cell (GE) and is covered by cytoplasmic extensions arising from the microvilli-bearing cell
(MV). Also shown are proximal (P) and distal (D) shell field cells, named in terms of their proximity to
the shell gland (SG). M = mitochondria. GR = granules. G = Golgi complex. RER = rough endoplasmic
reticulum. ICS = intercellular space. Not to scale.
404 L. S. EYSTER
21
FIGURE 2 1 . Schematic diagram of hypothetical arrangement of shell and early shell field cells, in
surface view. At the left the shell (S) covers the underlying cells and extends to the microvilli-bearing
cells (MV). The same region is redrawn (right) with the secreted shell material removed to reveal un-
derlying cells and pore of the shell gland (X). P = proximal cells. GE = growing edge cells. D = distal
cells. Not to scale.
1977). However, the shell gland of developing embryos is present for hours and goes
through several morphologically distinct stages or shape changes. Only Kniprath
(1977, 1979, 1980b) has specifically examined the morphological stage of the shell
gland at which shell secretion begins.
In Aeolidia papillosa the first organic shell material is secreted at least 10 hours
before the shell gland ""closure" stage (see Fig. 19) and at least 33 hours prior to
detection of shell mineralization. The electron dense granules believed to contain
components of organic shell material were present at least 1 3 hours prior to obser-
vation of first secreted shell material and 23 hours prior to shell gland closure. This
is in contrast to the findings of Kniprath (1981) who stated that in the species he
examined the cells of the shell field "do not synthesize anything for secretion before
the closure stage". It is uncertain whether the cells lining the shell gland lumen ever
become tightly apposed in Aeolidia papillosa.
In contrast, in the snails Lymnaea stagnalis (Kniprath, 1977) and Marisa corn-
uarietis (Kniprath, 1979), the shell gland lumen closes to a canal prior to shell
secretion. In the mussel Mytilus galloprovincialis (Kniprath, 1980b) the walls of the
shell gland seem to close so tightly that not even a narrow canal is detectable with
transmission electron microscopy. In that species, the shell gland lumen is apparently
completely gone prior to shell secretion. Thus closure of the shell gland lumen
precedes shell secretion in all three of these species, but not in Aeolidia papillosa.
Secretion of the first shell material while the shell gland is still open in Aeolidia
papillosa demonstrates that closure of the shell gland lumen is not requisite to
initiation of shell secretion. Instead, the size of the pore of the shell gland may be
the important factor, especially if the shell material is first secreted over the pore
rather than along the lining of the shell gland lumen. In all of the above species the
shell gland pore becomes smaller prior to shell secretion. Presumably a small pore
would be easier to seal over with shell material than would a large pore.
Identification of shell secreting cells
The embryonic shell of molluscs is often said to be secreted by the "shell gland"
(Fretter and Graham, 1962; Raven, 1966; Jablonski and Lutz, 1980), a term which
has different meanings to different authors. In many cases, general statements about
the "shell gland" are in fact references to the entire shell-secreting epithelium, (i.e.,
GASTROPOD EMBRYONIC SHELL FORMATION 405
the shell field) regardless of its morphology. Originally the term was applied by Ray
Lankester just to the invagination (Pelseneer, 1906), not to the entire shell field. The
term has more recently also been defined as the calcifying invagination of the ec-
toderm (Waller, 1978). As used in the present work the term shell gland strictly
refers to the invaginated region of the shell field without reference to function. Thus
the shell gland is the center of the early shell field. After shell gland evagination or
spreading, the cells that once lined the shell gland lumen are still shell field cells
(Kniprath, 1979, 1981).
The present ultrastructural evidence demonstrates that in Aeolidia papillosa the
first organic portion of the shell is secreted only by the non-invaginated shell field
cells around the shell gland pore. No substances resembling shell material were ever
observed within the lumen of the shell gland. Thus, it is clear that the shell gland
sensu stricto does not secrete the embryonic shell in Aeolidia papillosa.
These results support the electron microscopic work of Kniprath (1977, 1979,
1980a, 1980b) on Lymnaea stagnalis, Marisa cornuarietis, Helix aspersa, and My-
tilus galloprovincialis. Kniprath reported that the first shell material is secreted solely
by a ring of cells surrounding the shell gland pore while the invaginated cells of the
shell gland remain nonsecretory until calcium secretion begins. Possibly the invag-
ination of these cells while they are nonsecretory serves to prevent a large hole from
forming in the center of the shell, an idea suggested previously by several workers
(see Haas et al., 1979; Kniprath, 1979).
It should be noted that earlier work based on light microscopy (e.g., Gather,
1967, on Ilyanassa obsoleta; Demian and Yousif, 1973, on Marisa cornuarietis;
Raven, 1975, on Lymnaea stagnalis) produced results conflicting with later studies
utilizing TEM (e.g, Kniprath, 1977, on Lymnaea stagnalis; Kniprath, 1979, on
Marisa cornuarietis; present study on Aeolidia papillosa). These conflicting results
probably do not reflect biological differences. With light microscopy the shell ma-
terial was observed extending into the shell gland lumen and therefore was believed
to have been secreted there. With transmission electron microscopy the first shell
material is seen only outside the shell gland (lying over the shell gland pore). These
differences may reflect several factors. First, earlier authors may have been unable
to detect the very first shell material with light microscopy. The initial shell material,
because of its thinness, may not be detectable with light microscopy until it separates
from the underlying epithelium. After separation the shell material might then fold
down into the lumen, giving the impression that it was secreted there. Second is the
possibility that the earliest shell material might be dislodged from excapsulated or
non-encapsulated embryos during handling for fixation and dehydration. If the very
early shell material can be dislodged by handling then the shell would not be detected
when it is initially secreted. In the present study all embryos of Aeolidia papillosa
were fixed and dehydrated within their capsules. A third factor that might have lead
to these different conclusions concerning the timing and location of initial shell
secretion is sampling (fixation) frequency. In some species studied the frequency of
sampling may have been low relative to rate of shell development, so that the earliest
stages of shell formation may have been missed. Sampling more frequently relative
to developmental rate should help resolve some or all of these issues.
It is not clear whether the various cell types seen in the early shell field maintain
their respective functions throughout evagination or spreading of the shell gland.
In Aeolidia papillosa the cells at the growing edge are seen further and further from
the shell gland lumen as evagination proceeds, suggesting that these cells are merely
migrating. Whether or not the cells change function following evagination as the
shell field grows into a distinct mantle is yet to be documented.
406 L. S. EYSTER
Certainly, further studies of other molluscan species are required to determine
the range of shell gland morphologies and to elucidate the role of the shell gland
cells versus that of other cells of the shell field. If such studies demonstrate that the
invaginated cells have no role in secretion of either organic or inorganic shell com-
ponents, what is now called the shell gland might be better referred to as the shell
field invagination.
Site of early shell secretion
It is well known that regions where biomineralization proceeds are sealed off
from chemical influences of the surrounding environment (Wilbur, 1972; Clark,
1976). Clark (1976) reviewed three main approaches to marginal calcification in
post-larval invertebrates, two of which deserve further mention here. First, in some
invertebrates such as scleractinian corals, a marginal fold of tissue drapes over the
growing margin, isolating it from sea water. Secondly, in many molluscs and bra-
chiopods, periostracum is secreted in a marginal fold and isolates the underlying
region of shell mineralization. Also, the shell material in molluscs may be securely
anchored to the apices of the secreting cells (Chetail and Krampitz, 1982), thus
isolating the inner surface of the shell from the external medium. In molluscan
embryos the location of the growing edge and method(s) of sealing it off have not
been established.
Few authors have examined the location of the early growing edge in molluscan
embryos on an ultrastructural level. Humphreys (1969) briefly described embryonic
shell formation in the mussel Mytilus edulis, stating that the growing edge of the
shell was intracellular. He suggested that it undercut the cilia and microvilli of the
cell apices, all of which were subsequently sloughed off. However, Kniprath (1980b)
has determined that the first and outermost shell material of Mytilus galloprovincialis
is laid down extracellularly and seems to be protected from the surrounding medium
by a thick glycocalyx and by microvilli of the adjacent cells. He also reported that
the growing edge in M. galloprovincialis lies in an intercellular gap, sometimes down
to the desmosome, and that infoldings of the lateral plasma membranes in this
region seem to secrete materials that thicken the shell pellicle. This intercellular gap
may also serve to seal off the growing edge.
Ultrastructural observations on Aeolidia papillosa confirm that the shell is laid
down extracellularly. However, instead of forming in a lateral intercellular gap, the
embryonic shell of A. papillosa seems to be produced on the apical surface of the
GE cells. No infolding of the lateral plasma membranes was seen in this area, and
no shell material in addition to the two dense lamellae was observed in regions of
the shell distant from the growing edge. Thus additions to the shell in this species
seem to occur solely over the cell apices, where the growing edge is potentially
exposed to the surrounding environment.
Two factors may be involved in sealing off the shell edge of Aeolidia papillosa.
First, the shell in this region is closely applied to the underlying cells and seems to
be secured to the cell apices (Figs. 16, 18), while in regions away from the growing
edge the shell is often separated from the underlying cells by a gap. Secondly, the
MV cells may have a role in sealing off the growing edge. The microvilli of these
cells are angled towards the GE cells (Fig. 16), and long cytoplasmic extensions
arising from the proximal edges of the MV cells lie over the growing edge of the
shell (Figs. 9, 18). Haas (1976), Haas et al. (1979), and Kniprath (1980c) have ob-
served a similar situation and reached a similar conclusion for shell plate formation
in chitons. Haas (1976) suggests that the microvilli may form "a barrier which
GASTROPOD EMBRYONIC SHELL FORMATION 407
controls the growth of the tegmental crystals". Kniprath (1980c) provided support
for this idea by his observation that the first trace of mineral detectable under
polarizing microscopy was seen at exactly the stage of development where the large
flat villi from the distal edges of neighboring cells overlapped and closed off the
crystallization space. Although the MV cells may serve other functions (transport?)
than isolation of the growing edge of the shell in Aeolidia papillosa, it is doubtful
that they secrete the organic shell material since they lack the dense granules pre-
sumed to contain organic shell components and since the growing edge of the shell
lies over a different cell type.
Preliminary studies (Eyster, unpubl.) on the development of several other opis-
thobranch species show that the growing edge of the shell of these species also lies
over the apical surfaces of the GE cells and that it is covered by cytoplasmic ex-
tensions arising from the MV cells. Possibly, presence of the growing edge on the
potentially exposed cell apices of developing opisthobranch embryos reflects pro-
tection from the surrounding medium afforded by the embryonic capsules. Attempts
to mechanically remove the capsules surrounding young embryos failed although
the same procedures worked on embryos ready to hatch; embryos removed from
the egg mass but left within their capsules developed normal shells. Preliminary
attempts to examine the early growing edge in A. papillosa with scanning electron
microscopy have proved unproductive due to the obscuring of the embryonic surface
with precipitated components of the fluid held inside the capsule. Further studies
are required to determine whether these reported differences in early shell devel-
opment between opisthobranch gastropods and other taxa have phylogenetic sig-
nificance.
ACKNOWLEDGMENTS
Specimens were kindly collected by T. K. Van Wey and Dr. K. P. Sebens. I
express my gratitude to K. Porter and Dr. E. R. Rivera for demonstrating the
PA-TSC-SP stain technique, to E. Cole for typing the final drafts, to C. B. Galloway
and Drs. M. P. Morse, J. A. Pechenik, and R. D. Turner for critically reviewing the
manuscript, and to Drs. V. Fretter and N. Watabe for unknowingly sparking my
interest in molluscan shell formation more than six years ago.
LITERATURE CITED
BEVELANDER, G., AND H. NAKAHARA. 1969. An electron microscope study of the formation of the
nacreous layer in the shell of certain bivalve molluscs. Calcif. Tiss. Res. 3: 84-92.
GATHER, J. N. 1967. Cellular interactions in the development of the shell gland of the gastropod, Ily-
anassa. J. Exp. Zoo/. 166: 205-224.
CHETAIL, M., AND G. KRAMPITZ. 1982. Calcium and skeletal structures in molluscs: concluding remarks.
Proc. Seventh International Malacological Congress. Malacologia 22(1-2): 337-339.
CLARK, G. R., II. 1976. Shell growth in the marine environment: Approaches to the problem of marginal
calcification. Am. Zoo/. 16: 617-626.
DEMIAN, E. S., AND F. YOUSIF. 1973. Embryonic development and organogenesis in the snail Marisa
cornuarietis. (Mesogastropoda: Ampullariidae). IV. Development of the shell gland, mantle and
respiratory organs. Malacologia 12(2): 195-21 1.
FRETTER, V., AND A. GRAHAM. 1962. British Prosobranch Molluscs. Their Functional Anatomy and
Ecology. Bartholomew Press, Dorking, Great Britain. 755 pp.
HAAS, W. 1 976. Observations on the shell and mantle of the Placophora. Pp. 389-402 in The Mechanisms
of Mineralization in the Invertebrates and Plants. N. Watabe, and K. M. Wilbur, eds. Belle W.
Baruch Library in Marine Science, 5. Univ. South Carolina Press, Columbia, 461 pp.
HAAS, W., K. KRIESTEN, AND N. WATABE. 1979. Notes on the shell formation in the larvae of the
Placophora (Mollusca). Biomineralization 10: 1-8.
HAYAT, M. A. 1970. Principles and Techniques of Electron Microscopy. Biological Applications, I. Van
Nostrand Reinhold, Co., New York. 412 pp.
408 L. S. EYSTER
HUMPHREYS, W. J. 1969. Initiation of shell formation in the bivalve, Mytilus edulis. Proc. Electr. Microsc.
Soc. Am., Twenty-seventh Ann. Meeting: 272-273.
JABLONSKI, D., AND R. A. LUTZ. 1980. Molluscan larval shell morphology. Ecological and paleontological
applications. Pp. 323-377 in Skeletal Growth of Aquatic Organisms, D. C. Rhoads and R. A.
Lutz, eds. Plenum Pub. Co., New York. 750 pp.
KARNOVSKY, M. J. 1965. A formaldehyde-glutaraldehyde fixative of high osmolarity for use in electron
microscopy. J. Cell. Biol. 27: 1 37 A.
KNIPRATH, E. 1977. Zur Ontogenese des Schalenfeldes von Lymnaea stagnalis. Wilhelm Roux's Arch.
181: 11-30.
KNIPRATH, E. 1979. The functional morphology of the embryonic shell-gland in the conchiferous mol-
luscs. Malacologia 18: 549-552.
KNIPRATH, E. 1980a. Sur la glande coquilliere de Helix aspersa (Gastropoda). Arch. Zool. Exp. Gen 121:
207-212.
KNIPRATH, E. 1980b. Larval development of the shell and the shell gland in Mvtilus (Bivalvia). Wilhelm
Roux's Arch. 188: 201-204.
KNIPRATH, E. 1980c. Ontogenetic plate and plate field development in two chitons, Middendorffia and
Ischnochiton. Wilhelm Roux's Arch. 189: 97-106.
KNIPRATH, E. 1981. Ontogeny of the molluscan shell field: a review. Zool. Scri. 10: 61-79.
PELSENEER, P. 1906. Part V. Mollusca. In: 1906. A Treatise on Zoology, Edwin Ray Lankester, ed. Adam
and Charles Black, London. 355 pp.
PORTER, K. J., AND E. R. RIVERA. 1979. An ultrastructural cytochemical analysis of mucoid secretory
granules in nudibranch epidermis. Proc. Electron Microsc. Soc. Am., 37th Ann. Meeting: 302-
303.
RAVEN, C. P. 1966. Morphogenesis: The Analysis of Molluscan Development. Pergamon Press, New
York. 366 pp.
RAVEN, C. P. 1975. Development. Pp. 367-400 in Pulmonales I. Functional Anatomy and Physiology,
V. Fretter and J. Peake, eds. Academic Press, New York. 417 pp.
RICHARDSON, K. C., L. JARRETT, AND E. H. FINKE. 1960. Embedding in epoxy resin for ultrathin
sectioning in electron microscopy. Stain Technol. 35: 313-323.
SALEUDDIN, A. S. M. 1974. Ultrastructural studies on the structure and formation of the periostracum
in Helisoma (Mollusca). Pp. 309-337 in The Mechanisms of Mineralization in the Invertebrates
and Plants, N. Watabe, and K. M. Wilbur, eds. Belle W. Baruch Library in Marine Science,
5. Univ. South Carolina Press, Columbia. 461 pp.
SPURR, A. R. 1969. A low-viscosity epoxy resin embedding medium for electron microscopy. J. Ultra-
struct. Res. 26: 31-43.
THIERY, J. P. 1967. Mise en evidence des polysaccharides sur coupes fines en microscopic electronique.
J. Microsc. 6: 987-1018.
WALLER, T. R. 1978. Formation of a posterodorsal notch in larval oyster shells and the prodissoconch —
I/II boundary in the Bivalvia. Bull. Am. Malacol. Union, Inc., 1978: 55-56.
WEINER, S., AND L. HOOD. 1975. Soluble protein of the organic matrix of mollusk shells: a potential
template for shell formation. Science 190: 987-989.
WHEELER, A. P., J. W. GEORGE, AND C. A. EVANS. 1981. Control of calcium carbonate nucleation and
crystal growth by soluble matrix of oyster shell. Science 212: 1397-1398.
WILBUR, K. M. 1972. Shell formation in mollusks. Pp. 103-145 in Chemical Zoology, 7, M. Florkin,
and B. T. Scheer, eds. Academic Press, London. 567 pp.
WILBUR, K. M., AND L. H. JODREY. 1952. Studies on shell formation I. Measurement of the rate of
shell formation using Ca45. Biol. Bull. 103: 269-276.
YOUNG, S. D., M. A. CRENSHAW, AND D. B. KJNG. 1977a. Mantle protein excretion and calcification
in the hardshell clam Mercenaria mercenaria. I. Protein excretion in the intact clam. Mar. Biol.
41: 253-257.
YOUNG, S. D., M. A. CRENSHAW, AND D. B. KING. 1977b. Mantle protein excretion and calcification
in the hardshell clam Mercenaria mercenaria. II. Protein synthesis and excretion by the isolated
mantle. Mar. Biol. 41: 259-262.
Reference: Biol. Bull. 165: 409-415. (October, 1983)
POSTLARVAL GROWTH IN JUVENILE
RHITHROPANOPEUS HARRISII
JOHN A. FREEMAN1, TERRY L. WEST2, AND JOHN D. COSTLOW3
^Department of Biology. University of South Alabama, Mobile, Alabama 36688, 2Institute of Coastal
Marine Resources, East Carolina University, Greenville, North Carolina 27834, and 3Duke University
Marine Laboratory, Beaufort. North Carolina 28516
ABSTRACT
Eyestalk removal accelerated the molt cycles of megalopal and juvenile (first
through fifth crab instars) Rhithropanopeus harrisii. Eyestalkless crabs also dem-
onstrated a greater increase in size at each ecdysis. The growth rate of eyestalkless
crabs was approximately twice the rate measured in control crabs. Epidermal cell
density measurements showed that the cell density was the same in intermolt fifth
instar control and eyestalkless crabs. The results demonstrate that growth in juvenile
crabs is under the influence of eyestalk neurosecretory centers and that growth is
a result of epidermal cell proliferation and not cell enlargement.
INTRODUCTION
The growth rate of crustaceans is a function of both the molting rate and the
increase in size obtained at each molt. In adults, these aspects of growth are thought
to be regulated by hormones (see Passano, 1960; Kleinholz and Keller, 1979; Skin-
ner, 1983 for review). The molting rate may be controlled by molt-inhibiting hor-
mone (MIH) which is secreted by neuro-endocrine cells in the eyestalk. The eyestalk
may also contain a factor that restricts the uptake of water at ecdysis and, conse-
quently, the expansion of the new cuticle. The accelerated molting rate and the
greater incremental increase in size observed in eyestalkless animals is believed to
be a consequence of the absence of these two factors.
The action of endocrine factors in crustacean larvae and postlarvae, however,
is not clearly defined. In early studies it was found that eyestalk removal did not
result in a more rapid molting rate until the third post-larval instar in Callinectes
sapidus (Costlow, 1963) or the fourth post-larval instar in Rhithropanopeus harrisii
(Costlow, 1966). However, recent work in which the larvae were observed several
times a day, revealed that eyestalk removal did elicit a faster molting rate in R.
harrisii larvae (Freeman and Costlow, 1980).
In the present study, the effect of eyestalk removal during larval stages on molting
rate, incremental size increase, and epidermal cell density in early juvenile R. harrisii
is examined.
MATERIALS AND METHODS
Larval rearing
The larval development of R. harrisii consists of four zoeal instars and one
megalopal instar. Zoeae were hatched and mass reared in 25%o sea water maintained
at 20-2 1°C. The water was changed and freshly hatched Anemia were added daily.
Received 20 January 1983; accepted 19 July 1983.
409
410 J. A. FREEMAN ET AL.
Upon reaching the megalopal instar the larvae were maintained individually in
compartmentalized plastic boxes.
Eyestalk removal
Fourth instar zoeae were placed on a small glass disc (4 cm diameter) in a volume
of water that was sufficient to keep them moist (50-100 /ul), but small enough to
restrict their movement. An iris scapel was used to sever the eyestalk at the artic-
ulating membrane. The larva was returned to 25%o sea water immediately after the
operation. Sixty percent of the eyestalkless larvae lived to molt to the megalopal
instar. Of the larvae that molted to the megalopal instar, twenty five percent (9 of
36) lived to the sixth crab instar. Thirty nine percent (14 of 36) of the control crabs
lived to the sixth crab instar. No abnormal megalopae, or supernumerary larvae
were observed in either the intact or eyestalkless crabs.
Determination of molting and growth rates
Intact (control) and eyestalkless animals were observed twice daily for indications
of ecdysis (presence of shed exoskeletons), and/or for apolysis (retraction of the
epidermis from the cuticle). Apolysis indicates the initiation of the premolt phase
(Do) of the molt cycle (Drach and Tchernigovtzeff, 1967). Apolysis was determined
through microscopic observation of the integument in the leg, rostrum, antennules,
and antennae. Due to the opacity of the cuticle in third through fifth crab instars,
apolysis was not followed in these crabs. The incremental growth at each instar was
determined by measuring the differences in carapace width (CW) between the shed
exoskeletons of that instar and the previous instar. The number of crabs observed
for each measurement is indicated in the figures. Analysis of variance was done
using the F-test. Significant difference between means was done with the /-test.
Measurement of epidermal cell density
Cell density measurements were done on whole mounts of hepatic or branchial
sections (see McLaughlin, 1980) of dorsal carapaces removed from both intact (con-
trol) or eyestalkless fifth instar crabs. The specimens were fixed in Bouin's fluid,
stained by the Feulgen method and mounted in toto. Cell counts were made from
photographs of the stained whole mounts and are reported in the Results section
as # nuclei/ 100 ^m2. Differences in cell density in control and eyestalkless crabs was
determined with the /-test after analysis of variance.
RESULTS
Eyestalk ablation during the late zoeal period accelerated the molt cycles of
subsequent megalopal and juvenile instars (Fig. 1). The period from ecdysis to
premolt (stage D0) in eyestalkless crabs was significantly shorter (P < .05) than those
of intact animals. The duration of the molt cycle (ecdysis to ecdysis) was also sig-
nificantly reduced in eyestalkless crabs. These findings suggest that the eyestalks of
the juvenile R. harrisii contain a factor that inhibits molting. The degree to which
eyestalk removal shortened the molt cycle, however, varied from instar to instar.
The molt cycles of the megalopal instar and fourth and fifth instar crabs underwent
a greater reduction in duration, compared to control crabs, than did the molt cycles
of the first, second, or third instar crabs.
POSTLARVAL GROWTH REGULATION IN CRABS
411
1 0
10 1 o
Q
1 0
MEGALOPA
(34)
(12)
( 8 )
-h
(20)
•4
I E S
2 CRAB
*
(20)
fcl'"
•b
I E S
(28)
(18)
I E S I E S
4 C R A B ' 2 ° ' I «
(it)
t
1 CRAB
( r )
(331(18)
4-
I ES
3 CRAB
I E S
(24)
(18)
I E S
5 CRAB , , 4)
1 0 -
z
m
O
5 Q)
2 5
-20
1 5
-1 0
5
E S
I E S
FIGURE 1 . Duration (days) from ecdysis to premolt (D0, stippled bars) and to the next ecdysis (open
bars) in intact (I) and eyestalkless (ES) Rhithropanopeus harrisii megalopae and first through fifth crab
(designated 1, 2, 3, 4, 5 crab, respectively) instars. Each bar represents mean ± 1 standard deviation.
Bars without S.D. lines indicate no variation. Sample size for each measurement indicated in parentheses
above the bar. Asterisk indicates significant differences (P < .05) between intact and eyestalkless groups.
Carapace widths of eyestalkless animals were always significantly larger (P
< .05) than those of controls (Fig. 2). The actual difference in carapace widths
between control and eyestalkless crabs was small during the megalopal and first two
crab molt cycles. The differences increased, however, in the third, fourth, and fifth
crab molt cycles. These data indicate that eyestalk removal affects the mechanism
that regulates size increases at each ecdysis.
While it may be suspected that a crab would have more potential for growth if
the molt cycle was longer, the results reported here show that just the opposite
occurred in eyestalkless crabs. When data from Figures 1 and 2 are combined to
yield a growth rate (mm carapace width/time, Table I) it can be seen that, even
though the eyestalkless crabs reached the fifth crab in roughly two-thirds the time
required by the control crabs, their growth was over twice that of control animals.
To find if the growth rate varied in different instars, the increase in carapace width/
instar was calculated (Table II). With the exception of the first crab, the eyestalkless
412
J. A. FREEMAN ET AL.
6
^
4
-
111)
5
E
4
f
.
E
(17)
A
I
(14)
. " ' :
I
. I.
1
• , ." '
i-
1 n t ;
1 C t
1
. - J
0
4
1
5
(17)
1
(31)
I
: 1 ••••
; ••/..
3
1
. t
LU
0
. "•»'•."
E y €
s t a
k 1 e
S S
4
It
(28)
"•';•
|
n
r 1
• . • •
• r .
4
1
(28)
(14)
1
1
• • , \
'- ' m ,
tr 2
«
.1-.
' i
.' * 1
(18)
<£
(28)
*.
i •_ *
, *
1
• • Y '
( "', «
o
1
- . •
. • '
•
1
«
. '^
'^ '. -
1
(23)
(19)
1 .|.. i
- - ..
*- * ' ' ;
• 1 . \
—
" \
* '
t f '
v : •
• ' • l
*" * i
*• • l
MEG
1 Crab
2 Crab 3 Crab
4 Crab
5 Crab
Instar
FIGURE 2. Carapace widths of intact and eyestalkless megalopae and first through fifth instar crabs
(designated 1, 2, 3, 4, 5 crab, respectively) R. harrisii. Each bar represents mean ± 1 standard deviation.
Width measurements were taken at end of instar by measuring shed exoskeleton. Sample size for each
measurement indicated in parentheses above the bar. Asterisk indicates significant difference (P < .05)
between intact and eyestalkJess groups.
crabs demonstrated more growth per instar than did control crabs. Since the control
and eyestalkless groups differed in molt cycle length at each molt cycle, the growth
rates were calculated in terms of carapace width increase/day/instar (Table II). These
calculations demonstrate that, in each of the molt cycles examined, there was a
greater growth rate in the eyestalkless crabs.
If the greater incremental size increase of eyestalkless animals was strictly a
function of excessive cuticular stretching caused by unrestricted intake of water at
ecdysis, then the epidermal cell density of eyestalkless animals should be less than
that of intact animals. When epidermal cell density determinations were made on
TABLE I
Growth in Rhithropanopeus harrisii juveniles
Total time (days) from
megalopa to fifth instar crab
Growth of carapace during period from
megalopa to fifth instar crab (mm/day)
control
eyestalkless
55.7*
34.1
.052**
.117
* From Figure 1 : Sum of mean durations for first-fourth crab molt cycles.
** From Figures 1 and 2: mean CW fifth crab minus mean CW megalopa/sum mean durations of
first-fifth crab molt cycles.
POSTLARVAL GROWTH REGULATION IN CRABS 413
TABLE II
Growth rates for juvenile Rhithropanopeus harrisii
Carapace width increase Carapace width increase
(mm/instar)* (mm/day/instar)*
\**
Instar Control Eyestalkless Control Eyestalkless
Megalopa
.65
.70
.076
.125
First instar crab
.45
.45
.080
.089
Second instar crab
.40
1.00
.059
.204
Third instar crab
.70
.90
.090
.172
Fourth instar crab
.70
1.00
.056
.145
* From Figure 2, mean CW of instar n + 1 minus mean CW of instar n.
** From Figures 1 and 2, CW instar n + 1 - CW instar n/mean number of days per instar.
regions of the dorsal carapace, there was no significant difference in cell density
between the control and eyestalkless crabs (Table III), even though the mean car-
apace widths of the two groups differed by nearly 25%. These findings suggest that
the observed size differences between the control and eyestalkless crabs were not
due to differences in cell size, but rather to enhanced cell proliferation in the eyestalk-
less crabs.
DISCUSSION
Our results show that eyestalkless juvenile crabs molt at a more rapid rate than
intact crabs demonstrating that the eyestalks of juvenile crabs are involved in reg-
ulation of the molt and growth rates. This is in keeping with earlier findings (Freeman
and Costlow, 1980) which showed that MIH is produced during the larval period.
The extent to which the molt cycle is accelerated in eyestalkless crabs varied
during the megalopa and juvenile period. This suggests that the eyestalks may pro-
duce a molt-inhibiting hormone (MIH) in differing quantities during each of the
zoeal, megalopal, and juvenile phases of the life cycle. It is possible that alterations
in the molting frequency may be an adaptation to the different environments ex-
perienced by the three phases. Molt-inhibiting hormone is apparently present in
reduced amounts during the zoeal period (Freeman and Costlow, 1980), thus al-
lowing the larvae to grow and complete postembryonic development in the shortest
period of time. While the plankton contains optimal amounts of food for zoeal
growth, the longer the larva resides in the plankton, the greater the chance that it
will be consumed by larger larvae or fish (see Morgan, 1981). Conversely, increased
levels of MIH during the megalopal instar would lengthen the molt cycle, allowing
the crab more time to take up a benthic existence and find a suitable habitat. Then,
TABLE III
Mean cell density of carapace from control and eyestalkless fifth instar crabs Rhithropanopeus harrisii
Nuclei/ 100 urn2
control 1.22 ± .27* (n = 16)
eyestalkless 1.10 ± .16 (n = 5)
* Mean ± 1 standard deviation.
414 J. A. FREEMAN ET AL.
during the early juvenile phase, minimal production of MIH would again permit
rapid molting, providing a mechanism for rapid growth and onset of reproductive
maturity, which occurs in the fifth crab instar (Payen et ai, 1969).
The results of the present study differ from those of earlier reports on molting
in juvenile Callinectes sapidus (Costlow, 1963) and R. harrisii (Costlow, 1966). The
discrepancy may be explained by two important differences in the experimental
protocols. First, in this study, the animals were reared at 21°C while, in the earlier
studies, the crabs were reared at 25 °C. The lower temperature shows the molt cycle,
thus making subtle differences between the intact and eyestalkless animals more
evident. Second, both apolysis (stage D0) and ecdysis were followed in the present
study, while only ecdysis was noted in the earlier work. The observation schedule
used here has been shown to be a more accurate means of assessing the rate at
which an animal passes through the molt cycle stages (Freeman and Costlow, 1980).
Eyestalk removal also resulted in large increases in carapace width at each ec-
dysis, in keeping with earlier findings on larval crabs (Costlow, 1966) and shrimp
(Little, 1969). Similar findings have been presented for adult Uca pugilator (Abra-
mowitz and Abramowitz, 1940), Cambarus (Scudamore, 1947), Carcinus (Carlisle,
1955), Homarus americanus (Mauviot and Castell, 1976), and other crustaceans
(see Passano, 1960). Enhanced growth in eyestalkless animals has been attributed
to loss of a neurosecretory factor that regulates, in some manner, the rate of water
influx at ecdysis (see Passano, 1960). Water uptake at ecdysis is a normal physio-
logical event which serves to increase hemolymph hydrostatic pressure, thereby
causing the rupture of the weakened old exoskeleton and unfolding of the epidermis
from a plicated to a planar form (Drach, 1939; Passano, 1960). An abnormal increase
in the influx could result in an actual stretching of the integument. Direct proof for
a mechanism involving neurosecretory-controlled increase in water uptake, however,
has not been forthcoming. Alternatively, as pointed out by Passano (1960), the
increased extensibility may be due to a thinner exoskeleton at the time of ecdysis.
This would occur if ecdysis took place earlier than normal during the premolt period
when fewer lamellae would have been secreted in the new exoskeleton.
Initially, enhanced integumental stretch would result in each epidermal cell
having an increased apical area. Findings obtained in this study show, however, that
the epidermal cell density is the same in intermolt crabs from both the control and
eyestalkless groups. For the cell density to be similar, while the growth rate was
greater, there would have to be more cell proliferation in the integument of the
eyestalkless crabs. It is, therefore, possible that the epidermis of eyestalkless animals
responded to the stretch by increasing the amount of cell proliferation, thereby
restoring the apical region of the epidermal cell to the normal area. At the present
time, it is unclear if this enhancement of epidermal cell proliferation has an en-
docrine basis or if it is a result of the stimulation of metabolic processes that char-
acteristically follow eyestalk removal (Kleinholz and Keller, 1979).
In the present study, we were able to maintain eyestalkless R. harrisii through
one megalopal instar and five consecutive crab instars. In contrast, eyestalk removal
in adult crabs often results in death after one molt. Seldom are several consecutive
molts obtained. Although we can not explain this difference from the results reported
here, it is possible that the relatively brief molt cycle duration of the juvenile R.
harrisii (4-12 days) may allow them to molt several times before the detrimental
effects of eyestalk loss become severe. Larger crabs often have molt cycle durations
that are much longer than several days. In fact, the molt cycles of mature adult R.
harrisii can last for two months (Freeman, unpublished observations), almost twice
the total time for the larvae to pass through the first five crab instars (see Table I).
POSTLARVAL GROWTH REGULATION IN CRABS 415
ACKNOWLEDGMENTS
The authors wish to thank Ms. Anita Walker for technical assistance, and anon-
ymous reviewers for their helpful comments on the manuscript. This work was
supported by contract No. NR- 104- 194 between the Office of Naval Research and
Duke University.
LITERATURE CITED
ABRAMOWITZ, R. K., AND A. A. ABRAMOWITZ. 1940. Moulting, growth and survival after eyestalk
removal in Uca pugilator. Biol. Bull. 78: 179-188.
CARLISLE, D. B. 1955. On the hormonal control of water balance in Carcinus. Publ. Staz. Zoo/. Napoli
27:227-231.
COSTLOW, J. D. 1963. The effect of eyestalk extirpation on metamorphosis of megalops of the blue crabs.
Callinectes sapidus Rathbun. Gen. Comp. Endocrinol. 3: 120-130.
COSTLOW, J. D. 1966. The effect of eyestalk extirpation on larval development of the mud crab, Rhith-
ropanopeus harrisii (Gould). Gen. Comp. Endocrinol. 7: 255-274.
DRACH, P. 1939. Mue et cycle d'intermue chez les Crustaces Decapodes. Ann. Inst. Oceanogr. 19: 103-
391.
DRACH, P., AND C. TCHERNIGOVTZEFF. 1967. Sur la methode de determination des stades d'intermue
et son application generale aux crustaces. Vie Milieu Ser. A 18: 595-610.
FREEMAN, J. A., AND J. D. COSTLOW. 1980. The molt cycle and its hormonal control in Rhithropanopeus
harrisii larvae. Dev. Biol. 14: 479-485.
KLEINHOLZ, L. H., AND R. KELLER. 1979. Endocrine regulation in Crustacea. Pp. 159-213 in Hormones
and Evolution. E. J. W. Harrington, ed. Academic Press, New York.
LITTLE, G. 1969. The larval development of the shrimp Palaemon macrodactylus Rathbun, reared in
the laboratory, and the effect of eyestalk extirpation on development. Crustaceana 17: 69-87.
MAUVIOT, J., AND J. D. CASTELL. 1 976. Molt- and growth-enhancing effects of bilateral eyestalk ablation
on juvenile and adult American lobster (Homarus americanus). J. Fish. Res. Board Can. 33:
1922-1929.
MCLAUGHLIN, P. A. 1980. Comparative Morphology oj Recent Crustacea. Pp. 177. W. H. Freeman, San
Francisco.
MORGAN, S. C. 1981. Larval spines as an anti-predator device in Rhithropanopeus harrisii (Decapoda,
Xanthidae). Am. Zoo/. 21: 518.
PASSANO, L. M. 1960. Molting and its control. Pp. 473-576 in The Physiology of Crustacea, Vol. 1,
T. H. Waterman, ed. Academic Press, New York.
PAYEN, G., J. D. COSTLOW, AND H. CHARNIAUX-COTTON. 1969. Mise en evidence experimentale de
1'independance de la realisation de sexe chez le Crabe Rhithropanopeus harrisii (Gould) a Tegard
du complexe neurosecreteur organe de Hanstrom-glande du sinus. C. R. Acad. Sci. Paris 269:
1878-1881.
SCUDAMORE, H. H. 1947. The influence of the sinus glands upon molting and associated changes in the
crayfish. Physiol. Zoo/. 20: 187-208.
SKINNER, D. M. 1983. Regeneration and Molting. In The Biology of Crustacea, Vol. 9, D. E. Bliss and
L. H. Mantel, eds. Academic Press, New York.
Reference: Biol. Bull. 165: 416-418. (October, 1983)
INTRASEXUAL AGGRESSION IN METRIDIUM SENILE
SAUL W. KAPLAN
Institute of Animal Behavior, Rutgers University, 101 Warren Street, Newark, New Jersey 07102
ABSTRACT
The dioeceous anemone Metridium senile reproduces both sexually (in summer)
and asexually (year round). Asexual reproduction yields genetically identical clones
via longitudinal fission or pedal laceration. Clonemates may form large aggregates,
ranked together in close order, and become aggressive against neighboring clones.
Interclonal aggression is frequently carried out with the use of hypertrophied ten-
tacles referred to as catch tentacles. The present study indicates that catch tentacles
do not maintain clonal segregation, and do not serve as aggressive appendages against
all nonclonemates. Laboratory pairings of nonclonemates and observations of their
movements in the field indicate that interclonal aggression in this species is mediated
by sex. Nonclonemates will become aggressive only against same sexed individuals,
males fighting males and females fighting females, while nonclonemates of opposite
sex may exhibit nonaggressive interaction, with or without the use of catch tentacles.
Interclonal/intrasexual aggression in this species may function to increase the prob-
ability of successful fertilization during sexual reproduction by increasing the prox-
imity of males to females and vice versa.
INTRODUCTION
Metridium senile is a dioeceous cold water anemone common on both the east
and west coasts of the United States. Metridium reproduces both sexually (in sum-
mer) and asexually (year round) by pedal laceration and longitudinal fission (Ste-
phenson, 1935). Asexual reproduction commonly produces clones ranging from a
few to many hundreds of genetically identical individuals. Color variation among
clones makes it possible to distinguish easily between clonemates and nonclonemates
in the field (Hoffman, 1976).
Most populations include individuals bearing large opaque tentacles surrounding
the mouth which are structurally and functionally distinct from feeding tentacles
(Purcell, 1977). Prior studies have shown that these tentacles, referred to as "catch
tentacles" may be used in aggressive encounters between nonclonemates. These
appendages, derived from feeding tentacles though not themselves used in feeding,
had originally been reported to function in the maintenance of interclonal bound-
aries, in much the same way as acrorhagi maintain interclonal segregation in An-
thopleura elegantissima (Francis, 1973). Intermingling of clones among Metridium
is not uncommon, and individuals of clearly distinct appearance, bearing catch
tentacles, are frequently found adjacent and in contact in the field (Purcell and
Kiting, 1982), with no sign of aggressive interaction.
Interclonal aggression may be initiated when an individual spontaneously ex-
tends catch tentacles and contacts a nonclonemate, or when movement within or
between clones bring two nonclonemates within feeding tentacle range. When ex-
tended, catch tentacles are longer than feeding tentacles, and may be as much as
four times longer than the diameter of the oral disc (Purcell and Kiting, 1982),
Received 18 March 1983; accepted 25 July 1983.
416
INTRASEXUAL AGGRESSION IN METRIDIUM SENILE 417
extending the effective territory of the anemone by a factor of eight. When a catch
tentacle finds a nonclonemate and nematocyst discharge occurs, the tentacle tip
may adhere, while the tentacle retracts, so that the tip breaks off and remains attached
to the victim, continuing to sting after the aggressor has withdrawn. Following one
or more bouts of aggressive interaction, one anemone will usually retract its tentacles
within its column, and bend or move across the substrate, out of range of further
attack. Interclonal contact without aggressive interaction may also involve catch
tentacle extension. Nonclonemates may contact one another with catch tentacles,
draw closer, make contact with feeding tentacles, and remain in close proximity
with no aggression or withdrawal, sometimes for days.
The present study tests the hypothesis that catch tentacles are used in aggressive
interaction exclusively between nonclonemates of like sex, and that nonclonemates
of opposite sex not only tolerate one another's presence but may engage in non-
aggressive interaction.
MATERIALS AND METHODS
Seventeen anemones from five clones with catch tentacles (two male and three
female clones) were collected from Monterey Harbor and the Elkhorn Slough, and
allowed to settle on individually marked glass discs in flowing sea water aquaria.
Each subject was anesthetised in an isotonic magnesium chloride solution to facil-
itate examination of the contents of the gastrovascular cavity. A glass pipette was
introduced into the oral opening, and its contents withdrawn. In some cases the wall
of the gastrovascular cavity was pierced and cellular material withdrawn from within.
All subjects contained either live sperm or well developed eggs. In this manner it
was possible to determine the sex of the individual without resorting to the more
conventional sectioning and staining techniques which make subsequent behavioral
testing difficult.
Two individuals from different clones were placed in contact with one another
in sea water-filled glass observation bowls. Trials lasted up to twelve hours. Aggressive
contact was clearly distinguishable from nonaggressive interaction as it was quickly
followed by marked, sharp withdrawal as if in response to pain. In some cases
aggressive interaction began immediately upon contact. In others it appeared only
after hours of intermittent contact and withdrawal. Aggressive behavior, i.e., first
catch tentacle erection, number of catch tentacles erect, and elapsed time until
separation, were recorded. Eighteen of forty-two pairings elicited catch tentacle ex-
pansion. Eleven of these were aggressive encounters.
Following forty-two laboratory pairings all subjects were allowed to settle on a
Plexiglas panel which was then suspended in Monterey Harbor, so that the anemones
could move freely, contacting clonemates and nonclonemates of both sexes on the
basis of "preference." The position and movement of each anemone was checked
and recorded daily for an eighteen day period.
RESULTS
Forty-two trials were conducted, in which twelve individuals showed aggressive
behavior. All instances of aggressive behavior were confined to trials between in-
dividuals of the same sex (Table I). In no case did a male attack a female or vice
versa. The probability of this occurring on the basis of chance alone, and not due
to the sex of the animals being tested is equivalent to one half to the twelfth, or
.00024. Observation of the suspended panel corroborated the findings of the labo-
ratory pairings. In eighteen days of free movement the seventeen subjects showed
no tendency to reaggregate as clones, but in four instances individuals paired off in
418 S. W. KAPLAN
TABLE I
Interclonal aggressive encounters
Clone
#1
#2
#3
#4 #5
Female
#5
0
0
2
3
Female
#4
0
0
2
Female
#3
0
0
Male
#2
5
Male
#1
direct tentacle to tentacle contact with nonclonemates of the opposite sex for the
duration of the eighteen day period. Nonclonemates of the same sex were never
found in contact.
DISCUSSION
Metridium senile, unlike Anthopleura elegantissima, does not spawn synchro-
nously throughout a colony (Abbot, Hopkins Marine Station, pers. comm., 1982).
When gametes are released by individuals in an asynchronous manner into the
marine environment, the problem of achieving successful fertilization may be con-
siderable if an animal is surrounded by individuals of its own sex and its gametes
are quickly dispersed. A mechanism enabling a clonal coelenterate to discover the
sex of its neighbors and cause like sexed nonclonemates to move away so that
opposite sexed nonclonemates can approach would greatly increase the probability
of successful fertilization. Interclonal/intrasexual aggression in Metridium senile may
be just such a mechanism. Following three active mixed sex trials, a second ex-
amination of the gastrovascular cavity of each of the three females yielded both live
sperm and well developed eggs, where previously only eggs had been found. In no
other case were the two found within one animal. Internal fertilization has not
previously been reported to occur in this species. It is possible that catch tentacles
are not only used for intrasexual aggression, but may function in "courtship" as
well, enabling two anemones to release their gametes at the most propitious moment.
This is the impression given by observation of the long slow catch tentacle inter-
actions occasionally seen in mixed sex pairs of Metridium senile, as they touch and
probe one another's oral surfaces, and draw closer and closer together.
ACKNOWLEDGMENTS
I wish to thank Dr. J. Oliver, Dr. S. Lenington, and Dr. J. Purcell. This research
was submitted in partial fulfillment of the requirements for the Ph.D. This is con-
tribution number 36 1 from the Institute of Animal Behavior. (Special thanks to C.
Banas.)
LITERATURE CITED
FRANCIS, L. 1973. Intraspecific aggression and its effects on the distribution of Anthopleura elegantissima
and some related sea anemones. Biol. Bull. 144: 73-92.
HOFFMAN, R. J. 1976. Genetics and asexual reproduction of the sea anemone Metridium senile. Biol.
Bull. 151: 478-488.
PURCELL, J. E. 1977. Aggressive function and induced development of catch tentacles in the sea anemone
Metridium senile (Coelenterata, Actinaria). Biol. Bull. 153: 355-368.
PURCELL, J. E., AND C. L. KITING. 1982. Intraspecific aggression and population distributions of the sea
anemone Metridium senile. Biol. Bull. 162: 345-359.
STEPHENSON, T. A. 1935. The British Sea Anemones 2, The Ray Society, London, p. 426.
Reference: Biol. Bull. 165: 419-428. (October, 1983)
SPERM CHEMOTAXIS IN OIKOPLEURA DIOICA FOL,
1872 (UROCHORDATA: LARVACEA)
RICHARD L. MILLER1 AND KENNETH R. KING2
1 Department of Biology, Temple University, Philadelphia, Pennsylvania 19122, and
2 Department of Oceanography and The Friday Harbor Laboratories,
University of Washington, Seattle, Washington 98195*
ABSTRACT
An alcohol extract of unfertilized eggs of the larvacean, Oikopleura dioica, can
attract sperm over a distance of at least 80 nm from an artificial source. The sperm,
which normally swim in wide circles or straight lines, alter their path to form small
loops between straight or slightly curved segments directed up the gradient. During
the first loop, the velocity of sperm increases 50%. The new velocity is maintained
as long as the cells are influenced by the attractant. Once sperm reach the center
of the gradient, the path alters to the form of enlarging concentric circles which
eventually attain the diameter of the circles made in sea water. O. dioica sperm and
sperm attractant are species-specific in tests against attractants and sperm of sessile
tunicates. It has not yet been possible to test the species-specificity against other
larvaceans. We estimate that sperm chemotaxis in O. dioica increases the chance
of sperm-egg collisions from 4 to 1 5 times. This is mainly due to an increase in
apparent diameter of the egg and also to an increase in the velocity of attracted
sperm. Rapid population increase is characteristic of O. dioica under appropriate
conditions. An increase in the probability of fertilization produced by sperm che-
motaxis may be an additional factor leading to decreased generation time for the
population as a whole.
INTRODUCTION
Larvaceans are adult planktonic urochordates which resemble the tadpole larva
of sessile urochordates (Tunicata). They are widely distributed in tropical and tem-
perate oceans (e.g., Forneris, 1957; Fenaux, 1967) and may be found in immense
numbers under certain circumstances (Seki, 1973; Wyatt, 1973). Larvaceans may
rapidly attain large population size because they take advantage of short term con-
ditions optimal for maximum growth of the population. They possess very rapid
development (Gait, 1972; Fenaux, 1976) coupled with rapid growth to sexual mat-
uration (Fenaux, 1976; Paffenhofer, 1976). Generation times of 10 days or less have
been measured in enclosed water columns (King et al, 1980; King, 1982).
In contrast to other larvaceans, Oikopleura dioica is dioecious. Spawning is
random and may be triggered by physical means, such as turbulence or contact with
another object (Gait, 1972). The completely transparent eggs are denser than sea
water and sink after spawning (Bienfang and King, unpub.). Little is known about
gamete interactions in these organisms. If it is advantageous to decrease development
time in order to react quickly to favorable environmental conditions, then shortening
Received 16 July 1982; accepted 28 June 1983.
* Present address: Division of Biology and Living Resources, Rosenstiel School of Marine and At-
mospheric Sciences, University of Miami, 4600 Rickenbacker Causeway, Miami, Florida 33149.
419
420 R. L. MILLER AND K. R. KING
the interval between the spawning act and the time of fertilization may be important
in situations where the presence of the opposite sex cannot be predicted. One method
for ensuring fertilization and decreasing the time that the eggs remain unfertilized
is sperm chemotaxis where sperm move closer to the egg from some distance away
by following a gradient of a substance released by the egg (Miller, 1973). In taxa
where sperm chemotaxis has been described (Miller, 1966; 1975; 1977; 1979a), it
has been noted that the attractant often increases sperm velocity, further decreasing
the time of sperm approach. Sperm of many planktonic hydromedusae (Miller,
1979a, b) and sessile tunicates (Miller, 1975; 1982) exhibit chemotaxis. Here we
describe this phenomenon in a planktonic urochordate, and speculate on its possible
impact on the population dynamics of O. dioica.
MATERIALS AND METHODS
Sexual specimens of Oikopleura dioica were gently removed from the ocean
with a large bore pipette as they drifted past the dock at the Friday Harbor Labo-
ratories. The animals were made visible using a submerged night-light (Woodland,
Inc.). Ripe males could be distinguished by the swollen brilliant white testes. The
ovaries of females were also swollen and white but somewhat less brilliant. Indi-
viduals were kept segregated by sex in small clean finger bowls and used immediately
after collection.
Gametes were obtained by pricking the gonad surface of individuals which had
been previously transferred through several changes of HA-Millipore (0.45 nm)
filtered sea water to remove supernumerary sperm. Eggs were permitted to settle
and the sea water removed. The damp eggs were extracted for 10-20 minutes in
95% ethanol to yield the active extract (Miller, 1979a). Aliquots of this were air-
dried, diluted into an equivalent volume of sea water and injected into a suspension
of actively moving sperm. The sperm suspension was placed on a standard micro-
scope slide within a 2.4 cm2 area previously covered with a thin layer of 1% bovine
serum albumin in distilled water, to create a flat puddle a few mm deep. The egg
extract was injected with an RGI micrometer syringe connected by thin polyethylene
tubing to a micropipette of 30 ^m tip diameter. Back pressure was controlled by
filling the tubing and syringe with mineral oil. The pipette was lowered into the
puddle and brought to the slide surface while under observation in dark-field illu-
mination with a 10X objective.
Like other invertebrate sperm, O. dioica sperm become thigmotactic on non-
sticky, smooth surfaces. This allows the objective to be focussed on the thigmotactic
cells, which remain on the microscope slide surface indefinitely. The rationale for
the use of thigmotactic sperm for observation of chemotactic behavior and the
probable artifacts inherent in this approach are discussed in Miller (1973). Sperm
behavior was observed and photographed at 12 fps, with 4X reversal film using a
Bolex 16 mm camera. The developed film was analyzed with a Kodak "Analyst"
projector by projecting the film onto tracing paper and plotting the path of the
sperm cells by hand. Sperm velocity was determined by measuring the distance the
sperm head traveled each frame.
We tested for species-specificity by confronting the sperm of O. dioica with egg
extracts from several sessile tunicates, and the various tunicate egg extracts with the
sperm of the larvaceans and tunicates. The numerical estimate of extract activity
used in this work is the titer, or the number of serial half-dilutions required for
complete loss of activity against homo- or heterospecific sperm (Miller, 1979a).
SPERM CHEMOTAXIS IN A LARVACEAN
421
RESULTS
Under the standard conditions of observation, Oikopleura dioica sperm swim-
ming in sea water make relatively straight (Fig. 1A) or circular paths (Fig. IB, 2).
The average velocity during these "control" trails is 75.6 nm/s (Table I). Infrequent,
B
FIGURE 1 . Paths of Oikopleura dioica sperm in the presence of a pipette injecting sea water. A.
Mainly straight or slightly curved trails. B. Mainly curved trails. Pipette diameter is 30 /im. Each interval
on the trail represents 0.08 s.
422
R. L. MILLER AND K. R. KING
FIGURE 2. Paths ofOikopleura dioica sperm in the presence of a pipette injecting sea water. Curved
trails with rare, random loops. Pipette diameter is 30
random turns in the form of sharp loops may occur in some trails within 4-5 frames
(approximately 0.35 s) (Fig. 2). The direction taken after these loops have been
completed is roughly 270° relative to the original path direction. The form of the
new path is the same as the original. Injection of sea water into the sperm suspension
produces no change in sperm motility or direction as long as the rate of injection
is slow enough to prevent physical shifting of the sperm.
If a sea water solution ofOikopleura egg extract is injected (experimental trails),
the sperm behave quite differently (Fig. 3). Sperm enter the field on a typical preat-
traction circular path at about average velocity (72.0 /um/s; Table I) but, about 130
nm away from the pipette tip, undergo a looping behavior which brings them closer
to the pipette tip. The average velocity during these trails is 96.2 p.m/s (Table I).
TABLE I
Average velocities along Oikopleura sperm trails before and during chemotaxis
Number
Number of
Mean
Trail type
of trails
measurements
(M/S)
SD
SE
P
Control trails
21
852
75.61
1.896
+.065 )
Experimental trails
11
681
96.22
2.502
±.096 j
<.001
Pre-attraction
148
71.97
1.565
+.129 )
Post-attraction
148
109.88
1.652
±.136 j
Control and Experimental trails refer to groups of trails in sea water and exposed to a gradient of
sperm attractant, respectively. Pre-attraction and Post-attraction refers to measurements made at the start
of the 1 1 experimental trails and the same number of measurements made at the end of the same set of
trails, respectively.
a, Mest, groups; control trails versus experimental trails.
b, /-test, pairs; pre-attraction versus post-attraction in experimental trails.
SPERM CHEMOTAXIS IN A LARVACEAN
423
B
FIGURE 3. Paths of Oikopleura dioica sperm in the presence of a pipette injecting O. dioica egg
extract with a liter of 9-10. Trails in A and B were obtained from an 18.7 s film sequence and trail positions
have been slightly adjusted for the best demonstration of their characteristics. Arrows indicate point of
acceleration of sperm in response to the attraction gradient. Pipette diameter is 30 j/m.
424
R. L. MILLER AND K. R. KJNG
TABLE II
Oikopleura sperm trail loop and circle characteristics before and during sperm chemotaxis
Loops0
Circles0
Diameter (a, b)
Length (1)
Diameter (d)
Circumference (C = d)
a X b
Before
9.4 X
3.4 Mm
(5)*
24.7
± 1.14 M
:m(5)
97.7
± 5.8
Mm (20)
306.8 ± 15.9 MI
m (20)
During
24.2 X
17.9 Mm
i(25)
66.5
± 4.4 MI"
M25)
42.4
± 5.5
Mm (9)
133.3 ± 17.4 MI
m(9)
* number of measurements.
0 refer to Figure 4.
Table II and Figure 4 present measurements of the loops and circles made in control
and experimental trails. The average loop is 2.7 times longer and 5 times broader
in the experimental (attraction) trails than during the control trails. During the first
looping maneuver, the velocity of the sperm increases significantly (paired Mest; P
< 0.001) (Table I) and the new speed (109.9 nm/s) is maintained for the rest of
the trail.
Once the attracted sperm arrive at the pipette tip, they begin to circle around
it (Fig. 5, 6 A). The circles of all the sperm become more or less concentric, with
an average diameter half of those made during normal swimming (Table II; compare
Figs. 1A, B with Figs. 5, 6A). The concentric circular paths enlarge in diameter,
resembling those seen prior to attraction (Fig. 5A, 6B). All sperm swim counter-
clockwise during this behavior as they did in the circles and loops made before
attraction. Their velocity remains high (109.9 ^m/s). The cells seem to have entered
a new, stable motility configuration and behave as though the attraction gradient
is no longer present. The result of this sequence of behaviors is a rapid shift of the
sperm population toward the pipette tip. By the end of the film sequence, few sperm
are found at the margins of the area of observation.
SPERM TRAIL LOOP AND CIRCLE PARAMETERS
angle of intersection
less than 180°
angle of intersection
at ~* 180°
'1
LOOP
CIRCEF
FIGURE 4. Diagrammatic representation of sperm trail loops and circles, with measurement parameters
•;.v:'d to determine loop and circle sizes.
SPERM CHEMOTAXIS IN A LARVACEAN
425
FIGURE 5. Two trails of attracted sperm showing the start of the concentric circling behavior that
is the result of sperm chemotaxis. 5A shows the characteristic progressive enlargement of the circles.
Pipette diameter is 30
It is evident that the sperm are directed toward the pipette tip when the O. dioica
egg extract is released. To confirm this, the pipette was moved about 0. 1 5 mm from
the outer margin of the old aggregation and a new injection made. The sperm move
from the old aggregation into the new injection area, where a new swarm is formed
of sperm swimming concentrically about the pipette tip. Therefore, not only is a
gradient of attractant required for sperm aggregation, but the same cells can be re-
attracted by the same egg extract.
In three cases we were able to follow the movement of very small particles
(~ 1 /urn in diameter) in front of the pipette as the attractant was injected. Each of
these cases differed in the force of the injection. In the first, enough force was exerted
to push the particles 90 ^m away from the tip before they came to rest. Sperm were
FIGURE 6. A. Another trail showing the transition to concentric circling behavior and the transition
from small to large circles. B. A portion of a plot of sperm circling around the pipette tip at the end of
the preparation (injection stopped). This is 5.3 s of a film sequence showing this behavior only.
426 R. L. MILLER AND K. R. KING
seen to respond 40 ^m further away from the tip. In the second case, the particles
came to rest 60 ^m away from the pipette tip and the sperm were seen to turn a
further 40 /j.m away. In the third case, no particle movement occurred. In this case,
the sperm turned 80 /*m away from the pipette tip. These three cases suggest that
sperm can respond to attractant which has diffused at least 40 to 80 nm beyond the
area of injection.
Tests of the effects of egg extracts from sessile tunicates on O. dioica sperm have
yielded complete species-specificities in all cases. The active egg extracts of Ascidia
callosa (titer =11) and Chelyosoma productum (liter = 5) do not attract O. dioica
sperm, whereas the behavior of the sperm of Corella inflata, Corella willmeriana,
Ciona intestinalis, Ascidia callosa, Chelyosoma productum, Styela montereyensis,
Styela gibbsi, and Halocynthia igaboja remains unaffected by the presence of a gradient
of O. dioica egg extract (titer = 8-9).
DISCUSSION
We have demonstrated that O. dioica spermatozoa, when confronted with a gradient
of an egg extract, are capable of sperm chemotaxis. The trails of attracted sperm
strongly resemble those of chemotactic sperm of other invertebrates (Miller, 1966;
1975; 1977; 1979a). They most particularly resemble chiton sperm trails (Miller,
1977) and those of asteroid and holothuroid sperm (Miller, 1981; in prep.). Sperm
chemotactic behavior is reversible and can be highly species-specific (Miller, 1979a).
Recent work has shown considerable specificity at the genus level in the ascidians
(Miller, 1982). Species-specificity between the larvaceans and the ascidians is therefore
to be expected, and evidence for it has been presented. Interspecific comparison of
sperm chemotaxis between two species of larvaceans has not been possible for lack
of suitable material.
When Oikopleura sperm chemotactic behavior is initiated, sperm velocity in-
creases 50% and remains at this level for the rest of the trail. Velocity increase has
been observed during cnidarian sperm chemotaxis (Miller, 1966). In contrast to
chiton, cnidarian, and Oikopleura sperm, the sperm of the sessile tunicates Ciona
and Styela and those of several echinoderms show no velocity increase during che-
motactic turning or subsequent movement up the gradient (Miller, 1975; 1981;
1982; in prep.). No further changes in velocity occur during subsequent reorien-
tations of larvacean sperm, suggesting that reorientation behavior and velocity in-
crease may be independent in O. dioica sperm, unlike chiton sperm, where small
velocity adjustments occur during every reorientation loop (Miller, 1977). The
source of the activation stimulus may be the sperm attractant itself, though it is
possible that the egg extracts also contain a motility activator (Hansbrough and
Garbers, 1981).
Unlike sessile tunicates, larvacean populations are not limited by availability of
settling substrate for the larvae (Grosberg, 1981), but rather by food supply and
predation (King, 1982). The ability of larvaceans to rapidly increase population size
under certain conditions has been documented (King et al, 1980; King, 1982).
Quantitatively, the relative magnitude of the factors which aid in this increase are
uncertain. Any factor which shortens the developmental time from spawning to
sexual maturation would be of importance, particularly in this instance, where sexual
aggregation may not occur prior to spawning.
Larvacean eggs have a density greater than sea water and sink at about 25 m/
day (300 nm/s) (Bienfang and King, unpub.). The sperm velocity is moderate
SPERM CHEMOTAXIS IN A LARVACEAN 427
(~80-1 10 /um/s) and the sperm is quite small (~ 20 /urn in length; Flood and Afzel-
ius, 1 978), though large numbers are shed by each ripe male. Spawning is cataclysmic
and asynchronous in both sexes (Gait, 1972) so there should not be a uniform
concentration of sperm or eggs in the water column. Both gametes have a fertilizable
life of about 24 h (Gait, 1972).
Assuming the sperm are swimming in random directions relative to the egg, the
chances of an egg being fertilized at a particular time depend directly on local sperm
concentration, average swimming speed of sperm, the average age of the sperm and
eggs, the egg surface area and its sinking rate (Rothschild and Swann, 1949; 1951).
However, in larvaceans, some of these factors are not constant. Sperm velocity
increases and sperm path direction is determined by the presence of eggs, once the
sperm arrive within a minimum distance of 80 /urn from the egg. The effect of this
is to increase the chances of a nearby sperm making contact with the egg surface
by enlarging the effective egg diameter from 80 ^m (Delsman, 1912; Gait, 1972)
to 240 /urn or more.
The number of collisions of sperm with a non-sinking egg per unit time (Z) is
a function of the number of sperm (n), their average velocity (c), and the square
of the egg radius (r): (Z = 7rr2nc; Rothschild and Swann, 1951). Increasing velocity
by 50% (from 70 to 110 jum/s) will increase the number of collisions by two-thirds.
Increasing radius of the egg by three (from 40 to 120 /um) increases collision rate
9 times. The estimated increase in collisions due to sperm chemotaxis compared
to the "standard" fertilization paradigm is approximately 15 times for O. dioica,
assuming no increase in sperm numbers.
Unfortunately, sperm cannot swim as fast as eggs can sink. However, both sperm
and eggs occur in an aqueous medium under a low Reynolds number regime where
viscosity effects are dominant, flow is laminar, and nearby water tends to move with
objects that are subject to an external force (Purcell, 1977; Koehl and Strickler, 198 1 ).
In such a situation the sinking egg (Reynolds number - 0.02) will have a layer of
hydrodynamically constrained water (a boundary layer) associated with it. As the egg
sinks, water at and beyond the boundary layer reaches a velocity, relative to the egg,
equivalent to the egg sinking rate. Within the boundary layer, a velocity gradient
exists such that, at 20 pm away from the surface of the egg, water velocity relative
to the egg and maximum measured sperm swimming speed are equivalent (White,
1974). Assuming that sperm attractant is continually released into the boundary layer,
the increase in effective egg diameter by 50% (beyond 20 um the sperm cannot catch
the sinking egg) and the concomitant increase in sperm swimming speed upon contact
with attractant, yields about a four-fold increase in successful sperm egg collisions.
This is probably a worst case estimate of the efficacy of sperm and sperm attractant
interactions for larvaceans in the pelagial. Oikopleura dioica is usually most abundant
in the surface mixed layer. Here, small scale turbulence in the water column may
provide long term residency in the mixed layer for sperm and egg by effectively altering
egg sinking rates.
ACKNOWLEDGMENTS
We wish to thank the staff of the Friday Harbor Laboratories for their hospitality
during our residence. J. Rodzinski patiently hand-measured the intervals on the
sperm trails. The manuscript was greatly improved by the comments of two anon-
ymous reviewers. This research was supported by a Temple University grant-in-aid-
of-research to RLM and by NSF(IDOE) grant OCE77-27224 to Karl Banse.
428 R. L. MILLER AND K. R. KING
LITERATURE CITED
DELSMAN, J. H. C. 1912. Beitrage gur Entwicklungsgeschichte von Oikopleura dioica. Verh. Rifksinst.
Ondez. Zee (Derde Deel) 2: 3-20.
FENAUX, R. 1967. Les Appendiculaires. In Faune de {'Europe et du Bassin Mediterranean. Masson et
Cie, Paris.
FENAUX, R. 1976. Cycle vital d'un appendiculaire Oikopleura dioica Fol, 1 872 description et chronologic.
Ann. Inst. Oceanogr. 52: 89-101.
FLOOD, P. R., AND B. A. AFZELIUS. 1978. The spermatozoon of Oikopleura dioica Fol (Larvacea,
Tunicata). Cell Tissue Res. 191: 27-37.
FORNERIS, L. 1957. The geographical distribution of the Copelata. An. Acad. Bras. Cienc. 29: 273-284.
GALT, C. P. 1972. Development of Oikopleura dioica (Urochordata: Larvacea). Ph.D. Thesis, University
of Washington, Seattle.
GROSBERG, R. K. 1981. Competitive ability influences habitat choice by marine invertebrates. Nature
290: 700-702.
HANSBROUGH, J. R., AND D. L. CAREERS. 1981. Speract: purification and characterization of a peptide
associated with eggs that activates spermatozoa. J. Biol. Chem. 256: 1447-1452.
KING, K. R. 1982. The population biology of the larvacean Oikopleura dioica in enclosed water columns.
Pp. 341-351 in Marine Mesocosms: Biological and Chemical Research in Experimental Eco-
systems, G. D. Grice and M. R. Reeve, eds. Springer- Verlag, New York.
KING, K. R., J. T. HOLLIBAUGH, AND F. AZAM. 1980. Predator-prey interactions between the larvacean
Oikopleura dioica and bacterioplankton in enclosed water columns. Mar. Biol. 56: 49-58.
KOEHL, M. A. R., AND J. R. STRICKLER. 1981. Copepod feeding currents: food capture at low Reynolds
number. Limnol. Oceanogr. 26: 1062-1073.
MILLER, R. L. 1966. Chemotaxis during fertilization in the hydroid Campanularia. J. Exp. Zool. 162:
22-45.
MILLER, R. L. 1973. Chemotaxis of animal sperm. Pp. 31-47 in Behavior of Microorganisms, A. Perez-
Miravete, ed. Plenum Press, London.
MILLER, R. L. 1975. Chemotaxis of the spermatozoa ofdona infest inalis (Protochordata: Urochordata).
Nature 254: 244-245.
MILLER, R. L. 1977. Chemotactic behavior of the sperm of chitons (Mollusca: Polyplacophora). J. Exp.
Zool. 202: 203-212.
MILLER, R. L. 1979a. Sperm Chemotaxis in the hydromedusae. I. Species and sperm behavior. Afar. Biol.
53:99-114.
MILLER, R. L. 1979b. Sperm Chemotaxis in the hydromedusae. II. Some chemical properties of the sperm
attractants. Mar. Biol. 53: 115-124.
MILLER, R. L. 1981. Sperm Chemotaxis occurs in echinoderms. Am. Zool. 21: 985.
MILLER, R. L. 1982. Sperm chemotaxis in ascidians. Am. Zool. 22: 827-840.
PAFFENHOFER, G.-A. 1976. On the biology of appendicularia of the southeastern North Sea, Vol. 2, Pp.
437-455 in Proceedings of the 10th European Symposium on Marine Biology, G. Persoone and
E. Jaspers, eds. Universa Press, Wetteren, Belgium.
PURCELL, E. M. 1977. Life at low Reynolds number. Am. J. Physics 45: 3-11.
ROTHSCHILD, L., AND M. M. SWANN. 1949. The fertilization reaction in the sea urchin egg. A propagated
response to sperm attachment. / Exp. Biol. 26: 164-176.
ROTHSCHILD, L., AND M. M. SWANN. 1951. The fertilization reaction of the sea urchin. The probability
of a successful sperm-egg collision. J. Exp. Biol. 28: 403-416.
SEKI, H. 1973. Red tide of Oikopleura in Saanich Inlet. La Mer (Bull. Franco- Japonaise Oceanogr.) 11:
153-158.
WHITE, E. W. 1974. Viscous fluid flow. McGraw-Hill Book Co., New York.
WYATT, T. 1973. The biology of Oikopleura dioica and Fritillaria borealis in the Southern Bight. Mar.
Biol. 22: 137-158.
Reference: Bio/. Bull. 165: 429-435. (October, 1983)
LOWER MARINE FUNGUS ASSOCIATED WITH BLACK LINE
DISEASE IN STAR CORALS (MONTASTREA ANNULARIS, E. & S.)
TALIA RAMOS-FLORES
Department of Immunology and Infectious Diseases, The Johns Hopkins University,
School of Hygiene and Public Health, 615 North Wolfe Street, Baltimore, Maryland 21205
ABSTRACT
A disease of corals called "black line" has become widespread in the Caribbean
reefs. Although its etiology has not been determined, a lower marine fungus was
found closely associated with the disease. Corals of the species Montastrea annularis
(star coral) were collected from scattered areas of the Venezuelan reefs. Histological
examinations of black line-diseased corals showed this unidentified fungus in and
nearby all of the diseased tissue. The branched fungal hyphae lacked septa and
ranged in size from 5 to 10 yum long and from 2.5 to 3.0 /urn wide. No hyphae were
found in black line disease-free areas. No fungi have been detected previously in
soft coral tissue. The study of this naturally occurring infection could yield important
information concerning pathological processes in corals.
INTRODUCTION
Diseases in marine animals appear to be a common feature in the aquatic en-
vironment (Kinne, 1980). However, disease processes in marine animals have been
rarely studied as biological phenomena. Not much is known about pathological
conditions in cnidarians, especially in "true" corals (Anthozoa: Scleractinia). In
1975, Garrett and Ducklow first reported a naturally occurring disease in the scler-
actinian corals of the Bermudian reefs. Personal observations of similar conditions
in the Venezuelan reefs prompted my study four years ago.
Black line-diseased corals have been found in several Atlantic reefs (Fig. 1):
Bermuda (Garrett and Ducklow, 1975); Barbados (Ducklow, 1977); Florida (Voss,
1973 and pers. comm. from W. Jaap, Florida Department of Natural Resources,
Marine Research Laboratory, 100 Eighth Avenue, S.E., St. Petersburg, FL 33701);
Saint Thomas (Coki Bay) and Saint Croix (East Point and Buck Island), U. S. Virgin
Islands (pers. ob., 1976); and Venezuela. No reports of this disease have been
published concerning Pacific reefs.
MATERIALS AND METHODS
A comparison of diseased tissue with normal tissue was made (Fig. 2). For this
purpose, small coral heads of the species Montastrea annularis (Ellis and Solander),
measuring about 3 cm in diameter, were collected from the Venezuelan reefs of
Morrocoy National Park (10° northern latitude, 68° western longitude) and Los
Roques National Park (11° northern latitude, 66° western longitude). These col-
lection sites were chosen on the basis of field observations.
Three major collection sites were established for the study: 1) an area with a
high occurrence of the disease (southern reefs of Cayo Norte, Morrocoy Park); 2)
Received 18 November 1981; accepted 28 June 1983.
429
430
T. RAMOS-FLORES
Distribution of 'Black Line' Disease in Atlantic Reefs
85' 80' 75' 70'
if "BLACK LINE" DISEASE (4) ST THOMAS , US VIRGIN ISLANDS
(T) FLORIDA KEYS
(T) BERMUDA
(3) BELIZE
(5) ST CROIX, US VIRGIN ISLANDS
(?) BARBADOS
(7) VENEZUELA
FIGURE 1 . Distribution of "black line" disease in Atlantic reefs.
areas with moderate occurrences of the disease, ranging from 5 to 200 meters from
the disease area (reefs west and northwest of Cayo Norte, Morrocoy Park); and 3)
areas free of observable disease, ranging from 8 to 200 kilometers away from the
^ . ' • ^^ •
^/j^r ; *-5>-W*!
^-&, "*Y ,.J«^^
rfS -..>«- - -^
, iiiju.- -v^yak -^ „ . y
FIGURE 2. Cross section of a healthy Monlastrea annularis, E. & S., polyp. Epidermis is free of
invading organisms. Stained with toluidine blue O, methylene blue and borax. 40X.
LOWER MARINE FUNGUS IN STAR CORALS
431
diseased area (reefs of Cayo Sombrero in Morrocoy Park and reefs of Cayo Mosquito
in Los Roques Park).
Coral heads were collected and fixed in solution for twenty-four hours. The
fixative used was a modification of the formula given by McDowell and Trump
(1977). The ingredients used were: 2 ml of 50% glutaraldehyde; 10 ml of 40%
formaldehyde; 50 ml of filtered sea water; and 39 ml of filtered tap water. Ambient
filtered sea water was used instead of the recommended buffer (sodium phosphate
monobasic). The pH was adjusted to 7.4 with NaOH. The tissues were stored in
alcohol until processed.
Small pieces of tissue were decalcified in Von Eber's decal (50 ml of 36% Nad;
42 ml of distilled water and 8 ml of concentrated HC1). Small coral pieces took three
days to decalcify, larger pieces took up to seven days and the decalcifying baths were
changed daily. After decalcification, the tissues were washed, dehydrated in graded
alcohols, and embedded in JB-4 (Polysciences), a glycol methacrylate polymer. Sec-
tions cut 1.5 microns thick were stained with toluidine blue O, methylene blue and
borax dissolved in distilled water. The solution was prepared by adding 250 mg of
toluidine blue O, 250 mg of methylene blue and 250 mg of borax to 100 ml of
distilled water.
Other histological stains (Periodic acid Schiff, Giemsa, alcian blue and PAS at
pH 1.0 and pH 2.5) were used also, as well as Grocott's method for fungi (GMS).
The procedures for these stains are described in Luna (1968). After staining, the
sections were mounted on plastic-coated slides, covered with a mounting medium
and a cover slip, and examined under the microscope.
RESULTS
Identification of the disease in the field
The gross appearance of the disease in the field is a dark (black) line separating
the dead from the living tissue in a coral head (Fig. 3). The upper coral skeleton
FIGURE 3. Gross appearance of star coral (Montastrea annularis, E. & S.) presenting "black line'
disease. 4X. (Picture was taken under the laboratory dissecting microscope.)
432 T. RAMOS-FLORES
remains mostly intact until it is overgrown by algae and other organisms. No living
tissue is observable within the circumference of the black line. Well beyond the
black ring, the coral appears healthy and maintains all of its zooxanthellae. Dis-
colored patches on the coral heads are often seen in affected areas. This discoloration
may indicate an early stage of the disease and may result from the loss of zoox-
anthellae. The most commonly affected coral genera are Diploria (brain coral) and
Montastrea (star coral).
For practical purposes, a healthy coral head and polyp were denned as being
free of visible lesions. Moderately affected heads and polyps showed few fungi and/
or filamentous algae near the affected tissue. Infection did not occur in all cases.
Heavily affected heads and polyps showed massive fungal infection and the coral
tissue was destroyed for the most part. Algal invasions were present in some cases.
Histological examination in the laboratory
One hundred and fifty-nine polyps from twelve different coral heads were ex-
amined histologically (Table I). The epidermis of all individual polyps presenting
the disease was penetrated by fungal hyphae (Figs. 4, 5) and in more advanced stages
of the disease the gastrodermis and mesoglea also were invaded. Within a single
coral head, those polyps situated directly below the black line were most affected.
Polyps 1 cm away from the disease ring showed less fungal invasion and polyps 5
cm away from the diseased ring had almost no invading hyphae. The tissue appeared
to be normal in these areas.
Histologic examination of the black line area in Montastrea annularis revealed
an ellipsoidal tangle of densely packed, parallel hyphae, filamentous cyanobacteria,
algal fruiting bodies, diatoms, released zooxanthellae, and rodophytes. In some in-
stances, there were mixed fungal and algal invasions of the polyp epidermis. How-
ever, although algae were present in both disease and disease-free areas, fungal
hyphae were found only in areas where the black line disease occurred. The fungal
hyphae were branched and non-septate, 5 to 10 /urn in length and 2.5 to 3.0 nm in
width. The branching fungal filaments were stained orthochromatically with tolu-
idine blue O and were positive for the PAS and for the GMS tests for fungal iden-
tification.
Since no sexual or asexual fruiting bodies were present, the fungus cannot be
identified at this time. However, Dr. Jan Kohlmeyer (Professor, University of North
Carolina, Institute of Marine Sciences, Morehead City, NC 28557) and Dr. Charles
E. Bland (Professor, Department of Botany, University of North Carolina, Chapel
TABLE I
Occurrence of "black line" disease in geographically separated Venezuelan reefs:
relationship to the presence of the fungus
Coral heads Coral polyps
Gross appearance
of coral heads
Distance from area
affected by
"black line" disease
# diseased/
# examined
# infected with
fungi/# examined
healthy
moderately affected
heavily affected
8 km; 200 km*
5 m; 200 m*
0 meters*
0/4
1/4
4/4
0/54
6/30
75/75
* See text for exact location.
LOWER MARINE FUNGUS IN STAR CORALS
433
FIGURE 4. Hyphae infecting coral epidermis. Moderate infection PAS. 200X. Arrow points to
infection site.
Hill, NC 27514) commenting on sample slides, believed the fungus probably be-
longed to the lower marine fungi.
DISCUSSION
Although fungi are very abundant in the marine environment (Kohlmeyer and
Kohlmeyer, 1979) and appear to be major pathogens in some higher aquatic in-
-
FIGURE 5. Closer view of infecting hyphae: a) hypha proliferation in the coral epithelium, and b)
misplaced zooxanthellae PAS. 400 x.
434 T. RAMOS-FLORES
vertebrates such as crayfish and crabs (Nyhlen and Unestam, 1975; Sparks and
Hibbits, 1975), very little is known about their pathogenicity in lower aquatic in-
vertebrates. In this study, histologic examination of black line disease in corals has
shown that an invasion by fungal hyphae is associated with obvious pathological
changes in the tissues. The possibility that this fungus may be a boring species is
indicated by the presence of hyphae growing throughout the hard parts of the corals
and within the septal invaginations. It is not possible at this time to determine
whether the fungus is a primary or a secondary pathogen.
Other investigators have hypothesized that this disease may be caused by bac-
teria. Garrett and Ducklow (1975) have suggested a gram-negative filamentous Beg-
giatoa and a sulfate-reducing anaerobic Desulfovibrio as plausible pathogens. An-
tonius ( 1 977) has suggested a filamentous cyanophyte, Oscillatoria submembranacea
(Ardissone and Strafforelo) as the causative agent of the same coral condition. Nev-
ertheless, no one has isolated the pathogen or reproduced the black line disease
under controlled conditions.
The regenerative ability of some polyps may be a protective mechanism which
prevents complete elimination of the reef. Nearly a century ago, Metchnikoff(1892)
remarked on the amazing regenerative powers of coelenterates. The susceptibility
of regenerating polyps to the disease is unknown, but some mechanism of differential
susceptibility is likely since the disease does not always pursue a destructive course.
Knowledge of individual polyp susceptibility to black line disease could lead to a
determination of how a coral reef copes with advancing pathogens.
A large number of coral colonies on the reefs of Bermuda, Venezuela, and other
Caribbean areas have dead patches. Since many of these patches may be disease
related, the black line phenomenon may be an important factor in coral ecology.
Knowledge of the etiology and pathogenesis of black line disease could, therefore,
yield important clues to the manner in which corals defend themselves against
parasites and other pathogenic agents.
ACKNOWLEDGMENTS
The field studies were made possible through the Marine Ecology Program in
Morrocoy National Park sponsored by C.O.N.I.C.I.T. and the Venezuelan Institute
for Scientific Research (I.V.I.C.).
I thank the late Frederik B. Bang for his direction as my advisor in the Master
of Science degree program at the Johns Hopkins University School of Hygiene and
Public Health; Humberto Diaz, Gilberto Rodriguez, and the staff at the Centre de
Ecologia/I.V.I.C.; Fundacion Los Roques. Freddy Losada (Universidad Central de
Venezuela) helped to locate the diseased corals in the Venezuelan reefs. I am in-
debted to Luis Burguillos (I.V.I.C.) and Peggy Pula (Johns Hopkins U.) for their
technical advice on histological techniques. I am grateful to Hermine Bongers for
her secretarial assistance and to Chester Reather for his photographic expertise. The
information in this paper is also contained in a thesis submitted to and accepted
by the Johns Hopkins University School of Hygiene and Public Health in partial
fulfillment of the requirements for the Master of Science degree.
LITERATURE CITED
ANTONIUS, A. 1977. Coral mortality in reefs: a problem for science and management. Proc. Third Int.
Coral Reef Symp., Miami, 2: 617-620.
DUCKLOW, H. W. 1977. Influence of Sublethal Pollutant Concentrations on the Microbial Ecology of
Living Corals. Ph.D. Dissertation, Harvard University.
LOWER MARINE FUNGUS IN STAR CORALS 435
GARRETT, P., AND H. W. DUCKLOW. 1975. Coral diseases in Bermuda. Nature 253: 349-350.
KINNE, O. 1980. Diseases of Marine Animals. I General Aspects, Protozoa to Gastropoda. John Wiley,
Somerset, NJ. 460 pp.
KOHLMEYER, J., AND E. KoHLMEYER. 1979. Marine Mycology: The Higher Fungi. Academic Press,
New York. 690 pp.
LUNA, L. G. 1968. Manual ofHistologic Staining Methods of the Armed Forces Institute of Pathology.
Third ed. McGraw-Hill, New York. 258 pp.
MCDOWELL, E., AND B. F. TRUMP. 1977. Practical fixation techniques for light and electron microscopy.
Comp. Pathol. Bull. IX: 3.
METCHNIKOFF, E. 1 892. Lectures on the comparative pathology of inflammation. Translated from French
by F. A. Starling and E. H. Starling. Dover Publications, New York, 1968. 224 pp.
NYHLEN, L., AND T. UNESTAM. 1975. Ultrastructure of the penetration of the crayfish integument by
fungal parasite, Aphanomyces astaci, Oomycetes. J. Invert. Pathol. 26: 353-366.
SPARKS, A. K., AND J. HIBBITS. 1979. Black mat syndrome: an invasive mycotic disease of the tanner
crab, Chioecetes bairdi. J. Invert. Pathol. 34: 184-191.
Voss, G. 1973. Sickness and death within the Florida reefs. Nat. Hist. 82: 41-47.
Reference: Biol. Bull. 165: 436-443. (October, 1983)
THE DEVELOPMENTAL APPEARANCE OF PATERNAL FORMS OF
LACTATE DEHYDROGENASE AND MALATE DEHYDROGENASE
IN HYBRID HORSESHOE CRABS'
HIROAKI SUGITA AND KOICHI SEKIGUCHI
Institute of Biological Sciences, University of Tsukuba, Sakura-mura, Niihari-gun, Ibaraki 305, Japan
ABSTRACT
Differences in electrophoretic mobilities of lactate dehydrogenase (LDH) and
malate dehydrogenase (MDH) existed between three Asian horseshoe crabs, Tachy-
pleus tridentatus, Tachypleus gigas, and Carcinoscorpius rotundicauda, used for
interspecific hybridization. After electrophoresis of extracts of hybrid horseshoe crab
embryos on starch gels, the paternal, maternal, and hybrid forms of the LDH and
MDH were detected with specific enzyme staining. In viable hybrids the paternal
form of the LDH was detected at stage 17 (immediately before the 1st embryonic
molt). Similarly, evidence of gene expression for mitochondrial MDH was seen at
stage 14 (stage of appearance of rudimental appendages). Gene expression for su-
pernatant MDH was seen at stage 17 (immediately before the 1st embryonic molt).
Regarding the onset of genome control in embryogenesis, it was suggested that prior
to the activation of the maternal gene of the LDH, the paternal gene of the LDH
was activated in horseshoe crab hybrids. Furthermore, there was evidence that the
maternal effects on early embryogenesis were due to enzymes present in the egg
prior to fertilization, not to continued synthesis directed by stable messenger RNA.
INTRODUCTION
Morphological studies on echinoderm, amphibian, teleost, and other species
hybrids show that, in general, only maternal characters are evident until gastrular
or postgastrular organogenesis. This conclusion is supported by many studies in
which enzymes and other proteins of paternal type are first observed at postgastrular
stages (Davidson, 1976). If two species with differences in specific enzymes form
viable hybrids, and meternal- and paternal-type enzymes can be distinguished in
the offspring, the paternal enzymes should not appear until after the new diploid
genome is activated in the embryo. Therefore, the viable hybrids offer an opportunity
for studying maternal and paternal contributions to development. To detect the
paternal form of the enzyme, techniques of zone electrophoresis and specific enzyme
staining have been applied to lactate dehydrogenase (LDH) in hybrids of frogs
(Wright and Moyer, 1966, 1968; Wright and Subtelny, 1971) and fishes (Hitzeroth
et al, 1968; Goldberg et ai, 1969), and to malate dehydrogenase (MDH) in frog
hybrids (Wright and Subtelny, 1971). In interspecific hybrids of arthoropods, how-
ever, no work has been carried out to detect the paternal forms of enzymes during
the development of the embryo.
In this paper we report the time of the expression of the paternal genes controlling
Received 1 April 1983; accepted 20 July 1983.
1 Contribution No. 415 from the Shimoda Marine Research Center, University of Tsukuba.
Abbreviations: LDH = lactate dehydrogenase; MDH = malate dehydrogenase.
436
HYBRID HORSESHOE CRAB LDH AND MDH 437
lactate dehydrogenase and malate dehydrogenase in hybrid embryos of Asian horse-
shoe crabs.
MATERIALS AND METHODS
The Japanese horseshoe crab, Tachypleus tridentatus, was collected from Imari
and Fukuoka, Japan, and the Southeast Asian horseshoe crabs, Tachypleus gigas
and Carcinoscorpius rotundicauda, were collected from the vicinity of Bangsaen,
Thailand, by Professor Smarn Srithunya (Zoological Museum and Marine Aquar-
ium, Srinakharinwirot University, Thailand).
To contrast the paternal influence of three horseshoe crabs, eggs obtained from
one female were divided into three groups, and each group was artificially insem-
inated by sperm from one of the three species and kept at 30°C. Cross-fertilizations
were made in all nine combinations among three Asian horseshoe crab species. The
developmental stage of the embryos was determined according to the normal plate
of the Japanese horseshoe crab, T. tridentatus, described by Sekiguchi (1973), be-
cause fertilized eggs of T. gigas, C. rotundicauda, and the interspecific hybrids de-
veloped into swimming larvae (the first-instar larvae) through a similar morpho-
logical process to those of T. tridentatus.
A single embryo at each stage was homogenized in one or two drops of distilled
water. Larval extracts were prepared from a single animal at the first-instar stage
(just after hatching). To prepare the adult tissue extracts, the hepatopancreas was
homogenized in a volume of distilled water approximately equal to the tissue vol-
ume, because all isozyme molecules of the LDH were included in horseshoe crab
hepatopancreas. Sample homogenates were absorbed on a small piece of Toyo No.
50 filter paper and inserted into slits cut in the starch gel. Electrophoresis was carried
out at 4°C with 11% or 12% gel horizontally for embryonic samples or vertically
for larval and adult samples. Horizontal gel electrophoresis for embryonic and larval
MDH was carried out using Davis' (1964) buffer system. Selander and Yang's (1969)
buffer system was used during vertical gel electrophoresis for larval and adult MDH.
Gel and electrode buffers for LDH isozymes were prepared according to the method
of Selander and Yang (1969). A 100 ml staining mixture for the LDH consisted of
0.025 M Tris-HCl buffer (pH 7.4), 50 mg nicotinamide adenine dinucleotide, 35
mg nitro blue tetrazolium, 3 mg phenazine methosulphate, 2.0 ml 60% Na lactate,
and 1.0 ml 0.5 M KCN (Shows and Ruddle, 1968). The staining mixture for the
MDH was identical to the LDH but 10 ml 1.0 MNa malate, pH 7.0, was substituted
for 2.0 ml 60% Na lactate.
RESULTS
Lactate dehydrogenase
Before we consider the developing enzyme patterns in hybrids, we must examine
whether the enzyme variants are present in adult and larval samples. Figure 1 shows
the electrophoretic patterns of the LDH from the hepatopancreas tissues and the
first-instar larvae of 3 Asian horseshoe crabs. The larval LDH from 3 species showed
only one enzymic band, while the LDH from the hepatopancreas tissues of T.
tridentatus and C. rotundicauda occurred in 3 isozymic forms on starch gel. Fur-
thermore, the LDH from each of interspecific hybrid larvae showed 3 enzymic
bands, suggesting that 2 peptides produced from paternal and maternal genes for
the LDH could form heterodimers in horseshoe crab hybrids (Fig. 1 B). The adult
LDH of T. tridentatus and C. rotundicauda was monomorphic and that of T. gigas
438 H. SUGITA AND K. SEKIGUCHI
A B
Cr Tt Tg Cr rt tr Tt gt Tg
FIGURE 1 . Electrophoretic patterns of horseshoe crab LDH from hepatopancreas tissues (A) and
from the first-instar larvae (B). Vertical starch gel electrophoresis was carried out at 4°C with 12% gel,
using the buffer system of Selander and Yang (1969. Gel buffer: 0.08 M Tris and 0.005 M citric acid,
pH 8.7. Electrode buffer: 0.3 M boric acid and 0.06 M NaOH, pH 8.2). Cr = Carcinoscorpius rotundicauda;
Tt = Tachypleus tridentatus; Tg = Tachypleus gigas; rt = hybrid between Cr 9 and Tt S; tr = hybrid
between Tt 9 and Cr <J; gt = hybrid between Tg 9 and Tt <?.
was polymorphic (Sugita and Sekiguchi, in prep.). Genetic variants of the LDH
could not be detected in larvae developed from eggs of a single female.
In hybrid progeny obtained from the interspecific crosses of all 6 combinations
among 3 species as well as in normal progeny from the control crosses, the early
embryos displayed only the maternal LDH pattern which could be detected in
unfertilized eggs (results not shown). The maternal LDH from these embryos had
similar relative mobility to the LDH from the first-instar larvae (Fig. 1 B).
In hybrid embryos between T. tridentatus 9 and C. rotundicauda 6, the paternal
form of the LDH was first detected at stage 17 (30 days after insemination, im-
mediately before the 1st embryonic molt) (Fig. 2A), but the paternal form of the
LDH was not observed even on the 41st day after insemination (stage 19, after the
the 2nd embryonic molt) in hybrid embryos between C. rotundicauda 9 and T.
tridentatus 3.
The LDH from hybrid embryo between T. gigas 9 and T. tridentatus 6 occurred
in 3 molecular forms at stage 20 (32 days after insemination, after the 3rd embryonic
molt), suggesting that the LDH of the hybrid embryo consisted of a maternal hom-
odimer, a paternal homodimer, and a hybrid heterodimer (Fig. 2B). This paternal
form of the enzyme was first observed at stage 18 (28 days after insemination, after
the 1st embryonic molt) in hybrid embryo T. gigas 9 X T. tridentatus $ (results not
shown). On the other hand, the LDH from T. tridentatus 9 X T. gigas $ hybrid
embryo showed only the maternal form on the 32nd day after insemination (Fig.
HYBRID HORSESHOE CRAB LDH AND MDH
439
B
Tt tr rt rt Cr Cr Tg Tg Tg gt gt tg Tt Tt Tt
18 17 15 17 18 U U 18 19 19 20 (6) 19 20 U
FIGURE 2. Electrophoretic patterns of the LDH in unfertilized eggs and developing embryos of 3
Asian horseshoe crabs and their hybrids. The hybridization experiments were carried out 2 times using
different sets of 3 pairs (3 species) of horseshoe crabs. Electrophoretic patterns of the LDH in the 30th-
day and 32nd-day embryos from the 2 experiments are shown separately in A and B, except for a column
Cr U in A. Horizontal starch gel electrophoresis was carried out at 4°C with 1 1% gel, using the same
buffer system as explained in Figure 1 . Cr, Tg, Tt, gt, rt, and tr are as described in Figure 1 . tg = Hybrid
between Tt 9 and Tg <5; U = unfertilized egg. Unfertilized eggs as well as fertilized eggs were cultured
in sea water at 30°C for 3 days (A) and 32 days (B). Numbers indicate the developmental stage of
Sekiguchi's normal plate (Sekiguchi, 1973). The number 6 in parentheses means that hybrid embryo used
was able to live on until the 32nd day after insemination, although the development had stopped at stage
6 (blastula stage).
2B). Until this day the hybrid was able to live on, although the development had
stopped at blastula stage, or stage 6 (Sekiguchi and Sugita, 1980; Sugita et al, 1982).
The hybridized eggs of C. rotundicauda 9X7". gigas 6 and the reciprocal cross
stopped their development at blastula stage (Sekiguchi and Sugita, 1980; Sugita et
al., 1982) and never expressed the paternal forms of the LDH (results not shown).
Malate dehydrogenase
There are 2 major electrophoretic forms of the MDH in the horseshoe crab,
Limulus polyphemus, as well as in most animals and higher plants. These isozymes
are controlled by separate genetic loci and are localized in different subcellular
fractions, a mitochondrial form and a supernatant form (Selander et al., 1970). On
a gel run with Davis' (1964) buffer system, the larval MDH from 3 Asian horseshoe
crabs showed the slower-migrating system (MDH-1), which was the mitochondrial
form, and the faster-migrating system (MDH-2), or the supernatant form as Selander
et al. (1970) reported with Limulus MDH using the buffer system of Selander and
Yang ( 1 969). When, in our laboratory, electrophoresis was carried out using Selander
and Yang's (1969) buffer system, the mitochondrial bands were very close to the
440
H. SUGITA AND K. SEKIGUCHI
supernatant bands on a gel. Therefore, we used Davis' (1964) buffer system to
examine the developing MDH patterns in Asian horseshoe crabs and their hybrids.
There were electrophoretic variants of the MDH in 3 Asian horseshoe crabs, but
genetic variants of the MDH were not detected in larvae developed from eggs of
a single female (compare the MDH-1 of columns Tg, gt, and gr in Fig. 3 with that
of columns Tg and gt in Fig. 4).
The early embryos displayed only the maternal forms of both MDH-1 and
MDH-2 in hybrid and normal progeny and these enzyme forms were detected in
unfertilized eggs of 3 species (Fig. 3). The paternal form of the MDH-1 was first
detected in the T. tridentatus 9 X C. rotundicauda 6 embryo at stage 14 (stage of
appearance of rudimental appendages, 28 days after insemination) (Fig. 4A), while
in the hybrid embryo of the reciprocal cross the paternal forms of the MDH-1 and
MDH-2 were not expressed even on the 41st day after insemination (stage 19, after
the 2nd embryonic molt).
The paternal form of the MDH-2 was displayed in the T. gigas 9 X T. tridentatus
<3 embryos at stage 19 (after the 2nd embryonic molt) and stage 20 (after the 3rd
embryonic molt) (Fig. 4B). This paternal form in the T. gigas 9 X T. tridentatus
<3 embryo was first observed at stage 17 (22 days after insemination, immediately
before the 1st embryonic molt, results not shown).
On the other hand, the hybrid embryos whose development was stopped at
blastula stage, that is, T. tridentatus 9 X T. gigas <3, T. gigas 9 X C. rotundicauda
B
MDH-1
MDH-2
I
tr rt Cr Cr rg gr Tg Tg gt tg Tt Ti
155U53 U 666 6U
FIGURE 3. Electrophoretic patterns of the MDH in unfertilized eggs and early embryos of 3 Asian
horseshoe crabs and their hybrids. The 3rd-day and 6th-day embryos from a hybridization experiment
were used in A and B, respectively. Horizontal starch gel electrophoresis was carried out at 4°C with
1 1% gel, using the buffer system of Davis (1964. Gel buffer: 0.38 MTris-HCl, pH 8.9. Electrode buffer:
0.005 M Tris and 0.038 M glycine, pH 8.3). Numbers indicate the developmental stage of Sekiguchi's
normal plate (Sekiguchi, 1973). Symbols are explained in Figures 1 and 2, except for symbols denned
below. MDH-1 = Slower-migrating system, or mitochondrial form; MDH-2 = faster-migrating system,
or supernatant form; gr = hybrid between Tg 9 and Cr <5; rg = hybrid between Cr 9 and Tg 6.
HYBRID HORSESHOE CRAB LDH AND MDH
441
B
MDH-1
MDH-2
&• «•»
is '*»*>
Tt tr tr rt rt Cr Tg g1 gt tg Tt Tt
17 14 15 14 15 18 19 19 20 (6) 19 20
FIGURE 4. Electrophoretic patterns of the MDH from the 28th-day (A) and 32nd-day (B) embryos
of 3 Asian horseshoe crabs and their hybrids. Electrophoretic patterns of the MDH in embryos obtained
from different sets of 3 pairs (3 species) of horseshoe crabs are shown separately in A and B. Horizontal
starch gel electrophoresis was carried out at 4°C with 1 1 % gel using the same buffer system as in Figure
3. Numbers indicate the developmental stage of Sekiguchi's normal plate (Sekiguchi, 1973). The number
6 in parentheses means that the hybrid embryo was able to live on until the 32nd day after insemination,
although the development had stopped at stage 6 (blastula stage). All symbols are explained in Figures
1, 2, and 3.
3, and C. rotundicauda 9 X T. gigas 6 embryos did not express the paternal forms
of the MDH-1 and MDH-2.
DISCUSSION
The LDH of horseshoe crabs is D-lactate specific and has a molecular weight
of approximately 70,000 (Long and Kaplan, 1968, 1973). This D-LDH occurs in
3 dimeric forms, not in 5 tetrameric forms as does the L-LDH of vertebrates with
a molecular weight of 140,000 (Selander and Yang, 1970; see columns Tt and Cr
in Fig. 1 A). Although each LDH from the larvae of 3 Asian horseshoe crabs shows
only one dimeric form with different electrophoretic mobility from one another, the
LDH from the first-instar larvae of hybrid horseshoe crabs is composed of 3 mo-
lecular forms: a maternal homodimer, a paternal homodimer, and a hybrid heter-
odimer (Fig. 1 B). This hybrid LDH heterodimer was detected with maternal and
paternal homodimers in T. gigas 9 X T. tridentatus 6 embryo at stage 20 (Fig. 2B).
However, the paternal LDH homodimer from this cross-fertilized embryo was first
observed without the hybrid heterodimer at stage 1 8 (results not shown). Similarly,
in T. tridentatus 9 X C rotundicauda 8 embryo the paternal homodimer of the
LDH was first detected at stage 1 7 (Fig. 2 A) without the hybrid heterodimer which
was observed at stage 1 9 (4 1 days after insemination, results not shown).
442 H. SUGITA AND K. SEKIGUCHI
Based on the findings that no hybrid enzymes were detected in androgenetic
haploid frog hybrids, Wright and Subtelny (1971) indicated that the degradation of
maternal (cytoplasmic) enzymes in vivo did not yield subunits capable of reaggre-
gation with newly synthesized subunits to form active enzymes. This means that
the hybrid forms of enzymes are expressed at the time when both maternal and
paternal genes for the enzymes are activated together. Therefore, the findings that
the paternal and maternal homodimers were detected without their hybrid hetero-
dimer indicate that, with regard to the onset of genome control in embryogenesis,
prior to the activation of the maternal gene of the LDH the paternal gene of the
LDH was activated in the horseshoe crab hybrids.
The time of expression of the paternal genes controlling the mitochondrial malate
dehydrogenase (MDH-1) and supernatant malate dehydrogenase (MDH-2) was ex-
amined, although they did not show clear, electrophoretic patterns. Evidence of
paternal gene expression for the MDH- 1 was seen in T. tridentatus 9 X C. rotun-
dicauda <5 embryos at stage 14 (Fig. 4A). Expression of paternal gene for the
MDH-2 was seen in T. gigas 9 X T. tridentatus $ embryos at stage 1 7 (results not
shown).
In early embryos only the maternal forms of the LDH and MDH were observed
until postgastrular organogenesis (stage 13, stage of the germ-band formation). The
active maternal forms of these enzymes were present in unfertilized eggs of 3 Asian
horseshoe crabs (Figs. 2, 3) and the steady state activity of the maternal enzymes
in unfertilized eggs did not change dramatically during the culture for 32 days at
30°C in sea water (Fig. 2B). These and other results present evidence that the
maternal effects on early embryogenesis are due to enzymes present in the egg prior
to fertilization, not to continued synthesis directed by stable messenger RNA (Wright
and Subtelny, 1971).
ACKNOWLEDGMENTS
We thank Professor Smarn Srithunya for collecting Tachypleus gigas and Car-
cinoscorpius rotundicauda in Bangsaen, Thailand.
This work was supported by the grants-in-aid for scientific research from the
Ministry of Education, Science and Culture of Japan.
LITERATURE CITED
DAVIDSON, E. H. 1976. Gene Activity in Early Development, 2nd ed. Academic Press, New York. 452
pp.
DAVIS, B. J. 1964. Disc electrophoresis — II. Method and application to human serum proteins. Ann.
N. Y. Acad. Sci. 121: 404-427.
GOLDBERG, E., J. P. CUERRIER, AND J. C. WARD. 1969. Lactate dehydrogenase ontogeny, paternal gene
activation, and tetramer assembly in embryos of brook trout, lake trout, and their hybrids.
Biochem. Genet. 2: 335-350.
HITZEROTH, H., J. KLOSE, S. OHNO, AND U. WOLF. 1968. Asynchronous activation of paternal alleles
at the tissue-specific gene loci observed on hybrid trout during early development. Biochem.
Genet. 1: 287-300.
LONG, G. L., AND N. O. KAPLAN. 1968. D-lactate specific pyridine nucleotide lactate dehydrogenase in
animals. Science 162: 685-686.
LONG, G. L., AND N. O. KAPLAN. 1973. Diphosphopyridine nucleotide-linked D-lactate dehydrogenases
from the horseshoe crab, Limulus polyphemus and the seaworm, Nereis virens. I. Physical and
chemical properties. Arch. Biochem. Biophys. 154: 696-710.
SEKIGUCHI, K. 1973. A normal plate of the development of the Japanese horse-shoe crab, Tachypleus
tridentatus. Sci. Rep. Tokyo Kyoiku Daigaku Sect. B 15: 153-162.
SEKIGUCHI, K., AND H. SUGITA. 1 980. Systematics and hybridization in the four living species of horse-
shoe crabs. Evolution 34: 712-718.
HYBRID HORSESHOE CRAB LDH AND MDH 443
SELANDER, R. K., AND S. Y. YANG. 1969. Protein polymorphism and genie heterozygosity in a wild
population of the house mouse (Mus musculus). Genetics 63: 653-667.
SELANDER, R. K., AND S. Y. YANG. 1970. Horseshoe crab lactate dehydrogenases: Evidence for dimeric
structure. Science 169: 179-181.
SELANDER, R. K., S. Y. YANG, R. C. LEWONTIN, AND W. E. JOHNSON. 1970. Genetic variation in the
horseshoe crab (Limulus polyphemus), a phylogenetic "relic." Evolution 24: 402-414.
SHOWS, T. B., AND F. H. RUDDLE. 1968. Function of the lactate dehydrogenase B gene in mouse
erythrocytes: Evidence for control by a regulatory gene. Proc. Nat. Acad. Sci. USA 61: 574-
581.
SUGITA, H., K. SEKJGUCHI, F. SHISHIKURA, AND Y. YAMAMICHI. 1982. An evolutionary aspect to
horseshoe crabs based on developmental capacity of the interspecific hybrids (in Japanese). Proc.
Jpn. Soc. Syst. Zool. (Tokyo) No. 22: 1-6.
WRIGHT, D. A., AND F. H. MOVER. 1966. Parental influences on lactate dehydrogenase in the early
development of hybrid frogs in the genus Rana. J. Exp. Zool. 163: 215-230.
WRIGHT, D. A., AND F. H. MOVER. 1968. Inheritance of frog lactate dehydrogenase patterns and the
persistence of maternal isozymes during development. J. Exp. Zool. 167: 197-206.
WRIGHT, D. A., AND S. SUBTELNY. 1971. Nuclear and cytoplasmic contributions to dehydrogenase
phenotypes in hybrid frog embryos. Dev. Biol. 24: 1 19-140.
Reference: Biol. Bull. 165: 444-449. (October, 1983)
REPETITIVE CYCLES OF BIOLUMINESCENCE AND SPAWNING IN
THE POLYCHAETE, ODONTOSYLLIS PHOSPHOREA
FREDERICK I. TSUJI AND ELIZABETH HILL
Marine Biology Research Division, Scripps Institution of Oceanography, University of California,
San Diego, La Jolla, California 92093, and Veterans Administration Medical Center
Brentwood, Los Angeles, California 90073
ABSTRACT
Spawning by large numbers of the marine polychaete, Odontosyllis phosphorea,
occurred at fortnightly intervals. The animals appeared at the surface of the water
shortly after sunset and luminesced and spawned for approximately 30 minutes.
The spawning was correlated with the monthly lunar and tidal cycles and lasted
from June through October.
INTRODUCTION
Polychaetes of the genus Odontosyllis from Bermuda show spawning swarms
throughout the year, with lunar periodicity: shortly after sunset, the bioluminescent
worms appear at the surface, where they pair and mate, for several days immediately
after full moon (Galloway and Welch, 1911; Huntsman, 1948; Markert et al, 1961).
Similar behavior has been reported for two other species of this genus: one from
Puerto Rico (Erdman, 1965) and another, Odontosyllis phosphorea, from British
Columbia (Potts, 1913); lunar periodicity in the latter species, however, has been
questioned by others (Eraser, 1915; Berkeley, 1935). Still other species of this genus
spawn only once a year (Haneda, 1971; Daly, 1975; Horii, 1982). We have observed
spawning swarms in O. phosphorea from southern California, which are similar to
those described in the Caribbean, except that the spawning peaks are strongly sea-
sonal and occur at fortnightly intervals: i.e., follow a semi-lunar rhythm rather than
a lunar rhythm, as reported for other species of this genus.
MATERIALS AND METHODS
Observations were carried out from a 3 X 24 m floating dock in De Anza Cove,
Mission Bay, San Diego. The dock is oriented north-south, with the east side facing
the shore. Observations were made on the shore-side, which comprised a maximum
area of 530 m2 of water surface. The dock is connected to shore by a raised walk.
The distance from the edge of the dock to the high water mark on shore was ~32
m and to the low water mark, ~ 10 m. The water depth on July 12, 1982, at 19:15
(low tide = 19:49) was 2.3 m at the south end of the dock and 3.0 m at the north
end. Luminescing Odontosyllis swimming at the surface were counted by two ob-
servers walking along the edge of the dock using hand tally counters. Each observer
monitored one-half of the water surface; and one observer also kept time with a
stopwatch. An individual Odontosyllis was recognized by the greenish luminescence
produced by the swimming animal. A complete count of the area could be made
in one minute, even during peak activity. Records were kept of the direction and
Received 21 March 1983; accepted 25 July 1983.
444
SPAWNING CYCLES IN ODONTOSYLLIS 445
strength of the wind, condition of the water surface, surface water temperature,
phase and position of the moon (when visible), condition of the tide, and overhead
cloud cover.
The times (PST) of sunset, civil twilight, moonrise, and moonset, and phase of
the moon were calculated for San Diego (32.46°N) using standard tables (Nautical
Almanac for 1982, U. S. Naval Observatory, 1980); plotted tides were predictions
for San Diego (Tide Table for 1982, NOAA, 1981). The tidal difference between
the Pacific Ocean entrance to Mission Bay and De Anza Cove is negligible. Normal
probability curves and standard deviations were calculated from the observed data
points (Alder and Roessler, 1968).
Specimens of Odontosyllis were collected with a 90 ml ladle or a fine mesh net.
Each individual was immediately placed in a separate container. To determine the
sex of the animal, the coelomic cavity and reproductive organ were dissected mi-
croscopically and the type of gamete was determined. The number of eggs, and, in
some cases, the number of eggs already undergoing cleavage, were determined in
water samples collected at the same time as the specimen. Counts were made within
one hour after collection, using a dissecting microscope. Control water samples were
also collected before and after each night of observation. Specimens were collected
when swarming activity was at its peak.
RESULTS
The first flashes of light after sunset were usually from males. They swam in a
relatively straight line while the posterior section of the body luminesced internally.
A bright burst of luminous secretion was produced intermittently, forming a lu-
minescent trail. This trail hung at the surface of the water for about a minute before
dispersing. Water samples collected with such worms often contained spermatozoa.
The females began flashing shortly after the males. They appeared at the surface of
the water swimming in tight wiggling circles. The body as well as the secretion it
discharged were brightly luminescent. Sometimes a male and a female were observed
swimming together in a small circle. The water collected with such females frequently
contained eggs, and the body was nearly devoid of eggs when subsequently examined.
A fully elongated adult Odontosyllis was 20 to 30 mm in length and was about
one millimeter in width. Eggs were ~15 /j.m in diameter. When maintained in
filtered sea water at room temperature (~21°C), fertilized eggs began cleavage and
reached the gastrula stage after ~ 1 2 hours. The ciliated gastrulae actively swam in
circles near the surface of the water; they developed into early trochophores after
two days and into full trochophores after four days. Each was characterized by a
well developed apical tuft, prototroch, growth zone, and pygidium. The trochophores
had four black eye spots and the body showed signs of segmentation. The trocho-
phore larvae did not luminesce when tested with MgCl2 and KC1. However, within
a month they reached 40 ^m in length, developed parapodia, and possessed the
ability to luminesce. During peaks of swarming in July and August, 1982, egg counts
ranged from 35 to 63/ml (8 counts); sex ratio (males/females) varied from 0.3 to
0.6 (3 samples).
Counts of luminescing worms on a typical high-intensity spawning date are
presented in Figure 1 . The worms appeared about 1 7 minutes after sunset, the last
was seen about 32 min later, and peak abundance was about 33 min after sunset.
The data show a reasonably good fit with a calculated normal probability curve.
Similar curves were derived for all observation dates on which worms were seen
between July and October. The centers of these spawning peaks (mean time relative
446
F. I. TSUJI AND E. HILL
200-
17
21
25
29 33 37 41
Minutes after Sunset
45 49
53
FIGURE \. Plot of the number of Odontosyllis appearing after sunset on 9 August 1982. Solid line
represents the normal probability curve calculated from the observed data points. Arrow indicates the
mean minutes after sunset (time of peak abundance of the worms) and standard deviation.
to sunset) showed a tendency to occur progressively earlier during July, to reach a
minimum in early August, and to occur progressively later thereafter, with a total
seasonal range of about 30 min (Fig. 2). The days were becoming shorter (i.e., sunset
occurred earlier) throughout this interval. The length of twilight (~24 to 28 min),
the time the full moon was in the sky (~9.5 to 1 1.2 hour), and weather conditions —
even strong wind and overcast sky — had no apparent effect on the daily or fortnightly
timing of spawning swarms.
The first sighting of Odontosyllis was made on 13 May (last quarter moon, 15
May) when 26 worms were counted. On 15 May, many more worms were seen,
but not counted. Subsequently, worms were observed on 27 May (not counted; first
quarter moon, 29 May) and on 16 June (30 counted; last quarter moon, 14 June).
Thereafter, regular counts were taken. Figure 3 shows a plot of number of Odon-
tosyllis observed, high and low water predictions for the tide, surface water tem-
perature, and phases of the moon against dates of observation. The observations
representing each fortnightly peak show a reasonably good fit with a calculated
normal probability curve. The results show that swarming by Odontosyllis follows
a semi-lunar cyclic pattern, with peak spawning coinciding approximately with the
time of minimum variation in tidal amplitude, as well as with the first and last
quarter phases of the moon. The cumulative difference between the days of peak
swarming and the corresponding days of the quarter moon over the interval from
late June to late October was —0.1 day; individual peaks varied from —2.9 to +2.7
days. The duration of bioluminescence and spawning ranged from 27.7 to 44.9 min
between July and October, with a mean of 34.9 min. As surface water temperature
rose in July, the Odontosyllis swarm populations increased to a maximum peak
between 3-15 August. Thereafter, as the surface water temperature gradually fell,
the peaks decreased in height, with the exception of a large peak between 6-12
October. No Odontosyllis were observed during regular searches, centered around
times of first and last quarter of the moon, between 1 November and mid-April,
SPAWNING CYCLES IN ODONTOSYLLIS
447
LU
CO
ZI
=>
CO
£
cr
I9;00
1800
UJ
co
CO
65 r
60
55
50
L±J 3U
I-
U.
< 45
CO
UJ
t: 40
35
30 -
25
10 18 26 3 II 19 27 4 12 20 28 6 14 22 30
JULY AUGUST SEPTEMBER OCTOBER
FIGURE 2. Plot of the mean minutes after sunset and time of sunset against dates of observation,
July-October, 1982.
1983, as the surface water temperature reached a low of 14.0°C on 4 February. Two
worms were seen on 19 April (first quarter moon) and the onset of more intensive
swarming was signalled by a count of 80 worms on 3 June, by which time water
temperature had risen to 22.0°C.
DISCUSSION
The data in Figures 1-3 indicate three rhythmic components in the reproductive
behavior of O. phosphorea: a seasonal cycle, with peak spawning in the warm-water
months of July to October; a fortnightly cycle, with spawning on dates corresponding
roughly with first and last quarters of the moon (and hence, with neap tides); and
a strong daily cycle, with spawning confined to less than an hour, beginning shortly
after sunset. It is conceivable that water temperature itself influences spawning on
a seasonal basis; the observations of Fraser (1915) suggest a much more seasonally
restricted spawning of O. phosphorea in the colder waters of British Columbia. We
cannot determine from the present data whether endogenous factors are involved
in the fortnightly and daily rhythmicities. The semi-lunar rhythm may be directly
evoked by the tidal regime, or it might represent an endogenous rhythm, perhaps
synchronized by moonlight. Neumann (1976, 1978), in his laboratory studies of the
reproduction of the marine midge, Clunio marinus (which also shows annual, semi-
lunar, and daily rhythmicity), demonstrated that it is possible to induce fortnightly
rhythms in the breeding by either artificial moonlight (a few days per month) or
simulated tides, together with a light-dark cycle.
The fact that O. phosphorea shows a semi-lunar rhythm of spawning, and not
a lunar rhythm, as reported for other species of this genus from the Caribbean
448
F. I. TSUJI AND E. HILL
20 JO 10 20 30 10 20 30 10 20 3O 10 20 X 10
' I Ill I I I lil I I I III I I I I I II I I I I I II I I I I I I I I I I I I III
1982
FIGURE 3. Plot of the number of Odontosyllis appearing, high and low water tide predictions for
San Diego, surface water temperature, and phases of the moon against dates of observation, June-
November, 1982. Solid line represents normal probability curves for each activity period calculated from
total counts taken each evening. The values given above each peak represent the calculated mean date
of peak abundance of the worms and standard deviation. The percentage of total worms (43,983), ap-
pearing in each activity period peaks, was as follows: 28 June-2 July, 0.53; 10-18 July, 5.1 1; 23-31 July,
16.76; 3-15 August, 40.87; 22-30 August, 9.06; 4-14 September, 8.41; 22-28 September, 3.82; 6-12
October, 12.71; and 20-26 October, 2.73.
(Markert et ai, 1961; Erdman, 1965) suggests that regardless of proximate factors,
the behavior of the worms in Mission Bay is an adaptation to tidal conditions which
recur at fortnightly intervals. In this connection, we note that during neap tides,
there is minimal tidal flushing of an enclosed embayment, meaning that the progeny
of spawning worms are able to complete their early larval development in near
proximity to the adult habitat.
ACKNOWLEDGMENTS
We are greatly indebted to the following individuals: Dr. Kristian Fauchald,
Smithsonian Institution, for identifying O. phosphorea; Dr. Terrance E. Meyer,
University of Arizona, for assisting with some of the worm counts; and Dr. James
T. Enright of our Division, S.I.O., for critically reading the manuscript. This work
was supported in part by research grants PCM79-21658 and PCM82- 15773 from
the National Science Foundation.
LITERATURE CITED
ALDER, H. L., AND E. B. ROESSLER. 1968. Introduction to Probability and Statistics. W. H. Freeman
and Company, San Francisco. 333 pp.
SPAWNING CYCLES IN ODONTOSYLLIS 449
BERKELEY, E. 1935. Swarming of Odontosyllis phosphorea, Moore, and of other polychaeta near Na-
naimo, B. C. Nature 136: 1029.
DALY, J. M. 1975. Reversible epitoky in the life history of the polychaete Odontosyllis polycera (Schmarda
1861). J. Mar. Biol. Assoc. U. K. 55: 327-344.
ERDMAN, D. S. 1965. Lunar periodicity in the swarming of luminescent worms, Odontosyllis octodentata
Treadwell (annelida) off La Parguera, P. R. Caribb. J. Sci. 5: 103-107.
FRASER, C. M. 1915. The swarming of Odontosyllis. Trans. R. Soc. Can. 9: 43-49.
GALLOWAY, T. W., AND P. S. WELCH. 1911. Studies on a phosphorescent Bermudan annelid, Odontosyllis
enopla Verrill. Trans. Am. Microsc. Soc. 30: 13-39.
HANEDA, Y. 1971. Luminous swimming polychaeta from the Banda Islands. Sci. Rep. Yokosuka City
Mus. 18: 34-35.
HORII, N. 1982. Observation on luminous polychaeta, Odontosyllis undecimdonta from Toyama Bay,
Japan Sea. Sci. Rep. Yokosuka City Mus. 29: 1-3.
HUNTSMAN, A. G. 1948. Odontosvllis at Bermuda and lunar periodicity. J. Fish. Res. Board Can. 7:
363-369.
MARK.ERT, R. E., B. J. MARICERT, AND N. J. VERTREES. 1961. Lunar periodicity in spawning and
luminescence in Odontosyllis enopla. Ecology 42: 414-415.
NEUMANN, D. 1976. Entrainment of a semilunar rhythm. Pp. 115-127 in Biological Rhythms in the
Marine Environment, P. J. DeCoursey, ed. University of South Carolina Press, Columbia, South
Carolina.
NEUMANN, D. 1978. Entrainment of a semilunar rhythm by simulated tidal cycles of mechanical dis-
turbance. J. Exp. Mar. Biol. Ecol. 35: 73-85.
POTTS, F. A. 1913. The swarming of Odontosyllis. Proc. Camb. Phil. Soc. 17: 193-200.
Reference: Biol. Bull. 165: 450-457. (October, 1983)
THE KARYOLOGY OF TEREDO UTRICULUS
(GMELIN) (MOLLUSCA, PELECYPODA)
R. VITTURI, A. MAIORCA, AND E. CATALANO
Institute of Zoology, University of Palermo, Italy
ABSTRACT
By counting spermatocyte and oocyte bivalents and mitotic metaphase chro-
mosomes in cleaving eggs, we have determined, both the haploid (n = 19) and the
diploid numbers (2n = 38) respectively, for the species Teredo utriculus. An XY
and XO sex-determining mechanism is absent in the species under study. Chro-
mosomes cannot be grouped into different classes according to length. It seems that,
for Teredo utriculus, a high number of chromosomes is not necessarily accompanied
by a high amount of chromosomal DNA.
INTRODUCTION
The available karyological data on the Pelecypoda (Patterson, 1969; Hinegard-
ner, 1974; Ahmed, 1976; Rasotto et al, 1981), although still very scanty, have
brought to light some interesting cytological problems: 1 ) in many species the male
bivalents break easily and aggregate in groups (Rasotto et al., 1981); thus both the
number and morphology of these chromosomes are quite difficult to determine; 2)
the presence of sex-chromosomes has been hypothesized for two species of the family
Mytilidae: Mytilus californianus (Ahmed and Sparks, 1970) and Mytilus gallopro-
vincialis (Rasotto et al., 1981).
So far as evolution within the Pelecypoda is concerned, Patterson (1969) main-
tains that, as in the other groups of molluscs, the "generalized" species of this class
possess lower chromosome numbers; Hinegardner (1974), on the other hand, asserts
that the families Ostreidae, Pectinidae, Pinnidae, Petricolidae, and Pholadidae, con-
sidered to be more evolved on the basis of their morphological characters, have a
low DNA content.
To clarify these problems we thought it useful to study the chromosomes of a
member of the Teredinidae. This family includes highly specialized species and
belongs to the order Eulamellibranchia cytologically not extensively analyzed. In
fact, only 5 of the 59 recognized families (Grasse, 1960), have been karyologically
studied (Table I).
This paper reports the analysis of male and female bivalents, and of mitotic
chromosomes in cleaving eggs of the species Teredo utriculus (Gmelin).
MATERIALS AND METHODS
For the study of spermatocyte chromosomes 30 sexually mature male specimens
of Teredo utriculus, collected in the Gulf of Palermo, were used. The chromosome
preparations were made using the well-known squashing technique (Colombera,
1970).
Received 29 June 1983; accepted 18 July 1983.
450
KARYOLOGY OF TEREDO
451
TABLE I
Chromosome numbers in the order Eulamellibranchia (Mollusca. Pelecvpoda)
Species name
Authors
Family Unionidae
Unio sp.
Family Cardiidae
Dinocardium robustum
Cardium edule
Cardium tuberculatum
Family Mictridae
Mactra sp.
Labiosa plicatella
Mulinia lateralis
Family Donacidae
Donax variabilis
Family Veneridae
Mercenaria mercenaria
Mercenaria campechiensis
Chione cancellata
Saxidomus giganteus
Saxidomus nuttalli
Venus gal Una
Venus verrucosa
Venerupis aurea
Venerupis decussata
Pitaria chione
16
12
20
20
18
18
18
18
19
19
19
19
19
15
19
19
19
19
Ahmed, 1976
Menzel, 1968
Rasotto et al, 1981
Rasotto et al, 1981
Kostanecki, 1904
Menzel, 1968
Menzel, 1968
Menzel, 1968
Menzel and Menzel, 1965
Menzel and Menzel, 1965
Menzel, 1968
Ahemd and Sparks, 1967
Ahemd and Sparks, 1967
Rasotto et al., 1981
Rasotto et al., 1981
Rasotto et al., 1981
Rasotto et al.. 1981
Rasotto et al.. 1981
Unfertilized eggs of 10 females, eggs immediately after fertilization, and embryos
at the 4-8 blastomere stage, obtained by fertilization in vitro, were treated by the
method used by Colombera ( 1 969) for the chromosome study of the species Botryllus
schlosseri (Ascidiacea).
Observations and microphotographs of the chromosomes were performed with
a Wild-phase contrast microscope.
The idiogram was constructed from photographic enlargements of the chro-
mosomes in 7 late meiotic-II prophase plates, while the karyogram was prepared
from 5 mitotic metaphase plates in embryos at 4-8 blastomere stage.
The mitotic chromosomes were interpreted according to the classification of
Levan et al. (1964).
RESULTS
Meiotic chromosomes
From analyses of spermatocyte bivalents at diakinesis (Fig. la, b), the haploid
number was n = 19 (Table II). The count was not difficult as broken elements were
lacking.
The bivalents appeared well spaced, and intensely and homogenously stained.
In Figure la, the presence of chiasmata allowed different types of bivalents to
be distinguished: cross-shaped with two probable sub-terminal chiasmata, one ring-
shaped element with two terminal chiasmata, and rod-shaped elements in which the
presence and the position of chiasmata could not be hypothesized.
452 R. VITTURI ET AL.
TABLE II
Number of chromosomes found in the plates observed for Teredo utriculus
n
17
18
19
20
21
Spermatocyte bivalents
Oocyte bivalents
frequence
2
2
42
3
1
frequence
2
1
35
Late meiotic-II prophase
chromosomes frequence 2 18
Metaphase mitotic
chromosomes in 2n 36 37 38 39 40
cleaving eggs frequence 3 25 2
The dimensions of these chromosomes varied from a maximum of 2.7 ^m to a
minimum of 1.4 ^m.
At late diakinesis (Fig. Ib) the cross-shaped bivalents were still present. Owing
to the higher contraction of these chromosomes, the dimensions varied from 1.8
^m to 0.9 nm.
The oocyte bivalents at metaphase-I (Fig. 2) appeared well separated on the
squashing plane, thus allowing an easy count (n = 19) (Table II).
In addition to the numerous cross-shaped elements with two sub-terminal chias-
mata, bivalents with one terminal and one sub-terminal chiasma (Fig. 2, see arrows),
and apparently achiasmatic rod-shaped elements were also visible.
The dimensions of these chromosomes varied from 3.4 ^m to 1.8 jim.
In fertilized but uncleaved eggs, 20 plates, interpreted as advanced prophase at
the second meiotic division, were analyzed (Fig. 3). The 19 chromosomes observed
in these spreads (Table II) were rod-shaped, occasionally slightly bent, elongated,
and homogenously stained. A lighter, thinner area, explained as the probable cen-
tromere position, was present in a few elements (Fig. 3, see arrows).
An average idiogram was obtained (Fig. 5) (Table III) by measuring the chro-
mosomes of 7 plates and arranging them by length (Fig. 4, one plate is represented).
Mitotic chromosomes
Mitotic chromosomes at metaphase were observed in embryos at the 4-8 blas-
tomere stage (Fig. 6). The chromosomes displayed different contractions in the
various plates examined, and were arranged randomly on the squashing plane; from
their count the diploid number resulted as 2n = 38 (Table II). In these chromosomes
the kinetochore position could be identified. In fact, in some elements the sister
chromatids of each chromosome were visible, while a thinner area was present in
others.
An average karyotype (Fig. 8) (Table IV) was constructed by measuring and
arranging the chromosomes of 5 plates (Fig. 7, three plates are represented) according
to their length and to the centromere position. From its analysis it resulted that the
38 elements could be grouped into 19 pairs of autosomes, 3 of which were meta-
centric, 2 sub-telocentric, and 14 telocentric.
DISCUSSION
This study has determined the haploid number n = 1 9 and the diploid number
r 38 (Table II) for the species Teredo utriculus. The values which vary slightly
from n = 19 and 2n = 38 are to be attributed to the squashing technique.
KARYOLOGY OF TEREDO
453
1 *
iu.fi
5 6
8
9 10
12 ,3
) 7 )
i mi
14
15
16 17 18
( M
10 jum
10
FIGURE la, b. Diakinetic bivalents in male gonads of Teredo utricitlus.
FIGURE 2. Oocyte bivalents of Teredo utriculus.
FIGURE 3. Late prophase chromosomes at the second meiotic division of Teredo utriculus.
FIGURE 4. Idiogram constructed from 1 late meiotic-II prophase plate of Teredo utriculus.
FIGURE 5. Idiogram constructed from 7 late meiotic-II prophase plates of Teredo utriculus.
454 R. VITTURI ET AL.
TABLE III
Mean length of the chromosomes and S.D. in 7 late meiotic-II prophase plates of Teredo utriculus
Chromosome I 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
Mean length
in microns 6.43 5.78 5.36 4.87 4.67 4.64 4.60 4.54 4.47 4.29 4.22 4.09 3.96 3.67 3.56 3.46 3.28 3.21 2.50
S.D. 1.47 1.28 1.32 1.22 1.22 1.20 1.01 1.22 1.17 1.17 1.10 1.08 1.02 1.03 0.99 0.94 0.87 0.77 0.75
The analysis of male and female bivalents showed no heterotypic element and
heteromorphism is absent in every pair of chromosomes in the karyotype. We there-
fore think that this species does not possess an XY or XO sex-determining mech-
anism.
Differentiated sex chromosomes have not been observed in any of the species
of Pelecypoda cytologically examined up to now (Patterson, 1969; Ahmed, 1976;
Wada, 1978; Rasotto et ai, 1981), apart from the species Mytilus californianus
(Ahmed and Sparks, 1970) and Mytilus galloprovincialis (Rasotto et al, 1981);
however in both cases this assertion was based on the observation of two sper-
matocyte bivalents which seemed to be joined together at diakinesis.
We observed chiasmata at meiosis in both sexes, but it is very unlikely that in
both spermatocyte and oocyte bivalents, all the chiasmata present were counted,
due to their terminalization and the overcondensation of these chromosomes.
Furthermore, comparison of the male and female bivalents revealed the greater
dimensions of the latter.
The mitotic chromosomes at metaphase appear to be arranged randomly on the
squashing plane, thus excluding somatic pairing of homologous chromosomes for
the species under study. However, these chromosomes appear to be peculiar for
their shape, which brings to mind the "colchicinized" chromosomes (leyama and
Inaba, 1974; leyama, 1975; Wada, 1978).
This characteristic, previously observed in mitotic chromosomes of spermato-
gonial metaphase of some Gastropods (Vitturi et al., 1982), has not been confirmed
in the Polyplacofora (Vitturi, 1982; Vitturi et ai, 1982).
The chromosomes in the idiogram and karyogram cannot be grouped into classes
according to length since their dimensions vary gradually from the largest to the
smallest (Tables III, IV).
If we consider the number of chromosomes, the haploid value n = 19, which
characterizes the species Teredo utriculus, is found to be one of the highest, not only
within the order Eulamellibranchia (Table I), but also within the class Pelecypoda
(Rasotto et al., 1981).
If, in agreement with Ahmed (1976), the basic haploid number for this class is
considered to be n = 15, or a value close to that, then it seems probable that
evolution within this group has proceeded not only with a decrease (Ahmed, 1976;
Vitturi et al., 1982) but also with an increase in the number of chromosomes (Pat-
terson, 1969).
Finally, it is interesting to note that many species of the family Pectinidae possess
19 spermatocyte bivalents of greater dimensions (Rasotto et al, 1981) than those
of the male bivalents in the species analyzed here.
The finding of a low nuclear DNA content in these species (Hinegardner, 1974)
leads to the supposition that there is a low DNA content in Teredo utriculus as well.
All this would indicate that specialization is, in this particular case, linked to
a decrease in the chromosomal DNA content, thus supporting Hinegardner's hy-
KARYOLOGY OF TEREDO
455
I
• it -
*•» *
J» *
•
t % %
* • 1 |/ % r
» m
t %^
10
1 9
7 1 2
3456789
M 1
< >i l| II II II II "
a
10 11
12 13 14 15 16 17 18 19
1 '
? < 3 4 5 6 789
)t l
1 il 11 II II l| IA l|
b
10 11
12 13 14 15 16 17 18 19
i& 4;
, II il It «t M tf •• •»
f ' 2
34567 89
t * 4 j
i till II fill till
^
cio n
12 13 14 15 16 17 18 19
II i|
M II II •••••• It ••
8 ,
t 10 Aim
IIinililMini
5 4im
FIGURE 6. Mitotic metaphase plate in cleaving eggs of Teredo utriculus.
FIGURE 7. Three arrangements of mitotic metaphase chromosomes in cleaving eggs of Teredo
utriculus.
FIGURE 8. Karyogram constructed from 5 mitotic metaphase plates in cleaving eggs of Teredo
utriculus.
456 R. VITTURI ET AL.
TABLE IV
Mean length and arm ratio of the chromosomes of 5 mi l otic metaphase plates
in cleaving eggs oj "Teredo utriculus
Chromosome Mean length
pairs in microns
S.D.
Arm ratio
mean
Centromere
position
1 2.86
2 2.59
3 2.13
4 2.09
0.53
0.33
0.22
0.28
1
5.3
3.7
8.2
M
ST
ST
T
5 2.09
0.28
00
T
6
.93
0.21
00
T
7
.86
0.15
00
T
8
.82
0.12
00
T
Q
.82
0.12
00
T
10
.77
0.15
1
M
11
.72
0.19
1.7
M
12
.68
0.22
00
T
13
.66
0.18
00
T
14
.57
0.22
00
T
15
.43
0.15
00
T
16
.37
0.13
00
T
17
.29
0.15
00
T
18
.22
0.18
00
T
19
.11
0.25
00
T
pothesis that such a mechanism is present in all classes belonging to the phylum
Mollusca.
At any rate, as has already been suggested for the family Petricolidae (Pelecypoda)
(Rasotto et al, 1981), for the Polyplacofora (Vitturi, 1982) and for the Mesogas-
tropoda (Mollusca, Prosobranchia) (Vitturi and Catalano, in press) it appears that,
for Teredo utriculus as well, a high number of chromosomes is not necessarily
accompanied by a high amount of chromosomal DNA.
ACKNOWLEDGM ENTS
The authors are deeply indebted to Mr. G. Miceli for processing and printing the
microphotographs presented here.
LITERATURE CITED
AHMED, M. 1976. Chromosome cytology of marine Pelecypod Mollusc. J. Sci. (Karachi) 4: 77-94.
AHMED, M., AND A. K. SPARKS. 1967. Proc. Nat. Shellfish. Assoc. 58: 10.
AHMED, M., AND A. K. SPARKS. 1970. Chromosome number structure and autosomal polymorphism
in the marine mussels: Mytilus edulis and Mytilus californianus. Biol. Bull. 138: 1-13.
COLOMBERA, D. 1969. The karyology of the colonial Ascidian Botryllus schlosseri (Pallas). Caryologia
22: 339-350.
COLOMBERA, D. 1970. A squash method for chromosomes of Ascidians (Tunicata). Caryologia 23: 1 13-
116.
GRASSE, P. P. 1960. Traite de Zoologie, Anatomic, Systematique, Biologic. Tome V. Masson et C'e
Editeurs. Paris. 2.087-2.127.
HINEGARDNER, R. 1974. Cellular DNA content of the Mollusca. Comp. Biochem. Physiol. 47A: 447-
460.
IEYAMA, H. 1975. Chromosome numbers of three species in three families of Pteriomorphia (Bivalvia).
Venus. Jpn. J. Mai 34(1): 26-32.
IEYAMA, H., AND A. INABA. 1974. Chromosome numbers often species in four families pf Pteriomorphia
(Bivalvia). Venus. Jpn. J. Mai. 33(3): 129-137.
KARYOLOGY OF TEREDO 457
KOSTANECKJ, K. 1904. Cytologische studien an Kunstlich parthenogenetisch entwickeluden eies von
Mactra. Arch. Mikroskopische Anal. 64: 1-98.
LEVAN, A., K. FREDGA, AND A. A., SANDBERG. 1964. Nomenclature for centromeric position of chro-
mosomes. Hereditas 52: 201.
MENZEL, R. W. 1968. Chromosome number in 9 families of marine Pelecypod mollusc. Nautilus 82(2):
45-58.
MENZEL, R. W., AND M. Y. MENZEL. 1965. Chromosomes of two species of quahog clams and their
hybrids. Biol. Bull. 129(1): 181-188.
PATTERSON, C. M. 1969. Chromosomes of molluscs. Proc. Symp. Moll. Mar. Biol. Assoc. India 2: 635-
686.
RASOTTO, M., D. ALTIERI, AND D. COLOMBERA. 1981. I cromosomi spermatocitari di 16 specie ap-
partenenti alia classe Pelecypoda. Comunicazionepresentata al Simposio della S.M.I. (Soc. Mai.
Ital.)del 9.10/5/1981.
VITTURI, R. 1982. The chromosomes of Chiton olivaceus (Spengler) (Polyplacophora) Biol. Zbl. 101:
647-651.
VITTURI, R., M. B. RASOTTO, AND N. FARINELLA-FERRUZZA. 1982. The chromosomes of 16 molluscan
species. Boll. Zoo/. 49: 61-71.
VITTURI, R., AND E. CATALANO. Spermatocyte chromosomes in 7 species of the sub-class Prosobranchia
(Mollusca, Gastropoda). Biol. Zbl. (in press).
WADA, K. 1978. Chromosome karyotypes of three bivalves: the oysters, Isognomon alatus and Pinctada
imbricata, and the bay scallop, Argopecten irradians. Biol. Bull. 155: 235-245.
Reference: Biol. Bull. 165: 458-472. (October, 1983)
GAMETOGENESIS AND REPRODUCTIVE PERIODICITY OF THE
SUBTIDAL SEA ANEMONE URTICINA LOFOTENSIS
(COELENTERATA: ACTINIARIA) IN CALIFORNIA
STEVEN E. WEDI* AND DAPHNE FAUTIN DUNN**
^Department of Biology, San Francisco State University, San Francisco, California 94123, and
** Department of Invertebrate Zoology, California Academy oj Sciences, Golden Gate Park,
San Francisco, California 94118-9961
ABSTRACT
Sexual reproduction of the actiniid sea anemone Urticina (= Tealia) lofotensis
was studied for one year (1976-1977) in 105 specimens collected by hand monthly
at 7-16 m in Carmel Bay, California. Gametogenesis, evaluated by light microscopy,
is typical for an actinian. Oocyte maturation is asynchronous, even within a mes-
entery, whereas spermiogenesis of each male is synchronous. Each oocyte is asso-
ciated with a trophonema, and eggs may exceed 1200 pm in diameter. The study
population is dioecious, with a significant excess of females. Gonad indices and
histological data indicate that the period of maximum female ripeness ends in De-
cember as the male maximum begins. The spawning peak appears to occur then,
just as water temperature begins to fall from its annual high. Some females contain
large oocytes and seem to release eggs throughout the year. Greatest reproductive
quiescence is in April and May, when water temperature is at its minimum.
INTRODUCTION
Most studies of sea anemone reproduction have dealt with specimens collected
intertidally; subtidal studies have relied on dredged material. Although reproductive
studies on other subtidal coelenterates have been done with the aid of diving (e.g.,
Ostarello, 1973; Rinkevich and Loya, 1979), ours is the first published study of
subtidal sea anemone reproduction based on hand-collected specimens. It therefore
adds a new dimension to the growing body of research on sexual reproduction of
Pacific North American actinians begun two decades ago (e.g., Ford, 1964; Spauld-
ing, 1971; Siebert, 1974; Dunn, 1975; Siebert and Spaulding, 1976; Jennison, 1978,
1979;Sebens, 1981).
Urticina lofotensis (Danielssen, 1890) is a vivid crimson actiniid sea anemone
with white verrucae 1-3 mm in diameter that make it appear polka-dotted or, in
contraction, vertically striped (Fig. 1). Along the U. S. Pacific coast, its habitat is
almost exclusively rocky subtidal. Only rare animals are exposed by minus tides,
which probably accounts for the lack of biological information about the species.
One of five named members of Urticina in the northeastern Pacific (the others are
U. colurnbiana, U. coriacea, U. crassicornis, and U. piscivora), U. lofotensis ranges
from Alaska to the Channel Islands (Hand, 1955; Sebens and Laakso, 1978).
In using the name Tealia lofotensis for this organism, Hand (1955) identified
it with that which Danielssen ( 1 890) described from Norway as Madoniactis lofo-
Received 29 November 1982; accepted 1 July 1983.
* Current address 1444 Laguna Avenue, Burlingame, California 94010.
**To whom reprint requests should be addressed.
458
REPRODUCTION OF URTICINA LOFOTENSIS
459
FIGURE 1. Typical posture of Urticina lofotensis on rock substratum. Specimen is approximately
100 mm across.
tensis. The name Urticina has priority over Tealia, which is, in turn, senior to
Madoniactis (Williams in Manual, 1981). Manual (1981), following Stephenson
(1935), synonymized the European T. lofotensis with Bolocera eques Gosse, 1860,
which has been known as T. crassicornis, calling it U. eques. Manual (1981) ques-
tionably included Hand's T. lofotensis in the synonymy as well. It seems prudent
to maintain current usage of Urticina species names for animals of the north Pacific
pending further systematic study since anemones called U. lofotensis and U. cras-
sicornis are easily separable on the Pacific coast of North America (Hand, 1955;
Sebens and Laakso, 1978), and both seem to differ from the European U. eques as
per Stephenson (1935) and Manual (1981).
Stephenson (1935) and Manual (1981) summarized literature on, and mor-
phology of, European animals called Urticina lofotensis. Hand (1955) and Sebens
and Laakso (1978) described the anatomy of northeast Pacific anemones of the same
name. Data on reproduction are confined to remarks on size of gametes and dis-
tribution of gonads.
MATERIALS AND METHODS
Between 4 and 12 anemones were collected at four-week intervals from 9 No-
vember 1976, to 12 October 1977, by SCUBA diving from a boat off the rocky
north end of Carmel River State Beach, California (36°32'25"N, 121°55'53"W). The
bottom is characterized by rocky rubble interspersed with small sandy areas and
granitic boulders, most 3-5 m in diameter, some 7 m tall and rising to within 3 m
of the surface. Passages between boulders are subject to surge and scour, especially
in winter, due to ocean swells coming directly from deep water offshore. The study
area, approximately 200 by 100 m (Fig. 2), ranged in depth from 7 to 16m. Ane-
mones are scarce in water deeper than 16 m, and strong wave action and surge
made collecting in water shallower than 7 m difficult or impossible during most of
the year. An extension of the Carmel submarine canyon near the study area influ-
ences wave action, upwelling, and temperature fluctuation.
A different portion of the study area was sampled each month by two divers
swimming along a selected compass heading, arbitrarily removing anemones from
460
S. E. WEDI AND D. F. DUNN
Carmel
Riy«r
•ach
FIGURE 2. Map of southern Carmel Bay, California, indicating location of the study area. Depth
contours in meters.
the substratum. Measuring pedal disc diameter prior to collecting was attempted
but proved difficult in the surge, so was discontinued. Animals were carefully scraped
and peeled from the rocks using a dull knife blade. To prevent contact among
animals and to keep track of oocytes released after collection, each animal was
placed in a separate perforated plastic jar covered by a plastic screw cap. The con-
tainers, of minimal buoyancy, could easily be transported by the diver in a nylon
mesh bag. They were returned to the laboratory in a styrofoam cooler filled with
sea water. Water samples from the cooler were examined microscopically after re-
moving the jars. No eggs were found, but much undigested food expelled by the
anemones during transit was always evident.
In the laboratory, the animals (in jars) were placed in running sea water. The
next morning, after they had expanded fully in liter beakers of sea water, half the
fluid was replaced by 100-200 ml of 10% MgCl2 in sea water. Complete relaxation,
until a pinch on several of the by-then flaccid tentacles elicited no response and the
REPRODUCTION OF URTICINA LOFOTENSIS 461
oral disc was expanded and darkened, required several hours. If narcotization was
slow, additional relaxant was added; this was most often necessary with specimens
over 70 mm basal diameter.
Anemones were fixed in Bouin's solution (Humason, 1962) made with undiluted
sea water. Despite seemingly thorough relaxation, many contracted somewhat when
fixative was added. About 10% contracted violently, everting the actinopharynx,
which made dissection difficult. At least a week in Bouin's was allowed for complete
fixation.
Prior to dissection, pedal disc diameter was measured. The anemones were bi-
sected across the column, 10-20 mm distal to the base. Food objects and gonadal
tissue were removed with forceps under a dissecting microscope. Gonads were pre-
served in Bouin's solution. Mesenteries were counted in the basal section.
Each animal, minus its gonads, was dried for six days in a vacuum desiccator
at 60°C, and weighed immediately upon removal. Dry weight of gonad not set aside
for histological examination was determined after desiccation for 24 h. Four large
blotted pieces of gonad from each anemone were weighed prior to dehydration,
cleared, and embedded in paraffin. Their approximate dry weight added to that of
the desiccated pieces yielded the total dry gonad weight. The relation of dried gonad
weight to that of the entire animal, encompassing gonad as well as body, constituted
the gonad index (GI).
Seven /tin serial sections of gonad were stained with Hams' hematoxylin and
eosin (Humason, 1962). Fifty oocytes from each anemone were measured in sections
that included the nucleolus, which reduced the possibility of measuring the same
cell more than once. The two longest perpendicular diameters were averaged in
irregularly shaped oocytes. Eggs smaller than 25 nm were difficult to measure ac-
curately, so were not included in the count.
Maturity of male gonads was scored as follows: stage 1 — gonadal packets con-
taining only spermatogonia; stage 2 — packets with spermatogonia, spermatocytes,
and the first noticeable tailed sperm; stage 3 — fully mature packets containing pre-
dominantly sperm. Animals with follicles at a maturity level between stages 1 and
2 were placed subjectively in one or the other; those with packets between stages
2 and 3 were classified according to the relative abundance of sperm. For example,
a male with packets half full of mature sperm was at stage 2'/2.
Surface water temperatures were obtained from the California Department of
Fish and Game's Marine Culture Laboratory at Granite Canyon, south of the study
area. Water temperatures taken at depth on several collecting dives during the year
generally agreed with the data from Granite Canyon.
RESULTS
Sexuality and morphology
Urticina lofotensis is dioecious: 54 females, 34 males, 1 7 animals lacking gonads,
and no hermaphrodites were collected. Sex determination was not possible exter-
nally. Even under low magnification, immature gonads of both sexes appeared sim-
ilar, but at later developmental stages were distinguishable by color and form.
Male gonads were bright red, the color dulling considerably after fixation. The
greatly elongated, pleated gametogenic portion along the inner mesentery edge (Fig.
3a) was easily located and removed during dissection.
Female gonadal tissue was less convoluted, the oocytes were contained within
indistinct clusters along the mesentery edge. Mature clusters resembled bunches of
grapes. In Bouin's fluid this tissue was generally yellow or brown, and loose eggs
462
S. E. WEDI AND D. F. DUNN
l>' *rf '-Mr 3s*!*1
1?^* ^{
>. ..Vfy- "•£*
. «s^si;M«
:-:^
Uv Sfer^^i
*" * " i -*•*'. frSa^l35!- P**" '*A^r -'r&riV ^
v%i «!^ m »s,^ i M, ^
w|
e
m
FIGURE 3. a) Section through gametogenic portion of one mesentery from male anemone. Scale
bar = 100 nm. b) Section through early gonadal packets with spermatogonia (stage 1). Scale bar = 30
Mm. c) Section of stage 2 gonadal packet with numerous tailed gametes. Note layering. Arrow indicates
plug-like structure. Scale bar = 30 Mm. d) Mature gonadal packets (stage 2'A) with abundant immature
gametes. Scale bar = 50 pm. e = endoderm; g = gastrovascular cavity; gp = gonadal packet; m = mesoglea;
s = spermatozoa; sc = spermatocytes; sg = spermatogonia; t = sperm tail.
were yellow. Sometimes oocytes/ova were expelled during fixation and several fe-
males were collected with eggs among the tentacles and adhering to the oral disc;
diameter of these gametes was 700-800 /um. Spawning was never observed, and no
larvae were found in or on any anemone.
The number of mesenteries in Urticina lofotensis corresponds to the number of
tentacles and is the same distally and proximally. In 26 anemones of all sizes, it
ranged from 47 to 77 pairs, and did not correlate strictly with animal size as de-
termined by pedal disc diameter or dry weight. Generally, however, larger animals
had more mesenteries. Many weighing from 12 to 15 g had just over 50 pairs,
although a female with 77 pairs weighed only 8 g.
Oogenesis
The most immature germ cells observed were in the endoderm, ranged from 10
to 30 ^m, and contained a nucleus about half their diameter (Fig. 4a). Large con-
centrations of cells occurred near the junctions of germinal and non-germinal mes-
entery tissue, but some were scattered in the endoderm, many near mature oocytes
(Fig. 4b).
The smallest oocytes in the mesoglea were 20-50 /im in diameter. Previtellogenic
cells stained a characteristic deep blue with hematoxylin and eosin; yolk platelets
REPRODUCTION OF URTICINA LOFOTENSIS
463
&*«tfi3
fiff&SA
-*/\<
©•i oc . ^
^M5&W<
BSGSi&'Wt '•"!!•• »• - *'^-A'.^S
S!*>nv
>j
^%^fe%i
.-•A. ,,, -'•
m
FIGURE 4. a) Section of early female gamete in the mesenterial endoderm. Scale bar = 20 nm. b)
Section through a female gametogenic mesentery containing gametes in many stages of development.
Scale bar = 100 nm. c) Section of an oocyte with a trophonema. Scale bar = 50 ^m. d) Section through
a large, yolky oocyte with spines. Scale bar = 50 ^m. e = endoderm; g = gastrovascular cavity; m
= mesoglea; n = nucleus, ne = nucleolus; oc = oocyte; og = immature germ cell; sp = spines; tr
= trophonema; y = yolk granules.
took up eosin predominantly, giving larger cells a distinct pink color. During vi-
tellogenesis, the oocyte nucleus (germinal vesicle) moved peripherally to either side
of the cell, adjacent to the mesenterial endoderm, and did not increase appreciably
in size. Nuclei of larger cells therefore appeared relatively small. Oocytes of all sizes
contained one darkly stained, round nucleolus 10-20 pm in diameter.
In-oocytes undergoing vitellogenesis and some previtellogenic cells, a tubular
trophonema connected the cell through the mesoglea and endoderm to the gastro-
vascular cavity (Fig. 4c), its end flaring where it joined the oocyte. Attachment to
the gamete was always in proximity to the germinal vesicle. Trophonemata were
less prevalent in larger oocytes, but their remnants — small pieces of tissue adjacent
to the nucleus — were common. Spines 5-15 nm long covered the surface of most
larger oocytes. They were especially apparent where the mesoglea had pulled away
from the oocyte during fixation (Fig. 4d). Each oocyte within the mesentery had a
germinal vesicle.
Spermiogenesis
All mesenteries of an individual contained sperm follicles of uniform maturity.
Spermatogonia were not identifiable in the endoderm. The smallest sperm packets
in the mesoglea were round to ovoid 20-50 /*m across, and contained up to 30
spermatogonia, each approximately 3-5 nm in diameter, with an indistinct nucleus
half or less the diameter of the cell (Fig. 3b).
464
S. E. WEDI AND D. F. DUNN
Spermatogonia lined the periphery of the growing follicle while smaller sper-
matocytes (2-3 /urn) occurred centrally, layering becoming pronounced with in-
creasing numbers of cells. At a later stage, spermatids ( 1 nm diameter), in clumps
of four to eight cells, occupied the packet's center. By this stage the follicle was 33-
50% the width of the mesentery. Shortly thereafter the lumen of the packet opened
slightly, and tailed sperm with heads approximately 1 nm in diameter became ev-
ident (Fig. 3c).
Mature follicles expanded to nearly the full width of the mesentery (Fig. 3d).
They were lined with developing gametes, spermatogonia and spermatocytes at the
periphery, spermatids more centrally, and spermatozoa bundled with their tails
together in the lumen. A few mature follicles occurred in spawned-out males, sug-
gesting that all sperm are not always shed. Some spawned-out males also had im-
mature spermatogonial packets.
Gonad Cycles
Figure 5 indicates the relative size frequencies of oocytes measured in section
from the 54 female anemones collected. Although smaller oocytes were dispropor-
tionately represented due to their relatively large nuclei, changes in average gamete
size through the year were evident. Cells between 50 and 1 50 ^m predominated in
all animals, and very large oocytes (450-600 /um) were also present all year, although
in much smaller and varying quantities. Small oocytes made up a large percentage
of gametes during winter (November to February), and on into spring, but repre-
sented a much smaller proportion during later spring and summer. Oocytes in the
size classes of greatest frequency during winter averaged just under 100 nm. During
summer (June through September), smaller oocytes decreased in frequency but had
begun to increase again by October. Small quantities of large oocytes (350-500 ^m)
were present in November and especially December. By January, most had disap-
peared.
Proportions of large (pink-staining) oocytes in the 50 measured gametes are
displayed in Figure 6, which confirms their relatively high frequencies in some
anemones during November and December, and their generally low prevalence in
January and April. The increase from May through September is more evident in
Figure 6 than in Figure 5. Animals collected during September contained the greatest
20
1 600
2 500
t-
5 400
O 300
" 200
o
o
100
600
500
400
300
200
100
N
n= 4
D
5
1976
M
A
3
M
5
A
2
O
5
1977
FIGURE 5. Size-frequency polygons for diameters of 50 oocytes from each female specimen of
Urticina lofotensis. Each polygon indicates cumulative size frequencies for that month. One of the six
females collected in December and one of the seven collected in January contained only loose oocytes.
REPRODUCTION OF URTICINA LOFOTEN SIS
465
*S 70
-70
IL 60
60
o 12
>• H 50
-,
1
50
O >•
Z O 40
" -.
40
111 O
C
•
2 O 30
o
O
•
r
•
r
30
m U
r
-
u
•
•
Jr O 20
-i —
-
-
20
IL K
_ "
Tn r
"3
r
< 10
-, r
w
r
10
-" 5
1
Jl
0
c
I
N 0
i F M A M J J
A
s ' o
n: 4 5
65 3563265
1 976
1977
FIGURE 6. Histograms showing the proportion of large oocytes among the 50 cells measured from
each female specimen of Urticina lofotensis. Each histogram represents an animal. One of the females
collected in February and one collected in September contained only small oocytes.
proportion of large oocytes. Percentage of large cells in most females had dropped
considerably by October.
Maximum female GI was 16.4% in a 10.0 g animal collected in November. The
largest female, from the June collection, had a weight of 17.3 g and a GI of 15.1%;
the smallest female, taken in October, weighed 4.4 g and had a GI of 3.4%. Average
female GI gradually declined from November to its nadir in May, generally par-
alleling the pattern in males and surface water temperature (Fig. 7). Although it had
just begun to increase (Fig. 6), large oocyte frequency was also low in May, when
oocytes less than 200 ^m in average diameter were predominant (Fig. 5).
The gonad cycle of males is shown in Figures 7 and 8. In April, when the largest
male, weighing 19.3 g (GI 3.7%), was collected, and in May, when the smallest,
weighing 4. 1 g (GI 0.7%) was collected, gonad indices were low and males contained
only spermatogonia. The first sperm had developed by June when, as with females,
GI abruptly increased. From an August low (based on one male with 47 pairs of
mesenteries, several residual sperm packets in its immature gonads, and a GI of
1 .6%) almost equal to that of May, male GI increased through October, with a
3
Z
O
O
16
14
12
10
8
6
4
2
0
N
4
2
15
10
D
s
s
1976
7
3
1977
F M A M J J A SO
8 356327$ females
3 234(123 males
FIGURE 7. Monthly average gonad indices of female (--- O ---) and male (--- X ---) specimens of
Urticina lofotensis, with number of specimens indicated below. Surface water temperatures ( ) are
biweekly averages. No collection was made in March.
466
S. E. WEDI AND D. F. DUNN
S
Q.
o
Q
ik
O
111
o
<
r 2
3
-, r
2
C
0
u
1
^J r
-| r
-n
T-
-
O
u
0
C
N D
JFMAMJJASO
n= 2 S
33 2346123
1976
1977
FIGURE 8. Histograms showing maturity of male specimens of Urticina lofotensis through the year.
Each histogram represents an animal. See text for explanation.
concomitant sperm buildup. In September and October, gonadal packets were half
full of sperm. Males collected in winter had follicles filled with sperm. The highest
average monthly GI was in December (14.1%) when the five sea anemones had
predominantly stage 3 gonadal packets. Among them was the second largest male
collected (18.7 g), which had the highest individual GI, 22.7%. Although mature
follicles predominated through February, average male GI decreased from December
through May. GI generally increased with anemone size for both sexes (Table I).
Anemones with no gonads visible during dissection were not sectioned. All but
two of these animals had dry weights less than 7.6 g (Table I), indicating that they
were probably juveniles. The other two, both collected in February, were 8.0 and
13.5 g.
Natural history notes
Approximately 20% of dissected anemones contained shells, both empty and
with animals, of the small (10-20 mm) gastropod Calliostoma foliatum. Most con-
TABLE I
Sex and gonad indices (+/- standard deviations) o/Urticina lofotensis by weight class
Anemone
Number
of individuals
Average
gonad index
dry weight
(g)
Female
Male
Sterile
Female
Male
<3.9
0
0
2
4-5.9
4
3
5
3.6+/-3.0
1.4+/- 1.0
6-7.9
5
6
8
4.4+/-2.1
3.5+/-3.0
8-9.9
14
5
1
5.S+/-4.4
6.9+7-5.3
10-11.9
13
13
0
V.7+/-4.9
5.7+/-5.1
12-13.9
9
3
1
5.5+7-3.2
7.3+/-2.S
14-15.9
6
2
0
9.7+/-2.5
8.8+/-1.2
16-17.9
3
0
0
11.4+/-3.4
—
18-19.9
0
2
0
—
13.2+/-13.4
54
34
17
Anemone dry weight includes body and gonad.
REPRODUCTION OF URTICINA LOFOTENSIS 467
tained one shell, but a few had up to four. Other ingested objects included uniden-
tified gastropod shells, crustacean body parts, bryozoans, pieces of algae, one 1 5 mm
specimen of Corynactis californica, a 30 mm feather, a ctenophore 80 mm long, a
35 mm bat star (Patina miniatd), and a 100 X 50 mm flat abalone shell (Haliotis
wallalensis) that was lodged across the actinopharynx of an anemone with basal
diameter 60 mm.
Despite thorough searches of algal holdfasts, cracks, and caves with a diving
light during many dives in the study area and elsewhere in Carmel Bay, no anemones
less than 30 mm basal diameter were found. It is possible that very small animals
were overlooked because of low numbers, being covered with debris (large animals
often have material attached to their verrucae), or being hidden under algae.
DISCUSSION
Sexuality and morphology
Distribution of gametogenic mesenteries in Urticina lofotensis is characteristic
of the genus, the first ten pairs, including the directives, being sterile (Hand, 1955).
All other mesenteries may be, but are not necessarily, gametogenic. The maximum
of 77 pairs correlates well with Hand's (1955) data, but Sebens and Laakso (1978)
reported considerably more. The regular arrangement of mesenteries implies that
asexual reproduction does not occur in this species. [J. Brumbaugh (pers. comm.,
1982) observed an anemone of this species divide longitudinally in an aquarium at
Sonoma State University.]
Space for gonads should increase in larger anemones, and more mesenteries
should enhance fecundity. However, there was little correlation between amount
of gonadal tissue and number of fertile mesenteries in an animal. In fact, large
actinians with moderate numbers of mesenteries produced the greatest quantity of
gonad. Several anemones lacking visible gonads had more mesenteries than some
very fertile ones (a sterile 3 g individual had 54 pairs, as many or more than many
fertile animals weighing up to 9 g). Anemones add mesenteries as they grow, typically
to a species-specific maximum. They grow only if fed, though, and may shrink if
starved (Chia and Spaulding, 1972), so size, mesentery number, and age are not
necessarily interrelated. Sebens ( 198 1 ) found gonad as a percentage of body volume
to increase with gonad number which, in turn, increases with body size in Antho-
pleura xanthogrammica and A. elegantissima, the rise being more rapid in smaller
than larger anemones. GI of the sea urchin Strongylocentrotus purpuratus increases
with test diameter in small animals but not in large ones, despite internal space
expanding isometrically with size. Metabolic factors seem to be responsible for this
(Conor, 1972).
Laboratory raised Urticina crassicornis 40 mm in diameter are 18 months old
(Chia and Spaulding, 1972). Assuming a roughly comparable growth rate for U.
lofotensis in the field, the smallest anemones observed during this study are at least
a year old, and the smallest fertile ones at least a year and a half old.
Associated with an oocyte undergoing vitellogenesis is a trophonema. Recent
experimental evidence (Larkman and Carter, 1982) substantiated speculation (Ny-
holm, 1943; Loseva, 1971; Dunn, 1975) that this tube functions in nutrient transfer
from the gastrovascular cavity to the developing egg. It may also act as a channel
for egg release (Carter, pers. comm.). Trophonemata have been found, although not
always identified as such, in a cerianthid and many sea anemones belonging to
several families, but seem to be absent in other actinians (e.g., Nyholm, 1 943; Loseva,
468 S. E. WEDI AND D. F. DUNN
1971; Dunn, 1975, 1982; Riemann-Zurneck, 1976; Jennison, 1979, 1 98 1 ; Larkman
and Carter, 1982).
At the mesentery edge, some sperm packets have a plug-like structure (Fig. 3c)
that may be homologous with a trophonema. On the other hand, the convoluted
gametogenic mesentery of males has a large surface area that may facilitate nutrient
transfer from the gastrovascular cavity through the thin layers of endoderm and
mesoglea surrounding the gametes. Gamete release is probably facilitated for both
sexes by proximity to the mesentery edge.
Sex ratio of Urticina lofotensis is significantly different from 1:1 (chi square
= 4.54; 0.025 < P < 0.05). Although it is remotely possible that sampling error is
responsible, or that most of the 1 7 sterile individuals were male, the preponderance
of females is probably real. Such an excess is known in a variety of temperate and
tropical actinians (Dunn, 1982).
Oogenesis
Oogonia originate in the endoderm of anthozoan mesenteries. Dunn (1975) and
Jennison (1979) reported that after migrating into the mesentery's central mesogleal
layer, secondary oogonia cease mitosis and become oocytes. However, Loseva (1971)
failed to locate oogonia in Urticina crassicornis, and the smallest female germ cells
that Larkman (1981) identified in Actinia fragacea endoderm were oocytes.
Eggs of U. lofotensis grow to 700-800 /im (preserved diameter) before being
spawned. Dunn (1975) estimated that ova from Epiactis prolifera fixed in Bouin's
solution were approximately 65% of their actual diameter. Thus, oocytes of U.
lofotensis may actually exceed 1 200 pm in diameter. Ova in other actinians range
from 70 pm (Gonactinia prolifera; Gemmill, 1921) and 1 10 pm (Bunodosoma cav-
ernata; Clark and Dewel, 1974), to 750-800 /xm (Stomphia didemon; Siebert, 1973)
and 1 100 nm (Bolocera tuediae; Gemmill, 1921). Eggs of U. coriacea are reportedly
600 jim in diameter (Gemmill, 1921) and those of U. crassicornis up to 700 pm
(Gemmill, 1921; Chia and Spaulding, 1972).
Germinal vesicles of Urticina crassicornis, Epiactis prolifera, and Anthopleura
elegantissima are aligned on either side of the cell, as in U. lofotensis (Loseva, 1971;
Dunn, 1975; Jennison, 1979). In Actinia equina, by contrast, those of all oocytes
within each mesentery are arrayed on the same side (Chia and Rostron, 1970). The
trophonema abuts an egg adjacent to its nucleus, suggesting that one may influence
the position of the other. Germinal vesicles ofPeachia quinquecapitata, which seems
to lack trophonemata, are randomly oriented (Spaulding, 1974). Staining of cyto-
plasm in a large primary oocyte indicated protein synthesis and high concentrations
of RNA around the germinal vesicle (Dybas, 1973). Presumably the subunits for
these compounds reach the egg through the trophonema, as do other precursors
(Larkman and Carter, 1982). Large oocytes contain evenly distributed eosinophilic
yolk granules. The same is true of U. crassicornis (see Loseva, 1971).
In Actinia equina, oogenesis is synchronous within, but out of phase between
mesenteries (Chia and Rostron, 1970), whereas Urticina crassicornis, Peachia quin-
quecapitata, and Anthopleura elegantissima resemble U. lofotensis in being asyn-
chronous within mesenteries (Loseva, 1971; Spaulding, 1974; Jennison, 1979). As
in U. lofotensis, male gametes ofActinostola crassicornis ripen synchronously within
but not between individuals (Riemann-Zurneck, 1978). Gametes of both sexes in
all developmental stages occur in the same mesentery of hermaphroditic individuals
of Epiactis prolifera (see Dunn, 1975). Heterogeneity of gamete size is known in
such other marine invertebrates as hydrozoans (Kessel, 1968) and echinoids (Hol-
land, 1967; Gonor, 1973b). Spawned gonads of A. elegantissima contain residual
oogonia and previtellogenic oocytes that Jennison (1979) suggested either are pre-
REPRODUCTION OF URTICINA LOFOTENSIS 469
vented from maturing or comprise the first gametes of the next reproductive period.
Such hypotheses probably apply as well to U. lofotensis.
Spines reportedly range from 10 to 25 pm long in other actinians (Chia and
Spaulding, 1972; Spaulding, 1972, 1974; Siebert, 1973, 1974; Siebert and Spaulding,
1976; Jennison, 1979). Dunn (1975) suggested that surficial structures 1.5-4 ^m
long on oocytes of Epiactis prolifera may be fixation artifacts. Loseva ( 197 1 ) thought
that spines on Urticina crassicornis oocytes might function in nutrient absorption
from the mesoglea, while Siebert (1973) proposed that spines prevent polyspermy.
Oocytes are apparently released with the intact germinal vesicle containing a
single nucleolus, so final maturation divisions must occur during or after spawning,
perhaps even after fertilization, which must be external. Eggs of Urticina crassicornis
mature before being spawned (Chia and Spaulding, 1972).
Sperm iogenesis
Development of spermatogonia, which also arise in mesenterial endoderm, is
like that in other anthozoans (Chia and Rostron, 1970; Chia and Crawford, 1973;
Clark and Dewel, 1974; Dunn, 1975; Jennison, 1979). Discrimination of later sper-
miogenic stages is facilitated by layering of the gametes. Spermatids and the first
spermatozoa can hardly be identified individually. In an ultrastructural study of the
sea anemone Bunodosoma cavernata, Dewel and Clark (1972) reported that sper-
matocytes already possess a flagellum, making it difficult to distinguish between the
latter stages of spermiogenesis, a problem Jennison (1979) also had in a light mi-
croscopic study of Anthopleura elegantissima. The germinal portion of the mesentery
is resorbed after spawning of A. elegantissima, destroying the primary germ cells
that had occupied the mature follicle's periphery (Jennison, 1979). The same may
happen in Urticina lofotensis.
The 1 Aim sperm heads of fixed Urticina lofotensis are similar in size to those
of many other actinians (e.g., Chia and Rostron, 1970; Dunn, 1975; Jennison, 1979,
1981), but smaller than some (Frank and Bleakney, 1976). Live spermatozoa of U.
crassicornis have heads 1.5 X 2.0 nm (Chia and Spaulding, 1972), while those of
Peachia quinquecapitata are 5.5 X 6.5 nm (Spaulding, 1972) and those of two species
of Anthopleura are about 2 X 2-3 nm (Siebert, 1974).
Gonad Cycles
Gonad indices have been used to assess reproductive cycles of many marine
invertebrates (e.g., Pearse, 1970, 1978; Conor, 1972, 1973a, b; Rutherford, 1973),
but seldom sea anemones [Ford (1964) is an exception, and Sebens (1981) used a
modified volumetric index]. Actinian gonadal tissue, not being concentrated in dis-
crete organs, is not easily quantified. In addition, wet body weight is difficult to
assess, which is why dry weights were used in this study. Histological observations
acted as a check on GI (Giese and Pearse, 1974).
Data for Urticina lofotensis during 1976-1977 (Figs. 5-8) suggest an annual
reproductive cycle with prolonged gamete release. Male and female gonad indices
reached minimum values in May, when gametes of both sexes were immature and
water temperature was at its minimum. Male GI attained its maximum in December;
males were ripest December-February. Female GI had a high value in June and a
slightly lower one in November; the highest proportion of ripe eggs was July-
October.
Mesenteries of spawned-out female Anthopleura elegantissima were extensively
ruptured (Jennison, 1979). This was never apparent in Urticina lofotensis. Large
eggs of A. elegantissima disappeared after spawning, and several months later a new
470 S. E. WEDI AND D. F. DUNN
cohort began to grow (Ford, 1964). Most female U. lofotensis studied contained
histologically normal oocytes 600 /urn or more in diameter throughout the year.
These data explain the lower amplitude of female than male GI, and suggest that
U. lofotensis may release ova intermittently rather than massively. This is supported
by loose eggs in the enterons of many females. For example, such cells occurred in
three of five females collected during October; four of them contained proportion-
ately fewer large oocytes than any female taken the previous month. In contrast to
prior months, no loose eggs were found in females during May, when large oocyte
quantity was at a minimum.
The drop in male GI between December and January, just as water temperature
began to fall from its annual high, suggests a massive spawning, with continued
slower release until March, at the latest. A simultaneous decline in large oocytes
supports this as the main spawning period. Ripe spermatozoa during winter may
have been left after the major spawn. Residual sperm packets in Anthopleura ele-
gantissima can be maintained up to four months after spawning, but eventually are
resorbed (Jennison, 1979; Sebens, 1981). The actinian Halcampa duodecimcirrata
contains motile sperm both before and after female spawning (Nyholm, 1949). On
the other hand, except for April and May, some males always contained stage 2 and
riper sperm. Some sperm of Urticina crassicornis are released immature, with excess
cytoplasm around the head; their fertilization capability is unknown (Chia and
Spaulding, 1972). If further research determined that 1) this occurs in U. lofotensis,
and 2) the sperm can mature following release, a strategy for fertilization of the eggs
that seem to be continually dribbled out would be provided.
High GI did not always coincide with gamete ripeness. The abrupt increase in
indices during June may have been due to abundant food. The largest number of
Calliostoma foliatum shells were recovered from anemones collected then, when
three out of six females had loose oocytes in their coelenterons. For males, with only
immature sperm follicles, the increase was in gonad quantity but not maturity.
Sebens (1981) offered the same explanation for briefly increased volume of immature
gonads in Anthopleura elegantissima.
Low GI may indicate recent spawning, but handling might induce premature
gamete release, and individuals just developing gonads would presumably have a
low GI. [Animals lacking gonads were assumed to have either recently spawned
(several loose oocytes were found in two of them) or been sexually immature.]
Perhaps an anemone attains sexual maturity at a particular size, gradually coming
into phase with the rest of the population, as Gonor (1972) found in Strongylocen-
trotus purpurat us. Were this true, small samples containing animals of all sizes would
emphasize asynchrony of reproductive cycles. However, gonad indices varied even
among anemones of similar sizes collected at the same time. Sterile animals were
taken throughout the year.
Speculations on larval development and settlement patterns
Based on published reports and his own findings, Spaulding (1974) ventured
that internal brooding is facultative in Urticina crassicornis. Stephenson (1928)
noted that rather large, unwieldy planulae rich in yolk, such as those of U. lofotensis
in Europe, are rare except in viviparous forms. Whether he meant to imply that U.
lofotensis broods is unclear; no reference has been made to this habit by others.
There is no evidence that it does so in Carmel Bay.
In the study area, specimens of Urticina lofotensis occur more densely in shallow,
horizontal depressions of large boulders than on open substrata. Perhaps larvae or
adults are carried there by gravity or in eddies. If so, chances of survival should be
enhanced because food would be similarly concentrated and the depressions would
REPRODUCTION OF URTICINA LOFOTENSIS 471
provide protection from surge and scour. Alternatively, larval or adult anemones
might actively seek depressions [adults can creep on their pedal discs as can many
other anemones (Stephenson, 1928; Dunn, 1977)] for the shelter and abundant food
they provide.
ACKNOWLEDGMENTS
This paper is adapted from a thesis submitted to San Francisco State University
by S.E.W. in partial fulfillment of the requirements for the M.A. degree. S.E.W.
thanks Dr. Albert Towle for help during both research and writing phases of this
project, and Dr. Thomas Niesen for helpful editorial comments.
Collections for this study were frequently made under adverse conditions, and
would have been impossible except for the diving assistance of David Klise and Ben
Tetzner, both of whom must often have wondered what they had let themselves in
for. S.E.W's wife, Becky, was a valuable field assistant on many collections, often
made under less than ideal conditions, and was very understanding and supportive
during this research.
LITERATURE CITED
CHIA, F.-S., AND B. CRAWFORD. 1973. Some observations on gametogenesis, larval development and
substratum selection of the sea pen Ptilosarcus guerneyi. Mar. Biol. 23(1): 73-82.
CHIA, F.-S., AND M. A. ROSTRON. 1970. Some aspects of the reproductive biology of Actinia equina
[Cnidaria: Anthozoa]. / Mar. Biol. Assoc. U. K. 50: 253-264.
CHIA, F.-S., AND J. G. SPAULDING. 1972. Development and juvenile growth of the sea anemone, Tealia
crassicornis. Biol. Bull. 142: 206-218.
CLARK, W. H., JR., AND W. C. DEWEL. 1974. The structure of the gonads, gametogenesis, and sperm-
egg interactions in the Anthozoa. Am. Zool. 14(2): 495-510.
DANIELSSEN, D. C. 1890. Actinida. Norw. North-All. Exp. 1876-1878 19: 1-184.
DEWEL, W. C., AND W. H. CLARK, JR. 1972. An ultrastructural investigation of spermiogenesis and the
mature sperm in the anthozoan Bunodosoma cavernata (Cnidaria). J. Ultrastr. Res. 40(3/4):
417-431.
DUNN, D. F. 1975. Reproduction of the externally brooding sea anemone Epiactis prolifera Verrill, 1869.
Biol. Bull. 148(2): 199-218.
DUNN, D. F. 1982. Sexual reproduction of two intertidal sea anemones (Coelenterata: Actiniaria) in
Malaysia. Biotropica 14(4): 262-271.
DYBAS, L. C. 1973. A histochemical and ultrastructural study of the coelomic cells of Phascolosoma
agassizii Keferstein, with particular reference to correlating structure and function. M. A. thesis,
San Francisco State University. 175 pp.
FORD, C. E., JR. 1964. Reproduction in the aggregating sea anemone, Anthopleura elegantissima. Pac.
Sci. 18: 138-145.
FRANK, P., AND J. S. BLEAKNEY. 1976. Histology and sexual reproduction of the anemone Nematostella
vectensis Stephenson 1935. J. Nat. Hist. 10: 441-449.
GEMMILL, J. F. 1921. The development of the sea anemone Bolocera mediae (Johnst.). Q. J. Microsc.
Sci. 65: 577-587.
GIESE, A. C., AND J. S. PEARSE. 1974. Introduction: general principles. Pp. 1-49 in Reproduction of
Marine Invertebrates, A. C. Giese and J. S. Pearse, eds. Academic Press, New York and London.
GONOR, J. J. 1972. Gonad growth in the sea urchin, Strongylocentrotus purpuratus (Stimpson) (Echi-
nodermata: Echinoidea) and the assumptions of gonad index methods. /. Exp. Mar. Biol. Ecol.
10: 89-103.
GONOR, J. J. 1973a. Reproductive cycles in Oregon populations of the echinoid, Strongylocentrotus
purpuratus (Stimpson). I. Annual gonad growth and ovarian gametogenic cycles. /. Exp. Mar.
Biol. Ecol. 12(1): 45-64.
GONOR, J. J. 1973b. Reproductive cycles in Oregon populations of the echinoid, Strongylocentrotus
purpuratus (Stimpson). II. Seasonal changes in oocyte growth and in abundance of gametogenic
stages in the ovary. / Exp. Mar. Biol. Ecol. 12(1): 65-78.
GOSSE, P. H. 1860. Actinologia Britannica: A History of the British Sea-anemones and Corals. Van
Voorst, London. 362 pp.
HAND, C. 1955. The sea anemones of central California. Part II. The endomyarian and mesomyarian
anemones. Wasmann J. Biol. 13(1): 37-99.
472 S. E. WEDI AND D. F. DUNN
HOLLAND, N. D. 1967. Gametogenesis during the annual reproductive cycle in a cidaroid sea urchin
(Stylocidaris affinis). Biol. Bull. 133: 578-590.
HUMASON, G. L. 1962. Animal Tissue Techniques. W. H. Freeman and Co., San Francisco and London.
468 pp.
JENNISON, B. L. 1978. Effects of thermal effluents on reproduction in a sea anemone. Pp. 470-483 in
Energy and Environmental Stress in Aquatic Systems. J. H. Thorp and J. W. Gibbons, eds.
U. S. Dept. Energy Conf. 771 1 14.
JENNISON, B. L. 1979. Gametogenesis and reproductive cycles in the sea anemone Anthopleura elegan-
tissima (Brandt, 1835). Can. J. Zool. 57(2): 403-41 1.
JENNISON, B. L. 1981. Reproduction in three species of sea anemones from Key West, Florida. Can. J.
Zool. 59(9): 1708-1719.
KESSEL, R. G. 1968. Electron microscope studies on developing oocytes of a coelenterate medusa with
special reference to vitellogenesis. J. Morphol. 126(2): 21 1-248.
LARKMAN, A. U. 1981. An ultrastructural investigation of the early stages of oocyte differentiation in
Actinia fragacea (Cnidaria; Anthozoa). Int. J. Invert. Repro. 4: 147-167.
LARKMAN, A. U., AND M. A. CARTER. 1982. Preliminary ultrastructural and autoradiographic evidence
that the trophonema of the sea anemone Actinia fragacea has a nutritive function. Int. J. Invert.
Repro. 4: 375-379.
LOSEVA, L. M. 1971. Observations on oogenesis of actinians. II. Oogenesis in Tealia crass icornis(Mue\\.),
Metridium senile (L.) and Protanthea simplex (Carlg,.). [in Russian] Vestn. Leningr. Univ. Biol.
9: 22-29.
MANUAL, R. L. 1981. British Anthozoa. Synopses of the British Fauna No. 1 8. Academic Press, London.
241 pp.
NYHOLM, K.-G. 1943. Zur Entwicklung und Entwicklungsbiologie der Ceriantharien und Aktinien. Zool.
Bidr. Uppsala 22: 87-248.
NYHOLM, K.-G. 1949. On the development and dispersal of athenaria actinia with special reference to
Halcampa duodecimcirrata, M. Sars. Zool. Bidr. Uppsala 27: 465-506.
OSTARELLO, G. L. 1973. Natural history of the hydrocoral Allopora californica Verrill (1866). Biol. Bull.
145: 548-564.
PEARSE, J. S. 1970. Reproductive periodicities of Indo-Pacific invertebrates in the Gulf of Suez. III. The
echinoid Diadema setosum (Leske). Bull. Mar. Sci. 20(3): 697-720.
PEARSE, J. S. 1978. Reproductive periodicities of Indo-Pacific invertebrates in the Gulf of Suez. IV. The
chitons Acanthopleura haddoni Winckworth and Onithochiton lyelli (Sowerby), and the abalone
Haliotis pustulata Reeve. Bull. Mar. Sci. 28(1): 92-101.
RlEMANN-ZiJRNECK., K. 1976. Reproductive biology, oogenesis and early development in the brood-
caring sea anemone Actinostola spetsbergensis (Anthozoa: Actiniaria). Helgol. Wiss. Meeresun-
ters. 28: 239-249.
RlEMANN-ZiJRNECK, K. 1978. Actinaria des Siidwestatlantik IV. Actinostola crassicornis (Hertwig, 1882)
mil einer Diskussion venvandter Arten. Verqff Inst. Meeresforsch. Bremerhaven 17: 65-85.
RINKEVICH, B. AND Y. LOYA. 1979. The reproduction of the Red Sea coral Stylophora pistillata. I.
Gonads and planulae. Mar. Ecol. Prog. Ser. 1: 133-144.
RUTHERFORD, J. C. 1973. Reproduction, growth and mortality of the holothurian Cucumaria pseudo-
curata. Mar. Biol. 22: 167-176.
SEBENS, K. P. 1981. Reproductive ecology of the intertidal sea anemones Anthopleura xanthogrammica
(Brandt) and A. elegantissima (Brandt): body size, habitat, and sexual reproduction. J. Exp.
Mar. Biol. Ecol. 54(3): 225-250.
SEBENS, K. P., AND G. LAAKSO. 1978. The genus Tealia (Anthozoa: Actiniaria) in the waters of the San
Juan Archipelago and the Olympic Peninsula. Wasmann J. Biol. (for 1977) 35(2): 152-168.
SIEBERT, A. E., JR. 1973. A description of the sea anemone Stomphia didemon sp. nov. and its devel-
opment. Pac. Sci. 27(4): 363-376.
SIEBERT, A. E., JR. 1974. A description of the embryology, larval development, and feeding of the sea
anemones Anthopleura elegantissima and A. xanthogrammica. Can. J. Zool. 52( 11): 1383-1 388.
SIEBERT, A. E., AND J. G. SPAULDING. 1976. The taxonomy, development and brooding behavior of the
anemones, Cribrinopsis fernaldi sp. nov. Biol. Bull. 150: 128-138.
SPAULDING, J. G. 1971. The development and brooding behavior of the anemone Cribrinopsis williamsi
Carlgren. (Abstr.) Am. Zool. 11(4): 701.
SPAULDING, J. G. 1972. The life cycle of Peachia quinquecapitata, an anemone parasitic on medusae
during its larval development. Biol. Bull. 143(2): 440-453.
SPAULDING, J. G. 1974. Embryonic and larval development in sea anemones (Anthozoa: Actiniaria).
Am. Zool. 14(2): 511-520.
STEPHENSON, T. A. 1928. The British Sea Anemones, Vol. I. Ray Society, London. 148 pp.
STEPHENSON, T. A. 1935. The British Sea Anemones, Vol. II. Ray Society, London. 426 pp.
Reference: Biol. Bull. 165: 473-486. (October, 1983)
ECHINODERM IMMUNOLOGY: BACTERIAL CLEARANCE BY THE
SEA URCHIN STRONGYLOCENTROTUS PURPURATUS
MARY A. YUI AND CHRISTOPHER J. BAYNE
Department of Zoology, Oregon State University, Corvallis, Oregon 97331
ABSTRACT
Characteristics of bacterial clearance were investigated in the purple sea urchin,
Strongylocentrotus purpuratus (Echinodermata: Echinoidea). Primary clearance ki-
netics were determined for three bacteria, a marine Gram negative motile rod, a
marine Gram positive non-motile rod, and a Gram negative freshwater fish patho-
gen, Aeromonas salmonicida. Clearance kinetics differed for each of the three bac-
teria. Secondary clearance rates were not significantly different from primary clear-
ance rates for any of the three bacteria, regardless of the time interval between
inoculations (9-2 1 days), implying a probable absence of immunologic memory.
During primary clearance, total coelomocyte counts declined 93% by 90 min post
injection. All four coelomocyte types declined, however the relative proportions of
each type changed during the six-hour sampling period. In cell-free coelomic fluid,
viable counts of marine bacteria declined, with different kinetics for the two species.
Viable counts in sea water controls did not change. Declines in viable counts may
be due to bactericidal activity and/or agglutination, although bacterial agglutination
was not observed.
INTRODUCTION
Despite recent advances (reviewed by Cooper, 1976; Manning and Turner, 1976;
Marchalonis, 1977; Hildemann et al, 1981), the phylogeny of immunity remains
obscure. In particular, the mechanisms of invertebrate immunity are diverse, and
many are poorly understood. The phylogenetic position of echinoderms makes them
pivotal to the understanding of the phylogeny of immunity and the evolution of
vertebrate immunity.
Allogeneic transplantation studies have shown that memory and specificity, two
important characteristics of vertebrate immune responses, are possessed by echi-
noderms (Karp and Hildemann, 1976; Coffaro and Hinegardner, 1977; Coffaro,
1980), as well as by members of other invertebrate phyla (sponges, Hildemann et
al, 1979; cnidarians, Hildemann et al., 1977; annelids, Cooper, 1970). Any role of
memory and specificity in invertebrate internal defenses to potentially infectious
agents has yet to be identified.
In this study, we sought to determine whether the echinoderm, Strongylocen-
trotus purpuratus, exhibits memory or altered reactivity on secondary contact with
biologically relevant antigens, namely bacteria. Additionally, because the coelomic
fluid of healthy echinoderms is generally aseptic (Bang and Lemma, 1962; Unkles,
1977; Kaneshiro and Karp, 1980), and few studies have been made on the char-
acteristics and mechanisms of bacterial clearance in echinoderms (Johnson, 1969a,
b; Johnson et al., 1970; Johnson and Chapman, 1971; Wardlaw and Unkles, 1978),
Received 1 1 April 1983; accepted 25 July 1983.
473
474 M. A. YUI AND C. J. BAYNE
we also studied some of the in vivo and in vitro primary interactions between echi-
noderm coelomocytes and cell-free coelomic fluid and bacteria.
MATERIALS AND METHODS
Collection and maintenance of animals
Strongylocentrotus purpuratus (80-150 g) were hand collected from intertidal
surge channels at Yaquina Head, Newport, Oregon where the species is abundant.
Care was taken to avoid damaging the urchins during collection and transportation
to the 22,700 liter recirculating sea water system at Oregon State University in
Corvallis, 70 km away. Water temperatures were 12-15°C, comparable to coastal
temperatures. Urchins were fed macroalgae ad libitum.
Isolation and culture of bacteria
One Gram negative motile rod and one Gram positive non-motile rod were
isolated from the coelomic fluid (CF) of moribund sea urchins. Selection was based
on growth characteristics (rate of growth, colony color and morphology). All bacteria
were grown in marine broth 2216E (Difco Laboratories), enriched with peptone
(5 g r1) and yeast extract (3 g 1~'). Nutrient agar (1%) in marine broth was used for
pour plate viable counts. For each experiment, bacteria were inoculated into fresh,
enriched broth in triple baffle, side-arm flasks and grown at room temperature (20-
23°C) on a rotary table (100 rpm). The bacteria were harvested during log phase,
centrifuged for 10 min at 300 g, and resuspended in cold (10°C) sterile sea water
to appropriate concentrations. Bacterial suspensions used for injections were serially
diluted and plated to determine actual inoculation doses.
Because the urchins may previously have encountered the marine bacteria, a
freshwater bacterium, Aeromonas salmonicida, was also used for some experiments.
A. salmonicida is a salmonid fish pathogen, a Gram negative non-motile rod gen-
erally not present in the marine environment (although it has rarely been found in
marine fish, Evelyn, 1971). These bacteria were grown in tryptic soy broth using
methods already described.
Coelomic fluid volume estimates
Estimation of CF volumes was required for determining the quantity of bacteria
to be injected for a specific concentration in the perivisceral coelom. The weight,
test diameter, test height, and peristomium diameter were measured for 28 urchins.
The CF was then drained through a cut in the peristomium and the volume mea-
sured. Each external parameter was regressed on the CF volume. Weight was found
to be the best indicator of CF volume (R2 = 0.96). The following equation defined
the relationship and was used in all experiments:
Coelomic fluid volume (ml) = 0.35 X weight (g) - 4.2.
Clearance experiments
Urchins were weighed, CF volumes estimated, and the dose of bacteria calculated
for a given final concentration of bacteria in the CF. During experiments, urchins
were kept in 20 liter plastic aquaria filled with aerated sea water at 10°C.
Before injections and sampling, the peristomium was washed several times with
cold sterile sea water. Care was taken to avoid tearing tube feet during handling.
The bacterial suspension (0.4-0.6 ml) was injected through the peristomial mem-
BACTERIAL CLEARANCE IN URCHINS 475
brane with a 26 gauge, 0.5-inch needle and 1 ml tuberculin syringe. Preliminary
experiments showed that 40-90 min were required for even dispersal of bacteria.
Because we wanted to sample within that time, half of the inoculum was injected,
with the other half injected 180° from the initial injection site. Coelomic fluid
samples (0.2-0.3 ml) were removed with a sterile 26 g, 0.5-inch needle and 1 ml
syringe. Samples taken before the injection of bacteria were directly plated for ste-
rility checks. Samples taken after injection of bacteria were serially diluted in sterile
sea water and plated by pour plate methods for viable counts. Distinctive colony
morphologies, colors, and growth rates were used to help ascertain that these bacteria
were those previously injected.
Total and differential coelomocyte cell counts
At various times after urchins were injected with the marine Gram negative
bacteria, or with an equal volume of sterile sea water, CF was removed with a 20
g, 1.5-inch needle into an equal volume of cold anticoagulant (30 mM EDTA in
0.3 M Hepes buffered sea water, after Bertheussen and Seljelid, 1978). Two differ-
ential counts were made and averaged using a Brightline hemacytometer. The four
major coelomocyte types found in Strongylocentrotus spp. (Johnson, 1969a; Ber-
theussen and Seljelid, 1978), phagocytes (leukocytes), vibratile cells, and red and
colorless spherule (morula) cells, were counted. Total counts were determined by
adding the counts for the four cell types.
Humoral factors: In vitro effects of cell-free coelomic fluid
on viable counts of bacteria
Coelomic fluid was removed with a 20 g, 1.5-inch needle or drained out through
a cut in the peristomium into a sterile, cold beaker. The CF was immediately filtered
by gentle passage through two Millipore prefilters then sterilized using a 0.22 ^m
Millipore filter. Since clotting of CF (30 min, 10°C) before filtration does not seem
to affect its bactericidal activity ( Wardlaw and Unkles, 1 978; Yui, 1 982), we routinely
filtered before clot formation.
Two ml of cell-free CF were placed in sterile glass vials. Sterile-filtered artificial
sea water (Instant Ocean) or Hepes sea water medium (Bertheussen and Seljelid,
1978) was used as control fluid. Twenty p\ aliquots of the bacterial suspension were
added to each vial, which was then held at 10°C. Samples (10 and 100 n\) were
removed and plated by pour plate methods for viable counts.
Humoral factors: Bacterial agglutinins
The CF (5-10 ml) was collected through a cut in the peristomium, filtered,
added to the first well of a microtiter plate and serially diluted by one-half with
10°C artificial sea water (Instant Ocean). The last well was a sea water-only control.
An equal volume of bacterial suspension in sea water was added to each well, then
the plates were covered and placed on a rotary table (60 rpm) at 10°C. After 30
min to 20 h plates were inspected for agglutination.
RESULTS
Primary clearance
Gram negative bacteria. Six urchins injected with 3.3 X 107 marine Gram neg-
ative bacteria ml"1 of CF rapidly reduced viable counts (v.c.) of bacteria in the CF
476
M. A. YUI AND C. J. BAYNE
in the first hour post-injection (p.i.) by 95.7%, followed by a period of slower clear-
ance (Fig. 1 ). Clearance continued after 6 h (Fig. 2), and bacteria were not detected
4-8 days p.i. To further characterize the initial 90 min of clearance, four or five
samples were removed from each of nine urchins previously injected with 105-107
bacteria ml"1 of CF. Clearance was exponential. Lines were fitted to the log-trans-
formed data using linear regression by calculating independent slopes for each urchin
and obtaining a mean and standard error for those slopes. The slopes are equivalent
to the "Phagocytic Index" or K value as defined by Biozzi et al. (1953) for particle
clearance kinetics in mammals, and Renwrantz and Mohr (1978) for particle clear-
ance in a land snail, Helix. For the first 90 min of clearance, a K value of -0.0179
± 0.0016 logic bacteria min"1 was obtained for the Gram negative (R2 = 0.963).
The slope was not strongly dose dependent (Yui, 1982).
Because of individual differences, primary and secondary clearance rates were
determined for the same individual urchins. Injection doses of 106-107 bacteria ml"1
of CF were selected for further experiments because, at those doses, bacteria were:
107H
" , 106H
DC
LLJ
\-
o
<
CD
10-
0
I I I I I
23456
TIME (hours)
FIGURE 1. Primary clearance of the Gram negative and Gram positive marine bacteria, and Aero-
monas salmonicida, sampled at 1, 3, and 6 hours post-injection (mean ± standard error). • Gram negative
(n = 6), A Gram positive (n = 3), • Aeromonas salmonicida (n = 3).
BACTERIAL CLEARANCE IN URCHINS
477
(a) cleared without mortality or obvious trauma, (b) detectable at high enough levels
at 6 and 24 h p.i. that a decline upon secondary injection was still quantifiable even
with small CF sample volumes, and (c) persistent in the CF for a period of 4-8 days.
This persistence of viable bacteria in the CF was considered advantageous because
sensitization and induction of memory may require a long period of exposure, as
with graft rejection. Furthermore, the need for booster injections was precluded.
Gram positive bacteria. A rapid decline (97.7%) in v.c. also occurred within 1
h after injection of Gram positive bacteria (Fig. 1). However, unlike the results with
the Gram negative bacteria, v.c. were slightly higher at 3 h with declines continuing
to 6 and 24 h p.i. (Fig. 2). Bacteria were not detected 8-12 days p.i. The K value
for the first 90 min was -0.0195 ± 0.0037 (n = 3), not significantly different from
K for the Gram negative bacteria.
Aeromonas salmonicida. At 1 h p.i., v.c. dropped only 61.5%, a much lower
initial rate of clearance than that of the marine bacteria (Fig. 1). Clearance continued
at the same rate to 3 h and more slowly thereafter.
Primary versus secondary clearance
Because the first and second inoculation doses could not be made identical, the
slopes of the lines were calculated from Iogi0-transformed v.c. at time = 0 to 6 and
I
6 12
TIME (hours)
18
24
FIGURE 2. Primary clearance of the Gram negative and Gram positive marine bacteria, sampled
at 6 and 24 hours post-injection, and Aeromonas salmonicida, sampled at 6 hours post-injection (mean
± standard error). • Grain negative (n = 1 1), A Gram positive (n = 12), • Aeromonas salmonicida (n
= 6).
478 M. A. YUI AND C. J. BAYNE
24 h. The difference in slope between primary and secondary clearance for each
urchin was then calculated and compared to zero (Table I).
Urchins injected with approximately 107 bacteria ml"1 of CF were challenged
with a similar dose of the same bacteria 9, 14, and 21 days after the first injection
of the two marine bacteria, and after 19 days with Aeromonas salmonicida. In no
case was the mean difference in slope between primary and secondary clearance
significantly different from zero. Even when data from the three injection times were
pooled, the mean difference in slope was not significantly different from zero.
Total and differential counts of coelomocytes
The average number of coelomocytes ml~' counted from uninjected urchins col-
lected in August and September was 1.0 ± 0.2 X 107 (n = 6). The majority of cells
were phagocytes (67.8 ± 4.4%), followed by vibratiles (16.8 ± 3.8%), red spherules
(10.5 ± 3.1%), and colorless spherules (5.0 ± 1.3%). These values are similar to those
reported for S. droebachiensis (Bertheussen and Seljelid, 1978). In October and No-
vember, coelomocyte counts were lower, due predominantly to fewer phagocytes
(Yui, 1982).
Sea water injected control urchins exhibited a sharp, brief decline in cell numbers
during the first hour p.i., followed by rapid recovery to pre-injection values (Fig. 3).
After injection of Gram negative bacteria, a sharp decline was seen during the first
1.5 h. Declines continued, more slowly, to about 5 h. The overall drop was from
8.8 X 106 to 6 X 105 coelomocytes ml"1, a 93% decline.
After injection, all cell types declined (Fig. 3), with a change in the relative pro-
portion of each cell type (Fig. 4). The percentages of phagocytes and red spherule
TABLE I
Mean difference in slope between clearance of primary (1°) and secondary (2°) injections of Gram
negative and positive marine bacteria and Aeromonas salmonicida*
l°-2°
Interval
(days)
Sample
time
(hours)
Mean difference
in slope (1° minus 2°) x
±SD(n)**
Gram negative
Gram positive
Aeromonas
salmonicida
9
6
24
0.083 ± 0.106
(4)
0.006 ±0.018
(4)
-0.027 ± 0.108
(4)
0.011 ± 0.011
(4)
—
14
6
24
0.038 ±0.190
(4)
-0.030 ± 0.051
(3)
0.027 ± 0.078
(4)
-0.008 ± 0.016
(4)
—
21
6
24
0.032 ±0.182
(4)
0.013 ±0.024
(4)
-0.063 ± 0.061
(4)
-0.020 ± 0.025
(4)
—
19
6
—
—
0.067 ± 0.090
(4)
* Approximately 1 X 107 bacteria ml ' of coelomic fluid were injected for each clearance rate
determination.
* A negative value indicates a more rapid 2° rate of clearance, a positive value, a less rapid 2° rate
of clearance. None of these values were significantly different from zero using a Student's Mest.
BACTERIAL CLEARANCE IN URCHINS
479
10
4
(hours)
24
FIGURE 3. Total and differential coelomocyte counts (mean ± standard error) after injection with
108 Gram negative bacteria mr1 of coelomic fluid (n = 3). The 24-hour sample was from the only urchin
sampled at that time. Also included are the total coelomocyte counts for the sea water-injected control
(n = 1). Total coelomocyte counts: O sea water injected, D bacteria injected; differential coelomocyte
counts: A phagocytes, • vibratiles, T red spherules, • colorless spherules.
cells declined while the percentage of vibratiles increased. The percentage of colorless
spherule cells did not change appreciably, although the values were at very low levels
throughout. In the one animal sampled at 24 h p.i., cell counts (Fig. 3) and relative
proportions (Fig. 4) were approaching pre-injection values.
Humoral factors: In vitro effects of cell-free coelomic fluid on
viable counts of bacteria
Gram negative bacteria. Coelomic fluids from six urchins, three injected 3 days
earlier and three uninjected, were tested for their effects on viability of the marine
480
M. A. YUI AND C. J. BAYNE
70H
24
TIME (hours)
FIGURE 4. Relative proportions (mean % ± standard error) of the four coelomocyte types after
injection of 107 bacteria ml"1 of coelomic fluid (n = 3). The 24-hour sample was from only one of the
three urchins. • phagocytes, V vibratiles, A red spherules, • colorless spherules.
Gram negative bacteria. Sea water served as a control. No significant differences
were noted between any of the three treatments at 1 5 and 45 min post-inoculation
(one-sided Student's /-test) (Fig. 5). Although v.c. were lower in CF from preinjected
than in uninjected urchins at 90 min or 5 h, the differences were not significant.
V.c. were significantly lower in the CF from the six urchins than in sea water at 90
min (P < 0.005) and 5 h (P < 0.01).
Gram positive bacteria. The change in v.c. of Gram positive bacteria in CF differed
from that of the Gram negative (Fig. 6). V.c. had declined 2.5 orders of magnitude
below those of the initial inoculum and the sea water control at 2 h post-inoculation.
Test fluids from each of five urchins exhibited this large decline in v.c. at 2 h while
none of the three controls exhibited a similar response. This result cannot reflect
bactericidal activity since counts returned to values not significantly different from
sea water controls by 5 h. At 20 h, v.c. in CF were lower than those in sea water (P
= 0.05). Viable counts in CF were quite variable relative to sea water controls. This
variability probably reflects differences in amount(s) of humoral factor(s) present in
the CF rather than differences due to sampling methods.
Bacterial agglutinins
Cell-free CF samples from eight urchins were mixed with the Gram negative
bacteria and CF samples from four urchins were tested with the Gram positive
bacteria, in suspensions ranging from 5 X 104 to 5 X 108 bacteria ml"1. Samples
BACTERIAL CLEARANCE IN URCHINS
48
2 3
TIME (hours)
i
5
FIGURE 5. Viable counts (mean ± standard error) for the Gram negative bacteria in sea water and
cell-free coelomic fluid from injected and uninjected urchins. • sea water (n = 3), • uninjected (n = 3),
A injected (n = 3).
were checked at various times from 30 min to 20 h. Neither of the two bacteria
were agglutinated.
DISCUSSION
Primary clearance
Clearance from echinoderm coelomic fluid has been reported after the injection
of bacteria (Wardlaw and Unkles, 1978; Kaneshiro and Karp, 1980; Bertheussen,
1981), bacteriophage T4 (Coffaro, 1978), red blood cells, latex beads and yeast cells
(Bertheussen, 1981), xenogeneic cells and carborundum (Reinisch and Bang, 1971),
and bovine and human serum albumin (Hilgard and Phillips, 1968). However, the
kinetics of particle clearance have not been followed, and the fates of these particles
in echinoderms remain unknown.
Strongylocentrotus purpuratus efficiently cleared all three bacteria from its coe-
lomic fluid. Viable counts were reduced 90-99% in 3-6 h after injection of doses
of approximately 106-107 bacteria ml"1 of CF. Clearance occurred in at least two
phases, with the rate of clearance of one of the bacteria different from that of the
others during each phase. Clearance was approximately exponential for the first 1 .5
h p.i. for the Gram negative bacteria. Both marine bacteria were cleared at similar
rates over the first 1.5 h while the initial rate of clearance of A. salmonicida was
slower than that of the marine bacteria.
The second phase of clearance was slower. However, by 6 h p.i., A. salmonicida
had been cleared almost as well as the marine Gram negative bacteria despite the
slower initial rate, suggesting a higher rate of clearance after 1 h compared to the
482
M. A. YUI AND C. J. BAYNE
104-
OC
LU
I-
o
102-
234
TIME (hours)
20
FIGURE 6. Viable counts for the Gram positive bacteria in sea water and cell-free coelomic fluid
(mean ± standard error). O sea water (n = 3), • cell-free coelomic fluid (n = 5).
other bacteria. The marine Gram positive, on the other hand, had a significantly
lower reduction in v.c. at 6 h p.i. compared with the marine Gram negative bacteria
and A. salmonicida. At 24 h the overall percent reduction of Gram positive bacteria
was still lower than that of the Gram negative.
The clearance we observed resembles the equivalent process in other inverte-
brates and in mammals. Generally, the first phase of clearance in mammals is rapid
and exponential, with 90-99.9% reduction in circulating bacteria (Rogers, 1960).
This phase is relatively independent of the nature of the microbe, the animal under
study, and the subsequent outcome of injection. During the second phase the mi-
crobe either persists at lower concentrations or is slowly removed over several hours
or days. This phase differs considerably with different bacteria.
Similarly, four mollusc species (Bayne and Kime, 1970; Pauley et al, 1971;
Bayne, 1973; van der Knaap et al., 1981) and a crustacean (Smith and Ratcliffe,
1980) cleared 90-99% of injected bacteria (106-109 bacteria per animal) in the first
2-3 h p.i., with slower subsequent declines after 3 h. In some species Gram positive
and Gram negative bacteria may be cleared equally well (Smith and Ratcliffe, 1980),
while in others, Gram positives may be cleared more rapidly than Gram negatives
(van der Knaap et al., 1981). Gram negative and Gram positive bacteria may be
recognized by different coelomocyte subpopulations in the marine annelid, Arenicola
marina (Fitzgerald and Ratcliffe, 1982). In Strongylocentrotus spp., Johnson ( 1969b)
noted more active phagocytosis of Gram positive than Gram negative bacteria, in
BACTERIAL CLEARANCE IN URCHINS 483
contrast with this study. This difference may be due to the different bacteria used
or because phagocytosis alone does not determine the overall rate of bacterial
clearance.
Total and differential cell counts indicated a 93% reduction in numbers of all
four coelomocyte types accompanying primary clearance of Gram negative bacteria.
Similar declines in circulating hemocyte numbers have also been observed in crus-
taceans (Cornick and Stewart, 1968; Smith and Ratcliffe, 1980), insects (Wittig,
1965; Gagen and Ratcliffe, 1976), molluscs (Bayne and Kime, 1970; Pauley el al,
1971; van der Knapp et al., 1981; Renwrantz el al, 1981), and in leukocyte numbers
in mammals (Rogers, 1960).
Clotting, due to aggregation of phagocytes (Johnson, 1969a), no doubt contrib-
utes to some of the observed decline in numbers of cells and bacteria. Cellular clots
were often observed in coelomic fluid samples which were taken after injection of
bacteria. Bertheussen (1981) noted clot formation in S. droebachiensis only after
injection of bacteria and not after injection of other particles (red blood cells, yeast,
latex). Cellular aggregation and/or attachment to epithelia may enhance phagocy-
tosis by providing a substrate for trapping bacteria. Vertebrate leukocytes trap and
phagocytose encapsulated bacteria more easily when attached to substrates
(Wood, 1960).
Cellular clotting may also explain the observation that red spherule cells declined
proportionately with phagocytes while vibratile cells increased proportionately. Red
spherules are relatively non-motile and they may therefore be passively caught with
aggregating phagocytes as observed in hanging drops (Johnson, 1969a). Although
the percentage of non-motile colorless spherules did not appear to change, actual
cell counts were very low so changes were difficult to detect. Vibratiles, being highly
motile, may extricate themselves from the clots or may be recruited more rapidly
from elsewhere, causing the observed increase in relative abundance.
Cell-free coelomic fluid contains one or more humoral factor(s) active against
both Gram negative and Gram positive bacteria. To the contrary, Wardlaw and
Unkles (1978) found that coelomocytes were required for bactericidal activity. Be-
cause the quantity of humoral factor(s) would be limited and not renewable in a
cell-free in vitro system, the number of bacteria per given volume of coelomic fluid
would be critical in the detection of activity. In addition, observable activity may
depend upon the species of bacteria used and the species of echinoderm from which
test fluids are obtained.
Declines of 6 to 9 X 104 v.c. ml"1 of CF from 6 urchins were observed in vitro
by 2 h after inoculation with Gram negative bacteria, enough to account for at least
some of the decine observed in vivo. It is possible that the decrease in v.c. was due
to agglutination rather than bacterial killing, although agglutination was not detected
in this study. Prior injection with bacteria did not accelerate the decline.
Results obtained with Gram positive bacteria in cell-free CF are difficult to
explain since agglutinins were not found. Large declines in v.c. were noted in the
2 h samples from the CF from all 5 urchins tested. These declines could not be
explained by bacterial killing because v.c. returned to control levels by 5 h p.i. and
these bacteria do not grow that quickly at 10°C. These results, however, do cor-
respond with the lower v.c. noted at 1 h than at 3 h in in vivo clearance of the Gram
positive. Although v.c. in cell-free CF in vitro were not significantly lower than in
sea water at 5 h, v.c. at 20 h were considerably lower than the sea water controls
and the initial inoculation dose.
Although bacterial agglutination was not observed, these experiments do not
rule out the possibility that agglutinins function in clearance of bacteria. In fact, the
484 M. A. YUI AND C. J. BAYNE
results of the bactericidal test using the Gram positive suggest that agglutination
may occur temporarily at 2 h p.i. Relative proportions of bacteria to concentration
of active molecules in the CF may have been inappropriate for the observation of
agglutination.
Differences in the rates of clearance of the three bacteria may depend on ( 1 )
differences in specific or non-specific cellular recognition and response, (2) the quan-
tity, rate of release and effectiveness of humoral bactericidal substances, agglutinins,
and/or opsonins, and (3) the rate of bacterial reproduction. Based on our studies
of in vivo and in vitro primary clearance, it appears that both cellular and humoral
factors are involved in bacterial clearance. Clotting and the overall decline of coe-
lomocytes paralleled bacterial clearance. Bacteria were observed in the cellular clots,
and within phagocytes after injection (Yui, 1982). Different quantities of CF factor(s)
or numbers of responding cells may cause the large individual variation observed
between urchins.
The two stages of bacterial clearance in S. purpuratus are probably the result
of a decline in active cells and/or molecules in the first few hours with further
clearance being dependent on the level of cell recruitment or activation and/or
release of active molecules.
Primary versus secondary clearance
Our failure to find accelerated secondary clearance of bacteria is consistent with
the few other attempts to induce elevated responses in echinoderms to foreign ma-
terials, other than grafts. Coffaro (1978) was unable to elicit accelerated clearance
of the bacteriophage T4 in Lytechinus pictus. No increase in the rate of foreign
protein uptake was detected in S. purpuratus after injection of four doses of protein
at 18 h, 3 days, 7 days, and 2 months after primary injection (Hilgard and Phillips,
1968). Asterias vulgaris was inoculated with 5 X 105 Arbacia cells six times, with
no change in the rate of cell disappearance (Reinisch and Bang, 1971).
Secondary responses to bacteria have been elevated in other phyla, including
both deuterostomes (e.g., vertebrates) and protostomes, in which molluscs (see Bayne
et al, 1980 for review; Bayne, 1980; van der Knapp, 1980), sipunculids (Evans et
al, 1969), crustaceans (Evans et al., 1968; Acton et al., 1969; McKay and Jenkin,
1969; Stewart and Zwicker, 1974) and insects (Boman et al., 1972; Boman et al.,
1974) have yielded positive results. However, in the invertebrate examples, either
enhancement has been non-specific or the extent of specificity has not been thor-
oughly tested, and mechanisms of enhancement are seldom known.
The lack of accelerated secondary clearance in this study can be explained in
the following ways:
( 1 ) There is no memory component to bacterial clearance, although memory does
appear to exist in responses to grafts and may exist in response to other infectious
organisms.
(2) Memory does exist but was undetected due to:
(a) previous exposure to the same or similar bacteria sharing certain surface
characteristics or molecules critical for recognition and response, or
(b) lack of sensitivity in detection of memory or sub-optimal enhancement of
secondary responses.
Despite such considerations, there is a need to conduct experiments to identify
the conditions under which echinoderms will respond optimally to immunologic
BACTERIAL CLEARANCE IN URCHINS 485
challenge. Without such studies, the mechanisms of echinoderm internal defenses
and the evolutionary history of immune responsiveness will remain elusive.
ACKNOWLEDGMENTS
We thank E. S. Loker and C. A. Boswell for suggestions on the manuscript, and
J. A. Longmate for assistance collecting urchins. The research was supported by
grants from the Oregon State University Zoology Department Research Fund.
LITERATURE CITED
ACTON, R. T., P. F. WEINHEIMER, AND E. E. EVANS. 1969. A bactericidal system in the lobster Homarus
americanus. J. Invert. Pat hoi. 13: 463-464.
BANG, F. B., AND A. LEMMA. 1962. Bacterial infection and reaction to injury in some echinoderms. J.
Insect Pathol. 4: 401-414.
BAYNE, C. J. 1973. Internal defense mechanisms of Octopus dofleini. Malacol. Rev. 6: 13-17.
BAYNE, C. J. 1980. Molluscan immunity: Induction of elevated immunity in the land snail (Helix) by
injections of bacteria (Pseudomonas aeruginosd). Dev. Comp. Immunol. 4: 43-54.
BAYNE, C. J., AND J. B. KIME. 1970. In vivo removal of bacteria from the hemolymph of the land snail
Helix pomatia (Pulmonata: Stylommatophora). Malacol. Rev. 3: 103-113.
BAYNE, C. J., T. SMINIA, AND W. P. W. VAN DER KNAAP. 1980. Immunological memory: Status of
molluscan studies. Pp. 57-64 in Phytogeny of Immunological Memory, M. J. Manning, ed.
Elsevier/North Holland, Amsterdam.
BERTHEUSSEN, K. 1981. Endocytosis by echinoid phagocytes in vitro. I. Recognition of foreign matter.
Dev. Comp. Immunol. 5: 241-250.
BERTHEUSSEN, K., AND R. SELJELID. 1978. Echinoid phagocytes in vitro. Exp. Cell Res. Ill: 401-412.
BIOZZI, G., B. BENACERRAF, AND B. N. HALPERN. 1953. Quantitative study of the reticuloendothelial
system. II. A study of the kinetics of the granulopectic activity of the RES in relation to the
dose of carbon injected. Br. J. Exp. Pathol. 34: 441-457.
BOMAN, H. G., I. NILSSON, AND B. RASMUSON. 1972. Inducible antibacterial defence system in Dro-
sophila. Nature 237: 232-235.
BOMAN, H. G., I. NILSSON-FAYE, K. PAUL, AND T. RASMUSON. 1974. Insect immunity I. Characteristics
of an inducible cell-free antibacterial reaction in hemolymph of Samia cynthia pupae. Infect.
Immunol. 10: 136-145.
COFFARO, K. 1978. Clearance of bacteriophage T4 in the sea urchin Lytechinus pictus. J. Invert. Pathol.
32: 384-385.
COFFARO, K. A. 1980. Memory and specificity in the sea urchin, Lytechinus pictus. Pp. 77-80 in Phy-
logeny of Immunological Memory, M. J. Manning, ed. Elsevier/North Holland, Amsterdam.
COFFARO, K. A., AND R. T. HINEGARDNER. 1977. Immune response in the sea urchin Lytechinus pictus.
Science 197: 1389-1390.
COOPER, E. L. 1970. Transplantation immunity in helminths and annelids. Transplant. Proc. 2: 216-
221.
COOPER, E. L. 1976. Comparative Immunology. Prentice Hall, Inc. 338 pp.
CORNICK., J. W., AND J. E. STEWART. 1968. Interaction of the pathogen Gqffkya homari with natural
defense mechanisms of Homarus americanus. J. Fish. Res. Board Can. 25: 695-709.
EVANS, E. E., B. PAINTER, M. L. EVANS, P. WEINHEIMER, AND R. T. ACTON. 1968. An induced
bactericidin in the spiny lobster, Panulirus argus. Proc. Soc. Exp. Biol. Med. 128: 394-398.
EVANS, E. E., P. F. WEINHEIMER, R. T. ACTON, AND J. E. CUSHING. 1969. Induced bactericidal response
in a sipunculid worm. Nature 223: 695.
EVELYN, T. P. T. 197 1 . An aberrant strain of the bacterial fish pathogen Aeromonas salmonicida isolated
from a marine host, the sablefish (Anoplopomafimbria), and from two species of cultured Pacific
salmon. /. Fish. Res. Board Can. 28: 1629-1634.
FITZGERALD, S. W., AND N. A. RATCLIFFE. 1982. Evidence for the presence of subpopulations of Ar-
enicola marina coelomocytes identified by their selective response toward Gram + ve and Gram
- ve bacteria. Dev. Comp. Immunol. 6: 23-34.
GAGEN, S. J., AND N. A. RATCLIFFE. 1976. Studies on the in vivo cellular reactions and fate of injected
bacteria in Galleria mellonella and Pieris brassicae larvae. /. Invert. Pathol. 28: 17-24.
HlLDEMANN, W. H., R. L. RAISON, G. CHEUNG, C. J. HULL, L. AKAKA, AND J. OKAMOTO. 1977.
Immunologic specificity and memory in a scleractinian coral. Nature 270: 219-223.
HILDEMANN, W. H., I. S. JOHNSON, AND P. L. JOKJEL. 1979. Immunocompetence in the lowest metazoan
phylum: transplantation immunity in sponges. Science 204: 420-422.
486 M. A. YUI AND C. J. BAYNE
HILDEMANN, W. H., E. A. CLARK, AND R. L. RAISON. 1981. Comprehensive Immunogenetics. Elsevier/
North Holland, Amsterdam. 368 pp
HILGARD, H. R., AND J. H. PHILLIPS. 1968. Sea urchin response to foreign substances. Science 161:
1243-1245.
JOHNSON, P. T. 1969a. The coelomic elements of sea urchins (Strongylocentrotus) I. The normal coe-
lomocytes; their morphology and dynamics in hanging drops. J. Invert. Pathol. 13: 25-41.
JOHNSON, P. T. 1969b. The coelomic elements of sea urchins (Strongylocentrotus). III. In vitro reaction
to bacteria. J. Invert. Pathol. 13: 42-62.
JOHNSON, P. T., P. K. CHIEN, AND F. A. CHAPMAN. 1970. The coelomic elements of sea urchins
(Strongylocentrotus) V. Ultrastructure of leukocytes exposed to bacteria. / Invert. Pathol. 16:
466-469.
JOHNSON, P. T., AND F. A. CHAPMAN. 1971. Comparative studies on the in vitro response of bacteria
to invertebrate body fluids. III. Stichopus tremulus (sea cucumber) and Dendraster excentricus
(sand dollar). / Invert. Pathol. 17: 94-106.
KANESHIRO, E. S., AND R. D. KARP. 1 980. The ultrastructure of cpelomocytes of the sea star Dermasterias
imbricata. Biol. Bull. 159: 295-310.
KARP, R. D., AND W. H. HILDEMANN. 1976. Specific allograft reactivity In Dermasterias imbricata.
Transplantation 22: 434-439.
MANNING, M. J., AND R. J. TURNER. 1976. Comparative Immunobiology. Wiley and Sons. 184 pp.
MARCHALONIS, J. J. 1977. Immunity in Evolution. Harvard University Press. 316 pp.
McKAY, D., AND C. R. JENKJN. 1969. Immunity in invertebrates. II. Adaptive immunity in the crayfish
(Parachaeraps bicarinatus). Immunology 17: 127-137.
PAULEY, G. B., S. M. KRASSNER, AND F. A. CHAPMAN. 1971. Bacterial clearance in the California
seahare Aplysia californica. J. Invert. Pathol. 18: 227-239.
REINISCH, C. L., AND F. B. BANG. 1971. Cell recognition: Reactions of the seastar (Asterias vulgaris) to
the injection of amebocytes of sea urchin (Arbacia punctulata). Cell. Immunol. 2: 496-503.
RENWRANTZ, L., AND W. MOHR. 1978. Opsonizing effect of serum and albumin gland extracts on the
elimination of human erythrocytes from the circulation of Helix pomatia. J. Invert. Pathol. 31:
164-170.
RENWRANTZ, L., W. SCHANCKE, H. HARM, H. ERL, H. LIEBSCH, AND J. GERCKEN. 1981. Discriminative
ability and function of the immunobiological recognition system of the snail Helix pomatia.
J. Comp. Physiol. 141: 477-488.
ROGERS, D. E. 1960. Host mechanisms which act to remove bacteria from the blood stream. Bacterial.
Rev. 24: 50-66.
SMITH, V. J., AND N. A. RATCLIFFE. 1980. Host defense reactions of the shore crab, Carcinus maenas
(L): Clearance and distribution of injected test particles. J. Mar. Biol. Assoc. U. K. 60: 89-102.
STEWART, J. E., AND B. M. ZWICKER. 1974. Induction of internal defense mechanisms in the lobster
Homarus americanus. Pp. 233-239 in Contemporary Topics in Immunobiology, Vol. 4, E. L.
Cooper, ed. Plenum Press.
UNKLES, S. E. 1977. Bacterial flora of the sea urchin Echinus esculentus. Appl. Environ. Microbiol. 34:
347-350.
VAN DER KNAAP, W. P. W. 1980. Recognition of foreigness in the internal defense system of the freshwater
gastropod Lymnaea stagnalis. Pp. 91-97 in Aspects of Developmental and Comparative 1m-
munology-l, J. B. Solomon, ed. Pergamon Press, Oxford.
VAN DER KNAAP, W. P. W., T. SMINIA, F. G. M. KROESE, AND R. DIKKEBOOM. 1981. Elimination of
bacteria from the circulation of the pond snail Lvmnaea stagnalis. Dev. Comp. Immunol. 5:
21-32.
WARDLAW, A. C., AND S. E. UNKLES. 1978. Bactericidal activity of coelomic fluid from the sea urchin
Echinus esculentus. J. Invert. Pathol. 32: 25-34.
WITTIG, G. 1965. Phagocytosis of blood cells in healthy and diseased caterpillars. I. Phagocytosis of
Bacillus thuringiensis in Pseudaletia unipuncta. J. Invert. Pathol. 1: 474-488.
WOOD, W. B. 1960. Phagocytosis, with particular reference to encapsulated bacteria. Bacteriol. Rev. 24:
41-49.
YUI, M. A. 1982. Bacterial clearance in the sea urchin, Strongylocentrotus purpuratus. M.S. Thesis.
Oregon State University, Corvallis.
Reference: Biol. Bull. 165: 487-542. (October, 1983)
ABSTRACTS OF PAPERS PRESENTED AT THE GENERAL
SCIENTIFIC MEETINGS OF THE MARINE BIOLOGICAL
LABORATORY AUGUST 16-19, 1983
Abstracts are arranged alphabetically by first author within the following categories:
cellular and molecular biology, cell structure; developmental biology; ecology, evolution,
and plant sciences; gametes and fertilization; microbiology; neurobiology, learning,
and behavior; parasitology, pathology, and aging; and photoreceptors, vision, and
rhythms. Author and subject references will be found in the regular volume index in
the December issue.
Ernest Everett Just (1883-1941): A Dedication. WILLIAM R. JEFFERY (University
of Texas at Austin).
This session of the General Meetings of the Marine Biological Laboratory (MBL) is dedicated to
the centenary of Ernest Everett Just, a distinguished Professor of Zoology at Howard University and
prominent investigator at the MBL and various European scientific institutes. The contributions of
E. E. Just, who received his Ph.D. under the direction of Frank R. Lillie at the University of Chicago,
are many and varied. His dissertation involved an experimental analysis of the generation of polarity in
the egg of the marine annelid Nereis, which he showed was determined by the relationship between the
position of the polar bodies and the point of sperm entry. After pursuing independent research at the
MBL on the nature of fertilization in Nereis and sand dollar eggs, he was among the first to recognize
and characterize the so-called fertilization wave, now known to be caused by a calcium-mediated exo-
cytosis of the cortical vesicles. Although E. E. Just is often remembered at Woods Hole for his genius
in the design of experiments and the handling of marine eggs, he has left an almost-forgotten legacy to
the modern field of cell biology. He correctly predicted that the cell surface was not simply a static
limiting membrane, but instead a dynamic, compound structure composed not only of the plasma
membrane but also of an underlying motile cortex which he called the ectoplasm.
E. E. Just's philosophy of these General Meetings, were he to preside over them this week, might
best be expressed by a quote from his last book The Biology of the Cell Surface published in 1939.
"Although we may deal with particulars, we return finally to the whole pattern woven out of these.
So in our studies of the animal egg; though we resolve it into its constituent parts the better to understand
it, we hold it as an integrated thing, as a unified system; in it life resides and in its moving surface life
manifests itself.
The aspirations of the present generation of MBL scientists, especially those of us who study the
isolated parts of cells or organisms, might well profit by carefully considering this thought of their long-
deceased colleague, Ernest Everett Just.
CELLULAR AND MOLECULAR BIOLOGY; CELL STRUCTURE
Studies of the isolation and calcium-induced fusion of fusogenic wild carrot proto-
plasts. NINA STROMGREN ALLEN (Dartmouth College) AND WENDY F. Boss.
Wild carrot suspension culture cells with the potential to undergo somatic cell embryogenesis have
been grown; these cells yield protoplasts which are fusogenic (Boss and Grimes 1983, submitted to
Protoplasmd). Morphological changes occurring during protoplast formation as well as during the fusion
process were studied using videoenhanced microscopy (Allen and Allen 1983, / Microsc. 129: 3-17).
Fusion of the protoplasts is calcium dependent and is inhibited by EGTA. The fusion process is
rapid, and is complete within 1 to 20 minutes after initial contact. When calcium is added, the fusogenic,
but not the nonfusogenic protoplasts, crenate. The fusion occurs as follows. First, there is contact rec-
ognition, then adhesion followed by fusion at the points of adhesion. The complete expansion of the
cytoplasmic connection and the mixing of the cell contents generally occurred in 20 minutes, but could
be enhanced by exposing the protoplasts to a hypotonic solution. Electron Spin Resonance studies of
fusogenic and nonfusogenic cells suggest that the glycerol backbone region of the membrane was less
fluid in nonfusogenic cells than in fusogenic cells.
487
488 ABSTRACTS FROM MBL GENERAL MEETINGS
These filamentous extensions (0.1 nm or less in diameter) were found on fusogenic protoplasts in 0.4
molal sorbitol or after digestion on 0.4 molal sorbitol and 2% driselase. These membranous extensions
(Hechtian threads) connected adjacent cells during wall digestion and as the cells broke apart, the threads
often terminated in knob-like structures. Such filamentous threads were not seen on nonfusogenic protoplasts.
Larger, more readily visible extensions were seen when cells were osmotically stressed with 0.8 molal sorbitol.
A videotape (available from N. S. Allen) demonstrates the fusion, plasmolysis, and digestion events.
Fusogenic carrot cultures provide an ideal system for the study of membrane fusion. Cell-to-cell
fusion has potential use in the study of membranes and membrane fusion as well as for genetic engineering.
Supported by grants from Monsanto Co. (N.SA.), Pioneer Hi-Bred International, Inc. and North
Carolina Agricultural Research Services (W.F.B.) and the generous loan of equipment from Carl Zeiss,
Inc., R. D. Allen, and Hamamatsu, Inc.
A proteinase inhibitor released from the Limulus amebocyte during exocytosis.
PETER B. ARMSTRONG (Department of Zoology, University of California, Davis),
JAMES P. QUIGLEY, AND JACK LEVIN.
The blood cell (amebocyte) of the horseshoe crab, Limulus, is packed with large oval granules that
can be stimulated to release their contents by exocytosis. Among the materials released are a system of
proteolytic enzymes involved in the clotting reaction (Levin 1979, Prog. Clin. Biol. Res. 29: 131). We have
identified, in addition, a potent proteinase inhibitory activity that is also released. In the standard preparation,
a uniform suspension of washed amebocytes is suspended in 10 ml of 0.5 M NaCl + 10 mM CaCl2.
Exocytosis is initiated by adding the ionophore A23187 (10 mM) and the preparation is incubated at room
temperature for 10-60 min. One ml of packed cells releases enough inhibitor to half-inactivate 4 mg of
bovine pancreatic trypsin (9100 BEAE units/mg). During the release reaction, cell lysis is negligible, as can
be ascertained by: ( 1 ) direct microscopic examination of amoebocytes adherent to microscope coverglasses
and (2) the absence of lactate dehydrogenase in the fluid phase recovered from a preparation of cells after
exposure to ionophore. (One ml of packed cells releases 30 units of LDH if lysed in distilled water but
releases none under the conditions of ionophore-induced exocytosis.) No inhibitory activity is released
from living cells that have not been stimulated to undergo exocytosis. The inhibitor preparation suppresses
activity against both high (casein) and low (BAPNA) molecular weight substrates, is relatively stable at low
pH (half inactivation occurs at pH 2.9, 1 h, room temp.), and high temperature (half inactivation occurs
at 100°C, 2 min), and both native and acid-treated releasate are active against trypsin, chymotrypsin, and
thermolysin. The inhibitor is also active against the clotting enzyme (a serine proteinase) that clots the
coagulogen in the Limulus amebocyte lysate reaction.
Supported by NSF Grant No. PCM80-24181.
Isolation and characterization of tubulin clones from Dictyostelium discoidium. MON-
ICA CARSON AND REX L. CHISHOLM (MIT).
The a- and 0-tubulin genes have been found in three types of genomic organization. In most higher
eukaryotes, the a- and 0-tubulin genes exist as unlinked elements dispersed throughout the genome. In
Leishmania enriettii, the a-tubulin genes are tandemly repeated but unlinked to the /8-tubulins which
are also tandemly repeated. Finally, in Trypanosoma brucei, the a- and |8-tubulin genes are linked and
this a- and /3-tubulin unit is tandemly repeated. These three types of genomic organization may be
necessary for differential transcriptional regulation.
Dictyostelium discoidium possesses distinct developmental stages capable of directed cell movement.
Cellular movement at each stage probably involves cytoskeletal elements. To investigate the structure
of the Dictyostelium tubulin genes and as a prerequisite to studies of their expression, a Dictyostelium
genomic library was constructed and screened using either a Chlamydomonas reihardii a- or /3-tubulin
probe. Five recombinant phage which hybridized to the Chlamydomonas a-tubulin probe and eight
recombinant phage which hybridized to the Chlamydomonas /3-tubulin probe were plaque purified.
Initial restriction mapping of each of the Dictyostelium a- and /3-tubulin clones suggests that both
the a- and /3-tubulin clones contain overlapping segments of DNA from the same or similar regions of
the Dictyostelium genome. Furthermore, both the Chlamydomonas a- and /3-tubulin probes hybridize
to the identical 4 Kb Eco RI fragment of the Eco RI digested Dictysostelium tubulin clones. The 4 Kb
fragment from both a Dictyostelium a- and /3-tubulin clone was purified and used to probe a genomic
Southern of Dictyostelium DNA. The 4 Kb fragment from both the Dictyostelium a- and /3-tubulin clones
hybridized to the same fragments of Dictyostelium DNA. Therefore the Dictyostelium a- and /3-tubulin
. . nes appear linked to each other as observed in Trypanosoma. However, the Dictyostelium and Chla-
mydomonas probes hybridize to genomic fragments which co-migrate with the predicted mobilities of
CELLULAR, MOLECULAR BIOLOGY, ETC. 489
ribosomal DNA. The Dictyosteliurn probes also appear to hybridize to both ribosomal RNA and to RNA
which migrates at the a- and /3-tubulin position. These results suggest that in addition to the a- and /3-
tubulin gene linkage, these genes also may be linked to the ribosomal RNA gene cluster. Alternately, the
observed hybridization could result from fortuitous cross hybridizations. Further experiments to distin-
guish between these possibilities are in progress.
This work was supported by NIH Training Grant GM-31 136-05.
Immunofluorescence of Allogromia reticulopodia. V. E. CENTONZE (Dartmouth
College) AND J. L. TRAVIS.
Allogromia laticollaris, a marine foraminifera, extends a radial reticulopodial network upon settling
on a solid substratum. Bidirectional streaming and saltatory particle movements are evident in the
spreading network. Previous studies (Travis and Allen 1 98 1 , J. Cell Biol. 90: 2 1 1 -22 1 ) of Allogromia on
both the optical and electron microscope levels show that particle movement coincides with the position
of microtubule bundles which are the major cytoskeletal elements of the reticulopodial extensions.
To determine tubulin antigenic crossreactivity, we probed gluteraldehyde fixed networks with an-
tibodies prepared against tubulin from widely divergent organisms. A polyclonal antibody prepared by
Miles against chicken brain tubulin produces an intense staining of the microtubule bundles. When
comparing phase and fluorescence light micrographs it becomes apparent that individual microtubule
bundles may be resolved, especially in flattened lamellipodial regions. A monoclonal antibody probe, 34
#10, prepared against yeast tubulin produced a similar staining pattern identifying only the fibrous
bundles. Another monoclonal antibody YL1/2, specific to the carboxy terminal end of tyrosylated <Y-
tubulin, also stained the microtubule bundles. Fluorescence staining produced by this antibody was
similar though less intense. Therefore, due to this monoclonal's specificity we propose the tyrosylated
form of tubulin is a subset of Allogromia tubulin.
We would like to specially thank Dr. John Kilmartin for providing both of the monoclonal anti-
bodies. We would also like to thank J. Rosenbaum, E. Stromboli, and the entire Marine Biological
Laboratory Physiology Course. Support for this work was NIH Training Grant GM-31 136-05.
Marginal band function in the dogfish erythrocyte. WILLIAM D. COHEN AND
JACQUELYN JOSEPH-SILVERSTEIN (Hunter College, NY).
Marginal bands (MBs) of microtubules in mature erythrocytes of all non-mammalian vertebrates are
believed to function universally in cellular morphogenesis (transformation from sphere to flattened ellipse)
but not in cell shape maintenance (Behnke 1970, J. Ultrastruct. Res. 31: 61-75; Barrett and Dawson 1974,
Dev. Biol. 36: 72-8 1 ). The primary supporting evidence is that, in mature erythrocytes of certain species
(e.g., chicken), the MB disassembles at 0°C while native cell shape is retained. Although the same observation
can be made with erythrocytes of the smooth dogfish (M. canis), we believe the interpretation to be incorrect.
Two methods were used to produce dogfish erythrocytes containing or lacking MBs under otherwise similar
conditions: (a) stabilization of the MB at 0°C by taxol, and (b) inhibition of MB reassembly at room
temperature by nocodazole or colchicine. Cells with or without MBs had normal shape. Anucleate ghosts
were prepared by osmotic lysis and shearing of cells at 0°C. Ghosts containing MBs generally retained a
flattened elliptical shape, while those without MBs buckled. Living cells contained MBs at 0°C and when
subjected to mechanical stress (fluxing in glass capillary tubes) similarly maintained normal shape, whereas
those lacking MBs did not. The same result was obtained using fluxed cells + and -MBs at room temperature.
How might the MB maintain cell shape under such conditions? If normal dogfish erythrocytes are incubated
at room temperature for long periods (approx. 5-24 h), abnormal pointed cells containing pointed MBs
appear. However, we found that cells lacking MBs do not form points, demonstrating that MB shape
determines cell shape. We propose that erythrocyte shape coincides with the shape of the cell surface-
associated cytoskeleton (SAC), within which the MB acts as a flexible frame. We conclude that MBs may
function to maintain erythrocyte shape in non-mammalian vertebrates, resisting deformation and/or rapidly
returning deformed cells to an efficient shape in the circulation.
Supported by Professional Staff Congress-City University of New York grant #13567 and #6-63177,
and by NSF #PCM-8107195.
Actin microfilaments are a major cytoskeletal component in squid axoplasm. KARL
R. PATH (Case Western Reserve University) AND RAYMOND J. LASEK.
The axoplasm of the squid (Loligo pealei) giant axon can be extruded from its sheath leaving a 10
thick cortical rim of axoplasm with the discarded plasma membrane. The extruded cylinder of
490 ABSTRACTS FROM MBL GENERAL MEETINGS
axoplasm contains 1.4 mg/ml actin-60% of which is assembled into actin microfilaments (MF) (Morris,
in press, J. Cell Biol.) principally as a polymer approximately 0.5 ^m in length.
Two to three n\ of axoplasm was extruded into 200-300 n\ buffer (designed to simulate the solution
conditions in the axon, Morris 1982, J. Cell Biol. 92: 192-198) containing 10 nAf phalloidin which binds
to and stabilizes polymeric actin. Potassium iodide (0.6 M) was then added to disperse the axoplasm by
denaturing neurofilaments and microtubules, but leaving the phalloidin-stabilized MF intact. The dis-
persed axoplasm was reacted on a grid with the myosin subfragment one (S-l) and negatively stained
for transmission electron microscopy.
Measurements of a total of 500 S-l decorated MF from four different axons revealed a distribution
of lengths with a mode at 0.45 ^m (40% were between 0.3-0.6 j/m) and a range from 0.2 to 3 nm. Control
preparations without phalloidin contained MF of similar lengths suggesting that actin polymerization
was not induced by drug treatment. Purified skeletal muscle actin when polymerized in our buffer and
processed in an identical manner were much longer than axoplasmic MF indicating that MF were not
sheared in our preparations.
Other studies have shown that intact MF are necessary for transport of membranous vesicles in the
squid giant axon. We feel that the relatively modest lengths of axoplasmic MF reported above may limit
the types of models we can build regarding a role of actin in motile mechanisms.
Characterization of Trypanosoma brucei tubulin genes. A. FLISSER, A. S. FAIR-
HELD, AND D. WIRTH (Harvard School of Public Health).
Microtubules are associated with many eukaryotic cell functions. Alpha (a) and beta (/3) tubulins are
the main proteins of microtubules. Tubulin genes have been identified in organisms such as Chlamydomonas,
Drosophila, and man, where they appear as distinct gene families which exist in dispersed multiple copies.
Recently, a and /3 tubulin genes have been identified in Trypanosoma brucei and Leishmania enriettii;
unlike other eukaryotes, however, the genes are arranged as tandem repeats. Several hypotheses to explain
a tandem gene arrangement have been proposed, of which the most likely explanation is that tubulin is a
major biosynthetic product (up to 10% of total cell protein) of the organism.
In the work reported here, the alpha and beta tubulin genes from T. brucei were identified by
Southern blot using heterologous a and /3 tubulin probes from Leishmania enriettii. Restriction cut (Pst
1) T. brucei DNA was cloned into the bacterial plasmid pBR322, and of the resulting genomic library
93% of the clones contained inserts. The library was screened by colony hybridization and four positive
clones were identified with the tubulin probes. Two of these clones were isolated, the plasmid purified
and analyzed by restriction mapping.
A strategy to differentiate mutants affecting voltage-sensitive sodium channels in
Drosophila. LINDA M. HALL (Albert Einstein College of Medicine).
A goal of this laboratory is to identify the genes involved in the production and regulation of voltage-
sensitive sodium channels found in excitable cells. We are interested in developing pharmacological
procedures which will allow us to distinguish between different classes of mutants affecting this ion
channel. Two general mutant classes of interest would be: (1) those which increase channel activity and
(2) those which decrease channel activity. The first class of mutants would include those which affect
channel regulation causing overproduction and those which affect the activation and inactivation pro-
cesses. These would have agonist-like effects. The second class would be antagonist-like and would include
those which block channel function as well as those which reduce the number of channels produced
without affecting function. We predict that mutants which increase channel activity should show increased
sensitivity to agonists such as veratridine and decreased sensitivity to antagonists such as tetrodotoxin.
In contrast, mutants which decrease channel activity should show decreased sensitivity to agonists and
increased sensitivity to antagonists. To test this hypothesis we have used the temperature-sensitive par-
alytic mutant napK which has a reduced number of sodium channels as revealed by 3H-saxitoxin binding
studies (Hall et al. 1982, Ciba Found. Symp. 88: 207-220). Flies were fed either the agonist veratridine
or the antagonist tetrodotoxin and the lethality at specific doses was compared with that of wild-type
flies. As predicted by our hypothesis, the nap mutant was resistant to veratridine and sensitive to tetro-
dotoxin. Thus, by screening for tetrodotoxin-resistant mutants and then identifying that subclass which
show increased sensitivity to veratridine, it should be possible to identify new classes of sodium channel
mutants with increased channel activity. It will be of interest to determine whether these mutants identify
new genes or coincide with those already identified on the basis of temperature-induced paralysis and
•ations in 3H-saxitoxin binding activity.
This work was supported by NIH grant 16204.
CELLULAR, MOLECULAR BIOLOGY, ETC. 491
Structure of the isolated and in situ giant smooth muscle fibers q/Mnemiopsis leydii
(ctenophora). MARI-LUZ HERNANDEZ-NICAISE AND GHISLAIN NICAISE (Univ-
ersite Claude Bernard, Villeurbanne, France).
The first example of a giant smooth muscle cell has been reported in the mediterranean ctenophore
Beroe ovata (Hernandez-Nicaise et al. 1980, J. Gen. Physiol. 75: 79-105). These cells have been suc-
cessfully isolated in a functional state (Hernandez-Nicaise et al. 1982, Proc. Natl. Acad. Sci. 79: 1884-
1888). The limited availability of beroids prompted us to seek another suitable species. The lobate
ctenophore Mnemiopsis leydii — which is common during the summer in Woods Hole — possesses such
giant fibers, grouped in 2 sagittal bundles. Functional isolated cells were obtained after a sequential
digestion of mesoglea in 0.3% hyaluronidase (type III Sigma) for 75-90 min, followed by 0.3% hyal-
uronidase + 0.05% trypsin (type III Sigma) for 20-30 min, at 30°C, in Ca-free artificial sea water.
Each bundle is made of 30 to 50 multinucleated cylindrical cells which may reach 35 ^m in diameter
and 2 cm in length. The nuclei and non-contractile organelles (mitochondria, golgi, granular endoplasmic
reticulum) are contained in a discontinuous axial core, surrounded by a thick sheath of myofilaments.
Thin (actin) filaments, 5.9 nm in diameter, form irregular rosettes around the thick (myosin) filaments,
16.1 nm in diameter. An actin:myosin filament ratio of 7.2 and a myosin density of 249 filaments per
unr were found in cross-sections of relaxed in situ cells. No dense bodies or attachment plates were
observed. From the coiled shape of contracted single cells and from the rearrangement of organelles in
such coiled cells, we propose that myofilaments are organized in thin long myofibrils attached upon the
cell membrane at both ends, and that the attachment sites follow two (sets of) enantiomorphic helices.
The sarcoplasmic reticulum builds up a longitudinally oriented 3-dimensional network of narrow tubules
among the myofilaments. Its relative volume, estimated from cross-sections, amounts to 0.9% of the
contractile cytoplasm. No peripheral couplings have been observed, nor any tubular or vesicular invag-
ination of the sarcolemma.
Supported by NATO grant #251-81, and the CNRS (L.A. 040244).
Opposite end assembly-disassembly of single microtubules. H. HOTANI AND J. L.
TRAVIS (Yale University, Dept. Biology).
Tubulin assembles onto both ends of a microtubule filament and the microtubule grows quickly at
its plus end and more slowly at the minus end. The critical concentration for the assembly of tubulin
is higher at the minus end than the plus end. Analysis of tritiated GTP incorporation into microtubules
at steady state has suggested that treadmilling of tubulin subunits through the microtubule occurs. There
is therefore a net addition of tubulin subunits at the plus end and a net loss from the minus end, yet the
microtubule remains the same length. If this treadmilling occurred in the living cell, it might cause the
microtubule to change position relative to a fixed structure. This could have great importance for the
mechanism by which microtubules function in mitosis, particle movement, and other microtubule-based
processes. We visualized the treadmilling in single microtubules by dark-field light microscopy and dynein
decoration.
Purified brain microtubule protein was assembled into microtubules, the microtubules were soni-
cated to break them into small pieces, and the pieces were then decorated with purified Tetrahymena
dynein ATPase. When the microtubules are decorated with dynein they can easily be distinguished from
undecorated ones in the dark-field microscope because the decorated ones appear fat, and the undecorated
ones quite thin. The small pieces of decorated microtubules were then incubated with brain tubulin at
a concentration that permitted elongation to occur at both ends of the microtubules; the system was
allowed to come close to equilibrium, (little change in microtubule length) and then a video recording
was made of the changes in lengths of the undecorated segments which had elongated from both the plus
and the minus ends of the dynein decorated piece of microtubule. The length of the undecorated mi-
crotubule at the plus end increased (4 ^m/h) and that at the minus end decreased (3 nm/h), while the
decorated portion did not change in length. Moreover, since the decorated microtubule section was
attached to the coverslip, we observed that the microtubule changed position relative to other fixed
structures in the field due to its head to tail assembly.
We would like to thank Dr. J. Rosenbaum and the entire Marine Biological Laboratory Physiology
Course.
Fully automated image analysis can be used to study intramembranous particle
(IMP) behavior during development in Tetrahymena. LINDA A. HUFNAGEL
(University of Rhode Island).
The cell surface of Tetrahymena is covered by three membranes, the plasma membrane (PM), and
outer and inner alveolar membranes (OAM and IAM). The OAM and PM are closely associated via frequent.
492 ABSTRACTS FROM MBL GENERAL MEETINGS
10 nm long, cross-linking fibers, and thus assembly of these membranes must be coordinated. Nevertheless,
freeze-fracture studies reveal that these membranes have unique structures, which respond differently to
reduced temperatures (c.f. Hufnagel 1 98 1 , / Protozool. 28: 1 92-203). Comparison of morphological changes
in these two membranes during membrane growth accompanying refeeding of starved cells would be of
considerable interest. Membrane structure can be described in terms of size, frequency, orientation, and
locations of IMPs, considered to represent transmembrane proteins. To hasten such an analysis, the Zeiss
IBAS analysis system, attached to a video camera and light box, is being used to record, process, digitize,
and measure IMPs, starting with EM negatives. Suitability of this fully automated system for IMP analysis
was previously reported (Hufnagel 1983, Proc. EMSA 41st Ann. Meeting, pp. 637-639). Cells starved
overnight in 10 mM Tris buffer, and starved cells refed for several hours (thus in early stages of cytokinesis)
were compared. Based on measurements of several thousand IMPs, particle frequency increased from 2324
± 377 IMPs/nm2 in starved cells to 5138 ± 108 IMPs/nm2 in fed cells. Area and diameter distributions
of IMPs were also obtained. Differences were detected in the relative increase in frequency of different size
classes of particles in starved versus fed cells. Visual inspection of digitized images revealed differences in
spatial arrangements of IMPs, in starved and fed cells. Analysis of distribution of asymmetric IMPs relative
to angular orientation suggests that particle asymmetry results partly from shadowing direction, but that
classes of similarly oriented IMPs may exist in the PM. These preliminary observations suggest that membrane
structural changes during development can be characterized quite effectively and efficiently by fully automated
image analysis of freeze fracture replicas.
Visualizing extremely low contrast images by digital enhancement of selected por-
tions of the image grey scale. SHINYA INOUE (Marine Biological Laboratory),
THEODORE D. INOUE, AND GORDON W. ELLIS.
Image contrast in the light microscope has been substantially improved over the past 50 years.
However, one could detect and measure smaller retardations, absorbances, fluorescence, etc., and uncover
finer structural details of the specimen, if contrast could be improved further. Two years ago, we (Inoue
J. Cell Biol. 89: 346-356) and Allen et al. (Cell Motility 1: 275-289, 291-302) reported the use of video
to enhance microscope image contrast. This summer, we developed an interactive digital image processing
system that enhances selected regions of the image grey scale. The system works with video cameras
providing standard video signals, is simple to use, and less expensive than other digitized image enhancing
systems with comparable potentials. We can average out statistical image noise, subtract noise-averaged
background, select the image grey level to be enhanced and the degree of contrast enhancement, display
the enhanced regions in pseudo-color with the unenhanced regions in natural grey scale or pseudo-color,
sharpen edges, generate differential contrast, detect motion, etc., all in real time, and provide image
convolutions in fractional seconds. The computer hardware, including three 512 X 512 X 8 bit frame
buffers, an analog processor, and an arithmetic logic unit, were acquired from Imaging Technology Inc.
of Woburn, Massachusetts. The computer program for interactive image manipulation was developed
primarily by Ted Inoue. Performance of the new system, which requires little experience with computers,
and which should be applicable to electron microscopy, radiography, astronomy, surveillance, and in-
dustrial applications, in addition to light microscopy, was demonstrated at the Meetings. The system
attached to the microscope was demonstrated the same evening.
Grant support: NIH 5R01 GM 31617-02, NSF PCM 8216301.
Composition and function of the cytoskeleton in "blood clam" erythrocytes.
JACQUELYN JOSEPH-SILVERSTEIN (Hunter College, NY) AND WILLIAM D.
COHEN.
Erythrocytes ofNoetia ponderosa and related species contain a marginal band (MB) of microtubules
and a cell surface-associated cytoskeleton (SAC). The MB is cold labile, disassembling at 0°C and reas-
sembling upon rewarming. When nucleated cytoskeletons are prepared from room temperature cells by
Triton lysis in microtubule-stabilizing medium and analyzed by SDS-PAGE, the major protein com-
ponents are tubulin and two proteins which comigrate with human erythrocyte a-spectrin and actin.
Disassembly of the MB at 0°C allows one to localize proteins to the MB. When protein components of
cytoskeletons lacking MBs (cells at 0°C) are compared to those from cytoskeletons with MBs (cells at
room temperature), a diminution of the tubulin doublet and a decrease in two minor proteins (~80K,
— 105K) is observed. No change is apparent in those proteins comigrating with human erythrocyte a-
spectrin and actin, suggesting that they are in the SAC. The possibility that the ~80K and ~105K
components are MB microtubule-associated proteins (MAPs) was examined further. Cells at room temp.
\ere prepared with and without MBs by inhibiting MB reassembly with nocodazole or colchicine. Nu-
cleated cytoskeletons with and without reassembled MBs were compared for protein content by SDS-
CELLULAR, MOLECULAR BIOLOGY, ETC. 493
PAGE. Cytoskeletons from cells with reassembled MBs were enriched for the ~80K and ~ 105K proteins
as well as for tubulin. The results suggest that the ~80K and ~105K proteins are MAPs which cycle
with the MB. The ability to produce erythrocytes with and without MBs at room temperature allowed
us to examine the role of the MB in cell shape maintenance under conditions in which cells are subjected
to mechanical stress. When erythrocytes with and without MBs were fluxed in 10 ^1 capillary tubes, those
with MBs were still flattened and elliptical (98%), while many of those without MBs were deformed (20-
45%). The MB may thus play a role in cell shape maintenance, effecting the rapid recovery of erythrocyte
shape following deformation.
Supported by PSC-CUNY grant #13567 and #6-63177, and by NSF #PCM-8107195.
Two-dimensional gel analysis of sea urchin ciliary tubulins. THOMAS KELLY, JOEL
L. ROSENBAUM, AND TlM HUNT.
We isolated cilia from Arbacia punclulata embryos according to the method of Stephens (1977, Dev.
Biol. 61: 311-329). Samples were analyzed by two-dimensional gel electrophoresis. a and /3 tubulins
migrated as single discrete spots on coomassie blue stained gels. In a separate experiment, developing
embryos were labeled for 20 minutes with 35S-methionine five and nine hours post-fertilization. Whole
embryos were analyzed by two-dimensional gel electrophoresis. The resulting fluorograms showed that
a and fl tubulins migrated as discrete spots in the same positions as the tubulins from mature detached
cilia. Thus, in contrast to flagellar a tubulin of Chlamydomonas reinhardii (L'Hernault and Rosenbaum
1983, J. Cell Biol. 97: 258-263) and a tubulin from cilia of Polytomella agilis (McKeithan and Rosen-
baum 1981, J. Cell Biol. 91: 352-360), there is no evidence for a posttranslational modification of a or
(8 tubulin which would change the isoelectric point or molecular weight of ciliary tubulins relative to
cytoplasmic tubulins in sea urchins.
This work was supported by N.I.H. training grant GM-31 136-05. The authors thank Emilio Strom-
boli for stimulating discussions.
Enhancement of the appearance of lateral projections on negatively stained microtubules
after glutaraldehyde — tannic acid fixation. GEORGE M. LANGFORD (University
of North Carolina, Chapel Hill, NC).
Methods for enhancing the visualization of microtubule-associated proteins (MAPs) on the surfaces
of reassembled neuronal microtubules (MTs) by negative staining were investigated. A drop of MTs,
diluted 10-20 fold, was placed on a carbon-form var coated grid for 8-10 s. The grid was rinsed with 1-
2 drops of buffer, stained with 8-10 drops of 1% uranyl acetate (UA), air dried, and examined in the
electron microscope. This staining procedure yielded light and dark staining populations of MTs. The
dark staining MTs had short, globular projections on their surfaces while the surfaces of the light staining
ones appeared smooth. The two populations of MTs resulted from differences in the staining reaction
of MTs suspended in a droplet of buffer on the grid and those MTs adsorbed to the surface of the grid.
Microtubules that were adsorbed to the grid surface were flattened and their MAPs were attached to the
carbon-formvar film in an extended configuration. The MAPs in this configuration were difficult to
visualize by the UA stain and only an indistinct band, 40-50 nm in width, of fine, filamentous material
was seen along the sides of the MTs Microtubules that were in suspension were "fixed" by the UA stain
and their MAPs coiled into short globular projections, 7-10 nm in length; UA had altered the length and
configuration of the MAPs. To enhance the visualization of the MAPs, MTs were fixed in 1% glutar-
aldehyde-0.2% tannic acid before staining. This method of fixation increased the diameter of the pro-
jecting MAPs, thereby enhancing their contrast, but causing them to shorten to 20-25 nm; a length
which is shorter than the expected extended length of the MAPs. These data demonstrate that glutar-
aldehyde-tannic acid fixation is a very useful method for enhancing the contrast of MAPs on reas-
sembled MTs.
Supported by NIH grant GM28107.
Structure and expression of tubulin genes in the protozoan parasite Leishmania enriettii.
SCOTT LANDFEAR (Harvard University).
In the gut of the insect vector, protozoan parasites of the genus Leishmania exist as highly motile,
flagellated, extracellular organisms called promastigotes. When promastigotes are injected into the mam-
malian host by a bite of the sandfly vector, the parasites are phagocytized by host macrophages and
develop into intracellular non-motile forms, called amastigotes, which possess only a residual flagellum.
Amastigotes synthesize low levels of tubulin proteins, but the biosynthesis of both «- and /3-tubulin is
greatly increased during the transformation of amastigotes to promastigotes.
494 ABSTRACTS FROM MBL GENERAL MEETINGS
Previously, we have used a genomic a-tubulin clone from Leishmania enriettii to show that the
chromosomal copies of the a-tubulin genes are arranged in a precise tandem repeat containing about 15
copies of the 2 kilobase repeat unit. We have now cloned a copy of the /3-tubulin gene. This 4 kilobase
fragment of genomic DNA contains single sites for the restriction enzymes Bam HI, Xho I, and Hind
III. If genomic DNA is cut with these restriction enzymes, run on a Southern blot, and probed with the
/3-tubulin clone, a single 4 kilobase fragment hybridizes in all three digestions. This result shows that
each restriction site within the 0-tubulin gene is bounded, in the chromosomal DNA, by another such
site 4 kilobases away. The /3-tubulin genes must therefore be arranged in a tandem repeat consisting of
4 kilobase repeat units.
Equal amounts of total RNA from amastigotes and promastigotes have been run on Northern blots
and probed with the a- or /3-tubulin clones. The hybridization of either a- or /3-tubulin mRNA is about
5 to 10 fold higher in RNA from promastigotes compared to RNA from amastigotes. This result dem-
onstrates that tubulin gene expression is controlled at the level of mRNA accumulation during the
Leishmania life cycle.
Voltage clamp studies of dispersed toadfish pancreatic islet cells. D. R. MATTESON
(Dept. of Physiol., Univ. of Pennsylvania).
Pancreatic islet cells isolated from toadfish were voltage clamped using the whole cell variation of
the patch clamp technique. The cells were dispersed by treating islets with 2 mg/ml of trypsin and 1 mg/
ml of collagenase for 20 min at room temperature. Giga seals were readily obtained on isolated single
cells, 10-12 ^m in diameter, with 3 to 5 Mohm patch electrodes. By measuring capacitive currents, total
cell capacitance was estimated to be 3.5 ± 1.4 pF (6 cells). In the presence of 130 mA/ Na + 10 mA/Ca
externally and 130 mA/K + 10 mA/Cs internally ( 1 30 Na 10 Ca//130 K 10 Cs), the voltage dependent
ionic current at 0 m V consisted of a rapidly activating inward current, followed by a more slowly activating
phase of outward current. The reversal potential of the fast, early current is close to the calculated Na
equilibrium potential, and the current is blocked by tetrodotoxin (TTX), indicating that it is generated
by Na channels. The outward currents were blocked when patch electrodes were filled with Cs+, indicating
that K channels carry this current component. In 130 Na 10 Ca//140 Cs, two patterns of inward current
were frequently seen. ( 1 ) In some cells, the Na current appeared to only partially inactivate to a maintained
level of inward current. After TTX block of the Na channels in these cells, the remaining inward current
activated more slowly, did not inactivate in 7 ms, and was larger in the presence of Ba+4 than in Ca++.
This TTX insensitive inward current is most likely carried by Ca channels. (2) In the other type of cell,
the Na current inactivated completely revealing no maintained inward current.
High molecular weight (380Kd) ATPase in axoplasm of squid giant axon. M. M.
PRATT (Univ. of Miami School of Medicine).
Vesicle and organelle transport in axoplasm is a dramatic example of microtubule-associated mo-
tility, however, the mechanism by which this movement is generated is unknown. The force for micro-
tubule-mediated movements in ciliary and flagellar axonemes is provided by dynein, a Mg++-ATPase
with unique enzymatic properties, and a protein composition which includes polypeptides of 300-400
Kd. Since a cytoplasmic dynein can be isolated from unfertilized sea urchin eggs by calmodulin (CaM)
affinity chromatography, this technique was used in the study to examine squid axoplasm for the possible
presence of a high molecular weight ATPase which associated with microtubules.
When a soluble extract of axoplasm was fractionated on a CaM affinity column, a portion of the
total ATPase bound to the column in the presence of CaCl2, and could be eluted with EGTA, a calcium
chelator. SDS-polyacrylamide gel electrophoresis showed that the EGTA-eluted ATPase activity was
associated with a polypeptide of 380 Kd, along with minor bands at approximately 80 Kd, 70 Kd, and
60 Kd. To examine the association of the 380 Kd protein and of ATPase activity with microtubules, a
soluble cytoplasmic extract was prepared in tubulin isolation buffer. Stable microtubules were polymerized
from this fraction (using 10 nM taxol and 1 mM GTP) and 85% of the ATPase cosedimented with the
microtubules, along with nearly all of the 380 Kd polypeptide. When these microtubules were extracted
with 0.35 M NaCl, both the 380 Kd polypeptide and about 90% of the ATPase activity were solubilized.
The ATPase activity in the NaCl extract was activated equally by Mg++ or Ca++. When assayed in 0.5
M KC1 and 2 mA/ EDTA, the enzyme exhibited less than half of the Mg++ activated activity, suggesting
that it is not myosin-like. The Mg++-ATPase activity was inhibited 50% by 0. 1 mA/NajVO.,, an inhibitor
of dynein ATPase, and only 20% by NaF, an inhibitor of non-specific phosphatase.
These results suggest that axoplasm contains an ATPase of 380 Kd which can be partially purified
by CaM affinity chromatography in a manner similar to cytoplasmic dynein. Cosedimentation experi-
CELLULAR, MOLECULAR BIOLOGY, ETC. 495
ments further demonstrate that both the 380 Kd polypeptide and Mg++-ATPase activity associate with
repolymerized axoplasmic microtubules.
This study was supported by a Steps Toward Independence Fellowship from the Marine Biological
Laboratory, and by NSF grant PCM 81-19156.
Characterization and isolation of a homologue of alpha-2-macroglobulin from the
plasma of the horseshoe crab, Limulus. JAMES P. QUIGLEY (Marine Biological
Laboratory) AND PETER B. ARMSTRONG.
A proteinase inhibitor detected in the plasma of the horseshoe crab, Limulus. displays the following
features diagnostic for «2 macroglobulin: ( 1 ) the inhibitor is active against a variety of endopeptidases of
differing catalytic mechanisms (trypsin, chymotrypsin, plasmin, elastase, subtilisin, thermolysin, and papain),
(2) it suppresses activity against high — but not low — molecular weight substrates, (3) it protects the active
site of trypsin against macromolecular active site inhibitors such as soybean trypsin inhibitor, and (4) its
activity is destroyed by methylamine and low pH treatment. The inhibitor has been isolated from the cell-
free, hemocyanin-free plasma by polyethylene glycol precipitation (5.5-12% cut), followed by two passages
over a Sephacryl S-300 column. The inhibitor elutes from the column corresponding to a molecular weight
of 520 X 103 d. On the same column, human «2 macroglobulin elutes at the expected molecular weight
of 720 X 103 d. Electrophoresis of the isolated Limulus inhibitor on 6% polyacrylamide gels under non-
reducing conditions yields a single band at approximately 500 x 103 d, using unreduced plasma fibronectin
(440 x 103 d) as the molecular weight standard. Under reducing conditions, a single major band is present
at approximately 180 x 103 d, close to the position of the human a2 macroglobulin subunit. These data
are consistent with the possibility that Limulus a2 macroglobulin is a trimer of a 180 X 103 d subunit, in
contrast to the tetrameric structure of mammalian and the dimeric structure of fish a2 macroglobulin.
Supported by NSF Grant No. PCM 80-24 181.
Inhibition of mitotic anaphase and cytokinesis and reduction of spindle birefringence
following microinjection of anti-calcium transport enzyme IgGs into Echinaracnius
parma blastomeres. ROBERT B. SILVER (Department of Biological Chemistry,
Univ. of Health Sciences, North Chicago, IL).
Monospecific antibodies to the calcium transport enzyme (a-Ca-pump) inhibit mitosis when microin-
jected into sand dollar (E. parma) blastomeres. Immunoglobulin Gs (IgGs) were raised against the calcium
transport enzyme (Ca-pump) of sarcoplasmic reticulum from both rat skeletal muscle and guinea pig ileum
smooth muscle. Specific IgGs were further purified from whole IgG preparations by immunoarnnity chro-
matography, using the electrophoretically purified SR-Ca-pump as the immobilized ligand. ELISA dem-
onstrated that common epitopes are shared by SR, SR-Ca-pumps from rat skeletal and guinea pig smooth
muscle, and isolated membrane containing, "native" mitotic apparatus (MA) from first cleavage Stron-
gylocentrotos purpuratus embryos. Preimmune sera gave negative results in identical control assays. Triton
X-100 extraction of MA removes the SR-Ca-pump antigens. These «-SR-Ca-pump IgGs inhibit ATP
dependent 45Ca-sequestration by purified calcium sequestering MA membranes (Silver et al. 1980, Cell 19:
505-516) in a concentration dependent fashion. Indirect immunofluorescence light microscopy of isolated
native MA demonstrated coincident localization of the MA-Ca-pump, sequestered calcium (Ca-7-chloro-
tetracycline chelates), and membrane vesicles (differential interference contrast). Fluorescent foci were non-
uniformly distributed throughout the volumes of the asters and spindle. The majority of the MA-Ca-pump
and sequestered calcium was found in aspherical zone from 3 to 8 micrometers from the mitotic poles.
The mitotic poles were devoid of fluorescence, and thus do not have the MA-Ca-pump or sequestered
calcium. Microinjection of the a-Ca-pump IgGs into one of the two sister blastomeres, at second metaphase,
resulted in mitotic arrest of the injected cell, accompanied by a rapid loss of spindle birefringence. Karyomeres
formed and fused to form nuclei at the site occupied by the chromosomes at the time of injection of the
IgGs. The cleavage furrow did not develop in cells injected at metaphase. The cleavage furrow arrested,
then relaxed in cell injected in anaphase or beyond. Noninjected sister cells, and neighboring blastomeres
continue normal mitotic cycling. Routine control injections of bioled immune IgG, pre-immune IgG,
Wesson oil, buffer, or goat-anti-rabbit-IgG did not affect mitosis, Br of the MA, or cleavage furrow activity.
From these data it is clear that the MA-Ca-pump plays a key part in the functioning of the MA and in
mitosis.
This work was supported by a grant from the American Cancer Society (#CD-128) and a Steps
Towards Independence Fellowship from the Marine Biological Laboratory, Woods Hole, Massachusetts.
496 ABSTRACTS FROM MBL GENERAL MEETINGS
Lactoperoxidase-tubulin interaction in ciliary membranes. R. E. STEPHENS (Marine
Biological Laboratory).
Rousett and Wolff (1980, / Biol. Chem. 255: 2514) recently demonstrated that lactoperoxidase
(LPO) binds to both brain microtubules and tubulin at ratios of 0.2-0.3 and 2 moles LPO per tubulin
dimer, respectively, with a binding constant of 2 X 106 M '. Based on their work, I am using LPO
binding to study the disposition of membrane tubulin in molluscan (scallop) gill ciliary membranes and
in membrane vesicles reconstituted by detergent removal/freeze-thaw (Stephens 1983, J. Cell Biol. 96:
68-75). LPO interaction with intact cilia results in vesiculation and partial membrane protein solubili-
zation but in only minimal LPO binding to the remaining membrane. In the case of reconstituted vesicles,
however, the binding approaches one mole of LPO per mole of membrane tubulin dimer, resulting in
a monodisperse vesicle population of uniformly increased density. Half-maximal binding occurs in the
micromolar range, implying an apparent binding constant of > 106 M ~'. Judged both by direct sedimen-
tation analysis and by a shift in the Soret spectrum of the LPO heme group (characteristic of LPO-tubulin
binding), the interaction is relatively slow, going to completion in about 30 minutes at 25°C. The
interaction is slowed further by salt but is not inhibited by colchicine at 1 mM. Similar observations
were made by Rousett and Wolff for LPO-brain tubulin interaction. When either whole cilia or ciliary
membrane vesicles are labeled with LPO and fixed with glutaraldehyde/tannic acid/osmium, no obvious
surface labeling is evident; subjectively, the LPO-labeled membranes simply appear more granular. Two
conclusions can be drawn from these results: 1 ) LPO interacts with ciliary membrane tubulin in the same
manner as with brain tubulin; and 2) membrane tubulin in intact cilia is less accessible to direct LPO
interaction than in reconstituted vesicles, implying either inside-out vesicles or random insertion of
membrane tubulin. These results also suggest that labeled LPO could serve as a useful probe for membrane
tubulin localization.
Supported by NIH Grants GM 20,644 and GM 29,503.
Calcium activated channels in the mechanically sensitive abfrontal ciliated cells of
Mytilus gill. ELIJAH W. STOMMEL (Marine Biological Laboratory).
Mechanical stimulation of the cilia of abfrontal gill epithelial cells elicits depolarizing generator potentials
which in turn can elicit regenerative potentials of up to 40 mV (Stommel 1983, Biophys. J. 41: 90a). Both
the mechanically sensitive and the voltage sensitive channels appear to be selective for Ca+ + . Perfusion
with Co++ sea water or Ca++-free sea water eliminates any depolarizing response to mechanical stimulation.
Depolarization with injected current towards the ECa at least in part diminishes the depolarizing response.
Use of high resistance electrodes does not permit accurate bridge recordings for depolarizations greater
than +20 mV. Na+-free sea water (TMA or Tris substituted) has no effect on the depolarizing response or
the occasional spontaneous regenerative potentials. Substitution of Ba++ for Ca++ causes a long lasting
depolarization upon mechanical stimulation. Substitution of nitrate for chloride often causes sustained
depolarization, suggesting a role for chloride in repolarization. K+ blockers (TEA or 4 AP) have no obvious
effect on the repolarization. lontophoretic injection of EGTA into cells before mechanical stimulation
causes steady depolarizations that return to the original resting potential in discrete steps. These steps might
reflect the repolarization of electrically coupled cells occurring at different times as a result of unequal
amounts of calcium loading. The levels of the steps are similar from one stimulus to the next. Depolarizing
the cells away from the driving force for the repolarizing current, should produce an undershoot. However,
none occurs. It has not been possible to elicit regenerative responses by depolarizing the cells. If one assumes
that the excitable membrane resides in the cilia alone, then any current injected might be shunted through
low resistance cell bodies, where the resistance is 35 ± 11 MQohms (n = 15). The conductance/area
ratio is 5.6 m mhos/cm2. Because of the high core resistance of the cilia, they would offer an unlikely cur-
rent path.
Supported by NIH Grant GM 29,503.
Studies of cytotoxic free radicals produced by some methoxy-quinones plus ascorbate
in the presence of Ehrlich ascites cells. ALBERT SZENT-GYORGYI, PETER
GASCOYNE, RONALD PETHIG, AND JANE MCLAUGHLIN (Marine Biological
Laboratory).
Previous reports from this laboratory by Gascoyne el al. (1982, Biol. Bull. 163: 399) and Pethig et
al. (1983, Proc. Natl. Acad. Sci. USA 80: 129-132) demonstrated that direct correlations exist between
the electrochemical potentials, generated semiquinone and ascorbate free radical lifetimes, and cytotoxic
action in Ehrlich ascites bearing mice of various methoxy-substituted p-quinones in the presence of
ascorbic acid. Spectroscopic measurements and electrochemical titrations support the concept that the
CELLULAR, MOLECULAR BIOLOGY, ETC. 497
observed cytotoxic properties of the 2,5- and 2,6-dimethoxy quinones were related to the production of
long lived free radicals as a result of one- rather than two-electron reductions by ascorbic acid.
We have extended the electrochemical studies to include 2,3,5-trimethoxy- and tetra-methoxy-p-
quinone and the redox potentials obtained (at pH 7.4 and 25°C) were 72 mV and 99 mV, respectively.
In vivo studies of the cytotoxic properties of these quinones in the presence of ascorbic acid against
Ehrlich ascites are currently in progress.
The semiquinone and ascorbate free radical lifetimes in ascitic fluid have been determined as a
function of the Ehrlich ascites cell concentration. Evidence has been obtained to show that the rate of
disappearance of the generated free radicals is directly proportional to the concentration of viable ascites
cells. Blocking of cell surface sulfhydryl groups by N-ethylmaleimide has indicated that -SH groups are
responsible for the free radical depletion. The cell surface -SH groups are found to be of the order 20
times more efficient as radical scavengers than an equivalent aqueous concentration of glutathione.
The quinones were prepared in the laboratory of Professor Gabor Fodor, and the work is supported
by the National Foundation for Cancer Research.
ATP-reactivated models ofctenophore comb plates. SIDNEY L. TAMM AND SHOGO
NAKAMURA (Boston University Marine Program).
Comb plate cilia of cydippid larvae of Pleurobrachia and Mnemiopsis undergo a Ca2+-dependent
reversal in beat direction, causing larvae to swim backwards (Tamm and Tamm 1981, J. Cell Biol. 89:
495-509). We now find that 5 pM A23187 + 10 mM Ca2+ in Ca2+-Mg2+-free artificial sea water (ASW)
also causes backward swimming, confirming the role of Ca2+ in regulation of ciliary beat direction in
ctenophores.
Mnemiopsis larvae placed in 150 mM KC1, 1 mM EGTA, 30 mM PIPES, 2% polyethylene glycol,
pH 7.0 for 10 min dissociate into single living comb plate cells which beat in the normal direction and
"swim" in circular paths. When transferred by Ca2+-Mg2+-free ASW containing 5 ^M A23187 + 10 mM
Ca2+, these solitary comb plate cells, free of nervous tissue, beat in the reverse direction and "swim"
backwards in circles (high-speed video recordings). Thus, Ca2+ directly activates the ciliary reversal mech-
anism, and may be required for synaptic triggering of reversal, since comb plate cells in intact larvae are
innervated by the nervous system (Tamm and Tamm 1981).
ATP-reactivated models of comb plates were obtained by extracting larvae in 0.005% Triton-X 100,
10% glycerol, 2% polyethylene glycol, 2.5 mM MgCl2, 150 mM KC1, 1 mM EGTA, 30 mM PIPES, pH
6.9 (ES) for 3 min at room temperature. When placed in 2 mM ATP, 2.5 mM MgCl2, 2% polyethylene
glycol, 150 mM KC1, 1 mM EGTA, 30 mM PIPES, 1 mM DTT, pH 6.95 (RS), comb plates beat in a
direction similar to that of living ones. The beat frequency and extent of reactivation depend on the
Mg2+-ATP concentration, and reactivation is inhibited by 30 \iM vanadate. No beating occurs in the
absence of ATP. 10"6 M Ca2+ in RS causes reversal in beat direction, so that each beat cycle starts with
an aborally-directed recovery stroke, followed by an effective stroke toward the mouth (high-speed video
recordings). In RS containing trypsin, initial beating is followed by ATP-induced sliding disintegration
of the axonemes, resulting in extrusion and looping out of doublet microtubules.
Taking advantage of the unique untrastructural markers for specific doublet microtubules in cteno-
phore cilia, we plan to determine the effects of Ca2+ on the pattern of microtubule sliding during ciliary
reversal.
Supported by NIH Grant GM 27903.
Intracellular fusion between reticulopodial networks in Allogromia laticollaris. J. L.
TRAVIS (University of West Virginia) AND V. E. CENTONZE.
Membrane fusion plays an important role in the formation and activity of foraminiferal reticulo-
podial networks. This fusion is most noticable during the anastomosis and fusion of pseudopodia forming
the interconnected reticulopodial network. Membrane fusion may also occur between experimentally
(or accidently) excised networks and an intact portion. The excised or "satellite" portions show normal
bidirectional streaming at first, but this gradually becomes less vigorous. In addition, the cytoplasm of
the satellites withdraws radially to form a droplet that becomes quiescent (Jahn and Rinaldi 1959, Biol.
Bull. 117: 100). Satellites may be "rescued" (Allen 1964, Primitive Motile Systems in Cell Biology,
Academic Press) by fusing with an intact portion of the reticulopodial network which results in the
reincorporation of the satellite into the original network. While confirming these earlier studies, we have
determined that Allogromia laticollaris can fuse with satellites excised from other organisms forming
hybrid reticulopodia. These results differ from Schwab's (Schwab and Schwab-Stay 1980, Protoplasma
102: 141) which suggested that fusion cannot occur between reticulopodia of different organisms.
498 ABSTRACTS FROM MBL GENERAL MEETINGS
We thank J. Rosenbaum, E. Stromboli, and the entire Marine Biological Laboratory Physiology Course.
This work was supported by NIH Training Grant GM-31 136-05.
Marine molluscan genomes contain sequences homologous to the octopine synthase
gene o/ Agrobacterium tumefaciens. ERIC R. WARD AND WAYNE M. BARNES
(Washington University, St. Louis).
Octopine, a conjugate of pyruvate and arginine, has been characterized as an anaerobic metabolite in
a wide variety of marine molluscs. Its synthesis is catalyzed by a monomeric 40 Kd NADH-dependent
oxidoreductase called octopine dehydrogenase. Octopine is also found in crown galls, which are neoplastic
growths incited in dicotyledonous plants by the soil-borne Agrobacterium tumefaciens. A large (~ 120 Md)
tumor-inducing (Ti) plasmid mediates virulence in this bacterium. During infection of a plant by Agro-
bacterium, a portion of the Ti plasmid containing various oncogenic functions (the T-DNA) becomes
covalently joined to the chromosomal DNA of the host plant. Within the T-DNA lies the gene encoding
octopine synthase, an NADPH-dependent oxidoreductase similar to octopine dehydrogenase. The regulatory
signals of the octopine synthase gene closely resemble those of most eukaryotic structural genes and its
expression depends on incorporation of the T-DNA into the plant chromosome. These two apparently
unrelated systems have previously been assumed to share their unique ability to synthesize octopine as a
result of convergent evolution. Should the Agrobacterium and marine molluscan genomes display homology,
a putative evolutionary relationship could be inferred. We examined this question by digesting genomic
DNA from the squid Loligo pealii and the clam Spisula solidissima with restriction endonucleases, elec-
trophoretically separating the resulting fragments, denaturing and binding the fragments to nitrocellulose
filter, and hybridizing to the filter a radioactively labeled cloned DNA fragment of the A6 Ti plasmid
containing the octopine synthase gene. Genomes of both molluscs selectively hybridized to the labeled
probe sequence under conditions permitting approximately 10% sequence divergence. The homologous
sequence occurs in approximately single copy per haploid genome in both molluscs. DNA from Escherichia
coli, the sea urchin Stronglyocentrotus purpuratus, and the slime mold Dictyostelium discoideum did not
hybridize to the probe sequence.
E.R.W. acknowledges the encouragement of the staff and students of the 1983 Physiology course,
especially Drs. R. Chisholm, P. Lefebvre, and C. Silflow. This work was carried out under the generous
support of the James S. Mountain Memorial Fund.
Mitochondria! and spherosomal movement along a filamentous network in the ma-
rine slime mold Gymnophrydium marinum. STANLEY W. WATSON (Woods
Hole Oceanographic Institution), BRUCE J. SCHNAPP, AND ROBERT V. RICE.
Over 60 years ago investigators became aware of a rapid transport of optically detectable organelles
in eucaryotic cells. Why and how these organelles move still awaits experimental verification. Electron
microscopic studies strongly suggest that some organelle transport is associated with microtubules or
microfilaments, and bidirectional movement of submicroscopic particles along linear structures in axons
has been elegantly demonstrated by Allen el al (1982, Biol. Bull. 163: 379).
Present studies concern the multidirectional movement of mitochondria, spherosomes, and other
unidentified particles along a cytoplasmic filamentous, undulating network (100-200 nm in diameter) at
velocities of 1-2 ^m s~' in the marine slime mold Gymnophrydium marinum. Movement along this
cytoplasmic network is observed with both phase-contrast and differential interference contrast (DIC)
microscopy but the DIC image is greatly improved employing video and computer enhancement.
The biochemical nature of these filaments has not been identified but it seems unlikely that they
represent bundles of microtubules since the network is not disrupted nor does the transport of cytoplasmic
particles cease or appear effected in 10^3 M concentrations of colchicine. Cytoplasmic particles move
along the filaments in this branching network and are most clearly observed in the flat lobopodia (less
than a /urn thick). Similar filaments are observed in the narrow dense rhizopodia, but in these pseudopodia
extensive branching of the filaments is not observed. The undulating nature of these filaments suggest
that they may be composed of contractile proteins which in an unknown manner interact with organelle
membranes resulting in a rapid transport of such particles.
Rise of free intracellular Ca2+ in mouse macrophage associated with y2b/yl Fc
receptor-ligand interaction. JOHN DiNG-E YOUNG (The Rockefeller University,
New York, NY 10021).
Binding of the mouse macrophage y2b/yl Fc receptor (FcR) by immune complexes triggers a
number of dramatic responses, which include secretion of inflammatory metabolites and phagocytosis.
Previous work has shown that FcR behaves as a ligand-dependent ion channel.
CELLULAR, MOLECULAR BIOLOGY, ETC. 499
A rise of free cytosolic Ca2+ concentration [Ca2+], is a key regulator of cell surface-activated responses.
We now report on [Ca2+], changes associated with FcR-ligand interaction. We used quin-2A/M to measure
[Ca2+],. J774 macrophages grown in spinner cultures showed a loading efficiency of 15-20% (or [quin-
2], of 0.1 1-0.13 mM) with 20 nM of quin-2A/M. The maximum signal-to-noise ratio was 3.5-4.0. The
resting [Ca2+]j was 87 nM (±15 SE; n = 9) which could be lowered to 29 nM (±9 SE; n = 6) after a 30
min incubation in Ca2+-free medium. Addition of A23187 (10 nM) raised [Ca2+], to >1 \iM. Addition
of the monoclonal antibody 2.4G2 IgG (5 X 10~7 M), which binds to a functional site of FcR, raised
[Ca2+], to ~400 nM within seconds. This response was transient (lasting 5-10 min) and showed dose-
dependence. The monovalent ligand 2.4G2 Fab (10~6 M) gave only a small response (~120 nM) and
was capable of blocking cell response to subsequent addition of 2.4G2 IgG. Ligands of higher valence
(soluble and precipitable immune complexes) were more effective in raising [Ca2+], at much lower con-
centrations. Incubation of macrophages with antibody-coated erythrocytes raised [Ca2+], to nM levels.
[Ca2+], changes were only partially inhibited by the absence of external Ca2+ or following incubation with
valinomycin or FCCP (10 nM, 30 min). Depolarizing cells with 50 mM KC1 raised [Ca2+], to ~ 180 nM.
Preliminary experiments show that buffering [Ca2+]j with 100 nM quin-2A/M in the absence of external
Ca2+ inhibited phagocytosis of antibody-coated erythrocytes by macrophages. Together, these data suggest
a rise of [Ca2+], following binding to FcR that is due to influx of external Ca2+ and release of Ca2+ from
internal stores. These Ca2+ stores are not limited to mitochondria. The Ca2+ response is associated with
receptor cross-linking and aggregation by ligands and cannot be explained solely by a membrane de-
polarization effect induced by ligands.
This work was supported by a fellowship from the Grass Foundation. Suggestions and advice from
Dr. Joel Brown are heartily acknowledged.
Ocular lens aging in the skate. SEYMOUR ZIGMAN, TERESA PAXHIA, BLENDA AN-
TONELLIS, AND WILLIAM WALDRON (University of Rochester School of Medicine
and Dentistry, Rochester, NY 14642).
Skate lenses were used to study age related changes in gross and microscopic morphology and in
protein state (degree of aggregation). Lens weight of fresh Raja erenacea and Raja eglanteria was plotted
against body weight. Lens weights increased colinearly until plateaus began at 1 10 mg (R. erenacea) and
160 mg (R. eglanteria). Lens weight increase relative to body length or wing span were less definite.
After capsule removal, skate lenses were homogenized whole or after separation into concentric
layers (outer cortex, inner cortex, outer nucleus, nuclear core) in PO4 buffer (pH 7.4). Soluble and
insoluble fractions were separated by centrifugation (100,000 x g for 30 min). Insoluble proteins were
extracted successively with 8A/ urea, 1% SDS, and SDS + 50 mM DTT. Extracts were examined by
Lowry analyses and SDS polyacrylamide gel electrophoresis (PAGE), and soluble proteins were subjected
to high pressure liquid chromatography (HPLC). Insoluble protein levels became equivalent with soluble
proteins in lenses weighing 125 mg to 145 mg. In lenses weighing 400 mg, insoluble levels were 33% in
excess. Such high ratios of insoluble to soluble proteins would cause opacities, so that the high urea level
of the skate lens may prevent them.
HPLC indicated a predominance of lens soluble crystallins of 20,000 d molecular weight exclusively
in their cores, and additional heavier crystallins in their outer layers. SDS-PAGE revealed soluble crystallins
with molecular weights between 18,000 and 22,000 d. Extracts solubilized only by SDS plus DTT contained
26,000 and 22,000 d plus traces of 45,000 d chains. SDS and urea extracted noncovalently-linked chains
similar in size to the soluble crystallins, leaving only intrinsic proteins. Insoluble protein in the cores was
50% greater than in outer layers of the lens; 67% of the insoluble protein of outer layers, but only 50% of
the cores, were urea and SDS soluble.
Skate lenses are thus useful in assessing both morphological and biochemical features of aging.
Support: National Eye Institute and Research to Prevent Blindness.
DEVELOPMENTAL BIOLOGY
Developmental studies of a major maternal mRNA in Arbacia punctulata. SARAH
BRAY (University of Cambridge, England) AND TIM HUNT.
Synthesis of a maternal mRNA which encodes a 4 IK protein is initiated at fertilization in Arbacia
punctulata (Evans et al. 1983, Cell 33: 389-396). This mRNA is present at the same levels in both
unfertilized eggs and early embryos, so the onset of synthesis must reflect a change in the ability of the
embryo to translate this RNA. We have isolated a cDNA clone to this RNA by hybrid selection and are
using it to study the message and its fate during development.
500 ABSTRACTS FROM MBL GENERAL MEETINGS
When Northern blots of total RNA are probed using this cloned sequence, it hybridizes to a single
RNA species of approximately 3Kb. This is considerably larger than would be predicted for an RNA
encoding a 4 IK. protein. However we were unable to detect any processing of this RNA at fertilization
or at later stages in development. The RNA persists throughout development as late as pluteus; however
at later stages it is a much less abundant component of the RNA population.
We fractionated the RNA from eggs and 4-cell embryos using oligo-dT chromatography and found
that this RNA is present in the poly-A containing fraction in both.
Southern blots of genomic DNA from sperm of an individual Arbacia cut with different restriction
enzymes show two bands of equal intensity hybridizing to the cloned sequence suggesting at least two
copies per genome. The clone also cross-reacts with Strongylocentrotus purpuratus DNA and RNA.
The protein encoded by the RNA binds quantitatively to an anti-yeast-tubulin affinity column
(Kilmartin et al. 1982, J. Cell Biol. 93: 576-582). We hope ultimately to elucidate how the RNA is
regulated and the role of the protein it encodes.
Supported by NIH grant no. GM-31 136-05 and the Science and Engineering Research Council of
Great Britain.
Induction of heat shock proteins in early embryos of Arbacia punctulata. SARAH
HOWLETT, JOHN MILLER, AND GILBERT SCHULTZ (University of Calgary).
We have re-examined the ability of early sea urchin embryos to synthesise heat shock proteins (hsps).
By raising the culture from 20-22°C to 31°C, a discrete set of new proteins are induced, the major
species being 70,000 d (70K) in molecular weight. The hsp 70 was detectable within 25 minutes following
heat shock, together with several minor polypeptides (118, 60K). Little, if any, reduction in total protein
synthesis was observed.
We have confirmed that synthesis of hsp 70 is not inducible in unfertilized eggs or early embryos.
The first time at which hsp 70 is inducible appears to be at about the 64 to 128 cell stage, and the
response remains through hatching blastula and gastrula stages. With the possible exception of 2 and 4
cell embryos, most (greater than 75%) heat shocked embryos continued to develop to form normal
hatched blastulae following an hour long heat shock.
Heat shock in the presence of 20 ng/m\ actinomycin D to inhibit mRNA synthesis confirmed the
heat shock response to be dependent upon novel transcription. Indeed, in vitro translation in a reticulocyte
lysate cell free system of RNA extracted from heat shocked hatched blastulae showed an abundance of
this hsp 70. Further, a Drosophila genomic DNA fragment complementary to the coding region of a hsp
70 gene hybridised to polyadenylated RNA from mid-cleavage (32-128 cells) and gastrula stages after
heat shock but not to mRNA from control or heat shocked embryos up to the 16 cell stage. Further
experimentation is aimed at determining the exact cell cycle at which the switch to an inducible state
occurs.
This work was supported (in part) by NIH Training Grant 5-T35-HD07098 awarded to the Em-
bryology Course, Marine Biological Laboratory, Woods Hole, MA.
Changes in histone synthesis during Arbacia development. P. E. KUWABARA, K.
GREER, S. MAEKAWA, AND E. S. WEINBERG (University of Pennsylvania).
Histone protein in early developmental stages of Arbacia punctulata was studied using in vivo labeling
of eggs and embryos. Embryos were pulse labeled with 3H-leucine and 3H-lysine after aliquots were
removed at five different times after fertilization. Total cell histones were extracted with H2SO4 and run
on either an 18% polyacrylamide-SDS gel or a Triton X-acid-urea gel. The HI synthesized in the first
30 min after fertilization co-migrates in the SDS gel with the cleavage stage H 1 of Strongylocentrotus
purpuratus. In the next 30 min, after nuclear membrane breakdown, the synthesis of an HI protein
which co-migrates with 5. purpuratus a-Hl is seen. A similar shift is observed among the subtypes of
the H2A histones resolved on a Triton X gel. Accumulation of newly synthesized histones after fertilization
and in unfertilized eggs was demonstrated by continuous labeling with 3H-leucine. Samples were removed
at 10 min intervals and prepared for analysis on SDS-polyacrylamide gels by trichloroacetic acid pre-
cipitation followed by an acetone wash. Synthesis of histone proteins is detectable at twenty min after
fertilization and is also seen in the unfertilized egg. This indicates that some histone message is available
for translation before nuclear envelope breakdown and even before fertilization. Continuous labeling was
also done in the presence of 5 Mg/ml of aphidocholin. In treated embryos, the synthesis of the histone
proteins occurs during the first 30 min after fertilization but further increase is inhibited. This effect could
be related to the inhibition of DNA synthesis by aphidocholin or to the prevention of nuclear membrane
breakdown and consequent prevention of histone mRNA release. Aphidocholin also prevented the normal
DEVELOPMENTAL BIOLOGY 501
disappearance at first cleavage of a protein believed to be a-cyclin. Other proteins, including /3-cyclin,
are synthesized as in the control embryos.
Supported by NIH grant GM-3 1 1 36-05 to the Physiology Course.
Accumulation oflateH2b histone mRNA in sea urchin embryogenesis. GARY LYONS
(University of Pennsylvania School of Medicine), SUSAN HALSELL, AND ROB
MAXSON.
Three distinct histone protein subtypes appear during embryogenesis of the sea urchin Strongylo-
centrotus purpuralus. The switches in histone synthesis occur as a result of changes in the activity of
different histone gene sets. As a first step toward characterizing the kinetics of this system, the levels of
transcripts as successive stages of development were measured.
A 183 nucleotide cloned late H2b histone gene probe was used in these experiments. This probe
hybridizes to other members of the late H2b gene family but not to early H2b genes under appropriately
stringent conditions. RNA gel blots and RNA dot hybridizations were hybridized with nick-translated
probe DNA to determine the relative amounts of histone mRNA at various developmental stages. The
absolute amount of mRNA present at a given stage was measured by hybridizing the probe to increasing
amounts of RNA. Total RNA of 72 hour pluteus larvae was titrated with single stranded DNA probe
synthesized on an M 1 3 phage DNA template containing the late H2b fragment. The specific activity of
the probe was 1 ) calculated from the known specific activity of the 32P-ATP and the known sequence
of the probe and 2) measured by titrating the probe with increasing amounts of cold DNA. The two
estimates were in close agreement and a value of approximately 2 x 108 cpm/^g was obtained. From
the RNA titration curve the probe was determined to be 75% hybridizable. The level of late H2b mRNA
was determined to be 4.7 X 105 RNA molecules/embryo. From this information and from the relative
amounts of mRNAs determined by scanning the Northern blots and dot hybridizations, it was possible
to determine the number of mRNA molecules present at each stage. It was found that late H2b mRNA
levels increase rapidly between 14 and 16 hours, reach a peak of 6.2 X 105 RNA molecules/embryo at
22.5 hours, and start to decrease at the pluteus stage.
The probe was also used to assay adult tissues for late histone mRNA. RNA was isolated from
gonad, coelomocyte, intestine, and tube foot tissue by guanidine thiocynate or phenol extraction. After
fractionation on a gel and transfer to nitrocellulose, the RNA was allowed to hybridize the radiolabeled
probe. The results show that molecules of the late H2b class are present in all adult tissues, though in
smaller amounts than in the embryo.
This work was supported (in part) by NIH Training Grant 5-T35-HD07098 awarded to the Embryology
Course, Marine Biological Laboratory, Woods Hole, MA.
A video time lapse study of cell behavior during notochord morphogenesis in ascidian
embryos. DAVID M. MIYAMOTO (Seton Hall University).
Ascidians are attractive organisms to study how cells form structures in living embryos. Video time
lapse recordings of notochord cell behavior during gastrulation and tail formation in dechorionated
embryos ofdona intestinalis were made using a perfusion chamber that maintained temperature at 18-
19°C and permitted the use of oil immersion Nomarski microscopy. The eight central notochord cells
become flask-shaped as they turn inwards to form the anterior lip of the blastopore. As the lip moves
posteriorly, these cells divide parallel to the embryonic axis, away from the lip. Internalized cells divide
out of synchrony with those that remain part of the lip. Lip cells lose their attachment to the blastopore
only as it becomes smaller and enclosed within the forming posterior neural tube. No distinctive surface
activity such as blebbing was evident in these cells during gastrulation.
After gastrulation the notochord is a mass of spindle and wedge-shaped cells that show rhythmic
back and forth movements as they interdigitate to form a strand of disc-shaped cells lined up end to end
(early tailbud, 9 h). These cells decrease in diameter as they increase in length as the tail elongates at a
rate of 1.3 ^m/min. Basal surfaces adjacent to surrounding tissues begin to bleb actively, probably an
indication of the formation of the extracellular sheath described by others. Intracellular vacuoles appearing
at this time show dynamic behavior, extending and retracting bulges in various directions, as they increase
in size. Separating partitions disappear at about 16 h as these intracellular vacuoles fuse to form the
intercellular vacuoles described by others. Blebbing activity begins to decline in intensity at this point
and at the time of normal hatching (18 h), cell surfaces are quiescent. Cytoplasm and nuclei shift toward
the periphery and intercellular vacuoles combine to form the internal matrix core as previously described
by other workers.
502 ABSTRACTS FROM MBL GENERAL MEETINGS
The author gratefully acknowledges the assistence of Dr. J. R. Whittaker and Robert Crowther (Boston
University Marine Program) in this work. The work was supported by a University Research Council Grant
from Seton Hall University and NSF Research Instrumentation Grant, RII-821021.
Rates of 5S RNA and tRNA synthesis in sea urchin embryos animalized by Evans
Blue. ANNE F. O'MELIA (Department of Biology, George Mason University,
Fairfax, VA 22030).
The application of certain chemicals to whole sea urchin embryos between early cleavage and the
blastula stage interferes with normal cell associations and interactions producing characteristic malfor-
mations in germ layer formation, termed animalization (ectodermalization) and vegetalization (endo-
mesodermalization). Animalization of Arbacia punctulata embryos was induced by culturing embryos
in Evans Blue continuously from the 2-cell stage. Previous research showed that the accumulation of the
newly synthesized nucleolar ribosomal RNAs (26S, 18S rRNAs) is strongly inhibited in Evans Blue-
animalized embryos (O'Melia 1983, Dev. Growth Differ. 25: 171-180). The present study determined
the effect of this animalizing agent on the synthesis of the third rRNA, 5S RNA, and of transfer RNA
(tRNA). Cultures of mesenchyme blastulae, plutei and corresponding animalized embryos each were
labeled for 3 h with [8-3H]-guanosine. Total RNA was extracted using the cold (4°C)-phenol-SDS method
and purified (NaCl-soluble) RNA preparations were fractionated by electrophoresis on 10% polyacryl-
amide gels. Rates of accumulation of newly synthesized 5S RNA and of tRNA in control and in animalized
embryos were calculated from the radioactivity coincident with the 5S RNA and with the tRNA absor-
bance peaks ( A260nm) on each gel, from the known GMP composition of sea urchin 5S RNA and tRNA,
and from the average specific radioactivity of the GTP precursor pool during the 3 h labeling period.
The results showed that the rates of accumulation of newly made 5S RNA and tRNA per embryo and
per cell are similar in control and in Evans Blue-animalized embryos at each stage. Therefore, the
alterations in cell associations, interactions, and germ layer formation induced by Evans Blue did not
affect the synthesis of 5S RNA and of tRNA in sea urchin embryos.
Support: Office for Research and Advanced Studies, George Mason University.
Reproduction in Haploplana and Stylochus: developmental and cytoskeletal research
possibilities. PAUL P. PALASZEWSKI AND BARBARA C. BOYER (Union College).
An unusual cytoplasmic blebbing coinciding with polar body formation in the determinative eggs
of Hoploplana inquilina and Stylochus zebra make these hermaphroditic polyclad turbellarians valuable
research animals for developmental studies. H. inquilina, obtained from the mantle cavity of Busycon
sp., and S. zebra, collected from Pagiirus shells, were maintained in finger bowls of sea water and remained
reproductively active for 8 weeks. Artificial fertilization, by manually releasing gametes, provided eggs
free from the impermeable membrane normally surrounding them.
Both blebbing and cleavage were inhibited by colchicine and taxol at approximate concentrations
of 10 Mg/ml and 1 Mg/ml respectively. H. inquilina was slightly more sensitive than S. zebra. H. inquilina
eggs in 0.25 Mg/ml taxol displayed normal bleb formation but their characteristic resorption was incom-
plete and the cleavages were abnormal. In both species blebbing and cleavage were unaffected by cyto-
chalasin B concentrations as high as 20 Mg/ml. These results suggest a microtubule associated cytoskeletal
mechanism for this blebbing, which may function in cytoplasmic localization and other early organi-
zational events of development.
Comparisons of gonad development as a function of body length indicate that testes and ovaries
develop concurrently in H. inquilina. However, 20% of the S. zebra specimens collected were male,
averaging 1.2 cm in length; the remainder were hermaphroditic and averaged 1.9 cm in length. Addi-
tionally, since testes appeared to mature before ovaries a protandric transition in S. zebra is hypothesized.
In culture, H. inquilina produced an average 0.20 ± .03 egg masses-animal" '-day" '. Peak laying
activity followed a consistent 3 day cycle and productivity decreased linearly as a function of time in
captivity.
This work was supported by a Research Corporation Grant to B. Boyer.
Cell-cell recognition and adhesion during embryogenesis in the sea urchin. E. GAYLE
SCHNEIDER (University of Nebraska Medical Center).
The object of the current study was to investigate species-specific recognition and adhesion between
dissociated embryonic cells of hatched blastulae of Arbacia punctulata, Lytechinus variegatus, and Stron-
gylocentrotus purpuratus using a quantitative reaggregation assay. The assay used is a modification of
one previously developed by McClay and Hausman ( 1 975, Dev. Biol. 41: 454-460) and involves collection
DEVELOPMENTAL BIOLOGY 503
of labeled single cells to preformed collecting aggregates. The aggregates are prepared by dissociation of
blastulae in calcium and magnesium-free sea water. The dissociated cells are allowed to reaggregate in
stationary culture; the aggregates are collected by centrifugation and washed. The labeled probe cells are
prepared by incubation of blastulae in [3H]leucine and dissociation of these as above. The assay consists
of mixing labeled probe cells (0.15-0.8 X 106 cells/ml) with various concentrations of aggregates in a
total volume of 2 ml for 2 h in suspension culture. The aggregates and adhered probe cells are separated
from unadhered probe cells by gentle centrifugation, and the percent reaggregation of probe cells to the
aggregates is determined. The results indicate no significant adhesion to aggregates for probe cells fixed
by glutaraldehyde or formaldehyde or disrupted by sonication. In addition, fixation of aggregates by
glutaraldehyde greatly diminished binding of probe cells. Finally, adhesion of probe cells to homospecific
aggregates was significantly greater than that to heterospecinc aggregates. The results demonstrate recip-
rocal species-specific adhesion between Arbacia punctulata versus Strongylcocentrotus purpuratus and
Arbacia punctulata versus Lytechinus variegatus. The results extend previous work with other species
and suggest that species-specific recognition is a universal phenomenon in the sea urchin. In addition,
this recognition, as measured by the present assay, requires living cells.
This work has been supported by a 1982-83 Steps Toward Independence Fellowship from the
Marine Biological Laboratory and by a grant from the National Science Foundation (#PCM-81 18503).
Leukotriene B4 promotes the calcium-dependent aggregation of marine sponge cells.
GERALD WEISSMANN, CATHLEEN ANDERSON, LESLIE B. VOSSHALL, ABBY M.
RICH, KATHLEEN A. HAINES, TOM HUMPHREYS, AND PHILIP DUNHAM (Marine
Biological Laboratory).
We have previously reported that aggregation of dissociated Microciona prolifera cells induced by Ca,
Ca-ionophores such as A 23187, and specific aggregation factor (MAF) resembles the active, stimulus-
response coupling of human neutrophils (Dunham el al. 1983, Proc. Natl. Acad. Sci. 80: 4756-4760). We
now report that sponge cells, stimulated by >5 mAf Ca added to Ca-free media take up 45Ca from the
medium. Uptake was saturable (Kl/2 approx. 10 mAf) and was not reversible by excess La or EDTA. MAF-
induced aggregation was accompanied by 40 percent enhancement of 45Ca influx. Cells pre-loaded with
chlorotetracycline (CTC) underwent prompt decrements in CTC fluorescence upon addition of Ca, indirect
evidence for mobilization of endogenous membrane-associated Ca. Aggregation, 45Ca influx, and decrements
in CTC fluorescence were inhibited by non-steroidal antiinflammatory agents (indomethacin, ibuprofen,
piroxicam; 50-100 fiM), usually considered to act by inhibiting arachidonate oxidation via the cyclooxygenase
pathway to stable prostaglandins. Yet Microciona cells were neither aggregated, nor prevented from aggregating
in response to MAF or ionophore, by cyclooxygenase products (PGE, , PGF,n, PGF2o, PGE2, PGI2, PGA, ,
PGB,, PGD2; 5 pM — 1 mA/ ± theophylline). In contrast, the 5-lipoxygenase product leukotriene B4, an
aggregant of neutrophils, provoked Ca-dependent aggregation (<1 nAi), and nordihydroguaretic acid, a
lipoxygenase inhibitor, inhibited MAF- and ionophore-induced aggregation. Other lipoxygenase products
(5-, 12-, 15-HETE; trans isomers or omega metabolites of LTB4; 14,15-diHETE; 2 nM) failed to influence
sponge cell aggregation. Colchicine ( 1 mA/, 20 min), which specifically inhibits LTB4 synthesis of ionophore-
stimulated neutrophils, inhibited sponge cell aggregation induced by MAF or ionophore, but had no effect
on 45Ca influx. The data suggest not only that sponge cells utilize leukotriene B4, or a similar product, as
part of a Ca-dependent, stimulus-response coupling sequence, but that non-steroidal antiinflammatory
agents cannot exert all of their biological effects by simply inhibiting extracellular release of stable pros-
taglandins.
Evidence for regulation of protein synthesis at the level of translational machinery in
the sea urchin egg. MATTHEW WINKLER AND BREWER SHETTLES (Department
of Zoology, University of Texas, Austin, TX 78712).
Fertilization of the sea urchin egg results in a 20-40 fold increase in the rate of protein synthesis.
This increase is mediated by the mobilization of stored maternal mRNA into polysomes. It is not known
if this process is regulated at the level of mRNA availability (masked message hypothesis) or at the level
of the translational machinery. We have used a novel in vivo assay to try to distinguish between these
two possibilities.
Unfertilized Arbacia punctulata eggs were incubated for 45 minutes in 10~4 A/ emetine, a potent
protein synthesis elongation inhibitor. These eggs and untreated controls were fertilized and, at various
time intervals aliquots, were removed and processed for sedimentation on high salt sucrose gradients.
Before sedimentation samples were treated with ribonuclease A. Under these conditions polysomes are
converted into SOS monomers which are resistant to high salt. Free 80S ribosomes will be dissociated
into 40 and 60S subunits. This procedure allows one to easily determine the percent of ribosomes in
504 ABSTRACTS FROM MBL GENERAL MEETINGS
polysomes. Stored mRNAs in emetine treated eggs would be expected to bind a single ribosome as they
move into polysomes. If mRNAs are being activated then there should be l/(average number of ribosomes
per polysome) amount of high salt resistant SOS ribosomes. If some component of the translational
machinery were being activated, then emetine treated eggs and the controls should have equal amounts
of high salt resistant SOS ribosomes. We find that emetine treated eggs have from 50-80% of the control
value of high salt resistant SOS ribosomes. This value is significantly larger than the 1 5-20% expected
if mRNA availability limited protein synthesis. This result indicates that protein synthesis is regulated
at the level of the translational machinery at fertilization, possibly by activation of ribosomes.
This work was supported in part by NIH training grant 5-T35-HD07098 awarded to the Embryology
Course, Marine Biological Laboratory, Woods Hole, MA and NIH grant HD 17722-01 awarded to M.W.
ECOLOGY, EVOLUTION, AND PLANT SCIENCES
Current flow around Zostera marina plants and flowers: implications for submarine
pollination. JOSEF D. ACKERMAN (SUNY, Stony Brook, NY).
Zostera marina L. (eelgrass) is a marine angiosperm possessing a unique flowering morphology,
which suggests a singular adaptation to submarine pollination. The physical aspects of this process were
examined on both a macro and micro scale, in an eelgrass bed in Great Harbor, Woods Hole, MA, and
in a flume (flow channel) in the laboratory. Dyes and Sephadex particles were used to mark the flow.
Flow rates ( 1-2 cm s~') observed in the eelgrass canopy were simulated in the flume. Current flow patterns
were photographed and videotaped using a Zeiss stereomicroscope. Individual particle trajectories were
tracked on the frame advance mode for videotaped sequences.
Flow velocity was reduced in the eelgrass bed by 10-fold compared to that observed outside the bed.
Vertical stratification of velocity within the canopy was noted, with flow decreasing towards the sea floor.
The rhipidia (flowering branches) were found in the top two-thirds of the canopy. The overlap of inflo-
rescences on a rhipidium further reduced current velocity 5-fold. The most marked change in current
velocity and flow occurred within 3 mm of a female flower, with velocity being reduced from 1.5 to 0.1
cm s~', and flow directed towards a focal zone downstream from the stigmata. There is an upward
movement of water from the blade towards the stigmata, and a net downward draw of water along the
length of an inflorescence. These changes in flow result from a 12-fold increase in the area of water
influenced by a female flower.
Density of mature female flowers was established, and combined with current flow measurements,
provided an encounter frequency of water particles with flowers. From these calculations, it has been
determined that under moderate flow conditions, any particle within a m2 will encounter a mature
stigmata within 5-20 minutes. From these observations it is concluded that the search time for pollen
must be short. This corresponds to the short exposure time for female flowers (4-6 h). Submarine
pollination in Zostera marina is a process mediated by floral morphology as it influences current flow.
Special thanks to Phillip H. Presley for providing Zeiss instruments.
Population ecology of the Caribbean bivalve Asaphis deflorata (Linne, 1 758). PHILIP
ALATALO AND CARL J. BERG, JR. (Marine Biological Laboratory).
Field studies of the Caribbean bivalve Asaphis deflorata were conducted at Gold Rock Creek, Grand
Bahama Island, to establish life history parameters. Asaphis deflorata lives intertidally among coarse rock
gravel, 5-15 cm below the surface. It feeds on phytoplankton and organic detritus using long, extendible
siphons. Population densities average 34 clams/.25 m2, but may reach 87 clams/. 25 m2.
The total of all monthly size distribution samples were fit to the von Bertalanffy growth curve using
the ELEFAN I computer program and predicted curves were corroborated with probit analysis of in-
dividual monthly samples. Mean sizes of 19, 33, 43, 52, and 58 mm were calculated for animals 1 year
through 5 years old respectively.
Seasonal analysis of body parameters revealed an increase in dry meat weight for all size animals
from January to June. Individuals greater than 40 mm shell length failed to gain dry tissue weight through
September, suggesting summer spawnings for these larger animals.
Asaphis deflorata is dioecious and becomes sexually mature at shell lengths greater than 25 mm.
Based on histological analysis of gonads, it appears that natural spawnings occur between July and
September at water temperatures above 25°C. Eggs spawned in the laboratory range in size from 60-65
^m diameter. Larvae are planktotrophic and exhibit a brief pediveliger stage 10-12 days after spawning
ECOLOGY, EVOLUTION, PLANT SCIENCE 505
at 25-29°C. Metamorphosis occurs at a size of 161-205 jtm shell length. No specific substrate is required
for metamorphosis. Growth rates in the laboratory were slower than those predicted from population
samples.
Supported by funds from the Wallace Groves Aquaculture Foundation, Freeport, Bahamas.
Genetic variation in the queen conch, Strombus gigas, across its geographic range.
Preliminary results. CARL J. BERG, JR., KATHERINE S. ORR, AND JEFFRY B.
MITTON (Marine Biological Laboratory).
The queen conch, Strombus gigas, is an economically important gastropod mollusk distributed
throughout the Caribbean Basin, the Bahamas, and in Bermuda. It has pelagic larvae that remain in the
water column a minimum of 2 1 days, so the potential exists for long distance dispersal in this species.
Little is known, however, about the actual distances that larvae travel before abandoning the planktonic
stage of the life cycle and becoming benthic juveniles. Therefore a study of geographic variation of protein
polymorphisms was conducted to determine the degree and pattern of genetic differentiation. Population
samples were collected from four keys in the Turks and Caicos Islands; Carriacou in the Grenadines;
Ambergris Cay in Belize; and from Bermuda. Starch gel electrophoresis was conducted upon proteins
in the digestive gland of the conch.
Preliminary analysis of four polymorphisms (malate dehydrogenase 1 and 2, phosphoglucomutase,
and aminopeptidase) revealed statistically significant differentiation in two of the four polymorphisms
among the different geographic areas. Similar analyses among the four keys within the Turks and Caicos
Islands did not reveal any significant differentiation. These preliminary results suggest that there is not
sufficient dispersal between these widely spread island groups to prevent genetic differentiation.
Supported by the World Wildlife Fund-U. S.
The acquisition of a collection of western north Atlantic fishes (Pisces) by the Gray
Reference Collection, Marine Biological Laboratory, Woods Hole, MA. ALAN H.
BORNBUSCH (Department of Zoology, Duke University, NC).
During the summer of 1983, the George M. Gray Reference Collection, a synoptic collection of the
flora and fauna of the United States Atlantic coast from the Gulf of Maine to Cape Fear, SC, received
approximately 1200 specimens of western north Atlantic fishes (Pisces). Included are representatives of 74
families and 126 species, collected between 1929 and 1973 by the Woods Hole Laboratory of the National
Marine Fisheries Service (NMFS) and previously kept in storage. Prior to accessioning into the Gray
Collection, all identifications were checked and unidentified lots were identified by A. Bornbusch. Several
specimens were found to be of interest, two of which are briefly mentioned here: a single specimen of
Eumicrotremus spinosus (Miiller) (Cyclopteridae) was collected in the area of 42° 43'N, 70° 20'W; and a
specimen of Caristius, Gill & Smith (Caristiidae) was trawled 40 miles east of Nantucket, MA at a depth
of 70 to 82 fathoms. The latter specimen is of a standard length of 17.0 cm.
•With the addition of the NMFS material, 1 10 families and 317 species of Atlantic fishes are now
represented in the Gray Collection, thereby forming an important regional reference collection of western
north Atlantic fishes.
This project was made possible by a National Science Foundation Predoctoral Fellowship and a
grant from the Stephen H. Tyng Foundation of Williams College to A. Bornbusch. The help and advice
of Dr. Louise Bush, Curator and C. Diane Boretos, Assistant at the Gray Collection, and Roger Theroux
(NMFS) are gratefully acknowledged.
Anaerobic chitin degradation as a carbon and hydrogen source for sulfate reduction
and methanogenesis in salt marsh bacteria. JOSEPH N. BOYER (Virginia Institute
of Marine Science) AND RALPH S. WOLFE.
The anaerobic pathway of chitin decomposition was studied with an emphasis on product coupling
to other bacteria present in the marsh. Sediment samples were taken from Great Sippiwissett Marsh and
plated onto selective media for the isolation of chitin degraders, sulfate reducers, and methanogens.
Inoculations were performed under strict anoxic conditions using modified Hungate technique. Chitin
degraders were characterized as forming a zone of clearing around colonies on chitin agar. Black colonies
formed by the precipitation of ferrous sulfide were indicative of sulfate reducers. Methanogens produced
methane when grown under hydrogen-carbon dioxide atmosphere. Isolates transferred to chitin broth
medium were assayed for degradation products. Gas chromatographic analysis of headspace and medium
after incubation at 30°C for five days indicated the presence of acetate, hydrogen, and carbon dioxide.
506 ABSTRACTS FROM MBL GENERAL MEETINGS
Neither sulfate reducers nor methanogens grew on chitin as sole carbon source. When chitin degraders
were combined with methanogens in chitin medium, 20 mmoles of methane and 50 mmoles of acetate
per 100 mmoles chitin were produced anaerobically. Cocultures of chitin degraders and sulfate reducers
resulted in 35 mmoles of acetate with positive sulfide production. The combination of all three metabolic
types in a single tube resulted in the production of 37 mmoles acetate, 5 mmoles methane, and positive
sulfate reduction. It is interesting to note that the acetate levels in cultures containing sulfate reducers
were significantly lower than those without. This is consistent with the fact that some sulfate reducers
can use acetate as primary substrate. It seems clear that chitin degrades anaerobically and that the products
are coupled via interspecies transfer to both sulfate reduction and methanogenesis.
Speciation in the brine shrimp Artemia: cross-breeding between sexual Mediterra-
nean populations. ROBERT BROWNE (Wake Forest University).
Due to implications with regard to studies on partitioning the environmental and genetic components
of life history traits (Brown el al. in press. Ecology) and on the biogeographic distribution of brine shrimp
populations (Browne and MacDonald 1982, J. Biogeogr. 9: 331-338), three sexual Mediterranean pop-
ulations of Artemia were tested for cross-breeding potential. Virgin adults from Tunis Tunisia (TUN),
Lanarca Lake Cyprus (CYP), and Santa Pola Spain (SP), were reciprocally test crossed using the following
format and number of test pairs: TUN-TUN (3), CYP-CYP (7), SP-SP (6), TUN-CYP (8), TUN-SP (6),
and CYP-SP (7). All offspring produced were cysts, with the ratio of cysts hatched/total cysts produced
as follows: TUN-TUN 19/49, CYP-CYP 12/186, SP-SP 18/121, TUN-CYP 66/366, TUN-SP 22/142,
and CYP-SP 1/13. Although hatch rates are low in some crosses, subsequent hydrations may increase
yields, as has been found in past studies. Crossing experiments are to some extent complicated by a small
number (approximately 5%) of parthenogenetic females in the SP population. However, for each of the
three population crosses, viable cyst(s) were obtained from an SP male sire. F, progeny from all crosses
have been successfully reared to 1 8 days of age. Low cyst production in CYP-SP crosses is attributed
primarily to abbreviated adult lifespan in that group. Although F, crosses need to be conducted to check
for hybrid infertility, it is suggested that the three populations be grouped within the A. tunisiana des-
ignation.
Supported by an Archie Fellowship and a Research and Publication Fund grant from Wake Forest
University and travel funds from the Southern Regional Education Board.
Iron and phosphorus cycling in a permanently stratified coastal pond. NINA CARACO
(Boston University Marine Program, Marine Biological Laboratory) AND IVAN
VALIELA.
Although uptake by algae is generally assumed to be the only important phosphate removal mech-
anism in aquatic ecosystems, chemical reactions may remove phosphate from solution before it reaches
the biota. Iron oxides, which form at oxic/anoxic interfaces (oxyclines), are known to rapidly remove
phosphate from solution. Such chemical scavenging in aquatic systems can be important in regulating
the supply of phosphate from anoxic bottom waters to algae in the photic zone. We examined this process
in Siders Pond.
Siders Pond is a salt-stratified, meromictic coastal pond. The bottom waters are anoxic and during
1982-83 the depth of the oxycline varied between 2.5 and 7 m. Growth of phytoplankton is limited by
the supply of phosphate to the euphotic surface waters. The permanently anoxic bottom waters are
extremely high in dissolved P (up to 80 nM). Vertical mixing in Siders Pond could, therefore, supply
large quantities of phosphate to surface waters and stimulate phytoplankton growth.
To determine if scavenging of phosphate by iron oxides at the oxycline occurred, we analyzed
suspended particulates from the surface of the pond to 15 m (maximum depth) throughout the year.
Paniculate iron and phosphorus profiles showed peaks at the oxycline during most samplings, indicating
the phosphorus was being scavenged by iron oxides. In addition, these peaks in iron and phophorus were
not usually associated with peaks in paniculate organic carbon (POC) indicating the paniculate phos-
phorus at the oxycline was not all bound to organics. Both above and below the oxycline the POC:P
ratios in the pond were 100 to 200, whereas, the ratio at the chemocline ranged from 30 to 100.
To help quantify the importance of iron oxides in removing phosphate from the water column, the
downward flux of particulates was measured using sediment traps suspended at the bottom of the euphotic
zone, at the oxycline, and at 13m. We found that phosphorus flux was usually highest in the trap at the
oxycline. This peak in phosphorus flux was associated with a peak in Fe flux. As with the suspended
particulates, sediment trap material had lowest POC:P values at the oxycline. We estimate that about
half of the phosphate fluxing through the oxycline is associated with iron oxides. Chemical scavenging
by iron oxides is, therefore, significant in controlling the supply of phosphate to the photic zone in Siders
Pond.
ECOLOGY, EVOLUTION, PLANT SCIENCE 507
Interactions of harpacticoid copepods and photosynthetic microbes in the salt marsh
ALAN W. DECHO (Louisiana State University) AND RICHARD W. CASTENHOLZ.
A sandy drainage channel at Great Sippewissett Marsh, West Falmouth, was investigated during
early July with respect to its photosynthetic microbial flora and meiobenthic harpacticoid copepods. The
channel exhibited three zones, based on the color patterns of the sediments generated by the resident
microbial populations. A diatom zone, located near the center of the channel, was composed primarily
of naviculoid diatoms (densities = 4.44 ± 0.49 X 106 := x ± SD-cirT3). A purple zone, adjacent to the
diatom zone, was composed of the purple sulfur bacteria tentatively identified as Thiocapsa sp. A clear
zone, at the periphery of the channel, showed no apparent mibrobial coloration. Spectrophotometric
pigment analysis of sediments in the diatom zone, using absolute methanol extraction, showed chloro-
phyll-a concentrations of 695.2 ± 78.4 ^g-cm~3 sediment. This was primarily due to the diatoms and
partly to the cyanobacterium Oscillatoria. A concentration of bacteriochlorophyll-a of 126.9 ± 57.2 ^g
was the result of Thiocapsa. In the purple zone the chlorophyll-a (206.4 ±71.3 ^g) was less, but bac-
teriochlorophyll-a (525.7 ± 98.6 ng), a result of Thiocapsa, was quite high. In the clear zone chlorophyll-
a (341.3 ± 126.2 ^g) was generated by several cyanobacteria (Microcoleus, Lyngbya, Spirulina) which
encroach from a nearby microbial mat. Bchl-a (234.5 ± 1 16.7 j/g) was again from Thiocapsa.
For meiobenthic analyses, fifteen replicate samples were taken from the surface (0-0.5 cm depth)
sediments of each zone. Preliminary sampling indicated almost no copepods below this depth. Seven
species of harpacticoid copepods were found. Distributional data of three species showed significantly
different abundances depending on the zone (P < 0.001). Stenocaris c.f. pristina was found in very high
densities (1032 ± 125.9 = no-cm"3) in the diatom zone and relatively low densities in the purple (18.8
± 8.3) and clear zones. Psuedomesochra c.f. divaricata showed high densities in the purple (367.8 ± 78.4)
and clear (631 ± 1 18.8) zones and low densities (127 ± 68.7) in the diatom zone. Harpacticus nipponicus
exhibited a similar distributional pattern. The distribution of these species was strongly related to high
densities of photosynthetic microbes in their respective zones.
Growth responses o/Zostera marina (eelgrass) to in situ manipulations of sediment
nitrogen availability. W. C. DENNISON AND R. S. ALBERTE (University of
Chicago).
Sediment nitrogen availability to rooted aquatic angiosperms could be an important factor controlling
growth and biomass of these plants in coastal marine habitats. Therefore we examined this parameter in
relation to growth of the temperate seagrass, Zostera marina (eelgrass). In situ manipulations of nitrogen
availability were conducted with plexiglass chambers containing eelgrass roots, rhizomes, sediments, and
a diffusion exchange reservoir for pore water. Plants from shallow (1.3 m) and deep (5.5 m) stations in an
eelgrass bed in Great Harbor, Woods Hole, MA, were placed into replicate rhizosphere chambers and
grown at the shallow and deep stations for 30 days in June and July, 1983. Nitrogen availability was
increased by additions of ammonia (NH3) to the pore water reservoirs (>5 mA/ NH3) and decreased by
periodic flushing of the reservoirs. In control chambers, NH3 was initially added to the reservoirs to approximate
concentrations in the pore water (0.2 mM NH3) without further manipulations.
At the shallow station, leaf production rates were increased by 70% with additional nitrogen and
decreased by 4% with reduced nitrogen availability with respect to the control. Rhizome production rates
were increased (+1 1%) by nitrogen addition and decreased (-8%) by nitrogen reduction. Shoot density
increases were higher than controls in the added nitrogen treatment with little change in the reduced
nitrogen treatment. At the deep station, leaf and rhizome production rates and shoot densities were
unaffected by manipulations of sediment nitrogen availibility. These results provide evidence for nitrogen
limitation of growth in shallow growing eelgrass while deep station plants do not appear to be nitrogen
limited. Previous findings demonstrated light limitation of eelgrass growth at depth. Therefore, two
different limiting environmental factors, daily light period and nitrogen availability, can function in
determining the growth and depth distribution of Zostera marina within a single community.
Research supported by NSF Grants DEB 8210322 and OCE 8214914.
The sensitivity of fresh waters of Cape Cod, Massachusetts to acid precipitation.
JENNIFER DUNCAN (Marine Biological Laboratory), BRUCE PETERSON, AND
SALLY MARQUIS.
The annual mean pH of rain on Cape Cod is approximately 4. 1 . Have these acid inputs decreased
the pH of Cape Cod ponds and lakes? Are these surface freshwaters susceptible to a decrease in pH in
the future as they receive additional inputs of acid precipitation? The limited data available can only
provide a conclusive answer to the second question. The only two relatively complete surveys of Cape
Cod ponds and lakes include a 1975 study by the Environmental Management Institute (EMI), which
508 ABSTRACTS FROM MBL GENERAL MEETINGS
reported pH, conductivity, and chemical composition; and another begun in 1983, the Acid Rain Mon-
itoring survey (ARM), which is reporting pH and alkalinity. The EMI study showed that the mean pH
of surface waters was 5.08 and that approximately 20% of the ponds had a pH lower than 5. The
preliminary results of the ARM study confirm the high acidity of many ponds, and further show that
the alklainity of most are in the critical to highly sensitive range.
A model devised by Norwegian scientist Arne Henriksen may be useful in predicting the degree of
acidification of Cape Cod ponds. The premise of this model is that non-marine calcium and magnesium
are balanced by a nearly equivalent amount of bicarbonate in unacidified waters. Although bicarbonate
may be consumed and replaced by sulfate ion during acidification, thereby lowering the alkalinity, calcium
and magnesium should still represent unacidified alkalinity. Calcium makes up the major portion of
cations that originate from weathering, and may well be adequate for the purpose of the model. A graph
of pH versus excess calcium constructed by Henriksen showed a well defined dividing line between the
unacidified waters of southern Sweden and the acidified waters of western and central Sweden. The EMI
Cape Cod data plotted on a similar graph shows that most of the ponds are in the "acidified" portion.
The Henriksen calcium-pH model suggests but does not conclusively prove that Cape Cod pond
chemistry has been changed by acid precipitation. The data do show that Cape ponds have low alkalinity
and are likely to be sensitive to acidification.
Growth and photosynthetic responses to temperature of two populations of Zostera
marina. ANN S. EVANS (Virginia Institute of Marine Science, College of William
and Mary).
The growth and photosynthetic responses to temperature of two disjunct populations of the seagrass
Zostera marina were investigated during July 1983. Plants were collected in sediment cores from a 24-
27°C coastal lagoon, Bourne Pond (Falmouth, MA), and a 20-22°C open coastal environment, Great
Harbor (Woods Hole, MA) for growth in aquaria at 1 5 and 25°C, and for growth in their natural habitats
(in situ plants). After 2 weeks, photosynthetic characteristics were determined at 15 and 25°C for all
growth conditions, and after 4 weeks growth was measured.
The in situ Great Harbor plants had greater shoot length and biomass, specific leaf weight, and total
production than in situ Bourne Pond plants. These growth differences were maintained when plants from
both populations were grown at 15°C. The 25°C growth condition resulted in complete mortality of both
populations. In situ and aquaria plants from both populations had higher maximal rates of photosynthesis
(Pmax) at 25 than 15°C. Pmax, in ^mol O2 mg chl~' min"1, for plants grown in aquaria and measured at
their growth temperature, was 0.43 ± 0.10 at 15°C and 0.49 ± 0.09 at 25°C for Bourne Pond, and 0.18
± 0.03 at 15°C, and 0.59 ± 0.1 1 at 25°C for Great Harbor. Similar patterns for Pmax were observed for
in situ plants. The Bourne Pond plants grown at 25°C had a Pmax of 0.15 ± 0.02 when measured at
15°C, which was 31% lower than that at 25°C. The Great Harbor population, when examined under
identical conditions, showed a decrease of 36% in Pmax. In contrast, plants grown at 15°C and measured
at 25°C showed 61% and 51% increases for Bourne Pond and Great Harbor plants, respectively.
The fact that both populations had higher Pmaxs at the temperature (25°C) at which mortality occured
indicates that photosynthetic performance does not necessarily predict ecological success. The data further
indicate that the Bourne Pond population is phenotypically more plastic than the Great Harbor population
to growth temperature. That growth differences were maintained under identical conditions at 15°C
suggests the two populations may be ecotypically distinct.
Photosynthetic activity o/Zostera marina L. epiphytes in relation to light regime and
substratum. L. MAZZELLA (Stazione Zoologica di Napoli, Italy), W. C. DEN-
NISON, AND R. S. ALBERTE.
The epiphytic algae colonizing the leaves of submerged angiosperms such as seagrasses can contribute
significantly to the productivity and biomass of these ecosystems. We sought to assess the photosynthetic
activity, biomass, and productivity of Zostera marina epiphytes in relation to light regime and substratum.
Plants were marked in situ and collected 15 days later during July 1983, at two stations (1.3 m and 5.5
m) in an eelgrass bed in Great Harbor, Woods Hole, MA. Light regime was manipulated with underwater
lamps and shade screens. In addition, artificial eelgrass (polyammide blades) were planted at the control
site of both stations and collected after 1 5 days.
At the shallow station, light saturated photosynthetic activity (^mol O2 dm~2 min~') of Zostera
epiphytes was 0.12, 0.06, and 0.01 for increased, control, and decreased light treatments, respectively.
A similar trend was found in epiphyte biomass; 25.7, 18.9, and 5.6 mg dm~2 at the increased light,
control, and decreased light treatments, respectively. At the deep station, epiphyte photosynthetic activity
is 0.06, 0. 10, and 0.0 j/mol O2 dm"2 min"1 at the increased light, control, and decreased light treatments.
ECOLOGY, EVOLUTION, PLANT SCIENCE 509
respectively. A parallel trend was observed in epiphyte biomass for the three light treatments (22.2, 20.0,
and 0.00 mg dm 2). Experiments with artificial eelgrass leaves showed a higher photosynthetic rate (0.35
compared to 0.21 ^mol O: dm 2 min ') and biomass (1 1 1.3 compared to 84.5 mg dm 2) at the shallow
station than at the deep station.
Epiphyte photosynthetic activity, biomass, and productivity accounted for 7%, 4% and 8%, respec-
tively, of eelgrass at the shallow station, and 1 2%, 5%, and 7%, respectively, at the deep station. Light
regime manipulations strongly affected Z. marina epiphyte photosynthetic activity, biomass, and pro-
ductivity. The potential for epiphyte growth, as demonstrated by the rapid colonization of artificial
substratum, is greatest at the shallow station however, epiphyte biomass on Zostera leaves does not differ
with depth. Therefore, we conclude that epiphyte colonization and growth of Zostera marina is controlled
by leaf characteristics and by light regime.
Sulfate reduction following marsh grass die-back. SUSAN M. MERKEL (Ecosystems
Center, MBL), JEAN M. HARTMAN, AND ROBERT W. HOWARTH.
Sulfate reduction rates were measured following Spartina alterniflora die-back in an effort to understand
better decomposition in salt marsh soils. Marsh sediments below the top centimeter are anoxic, and the
major decomposition occurs through sulfate reduction and related fermentations. Sulfate reduction is fueled
by simple organic compounds which are metabolic products of fermentation. The source of these compounds
could be rapidly used labile organics or more slowly decomposed refractory root material. Die-back was
induced in two short Spartina sites in Great Sippewissett Marsh (Cape Cod, MA). One site was covered
with a board for 12 months, the other was covered with Spartina rack for 6 months and uncovered for 20
months. Four months after coverage, the board site showed sulfate reduction rates 40% higher than the
control site. After this initial rise, the ratio of sulfate reduction rates at the dieback site versus controls
decreased to .85 after 6 months, .75 after 13 months, and .45 after 26 months. Spartina did not recolonize
the rack site, even after 20 months. Hydrogen sulfide concentrations (0.2 mA/) were an order of magnitude
below that found to cause living grass to die.
We suggest that "leakage" from live roots maintains a certain level of SO4 = reduction. Following
die-back, a pulse of labile organic compounds is released which fuels high SO4 = reduction rates. As these
labile compounds are used up, SO4 = reduction rates decrease. Refractory root material is slowly decom-
posed by fermentative bacteria whose products support low levels of SO4 = reduction.
Diel vertical movements of bacteria in intertidal streams of Sippewissett Marsh. KEN-
NETH M. NOLL (Dept. of Microbiology, University of Illinois) AND RICHARD W.
CASTENHOLZ.
Bacteria living in the intertidal streams of the Sippewissett Salt Marsh move vertically within the
sediment in response to daily changes in environmental conditions. Light intensity plays a major role
in these movements. Both field and laboratory experiments demonstrated that a species of cyanobac-
terium, Oscillatoria, migrates to the surface of the sand under low light conditions. At night Oscillatoria
is distributed between 3 and 5 mm below the surface. This Oscillatoria also migrates toward sources of
reduced sulfur suggesting a role for sulfide in its daily movements. Sulfur-containing filaments of the
sulfide-oxidizing bacterium Beggiatoa come to the surface at night and migrate down during the daytime.
The appearance of colorless filaments lacking sulfur granules during the day at 4 mm below the surface
suggested the Beggiatoa may use the stored sulfur during the day. Gliding green sulfur bacteria of the
genus Chloroherpeton apparently migrate upward at dusk. A gliding, filamentous, phototrophic bacterium
(Chloroflexus-like) was found in these sediments and may also migrate upward at dusk. Preliminary
studies involving in vivo pigment analyses allowed more sensitive determinations of the vertical distri-
bution of species than did microscopic observations.
Vertical movements of the hard clam, Mercenaria mercenaria, in response to changes
in barometric pressure. EUGENE C. REVELAS (Marine Sciences Research Center,
State University of New York, Stony Brook, NY 1 1 794).
The hard clam or quahog, Mercenaria mercenaria is a commercially important bivalve species
common along the east coast of the United States. Due to its economic value, Mercenaria has been well
studied, yet little is known about the clam's burrowing behavior. Commercial clammers contend that
hard clams move up and down in the substrate aperiodically. Also, there are scattered reports in the
literature of clams found at unusual depths in the sediment (two to three times their normal life position
of one to five cm).
510 ABSTRACTS FROM MBL GENERAL MEETINGS
To investigate the burrowing behavior of Mercenaria, a method was devised (involving the attach-
ment of nylon string to clams) by which vertical movements of clams within natural sediments could
be determined. The movements of 1 5 clams placed in sand-filled aquaria in a running sea water system
were monitored from 16 June to 18 August 1983. Simultaneously, water temperature and barometric
pressure were recorded. During this period, vertical movements of clams were correlated with changes
in barometric pressure (Kendall's rank test for association, P < .05). That is, relatively large decreases
in pressure (>.40"Hg) coincided with downward movement (~1 cm) of clams. Clams did not move
during periods of steady or slightly changing pressure. To further test this response, clams were placed
in a sealed aquarium in which air pressure could be manipulated. Preliminary results from this system
also indicate that Mercenaria burrows deeper as pressure drops.
Mercenaria'^ response to barometric pressure may be the result of strong selective pressure to remain
within the sediment, thereby avoiding predators. Decreasing barometric pressure is often associated with
increased wind and wave action; therefore clams may move deeper to avoid disinterment by sediment
scour. How Mercenaria detects changes in pressure is unknown. However, the mechanism may be tied
to an internal clock which enables the clam to filter out larger, but predictable wave- and tide-induced
hydrostatic pressure changes.
Effect of age and quality of detritus on growth of the salt marsh snail, Melampus
bidentatus. CAROL S. RIETSMA (State University of New York at New Paltz).
Quality (nitrogen content) of Spartina alterniflora detritus as a food source for detritivores can be
altered by chronic fertilization. Fertilization increases its nitrogen content.
Newly formed detritus has high nitrogen and ferulic acid contents. As detritus ages, available nitrogen
is rapidly lost. Ferulic acid, an abundant phenolic acid, is lost more slowly.
Quailty and age of detritus can affect its palatability to detritivores such as salt marsh snails, Me-
lampus bidentatus. Detritus that is newly formed and from fertilized plots is more palatable. However,
high ferulic acid reduces palatability (Valiela et al. 1979, Nature 280: 55-57). This inhibitory effect
can be decreased or eliminated by artificially increasing nitrogen content (Rietsma 1981, Biol. Bull.
161: 330).
This study tested the effect of detritus quality and age on growth of snails.
Snails were reared on Spartina alternifora detritus in four treatments: detritus aged for two weeks
or for eight months from both control and fertilized salt marsh plots. Shell lengths and wet weights of
snails were measured at two week intervals for eight weeks. Each treatment was replicated five times
with a total of 50 snails in each treatment. Detritus was analyzed for carbon, nitrogen, total soluble
phenolic acids, ash, fiber, and soluble carbohydrates.
Snails fed 8-month-old detritus from control plots grew faster than in all other treatments. This
detritus had a higher carbon:nitrogen ratio considered unfavorable for growth. However, it had the lowest
phenolic acid, highest ash, and lowest fiber contents. Snails fed 2-week-old detritus from control and
fertilized plots grew slowest. Growth differences seemed to be related to the phenolic acid content of
detritus. It appears that the phenolic acid content of detritus overrides the importance of nitrogen content
in snail growth.
Supported by a SUNY Research Fellowship and Grant-in-Aid.
The design and construction of a benchtop reactor to model an anaerobic /oxic waste-
water treatment system. GEORGE J. SKLADANY (Clemson University), BRIAN A.
WRENN, AND ROBERT R. HALL.
Exposing sludge to an initial anaerobic treatment in a zone wherein influent and recycled sludge
are initially contacted followed by an aerobic treatment results in efficient removal of phosphate from
the water and produces a sludge with excellent settling qualities (M. Timmerman Dev. Ind. Microbiol.
1979, 20: 285-298.) To understand the biological phenomena we have designed a plug-flow reactor able
to model the anaerobic/oxic treatment process.
The reactor is constructed of interlocking circular plastic Tupperware hamburger freezing containers,
2 cm deep and 10.5 cm wide. Passages were cut in the plastic allowing gas and liquid to flow through
the system. Nitrogen gas or compressed air introduced into the bottom of each chamber mixed the
contents and maintained anaerobic or aerobic conditions. Sampling ports made from bulkhead fittings
modified to hold a rubber septum allowed liquids to be added or removed from any chamber with a
syringe. A clarifier. constructed of similar plastic sections was added to the final stage of the reactor to
provide for settling and recycling of the biomass. Clarified effluent passed out of the reactor through a
wier cut in the wall of the clarifier. Settled sludge was removed from the clarifier through glass tubes
attached to bulkhead fittings in place at the top of the clarifier, with open ends extending down into the
ECOLOGY, EVOLUTION, PLANT SCIENCE 5 1 1
sludge. Flow rates were controlled with a peristaltic pump. The device was clamped between two rect-
angular pieces of plastic for support. The reactor allows flexibility in experimental design and can ac-
comodate experiments not possible at a wastewater treatment plant.
This work was supported by funds to the Microbial Ecology course from Air Products and Chemi-
cals, Inc.
Deforestation in the Amazon Basin measured by satellite: a release ofCO2 to the
atmosphere. T. A. STONE, R. A. HOUGHTON, J. M. MELILLO, AND G. M.
WOODWELL (Ecosystems Center, Marine Biological Laboratory).
Deforestation is contributing to the build-up of CO2 in the atmosphere, yet there exists a six-fold
difference in the estimated rates of deforestation in the tropics. Many countries do not know the rates
at which their forests are being converted to other uses, and their estimates may be biased by political
or economic objectives. Use of satellite data allows complete and repeated coverage of the land surface
and the opportunity for an unbiased approach.
A change detection method using LANDS AT data has been applied to an area of 185 X 185 km
in the Brazilian state of Rondonia on the southwestern edge of the Amazon Basin. Rondonia has been
rapidly deforested by extensive colonization since 1 970. A time series of 0.0045 km2 resolution LANDSAT
data was analyzed and showed rates of deforestation of 26,900 ha/yr from 1976 to 1978 and 55,200 ha/
yr from 1978 to 1981. These data, combined with data from the literature on above and below-ground
biomass were used with a model to calculate that between 3.7 X 10'2 g C and 5.5 x 10'2 g C was released
to the atmosphere in 1981 from the area of the LANDSAT scene.
To examine the entire Amazon Basin would require use of about 400 LANDSAT scenes unless a
sampling strategy is adopted. An alternative method would be to use the NOAA7 satellite with an image
swath of 2400 km and a resolution of 1 km2, to determine areas of intense deforestation which can then
be examined in detail with the LANDSAT satellite data.
This research was supported by the Department of Energy grant P8000014.
Age of first reproduction in Melampus bidentatus: the effects of overwintering degrowth
and repair. JAY SHIRO TASHIRO, MARK WILTSHIRE, AND CHARLES POHL (Ken-
yon College, Gambier, OH 43022).
Theories of life-cycle evolution lack substantive data bases quantifying relationships between a particular
reproductive effort and future reproductive potential. Such relationships involve trade-offs in physiological
allocation of resources to repair of somatic tissues or into reproductive products. Biological repair of integral
structures sustains the life of an organism, but there is selection for balances between increasing probability
of survival (repair) and diverting resources to fecundity.
Overwintering maintenance and emergency repair were assessed in the salt-marsh pulmonate gas-
tropod species Melampus bidentatus. Maintenance repair was examined in specimens of Melampus from
a population in Little Sippewisset salt marsh, just north of Woods Hole, Massachusetts. The life cycle
and early life-history of specimens from this population had been reported in an elegant study (Russell-
Hunter el al. 1972, Biol. Bull. 143: 623-656). Collections from the summer of 1983 had the same age
and size structure as that reported for this population a decade ago. Snails were placed in a diapause
state in the winter of 1981. Experimentally induced diapause (10°C) was maintained for four weeks, at
which time tissue protein, carbohydrates, and dry weight were analyzed for 2- and 3-year-old animals.
Only the younger snails showed significant changes in biomass constituents (protein loss relative to
prediapause controls).
Emergency repair was assessed in the winter of 1982 using diapausing specimens of Melampus from
a population near Weymouth, Massachusetts (maintained under conditions identical to those used for
the Little Sippewisset population). The right tentacle for 2- and 3-year-old animals was ablated after four
weeks in diapause, followed by injection of 3H-thymidine. Nine days later, regrowth and repair were
quantified (morphometric measurements and autoradiographic techniques). The data suggest that younger
snails had higher rates of emergency repair during diapause.
Maintenance repair is manifest as protein degrowth in overwintering specimens of Melampus, but
degrowth is age-specific. Emergency repair during diapause is also age-specific. Repair can enhance sur-
vivorship and residual reproductive capacity, but degrowth in younger snails could delimit age of first
reproduction by precluding partitioning of resources into reproductive efforts.
This work was supported by a Theodore Roosevelt Memorial Grant from the American Museum
of Natural History.
512 ABSTRACTS FROM MBL GENERAL MEETINGS
The role of freshwater wetlands in the ontogeny of a New England salt marsh. JOSEF
P. TREGGOR (Central Connecticut State University, New Britain, CT 06050).
Previous ontological investigation of the Great Sippewissett Marsh (Falmouth, MA) indicated that
freshwater wetlands preceeded the upland development of the saltmarsh. This conclusion was based on
the presence ofPhragmites communis roots and rhizomes in salt marsh core horizons. While P. communis
can be found in freshwater systems, its window -of salinity tolerance is far too great for it to be a reliable
indicator. Therefore a method of freshwater horizon identification subject to less error was necessary.
Cores were taken in contemporary freshwater systems. Parameters yielding positive correlations with
the "freshwater" horizons were: root/rhizome dry weight, water and organic content, and direct com-
parison of root/rhizome samples. These parameters provide sufficient evidence for identification of fresh-
water horizons in salt marsh cores.
Cores were taken directly in the salt marsh to qualify the type of wetland, to determine any succes-
sional patterns, and to estimate the extent of coverage. The deep cores provided well-defined horizons
ofSpartina alterniflora. short form (high marsh cord grass), S. patens (salt hay), Chamaecyparis thyoides
(atlantic white cedar), Typha sp. (cattail), and Quercus sp. (scrub oak). P. communis was found throughout
the core but never in distinct horizons. The C. thyoides was oriented longitudinally in the tube indicating
the presence of standing trunks. Depths of salt marsh peat overlaying the freshwater horizons were <1
m while in the adjacent marsh exhibiting no freshwater constituents, peat depths exceeded 2 m.
The results of these analyses provide a successional history of the freshwater wetlands: a. wet uplands,
b. freshwater marsh, c. cedar swamp, and finally d. direct colonization of the high salt marsh. These areas
were of substantial proportion and resulted in the considerable delay of the upland expansion of the
marsh. The presence of P. communis as stated suggests that it was transitional between periods of systemic
change.
Microbial selection in an artificial ecosystem. W. S. VINCENT (University of Dela-
ware) AND ROBERT M. HALL.
Commercial waste water treatment plants form an artificial ecosystem in which waste water con-
tributes nutrients in very dilute concentrations. These nutrients are then converted into bacterial biomass
which is removed or recycled through the system as sludge. The effluent, essentially nutrient-free, is then
discharged into streams, lakes, and ground water.
As sludge is concentrated by gravitational settling, only those bacterial cells which can form clumps
will remain in the system. As 20% to 50% of the sludge is recycled with each complete flow through of
the system, there is a powerful selection process for cells which cause clumping, as well as those which
can be clumped.
Twenty-one different bacterial isolates from many sources were tested for their ability to be clumped
by a clumping strain (CH-1) isolated from a sludge sample. Only 3 strains, all Acinetobacter-like, isolated
from sludge, were competent to be clumped. Four strains of Acinetobacter Iwoffii isolated from soil were
not competent.
Several aspects of the clumping process have been determined. The CH-1 strain is hydrophobic; the
competent strains are not. The competent strains do not form clumps in the absence of CH-1. Competent
strains adhere to a small clump of GH-1, and then are able to cause other cells to adhere. Neither
CH- 1 cell-free growth media nor heat killed CH- 1 cells will cause clumping.
Thus, the water treatment system acts as a highly selecting ecosystem which favors the retention of
oligotrophic, clumping bacteria. As clumping is a procedure by which oligotrophic bacteria increase their
ability to take advantage of scarce nutrients, the recycled sludge process conserves those cells which
clump. Others will be lost from the system.
This work was supported in part by a grant from Air Products and Chemicals, Inc. to the Microbial
Ecology Course.
GAMETES AND FERTILIZATION
Is there specificity in the induction ofpolyspermy in sea urchins by protease inhib-
itors? M. C. ALLIEGRO AND H. SCHUEL (Dept. Anat. Sci., SUNY at Buffalo).
Sea urchin eggs contain a trypsin-like proteolytic activity that is activated and secreted during
fertilization. Results obtained in several laboratories have implicated this activity in the cortical reaction,
elevation of the fertilization envelope, and the establishment of the block to polyspermy. These conclu-
GAMETES AND FERTILIZATION 513
sions are supported in part by studies using several trypsin inhibitors (reviewed by: Schuel 1978, Gamete
Res. 1: 299.). However, this interpretation has been challenged, and past results with inhibitors were
attributed to a non-specific protein effect (Dunham et al. 1982, Biol. Bull. 163: 420.). This hypothesis
was tested by comparing the potency of enzymatic inhibitors of the sea urchin egg protease — soybean
trypsin inhibitor (SBTI), ovomucoid, limabean trypsin inhibitor, antipain (AP), leupeptin (LP), and tosyl
lysine chloromethyl ketone (TLCK.) — with their ability to promote polyspermy. Proteolytic activity was
isolated from unfertilized Strongylocentrotus purpuratus eggs by SBTI-affinity chromatography (Fodor
et al. 1975, Biochemistry 14: 4923.) in the presence of benzamidine to reduce autodigestion (Baginski
et al. 1982, Gamete Res. 6: 39.). The ability of the protease inhibitors to promote polyspermy in Arbacia
punctulata eggs coincides with their potency as inhibitors of the purified Strongylocentrotus egg protease.
According to Spearman's rank correlation, rs equals 0.93 (P < 0.01). Since AP, LP, and TLCK are not
proteins, it is unlikely that their action on eggs is due to a non-specific protein effect. Furthermore,
inactivation of SBTI by treatment with acid or alkali (Kunitz 1947, J. Gen. Physiol. 30: 291.) abolished
its ability to inhibit the egg protease or to cause polyspermy. We conclude that induction of polyspermy
in sea urchin eggs by protease inhibitors is indeed due to inhibition of the egg protease.
Supported by NSF grant #PCM-82-01561 to H.S.
Superoxide dismutase biomimetic compounds prevent fertilization in Arbacia punc-
tulata eggs. FREDRIC BLUM, MARGARET NACHTIGALL*, AND WALTER TROLL.
(N.Y.U. Medical Center, New York, NY).
Superoxide dismutase (SOD) is the major catalyst for the formation of hydrogen peroxide (H2O2)
from superoxide (O2). Fertilized sea urchin eggs excrete H2O2 promptly after fertilization thus inactivating
other sperms about to enter the eggs. Entrance of more than one active sperm disturbs normal devel-
opment and causes polyspermic fertilization. Destruction of H2O2 by catalase also results in polyspermy.
However SOD, which we expected to cause a burst of H2O2 production, did not disturb fertilization
(Colburn et al. 1981, Dev. Biol. 84: 235). A possible reason for SOD's lack of activity is that it does not
cross cell membranes and the egg does not excrete O2 . Thus enzyme and substrate may never meet.
SOD-biomimetic compounds (SODB) that are ether soluble and can cross cell membranes have recently
been described (Kensler et al. 1983, Science 221: 75). We noted that application of a SODB, e.g., Cu(II)-
diisopropyl salicylate, resulted in inhibition of fertilization presumably due to a burst of H2O2. Addition
of catalase, which converts H2O2 to O2, at a concentration of SODB where fertilization is completely
blocked resulted in 100% fertilization. This supports the proposed role of H2O2 in the inhibition of
fertilization. Sperms treated with SODB were not damaged. An unexpected finding was that addition of
SODB 30 s after fertilization caused polyspermy. This was significantly reduced by catalase and suggests
that sperms attempting to enter fertilized eggs 30 s after fertilization are prevented by a mechanism,
perhaps a physical membrane, vulnerable to H2O2.
Supported by NIH-CA- 16060 to W.T. and N.Y.U. Honors Program to F.B.
* Princeton University.
What makes cyclin cycle? RICHARD CORNALL, ELAYNE BORNSLAEGER, AND TIM
HUNT (Department of Biochemistry, University of Cambridge, England).
Cyclin is a 55,000 M.W. protein which is synthesized at a high rate after fertilization of Arbacia
punctulata eggs. It is destroyed at a certain point in the cell cycle (Evans et al. 1983, Cell 33: 389-396);
because its synthesis is continuous its level oscillates in a saw-tooth pattern. We wished to determine the
exact point of cyclin disappearance, and to investigate the effects of various inhibitors of the cell cycle
on its behavior in order to understand its role in early development.
35S-methionine was added to suspensions of fertilized eggs at 20°C. Samples were removed at in-
tervals and analyzed by autoradiography of SDS polyacrylamide gels. Parallel samples were fixed to
determine cleavage index, or orcein stained to visualize chromosomes. The precise point of cyclin break-
down was most clearly defined when 10~4 M emetine was added to the suspensions 50 minutes after
fertilization to prevent further cyclin synthesis during the period of rapid degradation. When added at
this time, emetine does not inhibit cleavage, and abrupt degradation of cyclin occurred at a time cor-
responding to the metaphase-anaphase transition. Early addition of emetine (20 minutes post-fertiliza-
tion), however, prevented both cleavage and cyclin degradation. Parthenogenetic activation of protein
synthesis by 10 mM NH4C1 also precluded cleavage and cyclin breakdown, as did the DNA synthesis
inhibitor aphidicolin (5 Mg/ml), provided that it was added before completion of DNA synthesis. The
motility inhibitors colchicine (100 nM), taxol (10 j/g/ml), and cytochalasin D (2 Mg/ml) prevent cleavage
514 ABSTRACTS FROM MBL GENERAL MEETINGS
but do not block cyclin breakdown. However, they delay the onset and reduce the rate of cyclin deg-
radation.
Our results strongly suggest, but by no means prove, that cyclin has a role in mitosis and cell division,
and that its disappearance is necessary for normal completion of the process.
We acknowledge the support of NIH training grant GM-31 136-05.
Isolation of cytoskeletons from Chaetopterus eggs. WILLIAM R. ECKBERG (Howard
University) AND GEORGE M. LANGFORD.
On cytological and experimental evidence, we indicated that 1) ooplasmic components in Chae-
topterus are embedded in a cytoskeletal matrix and that 2) this matrix is responsible for ooplasmic
reorganization in development and differentiation without cleavage. We report the initial results on
isolation and characterization of that matrix.
To visualize the surface structure of the cytoskeletons, we developed a procedure for rapid quantitative
vitelline layer removal. We added an equal volume of 0.5 M sucrose in 0.125 M EDTA, pH 8, to a 50%
egg suspension in MFSW, and centrifuged eggs out by hand after 20-30 seconds. Eggs treated in this way
could be fertilized, developed a rapid partial block to polyspermy, and developed to trochophore larvae.
Cytoskeletons were isolated by suspending vitelline layerless oocytes in 10-20 vol of 10 mM PIPES,
pH 6.8; 300 mM sucrose; 100 mM KC1; 5 mM MgCl2; 1 mM EGTA; 100 nM PMSF and 1% NP-40.
Oocytes were suspended for 30 min during which they cleared, beginning at the surface and moving
inward. Cytoskeletons were centrifuged by hand, washed with the above buffer minus NP-40, and re-
centrifuged.
Cytoskeletons contained 20-30% of the total cellular protein. SDS-polyacrylamide gel electrophoresis
indicated that the polypeptide composition of isolated cytoskeletons was qualitatively similar to that of
whole vitelline layerless eggs. However, several polypeptides were quantitatively reduced in isolated cy-
toskeletons and others were quantitatively increased. These latter polypeptides had apparent molecular
weights of 45K, 54K, 63K and 85K.
Cytoskeletons were fixed and processed for scanning electron microscopy which revealed that 1 )
cytoskeletons isolated by this procedure were entirely cortical and subcortical and 2) the inner surface
of the cytoskeleton showed granules of various sizes embedded in a filamentous network. These results
greatly extend and confirm our previous observations.
S. Johnson and D. Rogers provided technical assistance and were supported by the Rockefeller
Foundation/Howard University/MBL Careers in Science Program.
Calcium transients during fertilization in single sea urchin eggs. A. EISEN (Children's
Hospital, Philadelphia), G. T. REYNOLDS, S. WIELAND, AND D. P. KIEHART.
Two events associated with a putative transient increase in cytoplasmic free calcium include: acti-
vation of the starfish oocyte with the maturation hormone 1 -methyl adenine (1-MA), and fertilization
of the starfish egg with sperm. These events were investigated in single oocytes and eggs by the detection
of calcium specific luminescence from single cells injected with an acetylated form of the photoprotein
aequorin(10mg/mlin lOmA/HEPES, 0.2 mA/EGTA, pH 7.0 tp 3% of cell volume). Using a microscope-
photomultiplier and a microscope-image intensifier-SIT vidicon detector sensitive to <10~7 M Ca++ we
found: 1) a barely detectable (=S10~7 A/) change in free calcium from oocytes in response to 1-MA (final
concentration ca. 150 nM\ and 2) a large (ca. 10~6) increase from eggs fertilized with sperm 15 minutes
after application of 1-MA and 5 minutes after general vesicle breakdown (17°). The calcium-aequorin
luminescence increases as it propagates over 30-40 s and decays uniformly over 200-300 s. The absence
of a calcium transient in the Asterias forbesi differs significantly from the large (ca. 10 6 A/) transient
reported in the M. glacialis oocyte and is suggested as being a common feature of starfish oocyte activation.
The calcium transient at fertilization in Asterias eggs is similar to that described in several species of
sea urchin (A. punctulata and L. variegatus) although the propagation time is much longer in the
starfish egg.
We thank Dr. O. Shimomura for the gift of acetylated aequorin. We thank Dr. A. J. Walton for the
use of his microscope objectives and assistance in the experiments. This work was supported by DOE
Contract EY-76-S-02-3120 to G.T.R.
Calcium transients during fertilization in single sea urchin eggs. A. EISEN (Children's
Hospital, Philadelphia), G. T. REYNOLDS, S. WIELAND, AND D. P. KIEHART.
A transient increase in cytoplasmic free calcium occurs in the eggs of the sea urchins Arbacia
punctulata and Lythechinus variegatus at fertilization. This transient has been detected from the lumi-
GAMETES AND FERTILIZATION 515
nescence originating within single eggs injected with the calcium specific photoprotein aequorin. We used
the native protein and an acetylated form (10 mg/ml in 10 mM HEPES, 0.2 mM EGTA, pH 7.0, injected
to 3% cell volume), in conjunction with a microscope-image intensifier-SIT vidicon, or a microscope-
photomultiplier to determine the spatial distribution and time course of the calcium transient. In the
Arbacia egg the transient begins 26 ± 4 s after membrane depolarization. The Ca-aequorin luminescence
increases over 6-12 s, persists behind the advancing wave front, remains at its peak for ca. 25 s, and
decays uniformly over 100-120 s. The onset and peak of the luminescence occurs long before the onset
of fertilization membrane elevation, which occurs 54 ± 9 s after membrane depolarization. Observations
of the centrifugally organelle-stratified Arbacia egg indicate a possible biphasic release of Ca++ from two
sources, with the majority of the Calcium coming from a mitochondria associated source and going into
the mitochondria.
Similarly, in the L. variegatus egg, a Ca++ wave quickly traverses the egg, persisting behind the
advancing front, and decaying uniformly.
We thank Dr. O. Shimomura for the gift of native and acetylated aequorin. We thank Dr. A. J.
Walton for the use of his microscope objectives. This work was supported by DOE Contract EY-76-S-
02-3120toG.T.R.
A major maternally encoded 4 IK protein in both Spisula and Arbacia binds to an
anti-tubulin affinity column. ELIZABETH L. GEORGE, SARAH BRAY, ERIC T.
ROSENTHAL, AND TIM HUNT (Department of Biochemistry, University of Cam-
bridge, England).
The pattern of protein synthesis changes at fertilization in both Spisula solidissima and Arbacia
punctulata (Rosenthal et al. 1980, Cell 20: 487; Evans el al. 1983, Cell 33: 389). This rapid and specific
control of translation after fertilization suggests that the set of proteins activated in these organisms may
play a role in cell division during early development. Protein C in Spisula and protein B in Arbacia both
have a molecular weight of 41,000 and accumulate during early development rather than cycle with each
cell division. Both proteins also bind specifically to an anti-tubulin affinity column (rat monoclonal anti-
yeast alpha tubulin, provided by John Kilmartin). We wished to determine whether protein C in Spisula
bound directly to the tubulin antibody, or whether its binding was indirect via endogenous tubulin which
was present in the cell extract. Oocytes were activated by KC1 and newly synthesized proteins were labeled
with 35S-methionine. The post-ribosomal supernatant of the cell homogenate was then treated with Taxol
(40 ftM) in orderto removeendogenous tubulin . Protein C did not precipitate with the stabilized microtubules,
but remained in the supernatant. This supernatant was then passed over the anti-tubulin column, and
protein C was still qualitatively retained in the bound fraction as judged by SDS-polyacrylamide gel
electrophoresis. This suggests that protein C in Spisula shares an antigenic determinant with tubulin,
rather than having affinity for tubulin itself. The anti-tubulin affinity column system is a specific method
of purification of protein C, and preliminary evidence suggests that a coomassie-stained protein corre-
sponds to the labeled band. This should permit raising antibodies against protein C, which would be a
powerful tool in elucidating its cellular role.
This work was supported by NIH Training Grant GM-361 16-05.
An organelle complex responsible for mRNA localization in the cortex o/"Chaetop-
terus eggs. WILLIAM R. JEFFERY (University of Texas at Austin).
The mechanism of cortical mRNA localization in the egg of Chaetopterus pergamentaceus was examined
by a combination of in situ hybridization, centrifugation, and electron microscopy. The egg contains three
cyt&plasmic regions; the hyaloplasm, the endoplasm, and the cortical ectoplasm. The hyaloplasm consists
of clear cytoplasm derived from the germinal vesicle (GV), the endoplasm contains lipid and yolk particles,
and the ectoplasm is composed of very electron-dense particles embedded in a granular-fibrillar matrix. In
situ hybridization with poly(U), actin DNA, and histone DNA probes showed that more than 95% of the
poly(A)+RNA, actin mRNA, and histone mRNA was localized in the ectoplasm of the mature egg, although
this region represents less than 25% of the total egg volume. The mRNA appeared to co-distribute with
the ectopiasmic organelle complex (EOC) during early development. Both entities were present in the cortex
of mature eggs and zygotes, entered the endoplasm just prior to the first cleavage, and returned to the
cortex as the astral rays elongated during early cleavage. The ectopiasmic mRNA localization and the EOC
were divided into animal and vegetal fields shortly before the first cleavage. The animal field entered the
AB and CD blastomeres while the vegetal field was localized in the polar lobe of trefoil embryos and was
shunted primarily to the CD cell. In situ hybridization was conducted on eggs centrifuged through Ficoll
step gradients to determine whether the ectopiasmic mRNA is associated with the EOC. Centrifugation at
500 x g for 5 min at 18°C caused the EOC to be displaced to the centrifugal pole of the egg. Centrifugation
also caused quantitative displacement of the poly(A)+RNA, actin mRNA, and histone mRNA to the
516 ABSTRACTS FROM MBL GENERAL MEETINGS
centrifugal pole of the egg. These results suggest that maternal mRNA may be localized in the egg cortex
and differentially segregated to the AB and CD blastomeres by an association with the EOC.
This work was supported by NIH Training Grant 5-T35-HD07098 awarded to the Embryology
Course, Marine Biological Laboratory, Woods Hole.
The fertilization potential of eggs of the nermertean, Cerebratulus. DOUGLAS KLINE
(Univ. of California, Davis, CA) AND LAURINDA A. JAFFE.
Some electrical properties of the egg of the nemertean, Cerebratulus lacteus were studied before,
during, and after fertilization using intracellular microelectrodes. The membrane potential of the unfer-
tilized egg in sea water is -67 ± 15 mV (SD, n = 10). A long-lasting action potential can be elicited by
a depolarizing current injection. A peak amplitude of +43 ± 6 mV (n = 6) is reached within one second;
then the membrane potential reaches a plateau of about +20 mV and gradually returns to a negative
resting potential. The duration of the positive phase of the action potential is 8 minutes. At fertilization
the membrane depolarizes to +43 ± 9 mV (n = 10) and reaches a plateau potential of +21 ±6 mV
(n = 10) during the first 30 minutes following fertilization. The potential stays positive for 74 ± 22
minutes (n = 10). Development of 7 of these eggs was followed through first cleavage, and cleavage was
normal.
When the sodium concentration of sea water is reduced to one-tenth of the normal (500 to 50 mM,
choline substituted) the amplitude of both the fertilization potential and the action potential is reduced,
indicating that a large part of the potential changes may be due to sodium flux. The average potential
in 1/10 Na"1" sea water for the first 30 minutes following fertilization is -30 ± 22 mV (n = 4). Eggs
inseminated in 1/10 Na+ sea water become polyspermic. Eggs also become polyspermic when transferred
from normal sea water to 1/10 Na+ sea water as long as 15 minutes after insemination. This suggests
that the long positive phase of the fertilization potential is important in preventing entry of supernumerary
sperm until a permanent block to polyspermy is established.
This work was, in part, supported by an NIH training grant (5-T35-HD07098) awarded to the
Embryology Course, Marine Biological Laboratory, Woods Hole.
FPL-55712, a leukotriene antagonist, promotes polyspermy in sea urchins. R. Moss,
R. SCHUEL, AND H. SCHUEL (Dept. Anat. Sci., SUNY at Buffalo).
Sea urchin eggs release H2O2 during the cortical reaction at fertilization to inactivate excess sperm
at their surfaces thereby helping to prevent polyspermy (Boldt el al. 1981, Gamete Res. 4: 365.). This
process resembles the peroxidatic killing of bacteria by phagocytic leukocytes during inflammation. As-
sociated with these reactions in leukocytes, arachidonic acid can be oxidized via the cyclooxygenase
pathway to produce prostaglandins or via the lipoxygenase pathway to produce leukotrienes. Cycloox-
ygenase products have been implicated in the prevention of polyspermy in sea urchins (Schuel el al.
1982, Biol. Bull. 163: 377.). We now report that FPL-55712, a well known antagonist for leukotrienes
C4 and D4, causes a dose (1-10 nM) and sperm density dependent induction of polyspermy in Arbacia
punclulala if added before the eggs complete the cortical reaction (elevation of the fertilization envelope).
The dose at which 50% of the eggs become polyspermic upon insemination with excess sperm (4.0 ± 2.2
X 107/ml) is 2.5 ± 0.8 ^M. To determine which gamete is affected by the drug, eggs and sperm were
pretreated with 50 ^M FPL-55712 which was removed by dilution at fertilization. Eggs pretreated with
FPL-55712 become polyspermic upon insemination with control sperm. Sperm pretreated with the drug
do not cause polyspermy. These results suggest that: (1) leukotrienes may have a role in preventing
polyspermy in sea urchins; (2) leukotrienes may modulate the egg's receptivity to sperm during the
cortical reaction; and (3) both cyclooxygenase and lipoxygenase products derived from the arachidonic
acid cascade may help assure monospermic fertilization in sea urchins.
Supported by NSF (#PCM-82-0 1561) and NIH (#HD- 17087) grants to H.S.
Binding of'4C-gossypol by Arbacia sperm. EIMEI SATO (The Population Council),
N. MATSUO, M. H. BURGOS, S. S. KOIDE, AND S. J. SEGAL.
Gossypol, a phenolic aldehyde, inhibits sperm motility. This action has been attributed to suppression
of synthesis and utilization of ATP, possibly by blocking the activities of mitochondrial enzymes (Mg2+-
dependent ATPase, Na+,K+-dependent ATPase, pyruvate dehydrogenase) [Adeyemo el al. 1982, Arch.
Androl. 9: 343] and dynein ATPase [Mohri el al. 1982, Biol. Bull. 163: 374].
In the present study, binding of 14C-gossypol by Arbacia sperm was studied. The gossypol, radiolabeled
on the aldehyde group with sp. act. of 3.33 x 105 dpm/micromole, was prepared by Dr. K. Watanabe and
Dr. Y. F. Ren of Sloan Kettering Institute.
GAMETES AND FERTILIZATION 517
Uptake of 14C-gossypol by Arbacia sperm reached saturation rapidly. Within one min of exposure
to 10 nM radiolabeled gossypol (t = 22 °C), 1 ml of sperm suspensions at densities of 7 X 107/ml and
7 X 108/ml incorporated 58% and 74% of the labeled gossypol, respectively. Immobile sperm prepared
by heating at 60°C for 10 min or by suspending in Ca2+, Mg2+-free ASW incorporated the same amount
of radiolabeled gossypol as motile sperm. The incorporated 14C-gossypol resisted extraction by repeated
washing with ASW or with 7 M guanidine • HC1. It was not hydrolyzed under acidic or alkaline conditions
and was not displaced by unlabeled gossypol.
The amount of l4C-gossypol bound to Arbacia sperm and eggs was 28.2 and 16.3 nmoles/mg (dry
wt), respectively. Uptake of '4C-gossypol at 5 min was greater at 22°C than at 4°C. Binding was slightly
higher under acidic conditions. Addition of unlabeled gossypol prevented competitively the binding of
14C-gossypol. Specific binding sites for gossypol per individual spermatozoan or eggs were calculated to
be about 8 X 108 and 6 X 10", respectively. The 14C-gossypol-protein complexes were solubilized by
incubating radiolabeled sperm in three different media: 0.1% Triton X-100; 1 mM urea, 5 mM EDTA;
and 20% sodium dodecyl sulfate, 0.1 M 2-mercaptoethanol. The amount extracted was 15, 23, and 74%,
respectively. The sp. act. of the extracted complexes were 3.7 X 103, 4.4 x 103, and 0.5 x 103 per mg
protein.
The results suggest that there are specific binding sites for gossypol on the sperm surface and in the
cytoplasm. The interaction of gossypol with sperm proteins is strong, indicating covalent linkage.
E. Sato is a post-doctoral fellow of the Rockefeller Foundation.
Fertilization-induced ion conductances in frog eggs. LYANNE C. SCHLICHTER AND
LAURINDA A. JAFFE (Physiology Dept., Univ. of Connecticut Health Center,
Farmington, CT 06032).
Fertilization of the frog egg (Rana pipiens) elicits a membrane depolarization (fertilization potential,
FP) that lasts many minutes and functions as a fast block to polyspermy (Cross and Elinson 1980, Dev.
Biol. 75: 187-198). The FP is caused in part by opening Cl channels. We explored two main questions.
1) What are the ion conductances underlying the FP and how do they change with time? 2) Do the ion
channels pre-exist in the plasma membrane or are they inserted during cortical vesicle exocytosis?
We used the voltage-clamp technique to measure ion currents and conductances (g) before and
during fertilization or artificial activation. Before fertilization a voltage-sensitive gNa is present (Schlichter
1983a, b, Dev. Biol. 98: 47-59 and 60-69). On fertilization two new conductances (gK and go) appear,
reach a maximum in 1-2 min, then decrease more slowly. After fertilization gNa disappears. The time
course of the conductance changes is not affected by voltage clamping. gK and go were separated by
blocking gK with external tetraethylammonium. go is voltage dependent. The same conductance changes
are elicited by monospermy or by polyspermy or by artificial activation; therefore, the opening of fer-
tilization channels is an all-or-none event.
Simultaneous measurements of changes in membrane potential, conductance, and surface area (by
the AC capacitance method) were made during fertilization or activation. At fertilization the surface area
increases l'/2 to 2 fold because of cortical vesicle exocytosis. A significant increase in conductance precedes
the increase in surface area; therefore, cortical vesicle exocytosis is not the initial source of new ion
channels. Membrane area subsequently decreases, which might contribute to the loss of channels after
fertilization.
Supported by an NSERC postdoctoral fellowship to L.C.S., NIH grant 5 RO1 HD 14939 to L.A.J.,
and NIH training grant 2 T35 HD07098 to the Embryology Course at the Marine Biological Laboratory,
Woods Hole.
Ultrastructural changes characteristic of Arbacia sperm exposed to gossypol. S. J.
SEGAL (Rockefeller Foundation), M. BURGOS, AND S. S. KOIDE.
Gossypol, a yellow pigment extracted from the cotton seed, inhibits motility of Arbacia sperm. The
mechanism of this effect is not clearly understood, although inhibition of a series of mitochondrial
enzymes involved in ATP synthesis and utilization has been demonstrated.
Scanning electron microscopy reveals that the first change observed after sea urchin sperm are
exposed to gossypol (25 p.M) is a separation of the cell membrane in the region of the sperm head and
mid-piece. This appears to be due to an accumulation of fluid, possibly due to an alteration of cell
membrane permeability. Study of the cell membrane by freeze-fracture replicas reveals that after gossypol
exposure (25 nM/[Q min) there is a condensation of the small particles, normally distributed at random,
in the region of the head and mid-piece. Concurrently, small blebs, free of particles, appear in the P face,
leaving corresponding depressions in the E face. The particle-free blebs appear to coalesce so that large
regions of the cell surface can be affected. When these regions rupture, as is observed frequently, the
518 ABSTRACTS FROM MBL GENERAL MEETINGS
nuclear material can be seen below. The cell membrane overlying the tail appears to be most resistant
to these changes.
The most evident alterations observed by transmission electron microscopy are those affecting the
mitochondria. These structures lose their normal appearance and display a clear, watery matrix and
swelling of the cristae. An accumulation of round, dense bodies can be observed around the inner
mitochondria! membrane and also between the cell membrane and the mitochondria. These are inter-
preted as lipid droplets.
We conclude that gossypol selectively affects the cell membrane of sperm and that the substance
concentrates in the mitochondrial region.
In vitro transcription of histone genes in isolated nuclei from S. Purpuratus. KATHLEEN
SHUPE AND ERIC WEINBERG (Univ. of Pennsylvania).
Nuclei were isolated at various times (9, 13, 15, 17, and 20 hours) after fertilization and utilized in
a cell-free transcription system. Transcription per nucleus increased with time of development. The
average rate of incorporation was 30 pm UMP/108 nuclei/60 min in 9 hour nuclei, 45 pm/108 nuclei/
60 min in 15 hour nuclei and 75 pm/108 nuclei/60 min in 20 hour nuclei. RNA labeled during in vitro
transcription was isolated and the transcriptional products analyzed using dot blot hybridization. In vivo
labeled early histone H3 message was gel purified and used in all hybridizations as an internal control
and all 32P counts corrected for % homologous hybridization. Early histone transcription represented
approximately 1 1% of total transcription at 9 hours falling to 2% by 13 hours and rising again to 6% at
17 and 20 hours. This sharp fall in early histone mRNAs going from morula to blastula confirms in vivo
results, however early histone mRNAs are not seen to accumulate in vivo despite the observation that
transcription continues in vitro suggesting that in addition to strong transcriptional regulation there is
also a decrease in early message stability. Late histone gene transcription appears to begin at low levels
(0.1%) at 13 hours rising to 1.0% by 20 hours demonstrating that the late histone genes are also under
transcriptional regulation.
The in vitro system was demonstrated to faithfully transcribe from the plus strand only. Analysis
of count hybridized before and after RNAse treatment suggest that for each of the early genes a percentage
of the transcripts are terminating within the coding region. Initial experiments using -/-labeled nucleotides
plus/minus initiation inhibitors suggest no initiation is occurring in vitro although further work is war-
ranted.
Is there a developmental significance for mRNA localized in the cortex o/Chaetop-
terus eggs? BILLIE J. SWALLA (University of Iowa), RANDALL T. MOON, AND
WILLIAM R. JEFFERY.
An organelle complex containing the maternal complement of mRNA is localized in the cortex of
eggs and early embryos ofChaetopteruspergamentaceus. The organelle complex and its associated mRNA
molecules are quantitatively displaced to the centrifugal pole region of the egg by centrifugation. We have
employed a combination of egg fragmentation, in situ hybridization, and embryo culture methods to
investigate whether the cortical mRNA molecules are required for normal embryonic development.
Centrifugation of unfertilized eggs through sucrose step gradients results in their equatorial splitting and
separation into light and heavy fragments. The heavy fragments contain yolk particles, mitochondria,
the cortical organelle complexes, and all of the poly(A)+RNA that is detectible by in situ hybridization
with a poly(U) probe. The light fragments contain lipid droplets, mitochondria, hyaloplasm, and the
female pronucleus, but no detectible poly(A)+RNA. Using this technique eggs are separated into nucleate
fragments without mRNA and anucleate fragments with mRNA. To test their developmental capacity,
the egg fragments were washed in sea water, fertilized, and cultured. About 60% of the mRNA-containing
heavy fragments were able to cleave and form swimming larvae (presumably haploid). In contrast, about
90% of the mRNA-lacking light fragments arrested after the first or second cleavage and did not form
swimming larvae. These results show that egg fragments deprived of poly(A)+RNA develop abnormally
and are consistent with the possibility that maternal mRNA molecules are necessary for normal embryonic
development.
This work was supported by NIH Training Grant 5-T35-HD07098 awarded to the Embryology
Course, Marine Biological Laboratory, Woods Hole, MA.
GAMETES AND FERTILIZATION 519
Maturation of sea urchin and Chaetopterus oocytes results in a change in the pattern
of protein synthesis. ALBRECHT VON BRUNN (Albert-Ludwigs-Universitat Frei-
burg, W. Germany), RONALD A. CONLON, AND M. M. WINKLER.
We find that there are changes in the pattern of protein synthesis associated with oocyte maturation
in sea urchins and in the marine annelid Chaetopterus pergamentaceus. Recently changes in the pattern
of protein synthesis have been described in the surf clam Spisula solidissima and the starfish Asterias
forbesi. The similar changes in patterns of protein synthesis in such distantly related species suggest that
this phenomenon is a very general one and may indicate that a different set of specific translation products
are required for maintenance of the immature oocyte and the transition to a developing embryo.
Ovaries of Lytechinus piclus and Arbacia punctulata were dissected and oocytes were picked out
individually by mouth pipetting and washed with pasteurised MFSW. Only oocytes which were about the
same size as mature eggs were used. Immature Chaetopterus oocytes were collected by washing the animal
in Ca-free sea water (SW). Germinal vesicle breakdown (GVBD) was induced by transfer into MFSW. Sea
urchin oocytes could not be matured artificially by hypertonic, Ca/Mg-free SW, the ionophore A23187,
NH4C1. serotonin, or human chorionic gonadotropin. 35S-methionine was used as radiolabel at levels of
50-250 Ci/ml for qualitative and 70-75 Ci/ml for quantitative incorporation experiments. Label was applied
5 minutes postfertilization for 60 minutes ( 1 8°C). Protein synthesis was analyzed by 1 D SDS-PAGE. Several
bands are prominent before GVBD; their rate of synthesis decreases in mature eggs and their synthesis is
not detectable in fertilized eggs. Some bands appear only in immature oocytes. Others are present only in
mature and fertilized eggs. In uptake and incorporation experiments 3 to 10 times more label enters the
immature sea urchin oocytes as compared to mature eggs and 2 to 20 fold more label is incorporated into
TCA-precipitable proteins. In Chaetopterus the rate of incorporation does not seem to increase significantly
at GVBD or fertilization.
This work was supported in part by NIH Training Grant 5-T35-HD07098 awarded to the Em-
bryology Course, Marine Biological Laboratory, Woods Hole, MA and NIH grant HD 17722-01 awarded
to M.W.
Preliminary evidence indicating the existence of intermediate filament-like proteins
in unfertilized eggs of the surf clam, Spisula solidissima. KAREN M. YOKOO,
ANNE E. GOLDMAN, AND ROBERT D. GOLDMAN (Northwestern University
Medical School, Chicago).
Intermediate filaments (IF) are major cytoskeletal components of animal cells, but their existence
in oocytes has not been demonstrated conclusively. Unlike the other two major, highly conserved cy-
toskeletal components, microtubules and microfilaments, IF have subunit compositions which differ
significantly among various cell types. To explore the developmentally regulated basis for this IF diversity,
we attempted to determine whether IF are present in unfertilized Spisula eggs. Eggs were lysed in an IF
isolation/stabilization solution containing Triton X-100 which was developed for cultured baby hamster
kidney (BHK-21) cells (Zackroff and Goldman 1979, P.N.A.S. 76: 6226). Sodium dodecyl sulfate poly-
acrylamide gel electrophoresis (SDS-PAGE) revealed several proteins in the 40-70,000 molecular weight
(K) range with a major component at ~55K, which comigrated with the 54-55K subunits of cultured
BHK-2 1 cell IF. As in other IF systems, this Spisula preparation could be solubilized (disassembled) in
8 M urea, 5 mA/ Tris-HCl, 0.1% B-mercaptoethanol (BME), 0.1 mA/ phenylmethylsulfonylfluoride
(PMSF) (pH 7.4). Following ultracentrifugation at 55,000 rpm (Beckman 65 rotor) the supernatant was
dialyzed against assembly buffer (5 mM sodium phosphate, 0.1 mM PMSF, 0.1% BME, pH 6.6). This
cycle of disassembly followed by reassembly was repeated, and the resulting pellets were examined by
electron microscopy and SDS-PAGE. The latter analysis revealed great enrichment for a ~55K protein,
as well as several proteins in the 40-50K range and the >55K-70K range. By electron microscopy,
~ 10 nm diameter filamentous networks were observed. Peptide mapping by limited proteolysis revealed
that the 55K proteins of Spisula and BHK-21 differ significantly. However, immunoblotting analyses
showed that the major 55K band from Spisula reacted with BHK-21 IF antiserum. In addition, mouse
skin keratin antisera reacted with proteins in the >55K-70K range and the 40-50K range present in
both freshly isolated and reassembled Spisula IF preparations. These studies support the presence of both
mesenchymal and epithelial-like IF systems in unfertilized Spisula eggs.
Supported by NIH and NSF.
520 ABSTRACTS FROM MBL GENERAL MEETINGS
Hyperosmotic treatment inhibits cortical granule exocytosis in the sea urchin Lytechinus
pictus. JOSHUA ZIMMERBERG (NIH, Bethesda, MD 20205).
An osmotic hypothesis of exocytosis (Zimmerberg el al. 1980, Science 210: 901) was tested. If
osmotic swelling of exocytotic vesicles is an absolute requirement for fusion, hyperosmotic treatment of
secretory cells and the subsequent shrinkage of exocytotic vesicles should inhibit secretion. This is indeed
the case. Ninety-seven ± 1 .9% of sea urchin eggs treated with 40 nM calcium ionophore A23 1 87 raised
fertilization envelopes. If eggs were first placed for three minutes in sea water containing added sucrose
to a final osmolality of 2.42 Osm/kg (2.42 Osm SW) they shrank significantly. When ionophore (40 nAf)
was added to this mixture, only 6.3 ± 3.8% of the eggs formed fertilization envelopes. Use of 1.59 and
2 Osm SW in the above experiment led to 83% and 57% fertilization envelope elevation, respectively.
The inhibited eggs still had a full complement of intact cortical vesicles, as ascertained by direct micro-
scopic examination after fixing and clearing. The prevention of secretion was reversible. Seventy-one
± 2% of eggs placed in 2.42 Osm SW for three minutes, then returned to sea water for one or ten minutes,
and finally treated with 40 nM ionophore in sea water raised fertilization envelopes. Sucrose did not
pharmacologically interfere with exocytosis, as 85% of eggs treated with 40 \iM ionophore in 1 M sucrose
5 mM CaCl2 formed normal-looking fertilization envelopes.
Thus external hyperosmotic media reversibly inhibit exocytosis. These results are consistant with the
osmotic theory of exocytosis. It is known that the fertilization envelope elevation is due to increased colloid
osmotic pressure within the perivitelline space. This colloid is presumably contained within the cortical
granules in an inactive form. I propose that the sperm-induced rise in intracellular calcium triggers an
osmotic activation of previously inert cortical granule contents. These activated substances induce water
flow into the cortical granules, resulting in cortical granule swelling and fusion. The contents are thereby
secreted into the perivitelline space. They continue to induce water flow into the perivitelline space causing
fertilization envelope elevation.
This work was supported in part by NIH training grant 5-T35-HD07098 awarded to the embryology
course, Marine Biological Laboratory, Woods Hole, MA and in part by NIH NRSA 3-F32 GM09078-
01 to J.Z.
MICROBIOLOGY
Mutants o/Escherichia coli affected in "inducer exclusion. " E. B. ACKERMAN (De-
partment of Soil Science and Biometeorology, Utah State University, Logan,
Utah) AND H. L. KORNBERG.
Glucose, and non-catabolizable analogs such as 3-deoxy 3-fluoroglucose (DFG), inhibit the induction
of the lactose operon of Escherichia coli by preventing the initial entry of lactose into the cells; this
phenomenon is known as "inducer exclusion." Mutants altered in this property were selected by plating
samples of the K 12 strain HK 743 (ptsM umgC arg l/ir leu rpsL), pre-grown on glucose, on agar plates
that contained the required amino acids, salts, and 5 mM lactose as sole carbon source; in addition,
several drops of 0. 1 M-DFG were placed in the center. Growth of the organisms occurred initially only
at the periphery of the plates but, after 2-3 days' incubation at 41°C, a number of mutants appeared
within the zone of growth inhibition. These mutants were screened for their continued ability to grow
on glucose and to take up [MC]glucose and methyla-D-['4C]glucoside; this showed that they had lost
neither the Enzyme II for glucose uptake that is specified by ptsG* nor the factor IIIglc that is associated
with it. All the mutants tested also remained inducible for lactose utilization.
Like the iex mutants described by Parra et al. (1983, /. Gen. Microbiol. 129: 337-348) our mutants
readily induced the lactose operon when glucose-grown cells were allowed to grow further in media
containing 2.5 mM lactose and either 5 mM glucose or 5 mM N-acetylglucoseamine (NAG); the parent
strain did not do so. But, unlike the iex mutants previously reported, our mutants had simultaneously
lost "catabolite inhibition:" neither glucose nor NAG was used in preference to other sugars taken up
via the PEP-phosphotransferase system, such as fructose. In contrast, glucose 6-phosphate, which is taken
up as such by E. coli, excluded lactose and was used preferentially to fructose in our mutants just as it
was in the parent organisms.
Studies on manganese oxidizing, spore forming bacteria. H. O. HALVORSON (Bran-
deis University, Waltham, MA), A. KEYNAN, AND T. TIERNAN.
Manganese oxidizing spore forming bacteria were isolated from the Sippewissett Marsh by plating
heated (80°C for 30 min) marsh samples on Zobell sea water medium. Eight strains were selected by
MICROBIOLOGY 52 1
their ability to oxidize manganese verified by the Leucoberbelin reaction (Kumbein and Altmann 1973,
Helgol. Wiss. Meeresunter. 25: 347.) Strain Mn 8 was selected for further studies. Spores of this strain
were prepared as described previously (Wier et al. 1982, Biol. Bull. 163: 370). When examined by flame
photometry the spores of this species were found to have a significantly higher concentration of Mn+
and Fe++ (68.5 X 1CT8 and 32 X 1CT8 moles/mg dry wt.) than any of several well known soil or marine
spore forming bacteria. Further, sporulating cells concentrate Ca++ in a high Mg++ environment. Spores
of Mn 8 had a higher specific density than other marine spore formers investigated; they centrifuged
through 62.5-65% renographin, while most other spores pellet through 50-55% renographin. No signif-
icant germination occurred in Zobell or other nutrient media, with or without sea water. Over 80%
germination occurred when spores were incubated at 30°C for 80 min in medium containing 12.2 mM
glucose, 17 mM NaCl, 18.7 mM NH4C1, 0.4 mM L-alanine, 0.4 mM adenosine, 1% Tween-80 and 0.01
M Hepes buffer pH 8.2; choramphenical (50 ng/m\) was included to prevent outgrowth. Germination
did not require heat activation. During germination loss of refractility is accompanied by swelling with
a substantial increase in volume. No parallel decrease in optical density of the suspension was observed.
Although the germination requirements of strain Mn 8 seem to be more complex, they are similar to
those of a previously described marine spore former (Wier et al.) in their requirement for NH4+, Na+,
and relative high pH.
Numbers of symbiotic bacteria in the gill tissue of the bivalve Solemya velum Say.
TRICIA A. MITCHELL AND COLLEEN M. CAVANAUGH (Harvard University).
Symbiotic, sulfur-oxidizing, chemoautotrophic bacteria occur in Solemya velum Say, an Atlantic
coast bivalve found in reducing, muddy sediments (Cavanaugh 1983, Nature 302: 58-61). In this study
we investigated the effect of varying environmental conditions on the numbers of these symbionts.
Animals were collected from eelgrass beds at Hadley Harbor near Woods Hole, MA. Bacteria (rod-
shaped fluorescent cells) were enumerated using epifluorescent microscopy in tissue homogenates (from
formalin-fixed gills) after staining with acridine orange. All of the cell counts are reported as number of
cells per gram wet weight gill tissue; mean ± 1 S.D.
Animals, collected 2 June 1983, were kept for 10 days in MBL sea water tables at ambient tem-
perature ( 15°C) in mud from the collection site with running sea water and in running sea water alone.
During the course of the experiment, half of the sea water animals died whereas all of the animals in
mud survived. Direct counts of bacteria indicated that there were 2.56 ± 0. 1 1 X 109 (n = 3) in the freshly
collected animals, 1.63 ± 0.36 X 109 (n = 3) in those animals maintained in mud, and 0.54 ± 0.25
X 109 (n = 3) in the animals held in sea water. Bacterial numbers were significantly lower in those
animals kept in running sea water but not in those kept in mud.
In a subsequent experiment, animals collected 10 August 1983 were maintained under four different
treatments, all at ambient temperature (22°C): as above in mud or in running sea water, or in aerated,
filtered (0.45 ^m) sea water supplemented or unsupplemented with 0.4 mM thiosulfate. There was no
significant difference between bacterial numbers in animals harvested after 4 days from any of the
treatments (average cell count from all four treatments = 1.33 ± 0.63 X 109; n == 15) and in freshly
collected animals (1.23 ± 0.4 X 109; n = 3). This data suggests that the number of symbionts are resistant
to short term fluctuations in environmental conditions.
Qualitative observations indicate that the bacterial cells are larger in freshly collected animals and
in animals kept in mud, suggesting that biomass estimates, as well as CO? fixation activity measurements,
will provide a more accurate assessment than cell counts of the effects of varying environmental conditions
on the symbiotic bacteria in S. velum.
This research was supported by NSF grant DEB-8 1 0470 1 . We thank John Helfrich for helpful advice
and discussion and John E. Hobbie for the use of his epifluorescent microscope.
Sheath pigment formation in a blue-green alga, Lyngbya aestuarii, as an adaptation
to high light. LISA MUEHLSTEIN (Wright State University) AND RICHARD W.
CASTENHOLZ.
Lyngbya aestuarii is a predominant blue-green alga found in the top layer of the intertidal microbial
mats in the Great Sippewissett marsh. These mats are subjected to long exposures of high light intensities
which potentially cause photodynamic damage to the microorganisms exposed. L. aestuarii with heavily
pigmented sheaths is often found in these exposed areas. The pigment is yellowish brown and has been
called scytonemine. In order to study the sheath pigment further, axenic cultures were used. The cultures
were grown in high light outside, lower light outside, and fluorescent light inside. Sheath pigments were
formed only in the cultures grown in the high outside light. Methanol was used to extract the cell pigments
for spectrophotometric analysis. Dimethyl sulfoxide successfully extracted the sheath pigments which
remained after the methanol extraction. The absorbance of the sheath pigment is highest in the near
522 ABSTRACTS FROM MBL GENERAL MEETINGS
ultraviolet range (360-400 nm), which is thought to be the most damaging part of the spectrum normally
reaching the earth's surface. There is also heavy absorbance through the violet, blue, and blue-green
regions with a prominent secondary maximum at 495 nm. The absorbance is low at wavelengths greater
than ~540 nm. The sheath pigments have a much higher absorbance overall than cell pigments from
the same amount of culture material grown in the outside light, indicating that sheath pigment may
convey significant protection. Cell pigment regulation is another way that many photosynthetic organisms
protect themselves from photooxidative damage. In L. aestuarii the chlorophyll content of the cells also
appears to be regulated, decreasing as light intensity increases. The carotenoid to chlorophyll absorbance
ratio also decreases from high to low light. L. aestuarii appears to adapt to high light intensities inherent
to its environment, by regulating chlorophyll and carotenoids in the cells, as well as by the formation
of pigmented sheaths.
Factors affecting growth inhibition of enteric bacteria by methyl a-D-glucoside.
D. F. SUTHERLAND (Department of Biology, Creighton University, Omaha, NE
68178) AND H. L. KORNBERG.
It was reported previously (Schnell et ai, 1982, Biol. Bull. 163: 403) that the growth of some enteric
bacteria in media of low phosphate content is inhibited by methyl «-glucoside but that, after 2-4 h., the
organisms "escape" from inhibition; their subsequent growth is not affected by this and other glucose
analogs. Working with a number of strains of Escherichia coli and with Vibrio harveyi 392, we have
shown that:
(1) growth inhibition and "escape" occur also when cultures grow in media of high (50 mM)
phosphate content, even whent he phoA gene is deleted: alkaline phosphatase therefore plays no major
part in this phenomenon;
(2) the extent of growth inhibition depends on the amounts of methyl a-glucoside (phosphate)
accumulated inside the cells and retained by them. There was a loss of over 90% of the I4C taken up by
V. harveyi, whose growth on 10 mM-mannose had been inhibited by 2 mM methyl a-['4C] glucoside,
prior to "escape." Moreover, strains of E. coli that form the uptake system for glucose and methyl a-
glucoside constitutively (umgC) do not "escape" from inhibition;
(3) this "escape" from growth inhibition, by V. harveyi as by V. parahaemolyticus (Schnell et al.
1982), is associated with the appearance of a system that causes methyl a-glucoside taken up by cells to
be rapidly lost from them. Since the elaboration of this system is prevented by chloramphenicol (100
jig- ml"'), it probably involves de novo synthesis of protein; and
(4) the growth of cells subsequent to their "escape" is accompanied by repression of the Enzyme
II specified by ptsG* irrespective of the presence of the glucose analog in the medium.
Mechanical stimulation of bioluminescence in dilute suspensions of dinoflagellates.
G. T. REYNOLDS (Department of Physics, Princeton University) AND ALAN J.
WALTON.
Mechanical stimulation of individual dinoflagellates by means of a piezoelectric cylinder incorpo-
rated in a suction pipette has been reported (Reynolds, 1970, Biophys. Soc. Ann. Meet. Abstr. 10: 132A).
In this method the organism responds to a shock wave transmitted to the tip of the pipette, and the
stimulus may be pressure or membrane distortion. Following a suggestion and initial experiments by
James F. Case we have stimulated dilute suspensions of Gonyaulax polyedra and Pyrocystis lunula (ca.
1000/ml) by means of moving objects through the suspension. The container measured 10 X 10 X 30
cm3. A cone 2 cm high, 2.5 cm diameter, with a 1.5 cm. high, 2.5 cm. diameter cylinder attached, was
moved through the medium at velocities 5 cm/s. to 30 cm/s. The resulting patterns of bioluminescence
were recorded through a high gain image intensifier-SIT vidicon detector and stored on magnetic tape
for analysis.
Luminescence was observed at the top and sides of the advancing cone, and the side and trailing
edge of the cylinder. The Reynolds numbers in these experiments were low, but the sharp trailing edges
of the moving objects caused eddies. Bright luminescence was associated with the thread by which the
object was raised through the medium. Assuming the dinoflagellates attached to the thread while it was
at rest prior to upward motion, this indicated a response to shear as the thread moved through the
medium.
We thank Elijah Swift, Celia Chen, Donald Anderson, and David Kulis for preparation of the
cultures used in this work. The work was supported by DOE Contract EY-76-S-02-3120 and ONR
Contract N0014-83-C-0234.
NEUROBIOLOGY, LEARNING, BEHAVIOR 523
NEUROBIOLOGY, LEARNING, AND BEHAVIOR
Pseudostereoscopy allows direct visualization of the velocity distribution of particles
undergoing fast axonal transport. W. J. ADELMAN, JR. AND ALAN j. HODGE
(Laboratory of Biophysics, NINCDS, NIH, MBL)
The movement of particles undergoing fast axonal transport can be readily detected and their velocity
(speed) distribution visualized by a simple pseudostereoscopic viewing procedure utilizing pairs of images
derived from a videotape or other record, and separated by an appropriate time lapse. When such pairs
are examined stereoscopically, the parallax arising from particle motion results in the images of particles
being raised or lowered relative to an immobile background plane in proportion to their speed and
direction. In effect, the binocular optic axis serves as a velocity axis under these conditions.
The method is particularly useful when observing the simultaneous motion of large numbers or
swarms of particles and for the detection of small numbers of slowly moving particles. The technique
is generally applicable to a variety of situations, and can be made quantitative using standard photo-
grammetric procedures. It can also be readily adapted for on-line analysis, particularly in video imaging
systems where frame buffers can be utilized.
Transport of vesicles along filaments dissociated from squid axoplasm. ROBERT D.
ALLEN, DOUGLAS T. BROWN, SUSAN P. GILBERT, AND HIDESHI FUJIWAKE
(Dartmouth College).
It has been previously reported that fast axonal transport of vesicles could be observed in squid
axoplasm by Allen video enhanced contrast-differential interference contrast (AVEC-DIC) videomicroscopy
(Allen et al. 1982, Science 218: 1127-1129). Axoplasm extruded from axons displays similar transport
even when mechanically disrupted by stirring with a needle, so that its constituent linear elements have
been randomized in direction and shape (Brady et al. 1982, Science 218: 1 129-1 131).
We now report that squid axoplasm dissociated by gentle shear while diluted up to 1:5 in butter X
containing 1 mM of ATP (Morris and Lasek 1982, / Cell Biol. 92: 192-198) breaks up into linear
elements or filaments, some of which display unidirectional or bidirectional transport of vesicles. These
vesicles are in rapid Brownian motion in the vicinity of the filaments but adhere when they collide with
a filament, then move along the filament to one of its ends and are discharged into the medium. The
observations so far are consistent with the expectation that dissociated linear elements might comprise
neurofilaments (singly or in bundles) showing no motility and microtubules, either single or in bundles
showing unidirectional or bidirectional transport. Filamentous actin might be present in these filaments,
but would not be detected. The filaments themselves move about due certainly to Brownian bombardment
and possibly to motility as well. Different filaments display different degrees of Brownian deformation
consistent with the belief that they contain different numbers and/or types of cytoskeletal elements.
The observations provide evidence that the fundamental process in fast axonal transport can persist
in dispersed filaments and vesicles. They also suggest that reconstitution experiments involving bio-
chemically defined, interactive filaments and vesicles may shed some light on the mechanisms of fast
axonal tranport.
Presynaptic action ofbaclofen, a GABA analog, at the crayfish neuromuscular junc-
tion. SUSAN R. BARRY (Dept. of Neurology, Univ. of Michigan).
The action ofbaclofen, a GABA analog, was studied at the neuromuscular junction (NMJ) of the
crayfish Procambarus clarkii. Baclofen (Lioresal) is used clinically to treat spasticity. In the vertebrate
nervous system, the drug may bind to GABA receptors on presynaptic nerve terminals and produce a
decrease in transmitter release.
GABA mediates presynaptic and postsynaptic inhibition at the NMJ of the crayfish opener muscle.
The muscle is innervated by an excitatory and inhibitory axon. The inhibitory axon, whose transmitter
is GABA, also synapses on the excitatory nerve terminal. GABA acts postsynaptically by increasing
chloride conductance of the muscle and acts presynaptically by depressing transmitter release from the
excitatory nerve terminal (Dudel and Kuffler 1961, / Physiol. 155: 543-562). These two effects are
mediated by pharmacologically different receptors (Dudel 1965, Pflugers Archiv. 283: 104-1 18).
Baclofen was tested on the opener muscle junction to determine whether the drug mimicked GABA's
presynaptic or postsynaptic actions. 10~4 M Baclofen produced a 25% decrease in excitatory junction
524 ABSTRACTS FROM MBL GENERAL MEETINGS
potential (ejp) amplitude, but had no effect on the muscle input resistance. 10 4 M Baclofen also reduced
the frequency of spontaneous miniature excitatory junction potentials (mejp's) by 30% but did not alter
the size distribution of mejp's. Since baclofen reduced ejp amplitude and mejp frequency without affecting
muscle input resistance, it may act by depressing transmitter release from the excitatory nerve. Since the
size of the mejp's was not changed, baclofen probably did not alter the muscle's response to the excitatory
transmitter.
Thus, baclofen may mediate presynaptic but not postsynaptic inhibition at the crayfish NMJ. The
drug may bind selectively to presynaptic GABA receptors. Baclofen's action at the crayfish NMJ may
parallel its effect in the vertebrate nervous system.
I thank the Grass Foundation for their support and generosity and Dr.'s C. K. Govind, M. Goy,
J. Brown, and L. Rubin for technical assistance.
Slow rearrangements of membrane bound, halogenated jluoresceins produce altered
K+ currents in squid axon. RICHARD J. BOOKMAN (Dept. of Physiology, Univ.
of Pennsylvania).
The interaction of dyes with excitable membranes can be exploited for a variety of purposes. In
such studies it is important to distinguish between the actions of the dyes in the presence and absence
of light. In these experiments, halogenated fluoresceins (e.g.. Rose Bengal (RB), Eosin Y, Erythrosin,
& Phloxine B) have been shown to be specific and potent modulators of outward K+ current when applied
to the inside of the internally perfused, voltage clamped squid giant axon. This reaction, with 1 pM RB
inside, reaches completion very slowly (i.e.. minutes), modifies about 75% of the channels and is only
partially reversible. In the absence of light, K+ currents from such a stained axon exhibit a number of
interesting features: as measured at 7 or 35 ms after the application of a voltage clamp step to a positive
membrane potential, IK is diminished and has not reached a steady state. Long voltage clamp steps show
that these currents are still increasing after more than 100.0 ms. The ON kinetics are thus slowed by
more than an order of magnitude. However, once the K+ channels are open and conducting, they seem
to close with approximately normal kinetics upon returning to the holding potential of -70.0 mV. This
result is best demonstrated by using a double pulse procedure which also illustrates that recently closed
channels reopen with more nearly normal kinetics and that the full extent of the slowly opening behavior
is only re-established after many seconds. Repeatedly pulsing the axon to +80 mV leads to a frequency
dependent increase and speeding of the current — the recent history of the membrane can shift channels
into the rapidly opening state. The illumination of a previously stained axon specifically destroys K+
channels with bound dye. The decrease in IK proceeds exponentially and the remaining current has
almost normal kinetics.
These results show that halogenated fluoresceins are a new family of highly potent K+ blockers. The
nature of this block is similar to that which has been described for the aminopyridines (Yeh el al. 1976,
J. Gen. Physioi, 68: 519-535) and therefore suggests that squid K+ channels may have a site or sites
whose occupancy by either of these molecules can regulate channel function.
Supported by a Grass Foundation Fellowship.
Structure of the squid axon membrane as seen after freeze-fracture. DONALD C.
CHANG (Baylor College of Medicine, Houston, TX 77030), ICHIJI TASAKI, AND
TOM S. REESE.
A classical excitable membrane is the axolemma of the squid axon. We used freeze-fracture technique
to examine the morphology of this membrane to try to identify the membrane protein structures which
are thought to be conductance pathways for ions ("channels"). Many large pieces of membrane were
seen in the replicas of the intact fixed axon but fractures did not occur through the axolemma. Since
there are many layers of Schwann cells, most fracture planes tend to go through the Schwann cell
membrane rather than the axolemma. When the Schwann cells are removed, the axolemma is easily
recognized at the boundary between the external ice and the axoplasm. However, very little membrane
was seen in these replicas of desheathed axons because the axolemma was usually cross-fractured without
splitting it over any significant distance. The best results were obtained using axons with Schwann cells
chemically detached from the axon but not mechanically removed. In one particularly clear example the
fracture plane cut through a stack of Schwann cells and then exposed a large extent of axolemma.
Our first impression of the P-face of the axon membrane is that, unlike the Schwann cell membrane
or the membrane of myelinated nerve fibers, the squid axon membrane is marked by many small particles
(3 to 4 nm in diameter). It is also clear that there are large P-face particles distributed randomly in the
axon membrane. Judging from their size (between 10 and 18 nm) and density (1203 ±416 per Mm2),
some of these large particles are likely candidates for the intramembrane component of the "sodium
NEUROBIOLOGY, LEARNING, BEHAVIOR 525
channels." A peculiar structure was observed in one sample where a particularly large extent of axonal
membrane was exposed. Hemispherical blebs having a diameter ranging from 40 to 58 nm were distributed
randomly at the axon surface at a density of roughly 80 per ^m2, and the surface of these blebs lacked
intramembrane particles. These blebs are tentatively interpreted as contacts, presumably artifactitious,
between the axolemma and numberous underlying small vesicles.
Work supported partially by ONR Contract N00014-76-C-0100.
Fine structure of synapses and synaptosomes of the squid (Loligo pealei) optic lobe.
ROCHELLE S. COHEN, NASRIN HAGHIGHAT, AND GEORGE D. PAPPAS (Marine
Biological Laboratory).
Cephalopod optic lobes are a rich source of cholinergic endings (Dowdall and Whittaker 1973, J.
Neurochem. 20: 921-935). As a prelude to subsequent morphological and biochemical analyses of cho-
linergic transmission in the central nervous system (CNS), we describe the ultrastructure of synaptic
endings of the optic lobe of the squid and categorize them into distinct morphological types recognizable
in the squid optic lobe synaptosome fraction. Toluidine blue staining of epon-embedded thick sections
showed an outer cortex (consisting of four main layers) where the incoming nerve fibers meet the tangential
dendrites of second order visual neurons (Young 1974, Phil. Trans. B. 245: 263-302), and an inner
medulla, composed partly of radial columns and islands of different types of neurons. Photoreceptor
endings were seen within the plexiform layer of the cortex. Electron microscopy revealed that both
chemical and electrotonic synapses were present, the former being predominant and showing two basic
forms. One was an invaginated synapse between photoreceptor endings and spines; the second was a
typical chemical synapse, found in almost all layers except the upper portion of the first radial layer.
Most of the synapses in the medulla were of the second type although a few photoreceptor endings extend
to this region. Gap junctions were found where photoreceptor processes contact each other. Synapses
were categorized into five distinct types which corresponded to five types of synaptosomes recognized
in a synaptosome fraction derived from these lobes. E-PTA staining of synapses revealed a much thinner
layer of postsynaptic material than found at typical mammalian cortex synapses as postsynaptic densities.
Because of its high content of cholinergic endings and distinct synaptic types, the squid optic lobe may
provide an interesting model for the isolation of cholinergic synaptosomes and synaptosomal plasma
membranes from the CNS.
This work was supported by NIH (grants NS 15889 and NS 16610) and NSF.
Pathway tracing in the squid nervous system. SUSAN C. FELDMAN AND GEORGE D.
PAPPAS (Marine Biological Laboratory).
When germ agglutinin (WGA), a lectin which binds to sialic acid and N-acetyl-glucosamine residues,
has been shown to be axonally transported in the vertebrate visual system. In this study we demonstrate
the labeling of cells and fibers in the squid nervous system following injection of the lectin into the eye
and stellate ganglion. Squid received 2-5 n\ of a 30% solution of WGA or WGA conjugated to HRP
(HRP-WGA) into one eye or both stellate ganglia. Animals were allowed to survive up to 48 h in sea
water (1 1-16°C). WGA was localized immunocytochemically on 10 nM paraffin sections; HRP-WGA
was visualized using CoCl2 intensified DAB.
Injection of WGA into one eye resulted in a narrow patch of cells and fibers in the ipisilateral optic
lobe. With the more sensitive immunocyto-chemical procedure both labeled cells and fibers were seen
in the central ganglia and fibers were demonstrable in the contralateral optic lobe (30 h survival time).
Injection of WGA into the stellate ganglion resulted in labeling in the giant axons, in fibers within the
ganglion, and in a few small to medium-sized cells. No labeling was seen in the second-order fibers or
in or around the large ganglion neurons. In the giant axons the staining was restricted to parallel lon-
gitudinally arranged arrays with occasional labeled strands between them.
Ultrastructural studies are in progress to resolve the identity of the elements to which the lectin is
bound. The results of the present study demonstrate the potential usefulness of WGA, and other lectins,
as markers of specific pathways and in transport studies.
Supported in part by NIMH grant MH 38485 to S.C.F. and NS grant 16610 to G.D.P.
An infrared macrophotographic technique for quantifying the behavioral response to
rotation of the gastropod Hermissenda crassicornis. SERGE GART, IZJA LED-
ERHENDLER, AND DANIEL ALKON (Marine Biological Laboratory).
Positive phototaxis in Hermissenda is modified by repeatedly pairing light and rotation. Further
understanding of this associative learning depends on defining an unconditioned response to rotation. We
526 ABSTRACTS FROM MBL GENERAL MEETINGS
have developed infrared photographic methods to measure the foot muscle during rotation. This light, to
which Hermissenda is unresponsive can also provide high resolution negatives of these semi-translucent
animals, and clear negatives, without blur, of the subjects at high rpm.
The animals were placed in sea water filled tubes below a motor-driven Nikon FM 2 35 mm camera
with a 55 mm Micro-Nikkor lens, fitted with a Schott RG-630 and Tiffen polarizing and dichroic filters.
Four Vivitar 283 flash units with VP-1 varipower modules and two Tensor high-intensity contrast lights,
all fitted with Schott RG-665 and Tiffen polarizing and dichroic filters, were used to illuminate the
subject. The camera and flash units were attached to a modified motorized X-Y plotter in order to track
the moving animal. Shutter release was triggered manually or by a photoelectric cell linked to an electronic
delay device. We used Kodak Recording 2475 or High Speed Infrared 2481 film. Pictures were taken
at the rate of one per second, four prior to rotation, and up to 20 during rotation. The negatives were
projected onto an L-W Photooptical Digitizer for direct scaled measurements of the foot muscle.
The length of the foot decreased in all 20 animals tested. Average decreases were 14.8% one second
into rotation (N = 20), 13.1% after 3 s (N = 20), 11.1% after 6 s (N = 4), 3.7% after 9 s (N = 4), and
2.9% after 20 s (N = 4).
Thus foot shortening is greatest immediately after rotation starts and subsequently begins to recover.
Preliminary analysis indicates that area and width of the foot also change. The response to rotation may
thus involve several component elements in the foot muscle. Foot shortening is thus a reliable quantifiable
unconditioned response for use in conditioning studies.
Messenger RNA in squid axoplasm. ANTONIO GIUDITTA (Institute of General Phys-
iology, Via Mezzocannone 8, Naples, Italy), TIM HUNT, AND LUIGIA SANTELLA.
The axoplasm of the squid giant axon contains sizable amounts of tRNA (Black and Lasek 1977,
J. Neurobiol. 8: 229-237), while minor amounts of rRNA have been detected in the axoplasm of the
squid Loligo vulgaris (Giuditta et al. 1980, /. Neurochem. 34: 1757-1760). Furthermore, the axoplasm
of the latter species contains all soluble factors required for protein synthesis (Giuditta et al. 1977, J.
Neurochem. 28: 1393-1395). In the further search for additional components of the protein synthetic
machinery we have examined for mRNA in squid axoplasm. Our method of analysis was based on the
ability of the rabbit reticulocyte lysate to synthesize radioactive proteins using 35S-methionine in the
presence of exogenous mRNA. RNA was purified by phenol extraction from the axoplasm of the giant
axon, from the extruded stellate nerve, and from the giant fiber lobe of the squid Loligo paelii. A marked
stimulation of protein synthesis was obtained with all RNA preparations, including axoplasmic RNA.
In addition, radioactive translation products were separated by electrophoresis on SDS-polyacrylamine
gels and visualized by fluorography. Up to 50 different protein bands were found labeled when axoplasmic
RNA was used as template. Some of the bands were intensely radioactive. The overall pattern of labeling
was similar to that obtained with RNA extracted from the giant fiber lobe or from the extruded stellate
nerve, but several consistent differences were detected. Those present between axoplasm and extruded
stellate nerve appeared to exclude the possibility of contamination of axoplasmic mRNA by nerve material
intruding during the extrusion step. To examine the possibility that axoplasmic mRNA was originating
from mitochondria, several subcellular fractions were obtained from squid optic lobes, including a purified
fraction of synaptosomal mitochondria. RNA extracted from the latter fraction was essentially inactive
in the translation assay, at variance with the RNAs extracted from the other subcellular fractions. This
result suggests the extramitochondrial origin of axoplasmic mRNA. The functional role of axoplasmic
mRNA, i.e., its presence in an inactive or in an active form, remains to be established.
Supported by NATO grant no. 18781.
Phospholipid synthesis in the injected squid giant axon. ROBERT M. GOULD (Institute
for Basic Research in Developmental Disabilities), MARTHA JACKSON, AND ICHIJI
TASAKI.
Axoplasm extruded from the squid giant axon incorporates a variety of precursors into phospholipids.
In order to relate the giant axon's lipid metabolism with its excitable properties we injected small volumes
of labeled precursor in solution into giant axons and stimulated them for a variety of time courses. The
following precursors were used, 32P-inorganic phosphate, -y-32P-adenosine triphosphate, 3H-acetate, 3H-
choline, 3H-glycerol, and 3H methionine. After injection the axons were incubated in sea water for 30
min to 2 h. Some axons were stimulated at high frequency (50-100 Hz). The others were taken as control.
Following incubation and extrusion of axoplasm homogenates of axoplasm and sheath (containing cortical
axoplasm, axolemma, and glial cell layers) were extracted with acidified chloroform-methanol. With each
precursor, labeled lipids were found in both axoplasm and sheath. Excepting glycerol, the amount of
NEUROBIOLOGY, LEARNING, BEHAVIOR 527
recovered lipid was higher in the axoplasm than the sheath. Based on recoveries expressed in terms of
total lipid and aqueous (upper phase), radioactivity (total lipid formation) in axoplasm sheath was not
significantly altered by the stimulation. We will conclude with examination of the labeled lipids after
separation on TLC plates. Subsequent autoradiography and counting of specific lipids will reveal exact
distribution of the lipids formed and if the distribution is altered by stimulation.
This study was supported by a grant from the NIH NS- 13980.
Physiological activity of efferent vestibular neurons and their action on primary af-
ferents in the toadfish. STEPHEN M. HIGHSTEIN AND ROBERT BAKER (Marine
Biological Laboratory).
The efferent and afferent innervation of the vestibular semicircular canals in the toadfish (Opsanus
tan) are anatomically distinct both centrally and peripherally. This arrangement permits glass micro-
electrodes to be inserted into visually identified axons of efferent and afferent neurons. In order to study
the activity of efferent neurons and their physiological effect upon afferents, toadfish were spinalized, held
in a plastic tank, and perfused through the mouth with running sea water. Following stable axon pen-
etrations depolarizing pulses of 0.1-1 nA and 10-100 ms duration were passed through the microelectrode
to evoke action potentials. Efferent neurons characteristically responded with a single action potential
while all primary afferents generated a train of impulses equal to the duration of membrane depolarization.
Efferent neurons were spontaneously active (1-5 impulses/s) while afferents displayed the same spectrum
of regular and irregular activity seen in other vertebrate labyrinths.
When light punctate tactile stimulation was applied to particularly sensitive areas around the nose,
lips, and eyes all efferent vestibular neurons increased their level of spontaneous activity. Continuous
tactile contact — especially with moving stimuli — produced an "alerting response" characterized by eye
retraction, cessation of gilling, fin erection, and fanning. This behavioral arousal frequently culminated
in swimming. Since eye retraction was a sensitive measure of the onset and progress of the above behavioral
sequence it was employed to evaluate timing of neural activity. During all stages of the altering response
there was an increase in activity of both efferent and afferent neurons. The increase in discharge frequency
seemed to be positively correlated with the level of arousal and the changes were especially clear during
swimming. As expected, peripheral section of the efferent bundle abolished the behaviorally observed
activation of afferents. Electric pulse stimulation of either severed or intact efferent axons evoked pre-
sumably monosynaptic (latency 1-1.5 ms) EPSPs and action potentials in primary afferents. These data
suggest that the efferent vestibular system in the toadfish may terminate directly on primary afferents
with an excitatory action. In view of the neuronal correlates to behavior, we conclude that the physiological
role of the efferent vestibular system may be to enhance the sensitivity of afferents to motion both prior
to and during movement.
Supported by NS 21518.
Fast axonal transport is not affected by dimethyl sulj oxide (DM SO) used to facilitate
glycerination and/or glutaraldehvde fixation of squid axons. ALAN J. HODGE AND
W. J. ADELMAN, JR. (Laboratory of Biophysics, NINCDS, NIH, MBL).
Squid giant axons and smaller axons in fin and stellate nerves were examined under video-enhanced
differential interference contrast conditions during the application of DMSO-containing solutions de-
signed to facilitate (a) the formation of a glycerinated model axon system, and (b) rapid fixation using
glutaraldehyde as the cross-linking agent. Freshly dissected preparations were maintained in oxygenated
filtered sea water to establish the presence of vigorous fast axonal transport (FAT). Irrigation with sea
water containing 1 5% DMSO caused no change in the rate or character of the FAT over periods of
several hours. However, application of a glutaraldehyde fixative (Hodge and Adelman 1980, J. Ultrastr.
Res. 70: 220-24 1 ) containing 1 5% DMSO resulted in fairly rapid fixation as judged by cessation of
transport even deep in the axoplasm of giant axons within about a minute, and with no discernible
change in optical properties. Electron microscopy showed good preservation of axoplasmic structure,
including microtubules, comparable with that obtained by cannulation/irrigation fixation (Hodge and
Adelman 1980). Irrigation with a neutral buffered solution containing 15% of both DMSO and glycerol
caused quite severe blebbing and vacuolization within a few minutes, but with no apparent effect on
FAT. The blebbing receded and disappeared within a total time of about ten minutes. The axons now
were scarcely distinguishable from their original appearance, and continued FAT unabated. These pre-
liminary results support the notion that a model axon system suitable for the study of FAT may be soon
attainable.
528 ABSTRACTS FROM MBL GENERAL MEETINGS
The pH dependence of the tetrodotoxin-blockade of sodium channels. S. L. Hu AND
C. Y. KAO (Department of Pharmacology, State University of New York Down-
state Medical Center, Brooklyn, NY 1 1203).
Tetrodotoxin (TTX) and saxitoxin (STX) are important neurobiological tools because of their se-
lective and stoichiometric blockade of the sodium channel. Recently, some stereospecific, similar func-
tional groups have been identified in these different molecules, and a surface receptor is proposed as the
common site of their interactions with excitable membranes (see Kao 1983, Toxicon. Suppl. 3: 211-219).
Among these similarities are a cationic guanidinium moiety and a pair of adjacent -OH groups. In STX,
the -OH's are on C-12; in TTX, they are on C-9 and C-10. In STX, the C-12 -OH's are essential for
hydrogen-bonding to membrane components.
In TTX, the C-10 -OH deprotonates with a pK.a of 8.8, thereby permitting some manipulations of
its chemical form within physiological ranges of pH. Previous investigations on the pH dependence of
TTX-action relied on single, approximately equipotent doses of TTX at different pH's. Although lacking
in quantitative consistency, the earlier results show that TTX was more active in the cationic form at
neutral pH than in the zwitterionic form at alkaline pH.
We have reinvestigated the pH dependence of the TTX-blockade on the internally perfused squid
giant axon under voltage-clamped conditions. Dose-response relations have been obtained for the max-
imum INa at pH 8.80 and 7.80, at which the proportion of the protonated form of C-10 -OH is 0.5 and
0.9 respectively. Were the activity of TTX determined solely by the electric charge(s) of the whole
molecule, the relative potencies at these pH's should be close to 1.8 (0.9/0.5). We found an ED50 of 5.2
nA/at pH 7.80 and 14.2 nA/at 8.80. The potency ratio of 2.7 suggests that an important effect of alkaline
pH on the TTX-blockade is the loss of a hydrogen-bonding site, and not merely the abundance of a
zwitterionic form.
This work is supported in part by an NIH grant, NS 14551. S.L.H. is a Grass Fellow in Neuro-
physiology.
Hermissenda crassicornis: a disease complex. I. The normal animal. ALAN M. Ku-
ZIRIAN (NINCDS-NIH, Marine Biological Laboratory, Woods Hole, MA 02543),
Louis LEIBOVITZ, AND DANIEL L. ALKON.
The nudibranch mollusc, Hermissenda crassicornis, has been used for over a decade for neuro-
biological research with the aim of developing a model system for the study of associative learning. This
animal, endemic to the Pacific Coast of North America, is shipped to this laboratory weekly and is then
maintained under laboratory conditions consistent with short- and long-term animal husbandry. Nor-
mally, healthy animals have been cultured through three generations, and field collected adults have been
maintained in the laboratory for up to three months (Harrigan and Alkon 1978, Biol. Bull. 154: 430-
439). During the spring of 1983, the mean survival rate dropped precipitously to between 10 to 12 days,
with no change in the routine maintenance conditions. The nudibranchs also exhibited progressive
morphological changes consistent with some form of debilitating disease and infectious organism(s). The
onset of this problem coincided with naturally-occurring environmental perturbations experienced before
the animals were collected and sent to Woods Hole.
Due to the obvious pathology exhibited by the infected animals, a program of investigation was
implemented between the Section on Neural Systems, Laboratory of Biophysics, NIH, and the Marine
Animal Health Laboratory, to isolate and identify the causative agents infecting the nudibranchs. Con-
currently, a histological study of the non-neural organ systems of normal, non-infected Hermissenda was
also instituted to provide baseline data for comparison with the pathological material. Light, scanning,
and transmission electron microscopic observations of the oral tentacles, rhinophores, and cerata revealed
in normal individuals a similar ciliated epithelium with putative mechano- and chemoreceptors being
especially prevalent on the oral tentacles and rhinophores. There is an underlying, subepidermal complex
of mucous glands and nerve, muscle, and connective tissue fibers. The cerata contain hepatic tissue of
several cell types and the cnidosac with encapsulated nematocysts.
Associative learning in Hermissenda crassicornis (Gastropoda): evidence that light
(the CS) takes on characteristics of rotation (the UCS). I. IZJA LEDERHENDLER
(NIH Lab of Biophysics, Marine Biological Laboratory), SERGE GART, AND
DANIEL L. ALKON.
Associative learning in Hermissenda satisfies a host of criteria traditionally applied to vertebrate
learning. Reductions of positive phototaxis are produced by repeatedly pairing light and rotation stimuli.
A variety of such studies have shown that an associative change in behavior has been learned. The nature
NEUROBIOLOGY, LEARNING, BEHAVIOR 529
of the learned association between stimulus and response remained obscure however, because the re-
sponse^) to rotation (the unconditioned response, UCR) had not been specified.
Recent advances (see Cartel al., 19835/0/. Bull. 165) allowed a precise description and quantification
of the UCR. During rotation in the dark, all Hermissenda shorten the foot muscle. Presentation of light
alone causes an increase in foot length (68% of cases). After associative training, in 83.3% of cases, paired
animals shortened the foot in response to light. Control groups did not change.
We measured difference scores for each individual comparing length 3 s after light onset with length
in the dark. The mean score of the paired group (N = 6) was significantly reduced (P < .05). Neither
the random (N = 4) or the naive (N = 4) control groups were statistically different from pre-training
values or from each other. The pre-training minus post-training scores between paired and pooled control
groups were significantly different (P < .01).
An historic hallmark of classical conditioning (a special form of associative learning) is that the CS
take on some functional character of the UCS. Our data demonstrate that light, which originally evokes
foot-lengthening produces an opposite and, therefore, new response as the result of associative condi-
tioning. This new conditioned response resembles the unconditioned response to rotation.
Propagating calcium spikes in identified cells in the supraesophageal ganglion of the
giant barnacle, Balanus nubilus. LISA A. LEWENSTEIN (New York Medical Col-
lege, Valhalla, NY).
Reported here is a cell having sufficient TTX-insensitive calcium channels to generate a propagating
action potential, without the use of TEA to block voltage-sensitive potassium channels.
The cell is located on the posterior-medial margin of the ventral surface in each hemiganglion of
the supraesophageal ganglion of the barnacle. Its axon extends across the commissure to the contralateral
hemiganglion where it branches into a synaptic field and continues out the contralateral antennular
nerve. Only one such cell exists on each side.
Intracellular recordings were made from the soma using a 10-20 mesohm KC1 microelectrode. A
suction electrode was placed on the contralateral antennular nerve for recording extracellularly and
stimulating antidromically. Anatomical information was obtained by iontophoresis of Lucifer yellow into
the soma and cobalt backfills of the antennular nerve. Optical experiments were done by injecting
Arsenazo III, a calcium sensitive dye, into the cell and detecting absorbance changes at 660 nm in different
regions of the cell when the soma was stimulated.
In normal saline, the cell produced action potentials with an average amplitude of 80 mV, a 4 ms
duration at half-height and a 60 ms undershoot. Fifteen minutes after application of 3 X 10 7 M TTX,
action potentials could be elicited orthodromically and antidromically, while the extracellular recording
was devoid of other activity. Superfusion with saline in which all the sodium was replaced with choline
produced similar results.
Saline in which the normal 20 mM Ca was replaced with 2 mM Ca, 1 8 mM Co, or mM Ca, 30 mM
Mg allowed an action potential to propagate as well.
Finally, the addition of 3 X 10~7 M TTX to 2 mM Ca, 18 mM Co saline silenced all intracellular
regenerative activity as well as all extracellularly recorded activity. Activity returned after washing in
normal saline.
Optical experiments have confirmed calcium entry into the soma, along the length of the axon across
the commissure and through the contralateral hemiganglion when a propagating action potential was
stimulated.
In summary, here is a cell capable of producing a propagating calcium action potential along the
length of its axon, as well as its soma, for which either calcium or sodium is sufficient and neither calcium
nor sodium is necessary.
Visualization of depolarization-evoked presynaptic calcium entry and voltage de-
pendence of transmitter release in squid giant synapse. R. LLINAS (N. Y. U.
Medical Center), M. SUGIMORI, AND J. M. BOWER.
Data obtained using voltage clamp techniques in the squid giant synapse have suggested that pre-
membrane potential determines not only the amount of calcium that enters the preterminal during de-
polarization, but also directly influences calcium-evoked release of transmitter (Llinas 1981, Biophys. J.
33: 323-352; Simon 1983, Biophys. J. 41: 136). A further demonstration of this voltage sensitivity has
been performed utilizing direct intracellular pressure injections of calcium which induce large-amplitude
(10-20 mV) long-duration (10s of seconds) excitatory postsynaptic potentials (EPSPs). We have found that
following a calcium injection, a voltage change induced by current injection in the preterminal results in
an immediate and reversible increase in the amplitude of the injection-evoked release. This increase is
530 ABSTRACTS FROM MBL GENERAL MEETINGS
present even when entry of additional calcium through voltage-sensitive calcium or sodium channels is
blocked by cadmium and TTX or in low extracellular calcium medium (10~6 M).
We have also studied the spatial distribution of calcium entry during a presynaptic depolarization.
Using a double microchannel plate (ITT- Aerospace Optical Div.) coupled to an image-intensifying video
camera (Dage, MTI), we have achieved nearly single photon sensitivity and thus have been able to visualize
directly the spatial distribution of the light response of the photoprotein aequorin (Llinas 1975, Proc. Natl.
Acad. Sci. USA 72: 187-190) to increase in intracellular calcium concentration. Our results indicate that
calcium entry, even during prolonged (3 s) and large (50 mV) depolarizations of the preterminal, is restricted
to the region of synaptic contact of the preterminal, where morphological evidence suggests that calcium
channels are located and transmitter liberated (Pumplin 1981, Proc. Natl. Acad. Sci. USA 78: 7210-7213).
This calcium-evoked aequorin response is spatially restricted even though aequorin has diffused throughout
the presynaptic fiber.
Supported by NS 14014 from NINCDS.
Ordered assemblies of neurofi lament proteins isolated from squid giant axon. J.
METUZALS, D. F. CLAPIN (Faculty of Health Sciences, University of Ottawa,
Ottawa K1H 8M5, Ontario, Canada), P. A. M. EAGLES, AND G. J. FENNELY.
An investigation of the three-dimensional structure of the neurofilament network proteins and their
periodic supramolecular aggregates may enable us to model and understand the structure of the crystalline
arrays of filaments seen in brains of Alzheimer's patients. These preparations lend themselves to image
processing techniques through which the signal-to-noise ratio of the structural detail can be maximized.
A preparation of neurofilament proteins, obtained by extraction of extruded axoplasm, was solu-
bilized and reconstituted essentially according to the same procedure as that used for the crystallization
of tropomyosin (see Metuzals et al. 1982, Biol. Bull. 163: 387). The crystallization product was pelleted
by centrifugation at 10,000 X g for 10 min and was fixed and embedded according to standard procedures.
Analysis of the extracted axoplasm by SDS-PAGE showed that it consisted of 82% neurofilament proteins,
14% tubulin, and a small amount of actin. The pellet obtained following the crystallization procedure
had a similar composition.
Light microscopy of the precipitate showed a network of highly birefringent coiled strands and
numerous small birefringent crystals. Electron micrographs of thin sections of embedded pellets showed
sheets of intercoiled filaments 4-5 nm in diameter. The sheets are curved into tubes and rolled up to
form cylindrical scrolls. Densely packed, layered assemblies of filaments were also observed. The samples
of the pellets, stained negatively with 1% uranyl acetate, contain tubular networks and crystalline sheets.
The crystalline sheets consist of 2-3 nm wide filaments in a near orthogonal lattice with dimensions 4.6
nm X 5.9 nm. The filaments have a way or kinked appearance suggesting a helically intercoiled orga-
nization. Preliminary analysis of the arrangement of units in the sheets and of the computed diffraction
pattern of the tubes showed that the basic lattice is similar in the two structures.
The observed regular assemblies appear to be different motifs of assembly of neurofilament proteins
which may aid in the elucidation of the functional role of neurofilament proteins under normal and
pathological conditions.
This investigation was supported by grant MA- 1247 from the Medical Research Council of Canada.
Optical recording of action potentials from mammalian nerve terminals in situ.
A. L. OBAID, H. GAINER, AND B. M. SALZBERG (University of Pennsylvania
and N.I.H.).
A detailed understanding of the physiology of synaptic transmission in the vertebrates has been
delayed by our inability to monitor the action potential in nerve terminals because the small size of the
nerve terminal prevents a direct measurement of the presynaptic potential change. The vertebrate hy-
pothalamo-neurohypophysial system represents an excellent model for the study of excitation-secretion
coupling, but here also, the neurosecretory terminals are too small for microelectrode recording of the
electrophysiological events that affect release.
We report here the use of optical methods that employ voltage sensitive dyes to record action
potentials from populations of nerve terminals in the intact neurohypophysis of the CD-I mouse, and
the manipulation of the shape of the action potential by extracellular calcium and other agents known
to affect the release of neurohormones and neurotransmitters. A PDP 1 1/34 based system for Multiple
Site Optical Recording of Transmembrane Voltage (MSORTV) was used to record the absorption changes
from 124 regions of the posterior pituitary gland, stained with the potentiometric probe NK2367, and
stimulated with 0.5 ms electrical pulses to the infundibulum.
NEUROBIOLOGY, LEARNING, BEHAVIOR 531
We find that at 24°C, the action potential has a width at half height of 3-4 ms. This duration is
significantly increased in the presence of 4-aminopyridine at a concentration (50 pM) known to promote
exocytosis. Elevated Ca++, increased frequency of stimulation, and 4-aminopyridine enhance a slow
component of the optical response having a wavelength dependence characteristic of light scattering. This
signal appears in the absence of stain, does not reverse with wavelength, is present in white light, and
is blocked by 1 mMCd++. These results strongly suggest that the light scattering signal monitors secretion
and should be useful for resolving the kinetics of release.
We expect that direct optical measurement of transmembrane potential changes from the nerve
terminals of the mammalian neurohypophysis, when correlated with the light scattering changes that
appear to be associated with secretion of neurohypophysial peptides, will provide new insight into the
electrophysiology of transmitter and hormone release.
We are grateful to D. Langer for help during some of these experiments. Supported by U. S. Public
Health Service grant NS 16824 and a Steps Fellowship to A.L.O.
A relatively stable 100Kd protein is derived from the Ca2+ -dependent proteolysis of
neurofilament proteins in the squid axoplasm. HARISH C. PANT, PAUL E. GAL-
LANT, ROCHELLE S. COHEN, AND HAROLD GAINER (Laboratory of Preclinical
Studies, National Institute on Alcohol Abuse and Alcoholism, ADAMHA, Rock-
ville, MD 20852).
Previous studies on the kinetics of degradation of neurofilament proteins in squid axoplasm by an
endogenous calcium activated neutral protease (CANP) suggested a relatively stable lOOKj protein in-
termediate (Pant and Gainer 1980, J. Neurobiol. 11: 1-12). Further analysis using SDS-PAGE and gel
scanning of stained proteins on gradient slab gels confirmed these observations. These studies also in-
dicated that the neurofilament proteins which were found in the 100,000 X g supernatant and pellet from
axoplasm differed. The supernatant contained relatively more 200IQ neurofilament protein, whereas
pellet contained relatively more larger forms (>400K<1). Both forms were degraded by endogenous CANP,
but with different kinetics and peptide products. This was more clearly visualized by first phosphorylating
the neurofilament proteins using [7-32P]ATP and endogenous kinase, and then subjecting the labeled
proteins to CANP degradation. The labeled 200Kd proteins in both supernatant and pellet were degraded
to a major lOOKu protein product and several smaller (ca. 95KJ protein products. The labeled >400K<j
forms in both fractions did not produce these intermediates upon CANP degradation, but rather larger
(>250Kd) and smaller (ca. 37KJ protein products. This suggests that >400K<1 neurofilament protein is
not a simple oligomer of 200K<j protein, but may represent a separate gene product or a highly cross
linked form of the 2001^ protein.
Some morphological observations on the giant synapse of immature squid, Loligo
pealei. D. W. PUMPLIN (Univ. of Maryland Schl. of Med.) AND J. HARRIGAN.
Immature L. pealei (dorsal mantle lengths 3.5-20 mm) were captured in ongoing ecological studies
by trawling in the upper 30 m of the water column. As soon as possible after capture, specimens were
fixed with glutaraldehyde in phosphate buffer containing sucrose. Stellate ganglia were postfixed in os-
mium, dehydrated, and embedded by standard methods. Thin sections were taken at 30-micron intervals
transverse to the most distal giant fiber, proceeding from the center of the neuropil into the giant fiber
lobe. In squid with mantle lengths of 3.5-5.5 mm, pre- and post-synaptic giant fibers were identified by
their differential staining and electron density.
In immature ganglia, pre- and post-giant axons lay adjacent to each other separated by a thin glial
layer. The pre-axon was about 5 microns in diameter, comparable to or somewhat larger than the post-
axon. One pre-axon generally lay adjacent to two or more as-yet-unfused portions of the post-axon. In
one such case, a projection from one post-axon passed through the glial layer to appose the pre-axon
directly. This apposition was somewhat small (300 nm in length), but had the characteristic features of
active zones of the giant synapse in older squid. Thus at least some active zones develop prior to complete
fusion of elements of the post-axon.
Profiles of fibers contributing to the post-axon became more numerous and smaller in sections closer
to the distal part of the giant fiber lobe, but were always in a discrete bundle. Intercellular junctions were
found between some fibers. Junctions with widely-spaced (200 A) parallel membranes and intervening
electron density appear to be desmosomes; those with closely-spaced membranes (less than 100 A) suggest
invertebrate gap junctions, although both types should be characterized more fully.
D.W.P. is supported by grants from the NIH, Muscular Dystrophy Assoc., and the Bressler Fund of
the Univ. of Maryland. J.H. is supported by the NINCDS, NIH. We thank M. Volkman, C. Tyndale, and
the MBL EM lab for assistance.
532 ABSTRACTS FROM MBL GENERAL MEETINGS
Flight fuel utilization and flight energetics in the migratory milkweed bug, Onco-
peltus fasciatus. MARY ANN RANKIN AND LAURA L. MORROW (University of
Texas, Austin).
In order to identify the primary fuel for long distance flight in the migratory milkweed bug, On-
copeltus fasciatus (Lygaeidae: Hemiptera), total body lipid and carbohydrate were measured in animals
flown to exhaustion, animals stopped after 30 min of flight (regarded as potential migrants), and animals
that did not make a long flight. Total body lipid showed a significant decrease in long fliers compared
to short fliers, while carbohydrate was not significantly different in any of the groups tested.
Measurements were made of I4CO2 expelled during flight by animals injected prior to testing with
potential flight fuels. Injected 14C-palmitic acid was metabolized significantly above resting levels through-
out the flight period, while '4C-glucose was not oxidized by fliers to any greater extent than by non-fliers.
MC-proline was utilized significantly above resting levels during approximately the first 90 min of flight.
l4C-glutamic acid was utilized only during the first 15 min of flight, while l4C-glutamine was not oxidized
by fliers to any greater extent than by non-fliers.
Evidence from COj evolution and oxygen consumption during flight indicated that short flight or
the first hour of long-duration flight is energetically more expensive than subsequent hours of long
duration flight. Wingbeat frequency measurements by stroboscope showed a change from 67.8 ± 1.5
beats per second to 63.6 ± 0.2 beats per second during the first 30 min of flight. However, it is doubtful
whether this drop is sufficient to account for the sevenfold decrease in energy demand which occurs
during the first hour of flight. Measurements of thoracic temperature during flight are planned. It is
concluded that proline may be important as an energy source for metabolically expensive short or the
initial period of long-duration flight, while lipid is the primary fuel for long-distance flight in this species.
This work was supported by NSF grant # PCM-81 10568.
Calcium transients in voltage clamped presynaptic terminals. STEPHEN J. SMITH,
GEORGE J. AUGUSTINE, AND MILTON P. CHARLTON (University of Toronto).
We have used the indicator dye Arsenazo III to measure Ca transients in the giant presynaptic
terminal of Loligo pealei. A 2-microelectrode voltage clamp configuration was used to control the pre-
synaptic membrane potential, while a third electrode was inserted into the postsynaptic axon to measure
transmitter release. Preparations were treated with tetrodotoxin (10-6 M) and injected with 3,4-diami-
nopyridine and tetraethylammonium to block Na and K currents. Arsenazo III was injected to a final
concentration of 0. 1 to 1 mA/. Optical signals were acquired with a single optical fiber (20-60 ^m
diameter) placed over the presynaptic terminal and were detected with a multiwavelength microspectro-
photometer.
Depolarizing voltage steps produced transient changes in dye absorbance spectra consistent with a
change in intracellular Ca. Intracellular Ca rose linearly during 20-40 ms depolarizing pulses at a rate
dependent on the amplitude of the presynaptic depolarization. Transients could be detected for pulses
as short as 4 ms. Signals declined over several seconds following repolarization.
The Ca-Arsenazo signal had a bell-shaped voltage dependence, as expected from the Ca current-
voltage relationship. However, the precise form of the voltage dependence varied with the position of the
light pipe along the presynaptic terminal. Signals recorded from the distal portion of the terminal were
largest at command potentials of -10 to 0 mV and disappeared at depolarizations of +60 or +70 mV.
This relationship agrees with voltage clamp measurements of presynaptic Ca currents. Arsenazo signals
from more proximal portions of the terminal were skewed, with peaks at +10 to +20 mV and suppression
at potentials more positive than +80 mV. This suggests that the potential at the proximal portion is
different from the rest of the terminal, perhaps due to voltage decrement within the long presynaptic
axon. We tested this possibility by recording membrane potential at several points along voltage clamped
terminals and found that, during large depolarizations, the proximal portion could be 20-30 mV less
depolarized than the distal portion. We conclude that Arsenazo III can be used to measure Ca transients
in voltage clamped presynaptic terminals, but that Ca entry is influenced by presynaptic voltage gradients.
Supported by NIH grant NS- 16671 to S. Smith, NRSA Fellowship to G. Augustine, and Whitehall
Foundation and MRC (Canada) grants to M. Charlton.
Single amino acids stimulate lobster (Homarus americanus) behavior against am-
bient and modified amino acid backgrounds. MARILYN SPALDING AND JELLE
ATEMA (Boston University Marine Program, Marine Biological Laboratory,
Woods Hole, MA 02543).
Raw sea water contains free amino acid concentrations in the pico-to-nanomolar range; ammonia
occurs in micromolar quantities. This is the normal noise background for lobster chemoreception. Lob-
NEUROBIOLOGY, LEARNING, BEHAVIOR 533
sters have prominent populations of receptor cells which are narrowly tuned for single amino acids and
ammonia. Such cells are found both in smell and in taste organs. One might expect, therefore, that
elevating the normal background for one amino acid should raise its detection threshold to this new level,
but should not interfere with the reception of another.
We used eight one-year-old lobsters of about 13 mm carapace length in 50 ml centrifuge tubes. We
counted antennular-flicking rate to measure their responses to 2 ml stimuli injected into the sea water
background flow of 0.6 ml • s '. We obtained dose-response curves for L-proline, L-glutamine, and ammonia
from lO'12 M to 10~3 M in single log steps.
In normal sea water the proline and glutamine curves emerged from control levels at 10~9 M, the
ammonia curve at 10~5 M, i.e., just above the ambient sea water background for these three compounds.
With elevated sea water backgrounds of 10 8 M and 10 6 M proline, the proline threshold shifted up to
the new background levels, but the entire curve also dropped, indicating that even at high stimulus
concentrations responses were suppressed in elevated backgrounds. This was seen also for glutamine;
ammonia was not tested.
Similarly and unexpectedly, the glutamine and ammonia curves dropped in elevated proline back-
grounds; also, the proline response at 10~6 M was completely suppressed by a 10~6 M glutamine back-
ground. We conclude that the peripheral receptor cells converge centrally such that their narrow tuning
is not used behaviorally.
In elevated backgrounds lobsters responded to both higher and lower stimulus concentrations in-
dicating that sudden temporary dilution of only one amino acid in the whole background mixture cannot
only be detected but constitutes a behaviorally significant stimulus.
Depolarizing and desensitizing actions of glutaminergic and cholinergic agonists at
the squid giant synapse. E. F. STANLEY (Johns Hopkins Med. Sch.).
Studies on the squid stellate ganglion giant synapse have resulted in a detailed understanding of the
physiology of synaptic transmission. However, relatively little is known about the pharmacology of this
synapse, primarily because of the formidable diffusion barrier for drugs entering from the bathing medium.
This study used a technique of infusing substances through the arterial blood supply as described in
Stanley and Adelman (1982, Biol. Bull. 163: 403), to circumvent this barrier in order to examine the
actions of glutaminergic and cholinergic agonists at this synapse.
Abruptly switching the infusion solution from sea water to L-glutamate results in an immediate
(3 to 8 seconds delay for dead space in the cannula and artery), dose dependent (0.1 to 10 mA/) depo-
larization (maximum 1 3.5 m V) followed at higher concentrations of glutamate by a gradual repolarization,
presumably due to desensitization. Sterically restricted glutamate agonists also depolarized the postsyn-
aptic axon in a decreasing order (measured as maximum rate of initial depolarization) of: L-glutamate,
kainate, quisqualate, L-aspartate = D-aspartate, ibotenate, N-methyl-D-L-aspartate. The amino acids
taurine and glycine had no effect, whereas serine was a weak agonist.
Acetylcholine and its agonist carbamylcholine also depolarized the synapse in a dose-dependent
manner and with maintained infusion at a high concentration (10 mA/) desensitized to both themselves
and each other. The cholinergic receptor appeared to be distinct from the glutaminergic receptor since
desensitization by glutamate did not eliminate the carbachol response nor vice versa.
Glutaminergic agonists also desensitized the synapse to the endogenous transmitter as evidenced by
a gradual reduction and eventual disappearance of the EPSP evoked by stimulating the pre-nerve, whereas
the cholinergic agonists did not. This is consistent with the endogenous transmitter being a glutamate
agonist. Carbachol also reduced the EPSP but this effect was transient and the EPSP recovered during
carbachol desensitization. Since no change was detected in the intracellularly recorded pre-synaptic resting
potential these results suggest that the carbachol-induced reduction in the EPSP is a post synaptic phe-
nomenon, due perhaps to a fall in the input resistance. Such an effect, a depolarization combined with
a decrease in input resistance, is explained if carbachol activated a chloride channel, though this hypothesis
requires further study.
Functional and chemical characterization of squid neurofilament polypeptides.
R. V. ZACKROFF (Department of Cell Biology and Anatomy, Northwestern
University Medical School, Chicago, IL), W. D. HILL, M. TYTELL, AND R. D.
GOLDMAN.
Neurofilaments isolated from the optic lobe of the squid (Loligo pealei) by two cycles of in vitro
assembly-disassembly are composed of four polypeptides, with molecular weights 60K, 74K, 100K, and
220K (ZackrofFand Goldman 1980, Science 208: 1 152-1 155). Densitometric scans of sodium dodecyl
sulfate (SDS) gels indicate that >70% of neurofilament protein is represented in the 60K band, while 5-
10% of the protein is represented in each of the 74K, 100K, and 220K bands. When neurofilaments are
534 ABSTRACTS FROM MBL GENERAL MEETINGS
isolated by assembly-disassembly from the stellate ganglion, an additional protein of 65K is obtained.
We have investigated the ability of each of these proteins to form homopolymer intermediate filaments
(IF) in vitro after purification by SDS polyacrylamide gel electrophoresis followed by removal of the SDS.
We find that the 60K and 74K proteins can form homopolymer IF, while the 65K, 100K, and 220K
proteins do not. Peptide mapping of each of these proteins after digestion with S. aureus protease (Cleve-
land et al. 1977, J. Biol. Chem. 252: 1 102) results in a virtually identical pattern for each of the four
optic lobe proteins, while the 65K stellate ganglion protein exhibits a different peptide map. Since the
higher molecular weight (220K. and 100K.) proteins exhibit structural similarity to the 60K and 74K
proteins, these results suggest that all four of these proteins are IF structural polypeptides. It appears
likely that the 60K and/or 74K proteins are precursors which become covalently linked to form the 100
and/or 220K proteins, or, conversely, that the higher molecular weight proteins may be cleaved to form
the lower molecular weight (74K. and 60K) IF forming proteins. The properties of the 65K protein suggest
that it may be a neurofilament associated protein which is specifically localized in the neuronal cell body
and/or axonal hillock region.
Supported by NIH and NSF.
PARASITOLOGY, PATHOLOGY, AND AGING
Phagocytosis and intralysosomal killing o/Leishmania mexicana by Entamoeba his-
tolytica. L. F. ANAYA-VELAZQUEZ AND K.-P. CHANG (Experimental Pathology
Section, Center for Research and Advanced Studies of the National Polytechnical
Institute P. O. 14-740, Mexico D. F., 07000 Mexico).
The protozoa Entamoeba histolytica is one type of professional phagocytes, which can ingest and
digest other cells by unknown mechanisms. Leishmania spp. are unicellular protozoa, which normally
live in the lysosomes of the macrophages. Cellular interactions between these two organisms are, therefore,
of interest for investigations.
106 trophozoites and 2 X 107 leishmanial promastigotes were incubated in the amoeba culture
medium containing heat-inactivated fetal bovine serum at 35 °C in a stirring vial with or without cyto-
chalasin B (10 Mg/ml). Samples were taken at different times points for microscopic examinations and
for cultivation to check the viability of the promastigotes. In the absence of cytochalasin B, there was
a rapid uptake and degradation of promastigotes by the amoeba. The rate of uptake reached a maximum
of about 350 promastigotes/ 100 amoeba in 20 minutes. The rate of intracellular degradation of pro-
mastigotes reached 90% by 40 minutes. In the presence of cytochalasin B, there was neither uptake nor
degradation of the promastigotes by the amoeba.
We also prelabeled amoeba with FITC-Dextran overnight and then infected them with leishmanias.
UV microscopy revealed fluorescence in the leishmania-containing vacuoles, indicating lysosome-phagosome
fusion. Fluorescent intensity in these vacuoles was measured by using a photomultiplier at excitation
wavelengths of 350-450 nm and 450-495 nm, the emission wavelength being 550 nm. The intravacuolar
pHs were calculated from the ratio of 450-495 nm/350-450 nm against a standard plot of fluorescein
solutions at different pHs. The leishmania-containing vacuoles in amoeba were determined to have a low
pH of 4.5-5.0 consistent with that of the lysosomal compartment. Since E. histolytica does not utilize
oxygen for microbicidal action, its lysosomal killing of leishmanias must be based on a non-oxidative
mechanism.
IgE monoclonal antibodies produced from mice immunized with irradiated cercariae
0/Schistosoma mansoni. G. EKAPANYAKUL, A. FLISSER, A. Ko, AND D. HARN
(Harvard School of Medicine).
IgE antibodies have been implicated in the host immune response to several helminth infections
including schistosomiasis. The exact role that IgE antibodies play in these parasitic infections is not
understood. To help elucidate the functions in which IgE antibodies might be involved we generated
parasite specific IgE antibodies using hybridoma technology.
To generate parasite specific IgE antibodies mice were primed and boosted with irradiated cercariae
of Schistosoma mansoni. Spleens of immunized mice were used for fusions on days 4, 5, and 6 after the
final boost. Spleen cells were fused with NS-1 cells using polyethylene glycol.
Prior to spleen removal, immunized mice were bled and the sera were tested for IgE antibodies to
parasite antigens. IgE antibodies were detected by enzyme linked immunosorbent assay (ELISA) using
a purified rabbit anti-mouse IgE (epsilon specific) antibody or by passive cutaneous anaphylaxis in rats.
PARASITOLOGY, PATHOLOGY, AGING 535
Hybridomas were initially screened for IgE secretion and/or antibodies to parasite antigens by ELISA.
The frequency of IgE positive hybridomas ranged from 4 to 20 times that found in normal spleen cell
populations.
Putative IgE secreting hybridomas were also tested for surface binding to living schistosomula and/
or cercarial tails by indirect immunofluorescence. Several of the IgE antibodies were surface membrane
binding.
Ten IgE and parasite specific hybridomas were cloned by limiting dilution. Assay of the cloned
hybridomas allowed us to select and begin expansion of five clones which were still secreting IgE antibodies
as determined by ELISA.
Characterization of the major surface antigen of Plasmodium falciparum mero-
zoites. A. S. FAIRFIELD, D. A. E. DOBBELAERE, AND M. PERKINS (Rocke-
feller U.).
Invasion of erythrocytes by the merozoite stage of Plasmodium falciparum. a human malaria parasite,
requires specific recognition between the host cell and the parasite. The receptor in the red cell has been
identified as glycophorin, but the corresponding receptor on the merozoite is unknown. To date, the only
merozoite surface protein identified in P. falciparum by monoclonal antibody is a protein of MW 200
kd. Our project was concerned with the characterization of this protein with a view of understanding its
role in merozoite invasion.
Initially we were successful in purifying the 200 kd antigen using Affigel crosslinked to the mono-
clonal antibody. By labeling in vitro cultures of P. falciparum with 3H-glucosamine it was possible to
show that the protein is a glycoprotein.
Pulse-chase experiments show that during maturation and release of merozoites the 200 kd protein
is processed into two soluble (MW 1 30 and 1 50 kd) and two membrane-bound (50 and 80 kd) polypeptides.
To determine at which parasite stage the protease is present, 3H-proline-labeled substrate was incubated
with lysates of different stages of the parasite. Contrary to expectations, none of the parasite stages were
able to process the protein. However, schizonts solubilized with triton X-100 did permit proteolysis to
occur. It appears, therefore, that the protease is membrane-bound and closely associated with the 200 kd
protein to be effective. An identical experiment using red blood cells showed that they were not an active
factor in this process.
The possibility that this proteolysis was an artifact of parasite isolation also warranted investigation.
3H-glucosamine-labeled schizonts were allowed to mature and re-invade red cells in unlabeled medium.
Surprisingly, many of the major proteins, including the 200 kd protein, were conserved well into the parasite
ring stages.
In summary it appears that a major 200 kd surface protein of the mature stages of P. falciparum is
a glycoprotein based on 3H-glucosamine labeling and purification by monoclonal antibody. A putative
parasite protease cleaving the 200 kd antigen is produced as early as the schizont stage, is membrane-
bound, and is probably closely situated to its 200 kd substrate. The possibility that proteolysis of the 200
kd protein is a result of experimental manipulations was also raised.
An epizootic disease-complex of wild and laboratory-maintained Hermissenda cras-
sicornis. Louis LEIBOVITZ (Laboratory for Marine Animal Health, Marine Bio-
logical Laboratory, Woods Hole, MA 02543), ALAN KUZIRIAN, JUNE HARRI-
GAN, EDWARD F. SCHOTT, IZJA LEDERHENDLER, AND DANIEL L. ALKON.
Hermissenda crassicornis is an eolid marine nudibranch utilized in comparative developmental,
behavioral, and neurobiological laboratory studies. A highly fatal epizootic disease-complex in wild and
laboratory-maintained H. crassicornis is reported. Early signs of the disease are reduced photosensitivity,
decreased motility, and erosion and deformity (kinks) of the oral tentacles. More advanced gross signs
are pericardial clouding, cardiac arrest, increased cardiac eminence (hunchback), loss of cerata and oral
tentacles, and desquamation of the surface epithelium. Terminal signs are generalized body deformity
and depigmentation, rupture of the body wall, visceral prolapse, and death.
Histological and microbiological examinations revealed fungal infections of the surface epithelium
in early stages; generalized bacterial invasion in later stages. A ciliate, Licnophora sp., was found in
increasing numbers on surface appendages (tentacles and cerata) as the disease progressed. Work is in
progress to further characterize the disease and to develop methods of prevention, control, and eradication.
We wish to thank Dr. Jean Maguire, Ms. Catherine Tamse, and Dr. Eugene Copeland for their
assistance. This study was supported in part by a grant (No. 1-40-PRO 1333-03) from the Division of
Research Resources, National Institute of Health.
536 ABSTRACTS FROM MBL GENERAL MEETINGS
Trematode infection in Ilyanassa obsoleta: dependence on size and sex of the host,
and effect on chemotaxis. MATTHEW LIEBMAN (State University of New York
at Stony Brook).
Life cycles of trematodes parasitizing the common mud snail, Ilyanassa obsoleta, are well described,
but little is known about their effect on the population biology and behavior of host snails. I determined
infection rates of snail populations at Great Sippewisset salt marsh and Barnstable harbor in July and
August 1983 and tested the effect of parasitism on the alarm response to crushed conspecifics. Parasitism
was determined by dissection in most cases, or by release of cercariae.
Incidence of parasitism exhibited temporal and spatial variability. In an aggregation from the salt
marsh, the infection rate decreased from 71.6% on 3 July (n = 204), to 4.5% on 1 1 July (n = 200), to
3.1% on 16 July (n = 162), and then increased to 10.9% on 8 August (n = 1 10). On 3 July, Zoogonus
rubellus dominated the trematode community. On 8 August, Austrobilharzia variglandis was the dom-
inant parasite. On 3 July, an aggregation located fifty meters away showed 0% (cercarial release) infection
(n = 100). At Barnstable, an immobile aggregation of 100 snails showed 75.3% infection whereas only
2.7% (cercarial release, n = 110) of the main migratory population was infected. The dominant parasites
were Lepocreadium setiferoides and Himasthla quissetensis.
Female snails (mode = 24 mm) were larger than males (mode = 22 mm) and comprised 68% of
the population (n = 645) at Sippewissett. The incidence of parasitism was similar for males (27.1%) and
females (23.1%) and increased exponentially with size-class. Infection rates indicate snail age since in-
fections are permanent and older snails have been exposed to parasites longer. The largest size classes
of male and female snails had comparable infection rates, but the larger sizes of females indicate a faster
growth rate.
The effect of parasitism on the alarm response of snails from Barnstable was tested in a flowing sea
water aquarium. Snails were kept at the sediment surface, and then exposed to crushed conspecifics. No
significant differences in burrowing was observed between parasitized and non-parasitized snails.
These results indicate that incidence of parasitism 1 ) is spatially and temporally variable, 2) increases
with snail age, and 3) does not affect the alarm response.
Thanks to H. Stunkard, G. Brenchley, E. Shott, B. Crowther, and L. Leibovitz. Research supported
by a Founders Scholarship, MBL.
Mitogenic activity of extracts and supernates from Plasmodium falciparum. A.
PERCY (University of California, Los Angeles) AND M. PERKINS.
The malaria parasite Plasmodium falciparum has been shown to produce a mitogen which acts on
lymphocytes (Gabrielson and Jensen 1982, Am. J. Trop. Med. Hyg. 31: 441-448). The polyclonal ac-
tivation of lymphocytes is important in the disease, resulting in hypergammaglobulinemia and the pres-
ence of anti-self antibodies. As a first step in the characterization of this mitogen, we have examined
freeze-thaw extracts from various stages of the parasite for mitogenic activity.
Parasite extracts were prepared by the procedure of Gabrielson and Jensen from synchronous cultures
of P. falciparum. Dilutions of these extracts were incubated in microtiter plates for five days with 105
human peripheral blood lymphocytes. Parasites were maintained in erythrocytes from the lymphocyte
donor. After 1 16 hours of culture, five microcuries of tritiated thymidine was added to each culture,
incubated for four hours, and the cells were harvested. All experiments were performed in duplicate. The
human B cell mitogen, pokeweed mitogen, was used as a positive control, extracts from autologous
erythrocytes as a negative control, and extracts from heterologous erythrocytes and from another parasite
(Leishmania enriettii) as controls for antigen-specific stimulation.
Mitogenic activity (stimulation as strong as that seen with the mitogen control) was seen with 1:10,
but not 1:100 dilutions of extracts from ring forms, schizonts, and merozoites. However, both 1:10 and
1:100 dilutions of supernates from two hour cultures of schizonts and of merozoites showed high activity,
equivalent to that of the mitogen control. Pulse labeling experiments with tritiated proline have shown
that only two proteins synthesized by the parasite are released into the supernate, one of 130,000, the
other of 150,000 daltons.
Thus the Plasmodium mitogen is present in all stages of the parasite, but the highest activity is
obtained in the supernate of parasite cultures.
Intraparasitophorous vacuolar pH o/"Leishmania mexicana infected macrophages. L.
RIVAS AND K.-P. CHANG (Department of Microbiology, UHS/Chicago Medical
School, North Chicago, IL 60064).
Leishmania organisms are parasitic protozoa and agents of human leishmaniasis. They have de-
veloped an extraordinary mechanism of survival inside the macrophage lysosomes. One possibility pro-
PARASITOLOGY, PATHOLOGY, AGING 537
posed (Coombs 1982, Parasitology 84: 149) is that they may increase the lysosomal pH, thereby inac-
tivating the microbicidal function. We were unable to confirm this theory by quantitative fluorescence
measurement of individual vacuoles, labeled with FITC-dextran, with or without parasites.
The macrophages were infected at a ratio of 10 parasites per macrophage, as previously described
(Chang 1980, Science 209: 1240) in FITC-dextran at 1 mg/ml. The fluorescence was measured in sequence
with 2 excitation filters (350-450 nm and 450-490 nm). The emission filter was 550 nm. The ratio, 450-
490 nm/350-450 nm, was plotted against different pH standard solutions with FITC-dextran.
According to these data, in the early times of infection with either amastigotes or promastigotes
(1-3 h), a considerable proportion (30-40%) of the parasitophorous or nonparasitized vacuoles were
found in a weak acidic environment; later (24-48 h) after infection, they appeared in more acidic com-
partments of pH 4.5-5.5, very similar to the value previously reported for the lysosomes.
In other experiments, we have detected serine- and metalo-proteinases released by promastigotes
into the culture medium, as assayed by using I25l-casein at pH 6.5. The physiological role of these enzymes
in the survival of parasites in macrophages is not known.
We thank M. Bibee and P. Presley of Zeiss Co. and P. Olwell of Leitz Co. for their help with
fluorescent microscopy.
Detection of Leishmania Kinetoplast DNA using biotinylated DNA probes. GUIL-
LERMO ROMERO (Universidad Cayetano Heredia, Lima, Peru), YARA CSEKO,
AND DYANN WIRTH.
Recently, a method for the rapid identification of human pathogenic Leishmania was developed
(Wirth and Pratt 1982, PNAS19: 6999). The method uses 32P labeled Kinetoplast DNA (K-DNA) probes
hybridized to tissue blots and detected by autoradiography. To develop a more suitable technique to
work in endemic areas, we have examined the sensitivity of biotinylated K-DNA probes in the detection
of promastigotes and K-DNA of L. enrieltii.
L. enriettii K-DNA isolated from cultured promastigotes was nick-translated with biotinylated dUTP
(Enzo Biochemicals) and alfa-[32P] dATP. L. enriettii promastigotes and purified K-DNA were blotted
in nitrocellulose paper. The K-DNA probe (10 ng/ml) was hybridized as described (Wirth and Pratt
1982). The hybridized probe was detected by autoradiography and a variety of procedures based on the
recognition of biotin by streptavidin.
Two hundred promastigotes or 30 pg of K-DNA could be detected by autoradiography. A strep-
tavidin-horseradish peroxidase complex (Enzo Biochemicals) could resolve 105 promastigotes (15 ng of
K-DNA). Biotinyl-N-Hydroxysuccinimide ester was synthesized and used to biotinylate crosslinked al-
kaline phosphatase. The biotinylated enzyme allowed visualization of 104 promastigotes or 1.5 ng of
K-DNA (see Leary el al. 1983, PNAS 80: 4045). A fluorescence assay was also tested. Poly-1-lysine (M.W.
200,000 daltons) was consecutively biotinylated and labeled with excess fluorescein isothiocyanate (FITC).
The labeled protein was incubated at a 5 fold molar excess with streptavidin and the complex formed
was used to develop the blots. The biotinylated FITC labeled polylysine allowed the visualization of 1000
promastigotes ( 1 50 pg of K-DNA) with the aid of a hand held U V light. Extensive washing with 4 M
NaCl was used to eliminate nonspecific binding of the polylysine to DNA. Salmon sperm DNA was used
to verify the washing procedure.
The sensitivity of the procedures tested compares very well with other previously published reports.
The direct fluorescence assay may be useful in the field diagnosis of tegumentary leishmaniasis.
Antigens on both mechanical and lung stage schistosornula o/"Schistosoma mansoni
are masked by host molecules. L. D. SIBLEY, J. KRAKOW, A. FLISSER, AND D.
HARN (Harvard Medical School).
The acquisition of host antigens (Ag) by mechanical somula (MS) was studied using a monoclonal
antibody (Ab) to worm surface Ag and chronic mouse serum (CMS). Monoclonal 1C4 binds to the
surface of MS up to 96 h of culture in media containing FCS; culture of MS in normal mouse serum
(NMS) from CBA mice (H-2Kk) for 24 h results in elimination of 1C4 binding by immunofluorescence
assay (IFA). We conclude that host molecules present in serum mask worm surface Ag, and it was shown
by IFA with specific monoclonal Ab that these molecules are not H-2Kk or Iak. We have also shown that
IgG-depleted NMS eliminates 1C4 binding after 24 h in culture.
When lung stage schistosornula were examined by Sher et al. (J. Exp. Med. 1978, 148: 46-57) and
Gitter and Damian (Par. Imm. 1982, 4: 383-393). MHC Ag were detected on freshly harvested worms.
We confirmed the presence of H-2Kk and Iak on worms recovered from CBA mice at 6 h post-harvest.
However, presence of MHC by IFA on lung worms placed in culture diminished at 48 h indicating shedding
of these host Ag. This correlates with failure of CMS to bind freshly harvested worms, whereas worms
cultured 36 h showed strong binding of CMS.
538 ABSTRACTS FROM MBL GENERAL MEETINGS
To examine acquisition of MHC Ag, we cultured lung worms 72 h to allow shedding of original
host MHC molecules, and verified this by IFA. We then cultured MS or lung somula with washed
peritoneal exudate cells in PCS. While no transfer of MHC Ag was detectable up to 48 h, somula cultured
with CBA cells acquire H-2Kk and Iak after 72 h. Somula cultured with CD1 cells, shown to be negative
for H-2Kk and Iak, remain negative for these host Ag.
Inhibition of a surface binding monoclonal antibody to schistosomula o/Schistosoma
mansoni by lectins. LINDA SWISTON, ALBERT Ko, AND DON HARN (Harvard
School of Medicine).
Developing mechanical and lung somula of Schistosoma mansoni were analyzed for surface mem-
brane carbohydrates by a direct fluorescence assay using rhodamine conjugated (R) lectins. Lectins which
bound to the surface membrane of the schistosomula were also tested for their ability to inhibit binding
of a surface membrane specific monoclonal antibody. Our studies of lectin binding to mechanical somula
at various time points agree with those of Simpson et al. (1983, Mol. Biochem. Parasitol. 8: 191-205)
in that all lectins used [concanavalin A (Con A), lentil, peanut agglutinin (PNA), wheat germ agglutinin
(WGA), soybean agglutinin, and Ricinus communis agglutinin] showed a decrease in binding ability
related to an increase in time of culture. Our data on lectin binding to lung worms varies in two aspects
from the published data. We found no ability of RWGA, and only a slight ability of RPNA to bind to
our fresh and twenty-four hour cultured lung worms, whereas the findings of Simpson et al. and others
show binding of both RWGA and RPNA. However, these other studies did not examine cultured
lung worms.
Lectin inhibition of monoclonal antibody 1C4, developed by D. Harn and known to bind to the
surface of mechanical somula, was also studied. Worms were incubated with different lectins prior to the
addition of the monoclonal antibody. The parasites were then tested for rhodamine and fluorescein
fluorescence. Our results show that RPNA inhibits 1C4 binding up to ten hour post-transformation.
RConA showed a slight inhibition of 1C4 binding at a two hour post-transformation, but it seemed to
lose this quality quite rapidly. The other lectins showed no inhibition. Whether this inhibition was caused
by steric hindrance or direct binding of the lectin to the antigenic epitope was not addressed in this study.
Additional experiments involving inhibition with the respective sugars which will reveal the specificity
need to be carried out.
Host specificity of intestinal gregarines (Protozoa, Apicomplexa) in two sympatric
species o/Capitella (Polychaeta). GARY E. WAGENBACH (Department of Biology,
Carleton College, Northfield, MN), JUDITH P. GRASSLE, AND SUSAN W. MILLS.
Capitella species I and II were surveyed for parasites using field-collected and laboratory-reared
worms. Field-collected Capitella species I were from Wild Harbor (MA) and inbred laboratory-reared
strains originated from animals collected from Falmouth Harbor, from a culture maintained at the
Skidaway Institute of Oceanography, and from coastal California. Field-collected Capitella species II were
from New Bedford Harbor (MA) and laboratory strains originated from Woods Hole sewer outfall
animals. All laboratory strains were maintained under identical conditions: 15°C static culture in filtered
sea water using Sippewissett Marsh mud that had been freeze-thawed twice as substrate and food. Worms
were anesthetized in chloretone-sea water and mounted in 50:50 glycerol:ethanol for examination.
Gregarines were the most abundant parasites observed (field-collected Capitella occasionally had
trematode metacercaria and a possible coccidean). Notably all the Capitella species I samples had an
apparently identical unidentified gregarine, while both samples of Capitella species II contained a gregarine
of the genus Ancora. We propose that each Capitella hosts a unique gregarine even when the two Capitella
species are kept under identical conditions in the laboratory.
Highest densities of gregarines were found in the host gut between setigers 1 3 and 20-28 just posterior
to the stomach. Infection rates in the six samples varied from 27% to 100%. Five samples had a mean
gregarine population of <1 18 while one Capitella species I strain was 100% infected with a mean of 1645
gregarines/worm. The number of gregarines per individual in Capitella species I did not differ between
males and females. Heavily (mean = 1645 gregarines/worm) and lightly (mean = 9 gregarines/worm)
infected Capitella species I showed no difference in fecundity (number of eggs/individual in the first
brood). Capitella fecundity is strongly affected by genetic and environmental factors. We need to deter-
mine the probable effects of the gregarines on the population dynamics of these two sympatric Capitella
species.
The assistance of Dr. N. Levine is gratefully acknowledged.
PARASITOLOGY, PATHOLOGY, AGING 539
Structure oftubulin RNA from Leishmania enriettii. CLAIRE WYMAN (Johns Hop-
kins University School of Hygiene and Public Health) AND SCOTT LANDFEAR.
Protozoan parasites of the species Leishmania exist in two morphologic forms. The flagellated motile
promastigote lives extracellularly in the sandfly gut. Inside their mammalian host promastigotes are
ingested by macrophages and transform into nonmotile amastigotes. Tubulin expression during trans-
formation is a developmentally regulated process: promastigotes contain more tubulin mRNA and syn-
thesize more tubulin protein than amastigotes.
The a- and /3-tubulin genes of Leishmania enriettii are arranged as tandem repeats. The a gene
repeat contains approximately 15 copies of a 2 kilobase repeat unit. /3-tubulin genes are arranged in 4
kilobase repeating units. Given this array the tubulin genes could be transcribed from a single upstream
promoter to produce a very long (e.g., 30 kilobase) precursor that is processed down to the mature
message. Alternatively transcription could occur from promoters at the beginning of each repeat. The
4 kilobase /3-tubulin repeat unit also contains twice as much DNA as is needed to encode the mature
message. This extra sequence could either be a non-transcribed intergenic spacer or part of an initial
transcript that is cut out to form the mature message. The purpose of our experiments was to attempt
to detect precursors to mature a- or /3-tubulin mRNA.
RNA was obtained from promastigotes by both phenol and guanidium thiocyanate extraction.
Northern blot analysis was used to look for RNA precursors. Blots were probed with a- and /3-tubulin
clones and a fragment of the /3 clone containing sequence probably not present in the mature message.
Blots probed for a-tubulin RNA show several faint bands above the 2 kilobase message. Blots probed
for /3-tubulin RNA show three distinct bands other than the 2 kilobase message; one slightly above 2
kilobases, one slightly below, and one at about 4 kilobases. The two bands above and the one band below
2 kilobases also hybridize to the /3 fragment probe, but the 2 kilobase message itself does not hybridize
to this probe. These experiments provide preliminary evidence for longer length precursors for a- and
/3-tubulin mRNA, and suggest that at least part of the additional DNA in the /3-tubulin gene is initially
transcribed.
If the temperature of promastigote cultures is shifted from 27°C to 35°C the cells resorb their flagella
and begin to look like amastigotes. RNA was obtained from rapidly growing promastigotes and tem-
perature transformed pseudoamastigotes by guanidinium thiocyanate extraction. Northern blot analysis
shows that the promastigotes contain much more tubulin RNA than the pseudoamastigotes. Thus tem-
perature shifting may be a useful model for studying promastigote to amastigote transformation with
respect to tubulin gene expression.
PHOTORECEPTORS, VISION, AND RHYTHMS
Vision in Limulus mating behavior: tests for detection and form discrimination.
ROBERT B. BARLOW, JR., LEONARD KASS, VIVIAN MANCINI, AND JANICE L.
PELLETIER (Syracuse University).
Vision plays a role in Limulus mating behavior. During the mating season, these horseshoe crabs
move in from deep water, pair off, and build nests near the water's edge at high tide. Painted cement
castings of the female carapace and other forms placed in the nesting area attract males; the degree of
attraction depends in part on the visual contrast of the castings (Barlow et al. 1982, Nature 296: 65-66).
We tested the animal's ability to detect dark objects by observing male behavior in the vicinity of
a submefged black cement hemisphere (29 cm diameter). The hemisphere, which is about the size of an
adult female, was located 5 m below the high water line on the South side of Mashnee Dike, Cape Cod,
MA. Sixty-four percent of males (n = 285) moving within 1.2 m of the hemisphere oriented toward the
cement form and contacted it. No such behavior was observed for animals at distances greater than 1.2
m. It is interesting to note that at this distance the hemisphere can be seen by no more than four visual
receptors in the central portion of the lateral eye.
We also tested the animal's ability to discriminate among submerged objects of different form. Black
silhouettes of an adult female carapace (38 cm length), a hemisphere (29 cm diameter), and a square
( 1 8 cm/edge), all having equal surface area, were placed 1.1 m in front of the opening of a submerged
holding pen. The three silhouettes were located at 45°, 90°, and 135° with respect to the pen opening.
Seventy-seven percent of males (n = 349) leaving the pen approached and contacted one of the three
silhouettes without preference for form.
In sum, during the day and at night, male Limuli visually detected dark submerged objects at
distances of up to 1.2 m but did not discriminate small changes in the form of the objects.
540 ABSTRACTS FROM MBL GENERAL MEETINGS
We thank Heidi Howard, Maureen K. Powers, and George H. Renninger for their assistance. Sup-
ported in part by NIH grants EY-00667 and EY-05443 and NSF grant BNS 8104669.
Detection of membrane signals correlated with sensory excitation of phototactic
Halobacterium halobium. BARBARA E. EHRLICH, CATHY R. SCHEN, AND JOHN
L. SPUDICH (Albert Einstein College of Medicine, New York).
H. halobium, a bacterial species which lives in saturated brine, demonstrates both chemotactic and
phototactic behavior. While much is known about the molecular mechanism for sensory adaptation of
taxis, little is known about sensory excitation. The retinal-dependent phototaxis of halobacteria provides
a model system to look for signals related to sensory excitation. Wild type halobacteria have three known
retinal-containing pigments: bacteriorhodopsin (bR), halorhodopsin (hR), and s-rhodopsin (sR). The first
two hyperpolarize the cell membrane by electrogenic transport of H+ and CV respectively. The third
pigment, sR, may be a photosensory receptor because mutants lacking bR and hR retain phototactic
behavior. To examine the effects of photoexcitation on cells and membrane vesicles, we monitored light-
induced changes in fluorescence of the voltage-sensitive dye, diOC6(3). We were able to detect four types
of signals from cells and membrane vesicles. Red light-induced potential changes generated by bR were
seen only in wild type cells and were similar to signals described previously by Renthal and Lanyi (1976,
Biochem. 15: 2136). In cells lacking bR, signals generated by hR could be identified using four criteria:
wavelength dependence, Cl~ dependence, shunting by valinomycin and K+, and the absence of these
signals in hR-deficient mutants. In mutants lacking bR and hR, two additional signals were measured:
blue light caused a decrease and red light an increase in dye fluorescence. Both signals are retinal de-
pendent. These signals may represent localized potential changes (e.g., changes in surface charge due to
sR photocycling) rather than transmembrane potentials because the signal could not be shunted by
valinomycin and 1C. The behavioral response in cells and the fluorescent changes we detect in cells and
vesicles share two important characteristics: 1) the opposing effects of blue and red light and, 2) retinal
dependence. This correlation strongly suggests that these signals are generated during sensory excitation.
Supported by NY Heart Association and NIH GM 27750. B.E.E. is a NY Heart Association Young
Investigator.
Current clamp of photoreceptors and pacemaker neurons in eye o/ Bulla. JON W.
JACKLET (SUNY Albany, NY 12222).
In addition to photoreceptors, the eyes of certain gastropods such as Aplysia and Bulla contain
neurons that are circadian pacemakers. They are active, even in darkness, during the projected day, but
silent during the projected night, of a circadian cycle sending circadian information to central neurons
via optic nerve compound action potentials. A study of the Bulla eye (Jacklet and Colquhoun 1983, /
Neurocytology 12: 373-396) shows ca. 1000 large photoreceptors (30 x 100 /im) but only ca. 100 neurons
(15-25 Mm) tightly packed at the eye base. Both types have axons in the optic nerve. Gap junctions occur
between neuronal processes in the neuropil but not between juxtaposed somata. Current clamp with a
Dagan single electrode system shows membranes of large depolarizing photoreceptors have 30-80 ms
time constants and 50-100 mfi input resistances. Resistance decreases abruptly at ca. -35 mV indicative
of voltage dependent changes and decremented action potentials invading the soma of some cells. Brief
voltage and time dependent hyperpolarization follows release from hyperpolarizing pulses, indicative of
IA current. Initial voltage clamp shows a prominent IA current. Serotonin, known to phase shift the
circadian rhythm, hyperpolarizes photoreceptors and reduces their resistance, suggesting increased po-
tassium conductance. Neurons fire action potentials 1:1 with optic nerve compound action potentials.
Injected depolarizing current evokes decremented potentials and regenerative action potentials and also
optic nerve compound action potentials, suggesting neurons are electrically coupled. Neurons have several
time constants, a 12-20 ms time constant and a much longer one. Voltage and time dependent hyper-
polarization follows hyperpolarizing pulses. Thus both neurons and photoreceptors have time and voltage
dependent conductances including IA addition to expected light-evoked conductances.
Supported by NSF BNS 06245.
cAMP: a possible intracelhdar transmitter of circadian rhythms in Limulus photo-
receptors. LEONARD KASS, JANICE L. PELLETIER, GEORGE H. RENNINGER, AND
ROBERT B. BARLOW, JR. (Syracuse University).
At night a circadian clock in the Limulus brain transmits neural activity to the lateral eyes via
efferent optic nerve fibers (Barlow et al. 1977, Science 197: 86-89). The efferent input induces anatomical
and physiological changes that combine to increase retinal sensitivity (Barlow et al. 1980, Science 210:
PHOTORECEPTORS, VISION, RHYTHMS 541
1037-1039). Octopamine has been identified as a putative transmitter of the clock's action and exogenous
octopamine increases cAMP levels in the lateral eye (Battelle et al. 1982, Science 216: 1250-1252). In
this study, we investigated the possible role of cAMP as an intracellular transmitter.
Photoreceptor potentials were recorded from slices of retina maintained in an organ culture medium
(Bayer and Barlow 1978, J. Gen. Physiol. 72: 539-563). The intracellular records were characteristic of
those recorded in situ during the day in the absence of efferent input (Barlow and Kaplan 1977, J. Gen.
Physiol. 69: 203-220): they exhibit large spontaneous and light-evoked potential fluctuations (quantal
bumps up to 50 mV in amplitude), large resting potentials (~60 mV), and a plateau in the midrange
of the intensity-response function. Adding 8-bromo-cAMP (250 nM), a putative adenylate cyclase ac-
tivator (forscolin at 250 nM), or a potent octopamine agonist (naphazoline at 25 nM) to the bathing
medium induced physiological changes characteristic of those recorded in situ during the night (Kaplan
and Barlow 1980, Nature 286: 393-395): namely, spontaneous quantal bumps were reduced in frequency,
and the slope (gain) of the intensity-response function was increased. These were the most striking effects
but occasionally decreases in threshold were also observed.
In sum, our results are consistent with the following scheme: activity of a circadian clock in the
brain releases octopamine from terminals of efferent optic nerve fibers in the retina. The octopamine
increases cAMP levels in photoreceptors thereby changing their physiology, anatomy, and morphology.
Supported by NIH grants EY-00667 and EY-05443, NSF grant BNS 8104669, and a grant from
NSERC Canada.
Photoreceptors add at the anterior edge ofLimulus lateral eye. JENNIFER J. MARLER,
ROBERT B. BARLOW, JR., LESLIE EISELE, AND LEONARD KASS (Marine Bio-
logical Laboratory).
The lateral eye of Limulus polyphemus provides an interesting preparation for studying the devel-
opment of the visual system, by virtue of the continued addition of ommatidia to the eye during the
postembryonic growth of this animal. Given the recent description of a retinotopic map in Limulus
(Chamberlain and Barlow 1982, J. Neurophysiol. 48: 505-520), one may address the issue of how newly-
added units become organized in the optic nerves and make appropriate functional connections in the
optic ganglia. An important preliminary question, examined in this study, concerns where new ommatidia
are added to the eyes.
Sixth stage juvenile animals were collected, 2-3 retinal scars (each of which destroyed 5-10 om-
matidia) were made over their eyes in an array around the anterior edges, and the eyes were photographed.
The anterior margins of the eyes were chosen as scarring sites due to the observations that (a) the facet
diameters of anterior ommatidia are qualitatively smaller than more posterior ones in juveniles, (b) fault
lines exist in the hexagonal packing of ommatidia near the anterior edges, and (c) rows of small ommatidia
can occasionally be discerned beneath the carapaces of pre-molt juveniles. The animals were left to molt,
after which their eyes were rephotographed.
Comparison of pre- and post-molt eyes yielded the following results: (1) ommatidia are added, in
vertical strips, to the anterior edges of the growing lateral eyes of these juveniles, (2) the sizes of units
added show dorso/ventral differences (larger units are added ventrally), (3) diameters of existing om-
matidia increase during growth, and (4) the rate of ommatidial addition may vary between the two eyes
of a single individual. One consequence of this pattern of retinal growth is that the retinotopic organization
of the visual system changes as the animal grows. That is, the receptors seeing the anterior portion of
the visual field as juveniles will view the medial portion as adults.
Supported in part by NIH grants EY 00667 and EY 05443 and NSF grant BNS 8104669.
The effects of intracellular calcium/EGTA on the photoacti vation of Limulus ventral
photoreceptors. RICHARD PAYNE AND ALAN FEIN (Marine Biological
Laboratory).
Limulus ventral photoreceptors were impaled with two micropipettes, one containing 2.5 M KC1,
the other 0. 1 M K2EGTA [Ethyleneglycol-bis-(/3-aminoethyl ether) N,N,N',N' — tetraacetic acid] and suf-
ficient Ca(OH); to create buffered free calcium concentrations between 0.1 and 10 nM at pH 7.0. Ten
to 100 pi of the latter solution were pressure-injected into cells. Currents generated by 10 ms flashes were
recorded under voltage clamp at the resting potential. Each flash was estimated to produce approximately
200 discrete waves.
After injections of EOT A solutions containing 0. 1 \iM free calcium, the responses of 8 cells became
slower, with less abrupt rising edges, but the area under the responses was undiminished. After injection,
the responses could be modelled as the output of 7 cascaded, exponential stages of delay, having 6 time
constants, Ta, of 39 ± 11 ms and one, Tb, of 534 ± 128 ms. Injection of EGTA solutions containing
542 ABSTRACTS FROM MBL GENERAL MEETINGS
10 nM free calcium into 8 other cells resulted in a 100-fold reduction in response area. A 7-stage model
again described the time-course of the response, but with faster time constants, Ta = 10 ± 1 ms and Tb
= 89 ± 4 ms. Despite the reduction in peak amplitude, the average initial response after injection of 10
\iM calcium exceeded that after injection of 0.1 pM calcium. Calcium therefore increases both the rate
of production and the rate of decay of the photocurrent.
A considerable problem remains in relating the kinetics of the responses recorded before EGTA
injection to those recorded after. Dark-adapted responses recorded before injection of EGTA, or after
control injections of aspartate, exhibit too abrupt a rising edge to be modelled with 6-7 stages of delay.
One possible explanation is that EGTA buffers an early, local release of calcium which would normally
accelerate the initial generation of photocurrent.
Localization of calcium transients in the presynaptic terminals of a barnacle pho-
toreceptor detected using Arsenazo III. WILLIAM N. Ross AND N. STOCKBRIDGE
(New York Medical College, Valhalla, NY 10595).
The median photoreceptor of a giant barnacle, Balanus nubilus, with well separated cell body, axon,
and presynaptic terminal is a good preparation for studying calcium control of transmitter release at a
tonic synapse.
The supraesophageal ganglion, its connectives, and the median ocellus were dissected and mounted
on the stage of a Zeiss compound microscope. The preparation was imaged onto a 1 00-element photodiode
array with a 40X water immersion objective. With this lens, each element detected light from an area
of 25 X 25 p.m in the plane of the preparation. The calcium-binding dye, Arsenazo III, was iontophoresed
into the distal axon of 1 cm long photoreceptors about 100-200 nm from the end of the axon and allowed
to diffuse into the terminal arborization of the cell.
Absorbance changes were observed when the photoreceptor was depolarized electrically or by light.
These changes were consistent with those expected from calcium entering from outside the cell: increase
maximal at 660 nm, decrease maximal at 530 nm, and an isosbestic point at approximately 570 nm.
When calcium action potentials were elicited in 20-50 mM TEA, absorbance changes were much larger
than in normal saline. The absorbance changes were eliminated in saline in which cobalt replaced 90%
of the calcium.
Although dye was most clearly visible near the site of injection, absorbance changes were restricted
to the region of terminal arborization, about 50 ^m. Smaller signals recorded over the surrounding 50
^m area were consistent with scattering of light during passage through the tissue. The absorbance change
over the terminal was at least 50 times larger than that detected over the axon.
The afterhyperpolarization in the photoreceptor was TEA-insensitive and has a reversal potential
dependent on the extracellular potassium concentration (Edgington, unpub.). Its recovery time course
was well matched by the time course for calcium removal, suggesting that this hyperpolarization was due
to calcium-activated potassium channels. Since calcium enters only at the terminal, this conductance
must also be confined to the terminal region.
Continued from Cover Two
secutive Arabic numbers, on a separate sheet at the end of the paper. Footnotes should be limited to
authors' current addresses, acknowledgments or contribution numbers, and explanation of unusual ab-
breviations. All such footnotes should appear on the title page. Footnotes are not normally permitted
in the body of the text.
4. A condensed title or running head of no more than 35 letters and spaces should appear at the top
of the title page.
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 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, con-
forming 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 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 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 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 abbre-
viations invented by the author. For example, use Rit Visindafjelags Islendinga without abbreviation.
F. All single word 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 preferred forms rather than WORLD LIST or
USASI usage (e.g. Nature, Science, Evolution NOT Nature, Lond., Science, N.Y.; Evolution, Lancaster,
Pa.)
6. Reprints, charges. The Biological Bulletin has no page charges. However, authors will be re-
quested to help pay printing charges of manuscripts that are unusually costly due to length or numbers
of tables, figures, or formulae. Reprints may be ordered at time of publication and normally will be
delivered about two to three months after the issue date. Authors (or delegates or 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).
CONTENTS
AYLING, AVRIL L.
Growth and regeneration rates in thinly encrusting demospongiae from
temperate waters • • 343
BENAYAHU, Y., AND Y. LOYA
Surface brooding in the Red Sea soft coral Parerythropodium fulvum
fulvum (Forskal, 1775) 353
DUNCAN, THOMAS K.
Sexual dimorphism and reproductive behavior in Almyracuma proxi-
moculi (Crustacea: Cumacea): the effect of habitat 370
ECKELBARGER, KEVIN J., AND JUDITH P. GRASSLE
Ultrastructural differences in the eggs and ovarian follicle cells of Cap-
itella (Polychaeta) sibling species 379
EYSTER, LINDA S.
Ultrastructure of early embryonic shell formation in the opisthobranch
gastropod Aeolidia papillosa 394
FREEMAN, JOHN A., TERRY L. WEST, AND JOHN D. COSTLOW
Postlarval growth in juvenile Rhithropanopeus harrisii 409
KAPLAN, SAUL W.
Intrasexual aggression in Met r id i urn senile 416
MILLER, RICHARD L., AND KENNETH R. KING
Sperm chemotaxis in Oikopleura dioica Fol, 1872 (Urochordata: Lar-
vacea) 419
RAMOS-FLORES, TALIA
Lower marine fungus associated with black line disease in star corals
(Montastrea annularis, E. & S.) 429
SUGITA, HlROAKI, AND KOICHI SEKIGUCHI
The developmental appearance of paternal forms of lactate dehydro-
genase and malate dehydrogenase in hybrid horseshoe crabs 436
TSUJI, FREDERICK L, AND ELIZABETH HILL
Repetitive cycles of bioluminescence and spawning in the polychaete,
Odontosyllis phosphorea 444
VITTURI, R., A. MAIORCA, AND E. CATALANO
The karyology of Teredo utriculus (Gmelin) (Mollusca, Pelecypoda) 450
WEDI, STEVEN E., AND DAPHNE FAUTIN DUNN
Gametogenesis and reproductive periodicity of the subtidal sea ane-
mone Urticina lofotensis (Coelenterata: Actiniaria) in California . . . 458
Yui, MARY A., AND CHRISTOPHER J. BAYNE
Echinoderm immunology: bacterial clearance by the sea urchin Stron-
gylocentrotm purpuratus 473
ABSTRACTS OF PAPERS PRESENTED AT THE GENERAL SCIENTIFIC MEET-
INGS OF THE MARINE BIOLOGICAL LABORATORY
Cellular, molecular biology, etc. 487
Developmental biology 499
Ecology, evolution, plant sciences 504
Gametes and fertilization 512
Microbiology 520
Neurobiology, learning, behavior 523
Parasitology, pathology, aging 534
Photoreceptors, vision, rhythms 539
•
, ^ar^ -;*S Uoorston/ j
Volume 165 '. Number 3
-'AN 3 1983
BIOLOGICAL BULLETIN
PUBLISHED BY
THE MARINE BIOLOGICAL LABORATORY
Editorial Board
DANIEL L. ALKON, National Institutes of Health and MICHAEL G. O'RAND, Laboratories for Cell Biology,
Marine Biological Laboratory University of North Carolina at Chapel Hill
ROBERT B. BARLOW, JR., Syracuse University RALPH S. QUATRANO, Oregon State University at
Corvallis
WALLIS H. CLARK, JR., University of California at LlQNEL L R£BH University of Virginia
Davis
JOEL L. ROSENBAUM, Yale University
DAVID H. EVANS, University of Florida
DOROTHY M. SKINNER, Oak Ridge National
HARLYN O. HALVORSON, Brandeis University Laboratory
RONALD R. HOY, Cornell University . J°HN D- STRANDBERG, Johns Hopkins University
JOHN M. TEAL, Woods Hole Oceanographic
SAMUEL S. KOIDE, The Population Council, Institution
Rockefeller University
J. RICHARD WHITTAKER, Boston University
FRANK J. LONGO, University of Iowa Marine Program and Marine Biological Laboratory
CHARLOTTE P. MANGUM, The College of GEORGE M. WOODWELL, Ecosystems Center, Marine
William and Mary Biological Laboratory
Editor: CHARLES B. METZ, University of Miami
DECEMBER, 1983
Printed and Issued by
LANCASTER PRESS, Inc.
PRINCE & LEMON STS.
LANCASTER, PA.
THE BIOLOGICAL BULLETIN
THE BIOLOGICAL BULLETIN is published six times a year by the Marine Biological Laboratory, MBL
Street, Woods Hole, Massachusetts 02543.
Subscriptions and similar matter should be addressed to THE BIOLOGICAL BULLETIN, Marine Biological
Laboratory, Woods Hole, Massachusetts. Single numbers, $13.00. Subscription per volume (three issues),
$32.50 (this is $65.00 per year for six issues).
Communications relative to manuscripts should be sent to Dr. Charles B. Metz, Editor, or Pamela
Clapp, Assistant Editor, at the Marine Biological Laboratory, Woods Hole, Massachusetts 02543 between
May 1 and October 1, and at the Institute For Molecular and Cellular Evolution, University of Miami,
521 Anastasia, Coral Gables, Florida 33134 during the remainder of the year.
Copyright © 1983, by the Marine Biological Laboratory
Second-class postage paid at Woods Hole, Mass., 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). Very short
papers (less than 10 manuscript pages including tables, figures, and bibliography) will be published in a
separate section entitled "Short Reports." A limited number of solicited review papers may be accepted
after formal review. A paper will usually appear within four months after its acceptance.
The Editorial Board requests that manuscripts conform to the requirements set below; those manuscripts
which do not conform will be returned to authors for correction before review.
1 . Manuscripts. Manuscripts, including figures, should be submitted in triplicate. (Xerox copies of
photographs are not acceptable for review purposes.) The original manuscript must be typed in double
spacing (including figure legends, footnotes, bibliography, etc.) on one side of 16- or 20-lb. bond paper,
8'/2 by 1 1 inches. 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, 4th Edition (Council of Biology Editors,
1978) 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. Figures. Figures should be no larger than 8'/2 by 1 1 inches. The dimensions of the printed page,
5 by 73/4 inches, should be kept in mind in preparing figures for publication. We recommend that figures
be about I'/i 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 illustration 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. The author's name and an arrow indicating orientation should appear on the reverse side of all
figures.
3. 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 consecutive
Arabic numbers, on a separate sheet at the end of the paper. Footnotes should be limited to authors'
Continued on Cover Three
Reference: Biol. Bull. 165: 543-558. (December, 1983)
COMPETITIVE DISPLACEMENT OF NATIVE MUD SNAILS
BY INTRODUCED PERIWINKLES IN THE NEW
ENGLAND INTERTIDAL ZONE
G. A. BRENCHLEY1 AND J. T. CARLTON2
^Department of Ecology and Evolutionary Biology, University of California, Irvine, California 92717
and Marine Biological Laboratory, Woods Hole. Massachusetts 02543; 2 Department of Biology,
Woods Hole Oceanographic Institution, Woods Hole, Massachusetts 02543 and
* Williams College Program, Mystic Seaport, Connecticut 06355
ABSTRACT
During the nineteenth century the mud snail Ilyanassa obsoleta was abundant
on sand and mud flats, wood works, sea walls, salt marshes, eel grass beds, and cobble
beaches in New England. With the exception of sand and mud flats, these habitats
are now largely occupied by the introduced periwinkle, Littorina littorea. To determine
whether Littorina competitively displaces Ilyanassa, an experimental study was con-
ducted at a site in Barnstable Harbor, Massachusetts where the observed distributions
overlapped by 3% by Morisita's index.
Mark-recapture studies suggested that the distribution of Littorina was limited
by an abiotic factor, currents, through which this species realized its fundamental
niche. In contrast, density manipulations demonstrated that Ilyanassa emigrated from
areas where Littorina exceeded densities of 2 to 5 per 0.25 m2. Littorina limited the
upper and lower distribution of Ilyanassa and affected its microhabitat distribution
in the mid intertidal zone. Habitat displacement was 70% for Ilyanassa, calculated
as the difference between llyanasscfs observed distribution and its distribution during
littorinid removal experiments. The two species display reciprocal niche overlap with
each possessing an exclusive region from which the other is physically restricted. The
results suggest that the historical change in the distribution of Ilyanassa was due to
competitive exclusion by introduced Littorina.
INTRODUCTION
Introduced species are numerically dominant members of many marine and es-
tuarine communities in North America, having arrived as fouling organisms on ships
and with commercial oysters and by other means since the mid-eighteenth century
(Hanna, 1966; Carlton, 1979; Scheltema and Carlton, 1983). Introduced species have
been viewed traditionally from an economic perspective with most discussion oriented
towards their roles as pests and predators or for mariculture potential (e.g., Elton,
1958; Mann, 1979; Simberloff, 1981). Recently, studies on community structure in
marine systems have shown that invading species often competitively displace native
species (Farnham, 1980; Carlton, et al, 1982; Race, 1982). Displacement usually
involves resource partitioning whereby native species relinquish portions of their
habitats or microhabitats to introduced species. In most cases of introduced species
in marine systems, however, there are too few descriptions of the earlier community
to allow one to assess the degree of competitive displacerr jnt.
Received 25 April 1983; accepted 26 September 1983.
* Address for correspondence.
543
544 G. A. BRENCHLEY AND J. T. CARLTON
In this paper we report the results of experiments which demonstrate competitive
displacement of the mud snail Ilyanassa obsoleta (= Nassarius obsoletus), native to
New England, by the introduced periwinkle Littorina littorea. These species are the
most abundant, large intertidal gastropods of New England. Prior to the arrival of
Littorina, Ilyanassa was described by many naturalists (Say, 1822; Adams, 1839;
Gould, 1841; Stimpson, 1865; Perkins, 1869; Verrill and Smith, 1873; Rathbun,
1881). In a comprehensive survey of the Cape Cod region Verrill and Smith (1873)
ranked Ilyanassa "dominant" on marine and estuarine sand and mud flats, wood
works, sea walls, salt marshes, and eel grass beds; "common" on protected rocks,
cobble beaches, and pilings of wharves; and "present" in oyster beds. The habitat of
Ilyanassa has changed markedly since the arrival of Littorina although its geographical
range along the East Coast (Nova Scotia to northern Florida) has remained unchanged.
History of co-occurrence
Despite the presence of rare subfossil and fossil shells of Littorina littorea in
Newfoundland and Nova Scotia (Clarke, 1971; Wagner, 1977), the periwinkle was
absent from American shores south of Nova Scotia prior to 1860 (Morse, 1880;
Ganong, 1886; Kraeuter, 1976; Carlton, 1982) and was probably absent in modern
times from all of North America prior to about 1 800 (Carlton, in prep.). First recorded
in the Bay of Fundy in 1861, Littorina reached the north shore of Cape Cod by 1870.
It appeared south of Cape Cod at Woods Hole in 1875, in the New York region by
1879, and at Cape May, New Jersey by 1890 (Carlton, 1982; in prep.). To the south,
Littorina now occurs on isolated rock jetties in Delaware, Maryland, and Virginia,
but no populations are established south of about 38° north latitude. Littorina, although
usually associated with the rocky shore (e.g., Lubchenco, 1978; Carlton et ai, 1982),
is a predominant organism in marshes and eel grass beds that border soft bottoms,
and is also common on such "hard" substrates as wood, algae, rocks, and worm
tubes of soft bottom habitats (Rathbun, 1881; Pearse, 1914; Stauffer, 1937; Spooner
and Moore, 1940; Dexter, 1945, 1947; Bradley, 1957; Wharfe, 1977; Jiich and Boek-
schoten, 1980).
The historical progression of Littorina west and south along the Atlantic coast is
one of the best documented cases of the dispersal of a non-native marine species;
this, combined with the superb record-keeping of early naturalists, makes Ilyanassa
and Littorina an exceptional example in which the history of habitat overlap can be
reconstructed. North of Cape Cod, Littorina was reported to co-occur with Ilyanassa
on mudflats by Grabau in 1898 (near Boston, MA), on and among eel grass in 1912
by Pearse (1913, 1914; Nahant, MA), and in mud channels by Batchelder in 1915
(New Hampshire). Rathbun (1881), reporting upon observations made at Province-
town, MA in 1879, recorded Ilyanassa present on "the inner beaches, and extending
up to high tide level," and Littorina present "on the shore, on piles of wharves, and
... on the eel-grass in countless numbers," but did not specifically indicate direct
co-occurrence. South of Cape Cod, Balch reported Littorina to co-occur with Ilyanassa
along marsh edges and on mudflats in 1899 (Cold Spring Harbor, NY). Balch (1899),
noting the relatively recent arrival of Littorina on the New York shore, stated that
although it "does not appear as yet seriously to threaten Nassa obsoleta, the native
competitor for the mudflats," Ilyanassa was nonetheless "begin[ning] to yield room."
Recognizing differences in diet but without postulating a mechanism, Dimon (1905)
predicted that Littorina would displace Ilyanassa except "on the mud flats, from
which it is not likely to be crowded [out] by the newcomer." By 1923 Ilyanassa could
no longer be found in the Woods Hole region in two of the habitats, cobble and
COMPETITIVE DISPLACEMENT OF MUD SNAILS 545
wood pilings, in which it had been numerically dominant about 1871 (Verrill and
Smith, 1873; Allee, 1923). By 1930 Clench was able to report that, all along the
shores of bays and inlets of New England, Littorina "can be found everywhere between
the tide marks crawling over mud and on the blades and among the roots of Zostera."
Although juveniles are still found around marshes in New England (Brenchley, 1984b),
the adult Ilyanassa, as Dimon predicted, are generally confined to the soft sand and
mud flats (Burbanck, et al., 1956; Dippolito, et al, 1975), the remaining firmer
habitats generally being occupied by Littorina (Allee, 1923; Dexter, 1945; Burbanck,
et al., 1956; Dippolito, et al., 1975).
Life habit
The historical account strongly implies the displacement of Ilyanassa by Littorina
but the mechanisms of this displacement have not been previously known. Whether
competition for food exists among adult snails, although thought unlikely by Dimon
(1905) and by Dippolito et al. (1975), is not known. Littorina littorea is a facultative
omnivorous grazer, consuming both macroalgae and microalgae (Hylleberg and
Christensen, 1978; Lubchenco, 1978; Petraitis, 1983), invertebrate eggs (Brenchley,
1982), marsh detritus (Pourreau, 1979), barnacle cyprids in large numbers (Carlton,
pers. obs.), and a wide variety of other small encrusting or benthic animals (Blegvad,
1915; Hayes, 1929; Hylleberg and Christensen, 1978; Carlton, pers. obs.). Similarly,
Ilyanassa obsoleta's diet encompasses most of these types and other prey as well.
Ilyanassa has been described as a facultative herbivore/carnivore (Brown, 1969), as
an obligate omnivore (Curtis and Hurd, 1979), and as a grazer, deposit-feeder, and
detritivore (Connor and Edgar, 1982), ingesting sediment and a wide selection of
living and dead animal and plant material (Dimon, 1905; Gurin and Carr, 1971;
Atema and Burd, 1975; Haines and Montague, 1979; Abbott and Haderlie, 1980;
Curtis and Hurd, 1981; Connor and Edgar, 1982; Race, 1982; Brenchley, pers. obs.).
Dippolito et al. (1975) suggested that competition for space is also unlikely, Littorina
preferring solid substrates and Ilyanassa the softer substrata. During the reproductive
season, however, Ilyanassa move onto solid substrates to lay their egg capsules (Schel-
tema, 1962; Pechenik, 1978). Littorina occupy these substrates and graze attached
egg capsules (Brenchley, 1982). Laboratory studies demonstrate that during this process
Littorina physically interferes with Ilyanassa's egg laying behaviors (Brenchley, 1980,
1984a).
Alternatively, the change in Ilyanassa's distribution may be coincidentally rather
than causally related to Littorina's arrival. Chew (1981) has demonstrated that local
extinction and displacement of a native pierid butterfly in New England, believed to
be due to competitive exclusion by an introduced pierid, is actually the result of shifts
in land use and resultant changes in the flora. Alternative hypotheses that would
explain the observed shift in Ilyanassa's habitat utilization would therefore include
other possible physical or biological changes in the mud snail's environment in the
past century. We know of no physical (or chemical) change within Ilyanassa"?, former
or present habitat regime that could cause such shifts nor, in particular, any changes
that would affect Ilyanassa but no other species. Biologically, at least one other species
affecting Ilyanassa has also arrived recently in New England: the green crab Carcinus
maenas (reviewed by Vermeij, 1982a,b) which preys heav .y upon Ilyanassa's egg
capsules (Brenchley, 1982). Juvenile and adult Ilyanass. are prey for a variety of
species including birds (Recher, 1966), other snails (A/jma and Burd, 1975), crabs
(Stenzler and Atema, 1977; Brenchley, unpub. data) and sea stars (Peterson, 1979),
but the mud snail is believed to be a generally in .nor food item. Ilyanassa would
546
G. A. BRENCHLEY AND J. T. CARLTON
not be expected to respond dramatically to manipulations in the density of Littorina
if predation by other species or other factors were primarily responsible for the change
in Ilyanassa's distribution.
The present study was conducted on a sand flat located between a marsh and eel
grass bed, one of the few habitats where the two species still coexist. Through mark-
recapture experiments, factors controlling the distribution of Littorina were studied.
Density manipulations of Littorina were used to study its effect on distributions of
Ilyanassa. Elsewhere the behavioral components of the interactions (Brenchley, 1984a)
as well as juvenile distributions (Brenchley, 1984b) are detailed; here we focus on
patterns and factors controlling adult distributions.
MATERIALS AND METHODS
Patterns of distribution
The study was conducted in Barnstable Harbor, Massachusetts (41 °43'N, 70°20'W)
between June and September 1980, and between May and November 1981 on a
sandy intertidal flat between Indian Trail and Bone Hill Road. A census of adult
Ilyanassa and Littorina was conducted each month along three permanent transects
that extended through the intertidal zone from the marsh edge or high intertidal zone,
across a sand flat to an eel grass bed at the low intertidal zone, a distance of 1 50 to
250 m. Individuals on hard surfaces, on the sand, and buried 2 to 3 cm within the
sand were counted in 0.25 m2 quadrats (n = 4 to 8) every 5 to 10 m along the
transects. Additional transects 50 to 100 m in length and paralleling the edges of the
marsh and eel grass bed were also censused periodically.
Natural movements of individuals were studied by mark-recapture (Table I). Snails
were brought into the laboratory, kept in running sea water, marked, and returned
to the field within 3 days of collection. Ilyanassa shells were cleaned with a wire
brush and the apex was marked with a durable paint (Mark-Tex Corp., NJ). Each
shell was numbered with India ink. Littorina shells were marked in situ or in the
laboratory. During June 1981, marked snails were returned to their respective habitats:
Ilyanassa to the mid-intertidal sand flat, and Littorina to the eel grass bed, rocks on
TABLE I
Summary of research protocol, Barnstable Harbor. Massachusetts
Zones
Habitats
Mark-
recapture1
Manipulations
Fundamental2
niche
Realized
niche
Low
Eel grass bed
1 502 Littorina
None
August census
July census
Mid
Solid substrata
Sand
76 Littorina
781 Ilyanassa
Tide pools: littorine
removals, littorine
additions, controls
Littorine-removal
pools3
Littorine-control
pools3
High
Marsh
sediment
Marsh shoots
202 Littorina
Marsh edge: littorine
removal plots,
control plots
Littorine-removal
plots4
Littorine-control
plots5
1 Numbers of marked snails released in June 1981.
2 Calculated for Ilyanassa only.
3 Mean density on day 7.
4 Mean of peak density in each plot.
5 Mean of daily means in each plot.
COMPETITIVE DISPLACEMENT OF MUD SNAILS 547
the sand flat, and the marsh edge. Marked individuals were returned to the laboratory
for measurement every 6 weeks through November.
Experimental procedures
To determine their effect on the upper limit of Ilyanassa's distribution, all Littorina
were removed daily between 29 June and 7 July, 1980 from three replicate plots,
each 2 m long and extending 1.5 m into the marsh (Table I). Two unmanipulated
plots, each 1.5 m long and lying between test areas, served as control areas. Numbers
of both species were counted daily in 0.25 m2 quadrats in all areas; these were
approached from the marsh and comparably disturbed by the censuses. On the 4th
day of the experiment about 100 Ilyanassa were collected from both the marsh and
adjacent sand flat at low tide and measured (±0.1 mm).
Densities of Littorina were manipulated in the tide pools to determine their effect
on the distribution of Ilyanassa in the mid intertidal zone (see Table I). In June 1981
all littorines were collected from one pool (4 to 6 m2 area) and added to the center
of an adjacent pool that was similar in size and appearance. A third pool was left
undisturbed to serve as a control area. Densities of both species were counted prior
to these manipulations and also twice during the following week in 8 to 10 replicate
quadrats (each 0.25 m2) placed in the center of each pool. This experiment was
repeated three times in three weeks in different sets of tide pools. Treatments in 2
pools were reversed after one week by collecting all littorines in an addition-pool and
releasing them in a pool from which littorines previously had been removed (Ex-
periments Al, A2).
Censuses of snail distributions in the intertidal zone were used to calculate Morisita's
(1959) index of niche overlap. Despite modifications and alternatives to this original
index, it remains the least biased when sample sizes are small (Smith and Zaret, 1982)
and was appropriate for this study where five habitats were recognized (see Table I).
Because the eel grass lay limply at low tide, it was combined with sand into a single
habitat. Resource utilization for each species at each monthly census was calculated
from the mean density of individuals on each resource summed over the three tidal
zones (n > 16 quadrats per zone, see Table II).
Indices of niche overlap were similarly calculated to determine the extent of
habitat displacement (Table I). The "realized niche" (sensu Hutchinson, 1957) of
Ilyanassa and Littorina were derived from mean densities in experimental controls
and the July census of the eel grass bed when littorines were present. The "fundamental
niche" (sensu Hutchinson, 1957) of Ilyanassa was derived from densities resulting
in littorinid removal plots and the August census of the eel grass bed when littorines
were absent. Mark-recapture studies indicated that the fundamental niche was the
realized niche for Littorina; no additional calculations were made for this species.
RESULTS
Littorina distribution
Littorina littorea was most abundant in the upper intertidal zone in the marsh
at Barnstable Harbor (Table II) (as is typical of its distribution on soft sediments of
Europe and New England). Except during a period between ' ate July and September
its distribution extended through the mid intertidal zone, wKre it was locally abundant
on most firm substrates (wood, peat, pebbles, worm tu' js, algae), and into the low
intertidal eel grass bed where it crawled across sand a .d blades of grass at low tide.
Mark-recapture studies in the low intertidal zone demonstrated the transient
548 G. A. BRENCHLEY AND J. T. CARLTON
TABLE II
Mean densities per 0.25 m2 o/Littorina littorea on substrates in three intertidal zones in Barnstable
Harbor, Massachusetts, in monthly census in 1981
May June July Aug. Sept. Oct. Nov.
Upper Zone1
Marsh sediments
48.3
39.7
41.4
89.9
64.8
62.9
31.1
Marsh shoots
21.7
16.3
20.9
30.4
69.2
33.7
8.0
N
32
32
64
64
16
16
32
Mid Zone
Sand2
0.9
0.5
0.2
<0.1
0.4
0.1
0.7
Firm objects
3.8
2.3
2.1
0.8
2.3
3.6
3.9
N
32
64
64
64
60
16
16
Low Zone
Eel grass bed
9.8
9.4
9.4
0.1
0.1
5.3
6.7
N
32
16
16
16
32
16
32
"Sediment" includes bases of stalks; "shoots" refers to snails on blades of cord grass.
2 Combined areas with and without Ilyanassa; differences not significant (P > 0.05) by one-way
Analysis of Variance on pooled monthly data.
N, numbers of quadrats counted at low tide.
nature o/Littorina. Littorina were dislodged from "softer" surfaces (e.g., worm tubes,
sand, filamentous and "spongy" algae, eel grass) and were often seen rolling along
the bottom during incoming tides. These individuals crawled along the sand and
often followed mucous trails of conspecifics until a solid object was encountered. Of
1502 marked snails released in the eel grass bed in June 1981, only 40 were recovered
there after 1 week and only 3 were recovered after 1 month, all along the marsh edge.
Several lines of evidence indicated that widespread transport rather than mass mortality
was responsible for this low recovery rate. We also inspected thousands of empty
shells in the eel grass and marsh without finding any marked shells; most had been
bored by naticid gastropods.
Transport in the mid intertidal zone was documented in August 1980 when
approximately 100 unmarked periwinkles were released on each of four occasions in
sandy areas where solid substrates were rare and other littorines absent. After 3-4
days on each occasion less than 8 snails were found within a 20 X 20 m area (and
these were found on marking-stakes). This dispersion was a result of transport by
currents and not active movement, since littorines move only about 60 cm per day
on rocky shores (Dexter, 1943) and 20 to 25 m during the autumn on soft substrata
(Batchelder, 1915).
In higher intertidal areas where mucous dried during low tide, Littorina clung to
rocks and marsh grasses. A few individuals marked on rocks and in the marsh in
June were still present after 3 months. However, these individuals were also transient
as evidenced by rates of colonization. All periwinkles were removed from two rocks
and from a log in the mid intertidal zone every 3 to 4 days for a period of 4 weeks
in June 1981. Recolonization after 3-4 days ranged from 0 to 14 individuals per ca.
0. 10 m2, the unmanipulated density, with no change in numbers through time. Daily
colonization rates along the marsh edge were obtained from censuses of the littorinid
removal experiment, and ranged from 0 to 109 individuals per 0.25 m2 per day. In
this case colonization rates decreased steadily over the course of the experiment,
indicating local rather than widespread transport of individuals.
COMPETITIVE DISPLACEMENT OF MUD SNAILS
549
Ilyanassa distribution
The population of Ilyanassa obsoleta in the study area was estimated to contain
millions of snails. The population was dominated by adult-sized individuals (Fig. 1 A;
see Scheltema, 1964) with sparse recruitment in 1979, 1980, and 1981. Beginning in
May, as the water warmed, they moved onto solid objects of the mid intertidal zone
(Table III) to lay egg capsules, preferring isolated eel grass plants, drift algae, and
small islands of marsh peat and avoiding both the eel grass bed and marsh where
periwinkles were numerous.
The adult population moved about the 1 km stretch of shoreline of the study site
in the mid intertidal zone from March to November. Although Jenner (1956) reports
that mud snails in the Harbor aggregate after reproduction ceases, the study population
remained in dense aggregations throughout the year, foraging upon their own shells
(illustrated by Morse, 1921) and on each other's shell epiflora. Isolated individuals
were always rare although individuals moved freely between the aggregations (see
also Borowsky, 1979) of which there were usually two or three. General patterns of
movement were directed toward the marsh during spring tides from April through
July, and toward the eel grass during spring tides occurring in the summer and fall.
In a nursery area near the marsh there was a small group (2-5 thousand individuals)
comprised of fast growing, immature snails (<17 mm) which separated from the
adults in June and roamed about near the marsh until late August or September
when they rejoined the adult aggregations.
During winter months the population of Ilyanassa hibernated 5 cm in the sediment
in the mid intertidal zone. With littorines present in the low intertidal eel grass bed,
the mud snails did not migrate to the subtidal zone as has been reported for populations
in other areas (e.g., Batchelder, 1915; Sindermann, 1960; Scheltema, 1964; Stambaugh
and McDermott, 1969; Murphy, 1979).
Recapture rates of marked Ilyanassa were relatively high: of 781 snails released
in June 1981, 419 (53.6%) were recovered in September and 109 (13.9%) in August
T
5 10 15 20
shell length (mm)
25
FIGURE 1. Size frequency distributions of Ilyanassa obsoleta on t1 2 study site in Barnstable Harbor,
Massachusetts. A: Sizes of individuals in roving aggregations on the and flat in July 1981 (n = 500). B:
Sizes of individuals migrating into the marsh in littorinid remova' plots (thin lines) and on the adjacent
sand flat (heavy lines) on 3 July 1981 (n = 100 each).
550 G. A. BRENCHLEY AND J. T. CARLTON
TABLE III
Mean densities per 0.25 m2 of adult Ilyanassa obsoleta in three intertidal zones in Barnstable Harbor,
Massachusetts in 1981
May June July Aug. Sept. Oct. Nov.
Upper Zone1
Marsh sediments <1 <1 <1 <1 <1 0 0
Mid Zone
Sand flat2 89.3 81.6 93.5 61.1 82.6 67.6 81.8
Firm objects3 1.7 2.6 <1 <1 <1 <1 <1
»
Low Zone
Eel grass beds 0 <1 <1 12.0 1.1 0 0
1 Densities on marsh shoots always zero.
2 Densities in areas where aggregations of mud snails were present.
3 Excluding Ilyanassa shells.
Numbers of quadrats as in Table I.
1982. Of 200 new snails released in November 1981, 99 (49.5%) were recovered the
following August. Observations on recovered snails indicated that the decline in the
return rate was largely due to the mark, lost by snail grazing and overgrown by a
thick diatom layer; empty shells were always rare during the summers of this study.
Evidence of displacement of Ilyanassa
The roving groups of adult Ilyanassa rarely entered either the marsh or eel grass
bed during the summer migration (see Table III). As evidenced by density relationships,
the boundaries between mud snails and littorines were extremely abrupt and rarely
did the two species co-occur within a 0.25 m2 quadrat (Fig. 2).
Within 24 to 48 h after the removal of Littorina from the marsh edge, upwardly
migrating Ilyanassa moved into test areas of the marsh but they did not enter the
control areas (Fig. 3). Maximum densities of Ilvanassa were recorded 4 days (plot
"E"), 5 days ("A") and 6 days ("C") after the initial removal of littorines. This
experiment was performed during Ilyanassa's first advance toward the marsh in 1980
and prior to the separation of the immatures and adults. Samples collected on the
4th day showed that all Ilyanassa in the marsh were of adult size (Fig. 1 B). Adults
moved from the edge (lower 0.5 m band) into the marsh (upper 0.5 m band) (Fig.
3). As the neap tide approached, both adult and immature Ilyanassa retreated from
the general vicinity of the marsh. Observations made underwater at later dates revealed
that adult Ilyanassa occurred along the bases of marsh shoots but never up on the
grass blades as do immature Ilyanassa (Dimon, 1905; Brenchley, 1984b).
Use of microhabitats differed where the snails occurred in the mid intertidal zone:
Littorina were more common on solid objects (Table II) and Ilyanassa on sand (Table
HI). Manipulative experiments demonstrated that Littorina had three density-related
impacts on the microhabitat distribution of reproductively active Ilyanassa within
the mid intertidal zone (Table IV). (1) Densities of mud snails did not change following
the initial removal of littorines from tide pools, but (2) microhabitat distribution
changed within 3 days, and by the seventh day significantly more natives were found
on solid objects than in either the pre-manipulated, unmanipulated, or littorinid-
addition pools. (3) When numbers of periwinkles were doubled, densities of mud
snails had decreased by the third day and were significantly smaller than in control
COMPETITIVE DISPLACEMENT OF MUD SNAILS
551
CM
E
10
CM
d
V.
0
O
O
100 n
0
c
2,
50-
o
o
o
o
o
o
mm
Jk~* ..
B
GO O
0 50 100
Littorina littorea /0.25m2
FIGURE 2. Densities of adult Ilyanassa as a function of Littorina densities in 0.25 m2 quadrats in
Barnstable Harbor, Massachusetts, June-August 1981: A, upper intertidal marsh edge; B, lower intertidal
eel grass bed. Each point represents one quadrat.
pools by day 7. When experimental conditions were reversed in Experiment A2,
Ilyanassa immigrated into the pool from which they had previously emigrated, and
migrated from the new littorinid addition pool (Table IV). This result demonstrates
unequivocally that adult Ilyanassa avoided Littorina. Densities in unmanipulated
pools did not change significantly during any experiment.
The overlap in observed distributions of Ilyanassa and Littorina over the course
of the summer ranged from 2 to 5% and averaged 3% (Table V). The experimental
studies indicate that, had Ilyanassa not avoided Littorina, their distributions would
have overlapped 71%. The value is not 100% for three reasons: (1) in the absence of
Littorina, adult Ilyanassa did not move onto shoots in the marsh as did Littorina
(Fig. 3); (2) only a minority of mud snails on the sand flat moved onto solid substrates
to reproduce in the absence of Littorina (Table IV); and (3) densities of Littorina on
sand never matched those of Ilyanassa (Tables II and III).
DISCUSSION
Results of experimental manipulations of Littorina littorea demonstrate density
relationships that generally explain observed distributional patterns of Ilyanassa ob-
soleta in the study site. The results show the emigration of Ilyanassa from mid
intertidal areas when densities of littorines are manipulated (i.e., doubled) to match
the conditions found in the marsh and during most of the year in the eel grass beds.
Following manipulations to remove littorines from the marsh, Ilyanassa expands its
distribution upshore; a similar result occurs on cobble b .ches in Rhode Island (M.
Bertness, pers. comm.) In the one local marsh (Saler , MA) found to contain no
Littorina, Ilyanassa extended throughout to a retainir - wall. The results further show
that at densities below about 5 individuals per 0.25 rrr, Littorina alters the microhabitat
552
G. A. BRENCHLEY AND J. T. CARLTON
40
30
<M
E
20
m
CJ
d
: — ,
10
o
•»-
-2
0
"o
V)
•Q
30
O
0
tn
20
V)
0
c
o
10
^
0
removals
controls ' B D
01 2345678
5
0
30
20
10
DAY
01 2345678
7/7
FIGURE 3. Movement of Ilyanassa into the marsh in Littorina-remova] plots (A, C, E) and control
plots (B, D) in Barnstable Harbor, Massachusetts in 1980. Position of plots illustrated in upper right corner.
Numbers of Ilyanassa per 0.25 m2 are shown for two 0.5 m wide bands: open figure, lower edge; closed
figure, marsh side. Values are means of 4 (removals) or 5 (control) counts per day per area.
distribution of Ilyanassa. By excluding the indigenous species from firm substrata,
Littorina significantly affects the reproductive activity of Ilyanassa (Brenchley, 1981,
1984a). In the tide pool experiments, for example, Ilyanassa laid significantly more
egg capsules in the littorinid removal pools than in littorinid addition pools (Brenchley,
1981, 1984a).
Brenchley (1982) finds that Littorina littorea is a major predator on egg capsules
of Ilyanassa obsoleta in this harbor. Race (1982) documents a similar interaction in
San Francisco Bay between a native mud snail (Cerithidea californicd) and Ilyanassa
obsoleta, introduced about 1905 (Carlton, 1979). In both cases reproducing individuals
are more likely to contribute to the next generation if they avoid habitats occupied
by egg predators. If avoidance behaviors of this kind are genetic then only among
Cerithidea with nonplanktonic larvae are such traits inheritable within local popu-
lations. The larvae of Ilyanassa, by contrast, have broad dispersal capability (Scheltema,
1962; Gooch et al, 1972). Mechanisms responsible for generating the patterns observed
in this study are not likely to have a genetic basis.
Studies of conditioning in Aplysia californica by Carew et al. (1983) and others
have shown that snails can learn to discriminate between tactile stimuli even in a
single trial, and can demonstrate the response after several hours. Littorina provides
tactile stimuli by grazing on shell epiflora of Ilyanassa, a behavior which interferes
with foraging, locomotory, and reproductive activities of the native species (Brenchley,
1980, 1984a, in prep.). When either L. littorea or a native littorinid species (L.
saxatilis) is on its shell, Ilyanassa responds by twisting, a behavior which probably
is inherited (see McKillup, 1983, for the polytypic species Nassarius pauperatus) since
lead weights also elicit the response. Although twisting seldom removes the littorine
on the shell, it provides Ilyanassa with the opportunity to learn the littorinid scent,
or to reinforce prior learning given that the mud snail lives 8 or more years (Jenner
COMPETITIVE DISPLACEMENT OF MUD SNAILS
553
TABLE IV
Densities o/Tlyanassa obsoleta per 0.25 m2 (mean ± one standard deviation) in tide pools at Barnstable
Harbor, Massachusetts, having Littorina littorea removed, added, or unchanged in June 1981,
compared by one-way Analysis of Variance
Experiment:
Al
(Week 1)
A2
(Week 2)
B
(Week 2)
C
(Week 3)
Littorina removals
Overall density
Initial
Final
F,,4
On solid substrata
Initial
Final
26.8 ± 14.9
28.2 ± 29.8
0.014
1.2
3.8
16.21***
2.4 ±
31.2 ±
57.83***
4.1
9.2
3.1
48.22***
42.8 ± 29.0
35.7 ± 24.6
0.245
0.9
2.9
11.51**
43.0 ± 22.3
27.0 ± 19.0
3.091
1.0
3.5
29.98***
Littorina addition
Overall density
Initial
Final
23.7 ± 13.2
2.4 ± 4.1
16.66***
28.2 ± 29.8
7.1 ± 8.0
3.28
48.5 ± 32.4
5.9 ± 3.9
11.93**
92.4 ± 27.9
8.8 ± 7.4
58.39***
Unmanipulated
controls
Overall density
Initial
Final
39.5 ± 19.6
43.4 ± 20.4
2.911
see B
40.2 ± 20.9
45.6 ± 27.6
1.433
76.6 ± 31.5
83.0 ± 36.6
2.614
Each n = 8. Conditions of experimental pools in Al were reversed in A2.
** P < 0.005; *** P < 0.001.
and Jenner, 1977). Adult Ilyanassa responds to chemical cues, to carrion for example,
by extending its proboscis (e.g., Carr, 1967; Brown, 1969). This behavior is observed
when adult Ilyanassa responds to the littorinid species: after attempting to shake off
littorines by twisting, adult Ilyanassa will attack the littorine's foot with its proboscis
and radula (Brenchley, in prep.). Following sensitization to lead weights, a higher
proportion of immature mud snails show this behavior — evidence of a learned response;
but adults are slower to probe with their proboscis — evidence that the behavior can
be reconditioned.
TABLE V
Index of microhabitat overlap (Morisita, 1959) for Littorina littorea and Ilyanassa obsoleta at
Barnstable Harbor, Massachusetts, during the summer of 1981
Niche
Method
Index
Realized
Realized
Fundamental
Seasonal'
Experimental Controls2
Experimental Treatments2
0.031
0.021
0.711
1 Mean of six monthly censuses.
2 See Table I.
554 G. A. BRENCHLEY AND J. T. CARLTON
The proximity of the two species in space depends on the frequency with which
Ilyanassa encounter Littorina on their shells. This frequency depends on two main
factors: mobility of littorines, which is largely a function of substrata; and the epiflora
on the mud snail shells, which varies both with habitat and snail age. In habitats
where littorines are mobile, such as exists throughout the main study site, the critical
density is between 1 2 and 20 littorines per square meter; at higher densities encounters
are too frequent and mud snails emigrate. The critical density can be higher on cobble
beaches, for example, where Littorina are generally less mobile. Types of mud snail
shell epiflora correlate with habitat. In sand habitats the epiflora is thickest (up to 2
mm) with strands of Enteromorpha common in some regions of the harbor. Thus
in muddy habitats Ilyanassa can be found in the immediate vicinity of Littorina on
rocks; this situation rarely occurs in sandier habitats because Littorina will move
from rock edges to shells of passing mud snails to graze on shell epiflora. We have
observed the complete removal of Enteromorpha by Littorina from shells of a large
aggregation of mud snails during a 48 h migration across a rocky area. Finally,
Littorina rarely grazes on the shells of immature Ilyanassa in muddy or sandy regions
of the harbor (Brenchley, 1984b). In comparison to adults, juveniles of Ilyanassa
and Littorina have similar foraging behaviors, grazing on microflora on sand and
marsh plants. Immature Ilyanassa show no evidence of avoiding Littorina even when
resources are limiting. The change from exploitation to interference competition as
the snails age, which coincides with a change from inclusive to reciprocal niche overlap
(see Cowell and Fuentes, 1975), is a result of behavioral interactions discussed above.
Unlike the effects of pests and predators (Elton, 1958; Simberloff, 1981), com-
petition leading to displacement or niche partitioning in most cases is subtle. Com-
petition between Ilyanassa and Littorina, for example, becomes evident through
density manipulations but not by comparing the two species' habitat preferences,
physiological tolerances, or patterns of distribution (Dippolito et ai, 1975). We can
estimate from previous faunal descriptions the extent of habitat displacement of
Ilyanassa by Littorina, although we cannot as yet determine the extent to which the
abundance of Ilyanassa has been affected.
Our studies confirm Dimon's ( 1 905) prediction that the "struggle between [Littorina
and Ilyanassa] may result in a modification of the range" of Ilyanassa. Since Dimon's
observations in 1905, noting that Ilyanassa "act as scavengers for the coast," numerous
workers have attempted to elucidate, by observation and experimentation, the precise
nature of Ilyanassa's role in the economy of soft sediment shores (Grant, 1965; Mills,
1967; Sibert, 1968; Nichols and Robertson, 1979; Pace et ai, 1979; Hunt, 1981;
Connor and Edgar, 1982; Connor et ai, 1982; Levinton and Stewart, 1982). These
studies have demonstrated that Ilyanassa exerts significant effects upon community
structure, indirectly modifying resources (trophic, spatial, temporal, or otherwise)
required by other species, and directly by consuming or displacing potential members
of the community. Curtis and Hurd (1981) have speculated in particular upon the
full suite of potential impacts by Ilyanassa on community structure. We extend Curtis
and Hurd's rationale here. We suggest that Littorina^ displacement of Ilyanassa,
while significant to the mud snail, is secondary to the effects that this displacement
has had on the benthic community, released from mud snail perturbations. We suggest
that there have been major indirect effects in habitats where Ilyanassa has been
displaced by Littorina, whose sediment disturbance (=bioturbation) abilities are minor
compared to those of mud snails (Brenchley, pers. obs.); we predict that in habitats
from which Ilyanassa has been displaced the faunal and floral communities will be
similar to experimental manipulative studies that have removed Ilyanassa artificially.
The introduction of an exotic species has altered community structure not simply
COMPETITIVE DISPLACEMENT OF MUD SNAILS 555
by the modification of distributional patterns of a native species but more profoundly
by secondarily modifying the community interactions of the native species as well.
As one example, our predictions suggest that polychaete population explosions in
mud flats, whose rarity was linked by Levinton and Stewart (1982) to persistent snail
populations, would be more common in areas where Ilyanassa^ local distribution
has been contracted by competition. Furthermore, suspecting that the contraction in
Ilyanassd's distribution due to Littorina is associated with the aggregatory behavior
of Ilyanassa in our study site (Brenchley, 1980), we predict that the establishment
and subsequent destruction of dense beds of tubicolous amphipods by aggregations
of mud snails (Mills, 1967) will coincide with the abundance of Littorina in New
England. We conclude that the interpretation of factors controlling the structure of
many modern-day marine communities in North America must consider the dynamic
interactions of both ecological processes and historical impacts of introduced species.
ACKNOWLEDGMENTS
This research began while we were supported as Postdoctoral Scholars at the
Woods Hole Oceanographic Institution. Subsequent research was funded (for G.A.B.)
by Marine Biological Laboratory STEPS fellowships and faculty grants from the
University of California at Irvine and (for J.T.C.) by a National Science Foundation
National Needs Postdoctoral Fellowship. We thank Merryl Alber for field assistance
and Gayle Gray for preparing the manuscript. Lynn Carpenter, Mark Hixon, Sarah
Woodin, and an anonymous reviewer provided many useful comments on the manu-
script.
LITERATURE CITED
ABBOTT, D. P., AND E. C. HADERLIE. 1980. Prosobranchia: marine snails. Pp. 230-307 in Intertidal
Invertebrates of California. R. H. Morris, D. P. Abbott and E. C. Haderlie, eds. Stanford University
Press, Stanford.
ADAMS, C. B. 1839. Observations on some species of the marine shells of Massachusetts, with descriptions
of five new species. Boston J. Nat. Hist. 2: 262-288.
ALLEE, W. C. 1923. Studies in marine ecology. I. The distribution of common littoral invertebrates of the
Woods Hole region. Biol. Bull. 44: 167-191.
ATEMA, J., AND G. D. BURD. A field study of chemotactic responses of the marine mud snail, Nassarius
obsoleliis. J. Chem. Ecol. 1: 243-251.
BALCH, F. N. 1899. List of marine Mollusca of Coldspring Harbor, Long Island, with descriptions of one
new genus and two new species of nudibranchs. Proc. Boston Soc. Nat. His. 29: 133-162.
BATCHELDER, C. H. 1915. Migration of Ilvanassa obsoleta, Litorina litorea and Litorina rudis. Nautilus
29: 43-46.
BLEGVAD, H. 1915. Food and conditions of nourishment among the communities of invertebrate animals
found on or in the sea bottom in Danish waters. Rep. Danish Biol. Sta. 22: 41-78.
BOROWSKY, B. 1979. The nature of aggregations of Nassarius obsoletus in the intertidal zone before the
fall offshore migration. Malacologia 12: 89-90.
BRADLEY, W. H. 1957. Physical and ecologic features of the Sagadahoc Bay tidal flat, Georgetown, Maine.
Mem. Geol. Soc. Am. 67(Part 2): 641-682.
BRENCHLEY, G. A. 1980. Distribution and migratory behavior of Ilvanassa obsoleta in Barnstable Harbor.
Biol. Bull. 159: 456-457.
BRENCHLEY, G. A. 1981. Limiting resources and the limits to reproduction in the "mud" snail Ilyanassa
obsoleta in Barnstable Harbor, Massachusetts. Biol. Bull. 161: 323.
BRENCHLEY, G. A. 1982. Predation on encapsulated larvae by adults: effects of introduced species on the
gastropod Ilyanassa obsoleta. Mar. Ecol. Prog. Ser. 9: 255-262
BRENCHLEY, G. A. 1984a. Substrate selection for reproduction in Ilyan ssa obsoleta: avoidance of objects
used by introduced Littorina littorea. (In review.)
BRENCHLEY, G. A. 1984b. Competition between native and intr juced marine snails: juveniles exploit
resources. (In review.)
556 G. A. BRENCHLEY AND J. T. CARLTON
BROWN, S. C. 1969. The structure and function of the digestive system of the mud snail, Nassarius obsoletus
(Say). Malacologia 9: 447-500.
BURBANCK, W. D., M. E. PIERCE, AND G. C. WHITELEY, JR. 1956. A study of bottom fauna of Rand's
Harbor, Massachusetts: an application of the ecotone concept. Ecol. Monogr. 26: 213-243.
CAREW, T. J., R. D. HAWKINS, AND E. R. KANDEL. 1983. Differential classical conditioning of a defensive
withdrawal reflex in Aplysia californica. Science 219: 397-400.
CARLTON, J. T. 1979. History, biogeography, and ecology of the introduced marine and estuarine invertebrates
of the Pacific coast of North America. Ph.D. Dissertation, Univ. California, Davis.
CARLTON, J. T. 1982. The historical biogeography of Littorina littorea on the Atlantic coast of North
America, and implications for the interpretation of the structure of New England intertidal com-
munities. Malacol. Rev. 15: 146.
CARLTON, J. T., D. P. CHENEY, AND G. J. VERMEIJ, eds. 1982. Abstracts of a Symposium on the ecological
effects and biogeography of an introduced species: the periwinkle Littorina littorea. Malacol. Rev.
15: 143-150.
CARR, W. E. S. 1967. Chemoreception in the mud snail, Nassarius obsoletus. I. Properties of stimulatory
substances extracted from shrimp. Biol. Bull. 133: 90-105.
CHEW, F. S. 1981. Coexistence and local extinction in two pierid butterflies. Am. Nat. 118: 655-672.
CLARICE, A. H. 1971. Littorina littorea, native or introduced? The Biologist 53: 160-162.
CLENCH, W. J. 1930. Littorina littorea Linn. Nautilus 43: 105.
CONNOR, M. S., AND R. K. EDGAR. 1982. Selective grazing by the mud snail Ilvanassa obsoleta. Oecologia
(Berl.)53: 271-275.
CONNOR, M. S., J. M. TEAL, AND I. VALIELA. 1982. The effect of feeding by mud snails, Ilvanassa obsoleta
(Say), on the structure and metabolism of a laboratory benthic algal community. J. Exp. Mar.
Biol Ecol. 65: 29-45.
COWELL, R. K., AND E. R. FuENTES. 1975. Experimental studies of the niche. Ann. Rev. Ecol. Syst. 6:
281-310.
CURTIS, L. A., AND L. E. HURD. 1979. On the broad nutritional requirements of the mud snail Ilvanassa
(Nassarius) obsoleta (Say), and its polytypic role in the food web. J. Exp. Mar. Biol. Ecol. 41:
289-297.
CURTIS, L. A., AND L. E. HURD. 1981. Nutrient procurement strategy of a deposit-feeding estuarine
neogastropod, Ilvanassa obsoleta. Estuarine Coastal Shelf Sci. 13: 277-285.
DEXTER, R. W. 1943. Observations on the local movements of Littorina litorea (L) and Thais lapillus (L).
Nautilus 57: 6-8.
DEXTER, R. W. 1945. Zonation of the intertidal marine mollusks at Cape Ann, Massachusetts. Nautilus
58: 135-142.
DEXTER, R. W. 1947. The marine communities of a tidal inlet at Cape Ann, Massachusetts: a study in
bio-ecology. Ecol. Monogr. 17: 261-294.
DIMON, A. C. 1905. The mud snail: Nassa obsoleta. Cold Spring Harbor Monogr. 5: 50 pp.
DIPPOLITO, A. F., J. SlENKO, AND B. S. SMITH. 1975. Relationship between mud snail and periwinkle.
Proc. Malacol. Soc. Lond. 41: 529-532.
ELTON, C. S. 1958. The Ecology of Invasions by Plants and Animals. Methuen Co., London. 181 pp.
FARNHAM, W. F. 1980. Studies on aliens in the marine flora of southern England. Pp. 875-914 in The
Shore Environment, vol. 2, J. H. Price, D. E. G. Irvine and W. F. Farnham, eds. Academic Press,
New York.
GANONG, W. F. 1886. Is Littorina littora introduced or indigenous? Am. Nat. 20: 931-940.
GOOCH, J. L., B. S. SMITH, AND D. KNUPP. 1972. Regional survey of gene frequencies in the mud snail,
Nassarius obsoletus. Biol. Bull. 142: 36-48.
GOULD, A. A. 1841. Report on the Invertebrata of Massachusetts comprising the Mollusca, Crustacea,
Annelida, and Radiata. Folstom, Wells and Thurston, Cambridge.
GRABAU, A. W. 1898. Zoology: marine invertebrates. Pp. 67-96 ip-Guide to Localities Illustrating the
Geology, Marine Zoology, and Botany of the Vicinity of Boston, A. W. Grabau, and J. E. Woodman,
eds. Am. Assoc. Adv. Sci., 50th Anniv. Meeting, Boston, August 1898. vi + 100 pp.
GRANT, D. C. 1965. Specific diversity in the infauna of an intertidal sand community. Ph.D. dissertation,
Yale University, New Haven.
GURIN, S., AND W. E. CARR. 1971. Chemoreception in Nassarius obsoletus: the role of specific stimulatory
proteins. Science 174: 293-295.
HAINES, E. B., AND C. L. MONTAGUE. 1979. Food sources of estuarine invertebrates analyzed using I3C/
12C ratios. Ecology 60: 48-56.
HANNA, G. D. 1966. Introduced mollusks of western North America. Occas. Pap. Calif. Acad. Sci. 48:
108 pp.
HAYES, F. R. 1929. Contributions to the study of marine gastropods. III. Development, growth and behaviour
of Littorina. Contrib. Can. Biol. 4: 413-430.
COMPETITIVE DISPLACEMENT OF MUD SNAILS 557
HUNT, J. H. 1981. The importance of adult-larval interactions in determining abundance patterns of soft-
sediment infauna. M.S. thesis, University of North Carolina, Chapel Hill.
HUTCHINSON, G. E. 1957. Concluding remarks. Cold Spring Harbor Symp. Quant. Biol. 22: 415-427.
HYLLEBERG, J., AND J. T. CHRISTENSEN. 1978. Factors affecting the intraspecific competition and size
distribution of the periwinkle Littorina littorea (L.). Natura Jutlandica 20: 193-202.
JENNER, C. E. 1956. A striking behavioral change leading to the formation of extensive aggregations in a
population of Nassarius obsoletus. Biol. Bull. Ill: 291.
JENNER, C. E., AND M. G. JENNER. 1977. Experimental displacement of an Ilyanassa obsoleta (Mollusca:
Gastropoda) aggregation. J. Elisha Mitchell Sci. Soc. 93: 103.
JUCH, P. J. W., AND G. J. BOEKSCHOTEN. 1980. Trace fossils and grazing traces produced by Littorina
and Lepidochitona, Dutch Wadden Sea. Geol. Mijnbouw 59: 33-42.
KRAEUTER, J. N. 1976. Offshore currents, larval transport, and establishment of southern populations of
Littorina littorea Linne along the U. S. Atlantic coast. Thalassia Jugosl. 10: 159-170.
LEVINTON, J. S., AND S. STEWART. 1982. Marine succession: the effect of two deposit feeding gastropod
species on the population growth of Paranais litoralis Miiller 1784 (Oligochaeta). /. Exp. Mar.
Biol. Ecol. 59: 231-241.
LuBCHENCO, J. 1978. Plant species diversity in a marine intertidal community: importance of herbivore
food preference and algal competitive abilities. Am. Nat. 112: 23-39.
MANN, R., ed. 1979. Exotic Species in Mariculture. MIT Press, Cambridge. 363 pp.
McKiLLUP, S. C. 1983. A behavioural polymorphism in the marine snail Nassarius pauperatus: geographic
variation correlated with food availability, and differences in competitive ability between morphs.
Oecologia (Berl.) 56: 58-66.
MILLS, E. L. 1967. The biology of an ampeliscid amphipod crustacean sibling species pair. / Fish. Res.
Board Can. 24: 305-355.
MORISITA, M. 1959. Measuring of interspecific association and similarity between communities. Mem.
Fac. Sci. Kyushu Univ., Ser. E Biol. 3: 65-80.
MORSE, E. S. 1880. The gradual dispersion of certain mollusks in New England. Bull. Essex Inst. 12: 171-
176.
MORSE, E. S. 192 1 . Observations on Living Gasteropods of New England. Peabody Museum, Salem. Nichols
Press, Lynn, Massachusetts. 29 pp.
MURPHY, D. J. 1979. A comparative study of the freezing tolerances of the marine snails Littorina littorea
(L.) and Nassarius obsoletus (Say). Physiol. Zoo/. 52: 219-230.
NICHOLS, J. A., AND J. R. ROBERTSON. 1979. Field evidence that the eastern mud snail, Ilyanassa obsoleta,
influences nematode community structure. Nautilus 93: 44-46.
PACE, M. L., S. SHIMMEL, AND W. M. DARLEY. 1979. The effect of grazing by a gastropod, Nassarius
obsoletus, on the benthic microbial community of a salt marsh. Estuarine Coastal Mar. Sci. 9:
121-134.
PEARSE, A. S. 1913. Observations on the fauna of the rock beaches at Nahant, Massachusetts. Bull. Wisconsin
Nat. Hist. Soc. 11: 8-34.
PEARSE, A. S. 1914. Observations on the fauna of the rock beaches at Nahant, Massachusetts (continued).
Bull. Wisconsin Nat. Hist. Soc. 12: 72-80.
PECHENIK, J. A. 1978. Adaptations to intertidal development: studies on Nassarius obsoletus. Biol. Bull.
154: 282-291.
PERKINS, G. H. 1869. Molluscan fauna of New Haven. Proc. Boston Soc. Nat. Hist. 13: 109-163.
PETERSON, C. H. 1979. The importance of predation and competition in organizing the intertidal epifaunal
communities of Barnegat Inlet, New Jersey. Oecologia 39: 1-24.
PETRAITIS, P. S. 1983. Grazing patterns of the periwinkle and their effect on sessile intertidal organisms.
Ecology 64: 522-533.
POURREAU, C. N. 1979. Succession of five common salt marsh detritivores on Spartina alterniflora detritus
of decreasing particle size and increasing age. Biol. Bull. 157: 390.
RACE, M. S. 1982. Competitive displacement and predation between introduced and native mud snails.
Oecologia (Berl.) 54: 337-347.
RATHBUN, R. 1881. The littoral marine fauna of Provincetown, Cape Cod, Massachusetts. Proc. U. S. Nat.
Mus. 3: 116-133.
RECHER, H. F. 1966. Some aspects of the ecology of migrant snorebirds. Ecology 47: 393-407.
SAY, T. 1822. An account of some of the marine shells of the United States. J. Phil. Acad. Nat. Sci. 2:
221-248.
SCHELTEMA, R. S. 1962. Pelagic larvae of New England intertidal gastropods. I. Nassarius obsoletus Say
and Nassarius vibex Say. Trans. Am. Microsc. Soc. 81: 1-11.
SCHELTEMA, R. S. 1964. Feeding habits and growth in the mud-snail N ssarius obsoletus. Chesapeake Sci.
5: 161-166.
SCHELTEMA, R. S., AND J. T. CARLTON. 1983. Methods of disper among fouling organisms and possible
558 G. A. BRENCHLEY AND J. T. CARLTON
consequences for range extension and geographical variation. In Proc. Symp. Mar. Biodeterioration,
Bethesda, MD. Office of Naval Research, U. S. Naval Institute, Washington, DC. In press.
SlBERT, J. 1968. Interactions between detritus, microorganisms, and the mud-snail, Nassarius obsoletus
(Say). Ph.D. dissertation, Columbia University, New York.
SIMBERLOFF, D. 1981. Community effects of introduced species. Pp. 53-81 in Biotic Crises in Ecological
and Evolutionary Time, M. H. Nitecki, ed. Academic Press, New York.
SINDERMANN, C. J. 1960. Ecological studies of marine dermatitis-producing schistosome larvae in northern
New England. Ecology 41: 670-684.
SMITH, E. P., ANDT. M. ZARET. 1982. Bias in estimating niche overlap. Ecology 63: 1248-1253.
SPOONER, G. M., AND H. B. MOORE. 1940. The ecology of the Tamar Estuary. VI. An account of the
macrofauna of the intertidal muds. J. Mar. Biol. Assoc. U. K. 24: 283-330.
STAMBAUGH, J. E., AND J. J. MCDERMOTT. 1969. The effects of trematode larvae on the locomotion of
naturally infected Nassarius obsoletus (Gastopoda). Penn. Acad. Sci. 43: 226-231.
STAUFFER, R. C. 1937. Changes in the invertebrate community of a lagoon after disappearance of the eel
grass. Ecology 18: 427-431.
STENZLER, D., AND J. ATEMA. 1977. Alarm response of the marine mud snail, Nassarius obsoletus: specificity
and behavioral priority. /. Chem. Ecol. 3: 159-171.
STIMPSON, W. 1865. On certain genera and families of zoophagous gasteropods. Am. J. Conchol. 1: 55-
64.
VERMEIJ, G. J. 1982a. Environmental change and the evolutionary history of the periwinkle (Littorina
littorea) in North America. Evolution 36: 561-580.
VERMEIJ, G. J. 1982b. Phenotypic evolution in a poorly dispersing snail after arrival of a predator. Nature
299: 349-350.
VERRILL, A. E., AND S. I. SMITH. 1873. Report upon the invertebrate animals of Vineyard Sound and the
adjacent waters. Rept. U. S. Comm. Fish, and Fisheries for 1871-1872. pp. 295-537.
WAGNER, F. J. E. 1977. Palaeoecology of marine Pleistocene Mollusca, Nova Scotia. Can. J. Earth Sci.
14: 1305-1323.
WHARFE, J. R. 1977. An ecological survey of the benthic invertebrate macrofauna of the lower Medway
estuary, Kent. / Anim. Ecol. 46: 93-1 13.
Reference: Biol. Bull. 165: 559-568. (December, 1983)
THE BIOLOGY OF FISSURELLA MAXIMA SOWERBY (MOLLUSCA:
ARCHAEOGASTROPODA) IN NORTHERN CHILE. 2. NOTES
ON ITS REPRODUCTION
MARTA BRETOS1, ITALO TESORIERI, AND LUIS ALVAREZ
Centra de Investigations Mannas, Universidad Del None Sede Iquique.
ABSTRACT
For 14 months, monthly samples were collected to study reproduction in Fissurella
maxima at Huayquique. Results indicate that F. maxima is a dioecious species; no
sign of hermaphroditism has been observed. The sex ratio is 1:1 in the different size
classes analyzed. Ovaries are green and testis are median brown to yellowish white.
Eggs in the ovary measure from 120-280 /u without envelopes. The gonads are par-
asitized by adult digenea trematods of the genus Proctoeces. Some effects of parasitism
are discussed.
Variations in mean monthly gonadosomatic index suggest that there is a main
spawning period in late November-December (late spring-early summer) and a sec-
ondary period in July-August (winter). Fluctuations in mean gonad index show a
close correlation with sea water temperature variations.
The youngest mature specimens detected were about 5 cm in shell length (over
two years old), but the majority of mature animals were over 6.5 cm.
INTRODUCTION
Fissurella maxima Sowerby, 1 835, is the most conspicuous of the Chilean Fissurella
species, reaching sizes of about 12 cm in shell length at the Iquique region (20°14'S,
70° KW) and 14 cm at Los Vilos (31°55'S, 71°32'W). It lives throughout the low
intertidal and high subtidal levels, under Lessonia sp. leaves on exposed rocky shores.
F. maxima is a species with a life span of about 7-10 years (Bretos, 1982), and like
F. crassa it seems to form two shell growth rings per year (Bretos, 1980). Typical
commercial sizes vary from 60-85 mm in shell length at Iquique; these animals are
usually between 2 and 4 years old.
Although keyhole limpets of the genus Fissurella are abundant on Chilean coasts,
little information is available on their biology and there appears to be no published
studies on their reproduction. Some data has been found on reproduction of Fissurella
from other regions. The breeding cycle of a small sized Caribbean species, F. bar-
badensis Gmelin has been described by Ward (1966). This research was based on
collections, made at bimonthly intervals, analyzed by using histological study of the
gonads. Two principal spawning periods were recorded: from September to November
and from March to June. Spawning specimens were present in all but two samples
throughout the collecting period (early January and early April). The results of this
study indicate that there is no resting phase in F. barbadensis along the coasts of
Barbados.
Received 8 July 1983; accepted 29 August 1983.
1 Present address: Departamento de Ciencias Basicas, Universi^ad de La Frontera, Casilla 54-D, Te-
muco, Chile.
559
560 M. BRETOS ET AL.
Concerning European species, Boutan (1885) reported that F. reticulata spawns
from May to early July at Port Vendres. In F. (Cremides) nubecula (L.) spawning
occurs in May at Naples Port (Bacci, 1947).
The present study was undertaken as the first step in the analysis of reproduction
of F. maxima in Northern Chile.
MATERIALS AND METHODS
F. maxima samples were collected at Huayquique (20° 17'S, 70°08rW), in northern
Chile. Sampling took place at approximately monthly intervals, from July 1979 to
August 1980. The animals were collected by diving in shallow waters, from 0-2 m
below low-water mark, and intertidally. The sampling area was a rocky shore, open
coast habitat.
In the laboratory, each animal was weighed (wet weight) and removed from its
shell. Wet weight of gonad and soft parts were also determined by using a digital
Sauter balance to the nearest 0. 1 g. Shell length was measured by using vernier calipers
to the nearest 0. 1 mm. Sampling covered the available size range. Size of the specimens
was not selected in order to determine the size at which F. maxima attains first sexual
maturity.
The sex of the animal was determined when the gonad was exposed by gross
dissection. Gonads were observed under a stereo-microscope and notes were made
on their appearance. Egg diameters in the ovary were measured by using micrometric
eye lens.
Data were grouped in size classes of 5 or 10 mm. The general reproductive
condition of each sample was assessed by calculating the gonadosomatic index (GSI).
This was calculated by expressing the ratio of gonad wet weight to total wet body
weight as a percentage. Sexually undetermined animals were numerous in size classes
up to 60 mm in shell length. For this reason, data were analyzed mainly in animals
whose shell length was greater than 60.0 mm. Separate monthly GSI means were
calculated for each sex in animals over 60.0 mm in shell length.
Sexual maturity of each animal was estimated by considering its GSI, and the
size and appearance of the gonad. Sexual maturity of animals in each sample was
estimated.
Trematods were present in the gonadas ofF. maxima. The percentage of infection
was analyzed in sexually undetermined specimens.
Variations in monthly GSI means were related to sea surface temperature. It was
measured daily at 9:00 hours at the sampling locality.
RESULTS
A total of 1602 animals were examined whose sizes ranged from 21.5 to 98.6
mm in shell length (Table I). Only 24 animals were captured in May 1980 because
of strong seas.
The gonads
F. maxima is a dioecious species; no hermaphrodites were detected among the
animals studied. The sexes cannot be distinguished externally.
Animals classified as sexually undetermined had inconspicuous or no discernible
gonads, whitish or transparent, sometimes pinkish colored. The pinkish color was
due to parasites in the gonad. These parasites were identified as adult specimens of
the digenetic trematods Proctoeces Odhner, 1911 (Bretos and Jiron, 1980). Many
REPRODUCTION OF FISSURELLA MAXIMA
561
TABLE I
Material of F. maxima collected at Huayqiiique
Sexed animals
Females
Males
Date
Total N
N
%
N
%
Sexually
undet. animals
4-7-79
136
69
52.3
63
47.7
4
27-8-79
139
72
57.1
54
42.9
13
25-9-79
150
61
44.9
75
55.1
14
22-10-79
137
50
46.3
58
53.7
29
19-11-79
130
35
44.3
44
55.7
51
20-12-79
116
34
44.7
42
55.3
40
2-1-80
105
21
39.6
32
60.4
52
11-2-80
70
30
46.9
34
53.1
6
10-3-80
78
38
50.0
38
50.0
2
14-4-80
138
60
44.8
74
55.2
4
29-5-80
24
13
54.2
11
45.8
0
9-6-80
105
48
48.0
52
52.0
5
29-7-80
148
72
51.4
68
48.6
8
26-8-80
126
68
57.6
50
42.4
8
Total
1.602
671
695
236
Percentages of females and males are given for sexed animals.
young specimens of sexually undetermined F. maxima (73.7%) had as many as 17
adult trematods in their gonads (Table II).
The gonad is single. When developed or mature, the female gonad is green and
the male gonad varies from median brown to yellowish white.
In young specimens the small gonad is found next to the digestive gland; its
weight was under 0.1 g. The smallest female with a detectable gonad was 27.3 mm
TABLE II
Numbers of sexually undetermined specimens ofF. maxima from Huayquique. and quantity
of parasites in their gonads
Shell
length
(mm)
Infected
specimens
Intensity
Not infected
specimens
Total specimens
examined
Mean ± SD
Range
20.1-25.0
1
2
2
0
1
25.1-30.0
1
1
1
0
1
30.1-35.0
1
2
2
4
5
35.1-40.0
8
4.0 ±
3.0
1-10
10
18
40.
-45.0
25
3.6 ±
2.4
1-9
10
35
45.
-50.0
45
4.4 ±
2.9
1-17
17
62
50.
-55.0
63
4.1 ±
2.2
1-11
17
80
55.
-60.0
25
4.9 ±
3.4
1-16
2
27
60.
-65.0
3
5.0 ±
2.7
3-8
I
4
65.
-70.0
2
4.5
4-5
0
2
70.
-75.0
0
—
—
1
1
Total
174
62
236
%
73.73
26.27
100.00
562 M. BRETOS E T AL.
long, the smallest male 30.7 mm. Both sizes correspond to one-year-old animals
(Bretos, 1982).
As the gonad grows, it remains attached to the digestive gland by the connective
tissue envelope sheet. When the gonad is separated from the digestive gland by dis-
section, the gonad sac opens. Both gonads consist of a sac with a large lumen. Trabeculae
occur within the gonads. At mature stage, the gonads are filled with the gametes.
Eggs found in the lumen of the ovary measured from 120-280 /u in diameter without
envelopes. At least two sheets of a gelatinous matrix have been detected around the
eggs at observation under the microscope.
The mature gonad can attain a wet weight of 30.6 g in females and 17.4 g in the
reproductive season (November). The gonad is turgent and gametes can easily emerge
when the gonad is dissected. The sex cells are discharged into the sea water through
the right nephridiopore. It has been observed in males placed in aquaria, that the
sperms are liberated through the apical hole as a white jet.
Sex ratio
Sex ratio was calculated as a percentage in size classes without considering sexually
undetermined individuals (Fig. 1). Sexes were similarly represented in classes over
60 mm. Among sexed animals, 49.12% corresponds to females and 50.88% to males
(Table III)
Sexually undetermined specimens are abundant in size classes up to 60 mm in
shell length (two year old animals). Among 236 undetermined animals examined,
only 7 were longer than 60.0 mm (Table II).
Most of the sexually undetermined animals occurred in samples from October
to January (Table I). The lowest mean shell lengths (Table IV) were observed during
this period because many small animals were found in these months.
Estimated maturity
Some gonad characteristics and the gonadosomatic index were used as criteria to
classify sexed animals as mature or immature.
v.
100
50
1 A S 0~ ~N 5 ] F M~~A~ n=T ~ J A~
1979 1980
FIGURE 1. Sex ratio in F. maxima from Huayquique.
REPRODUCTION OF FISSURELLA MAXIMA
563
TABLE III
Se.\ distribution off. maxima in size classes from all samples
Sexed animals
Shell
length
(mm)
Females
Males
Females
+ males
Sexually
undet.
Total
N
%
N
%
20.
-30.0
1
0
_
1
2
3
30.
-40.0
1
—
2
—
3
23
26
40.
-50.0
25
48.1
27
51.9
52
97
149
50.
-60.0
110
50.9
106
49.1
216
107
323
60.
-70.0
177
48.9
185
51.1
362
6
368
70.
-80.0
228
48.7
240
51.3
468
1
469
80.
-90.0
114
48.5
121
51.5
235
0
235
90.
-100.0
15
51.7
14
48.3
29
0
29
Total
671
695
1366
236
1602
% in total
49.1
50.9
100
Two aspects were studied in male gonads: the color variation and the relative
abundance of ripe spermatozoa. An attempt was made to find a correlation between
the color and the maturity stage in male gonads. Testis with high GSI were creamy
or light olive green, but there was no clear color graduation nor a constant relationship
between the color and the GSI. In addition, male gonads were classified as milky,
semi-milky, or not-milky, according to the quantity of semen observed among the
testis trabeculae after dissection. Milky and semi-milky testis were usually present in
specimens with high or medium GSI values.
No color differences were observed in female gonads of animals with different
GSI values. Only very small ovaries had a lighter green color. Female specimens with
Size off. maxima
TABLE IV
Samples
N
Shell length (mm)
Mean ± S.D.
Range
Jul 79
136
74.9 ±
9.2
35.2-96.5
Aug 79
139
74.2 ±
13.1
35.5-98.6
Sep 79
150
72.7 ±
13.8
27.3-96.1
Oct 79
137
61.7 ±
10.7
34.3-86.3
Nov 79
130
60.9 ±
13.0
36.8-90.3
Dec 79
116
61.2 ±
9.4
41.4-89.9
Jan 80
105
56.3 ±
11.1
21.5-87.2
Feb 80
70
63.0 ±
11.6
30.7-95.7
Mar 80
78
72.5 ±
12.7
37.2-94.2
Apr 80
138
68.3 ±
11.2
37.4-98.5
May 80
24
78.7 ±
7.8
61.5-92.9
Jun 80
107
67.0 ±
12.5
25.9-97.9
Jul 80
148
66.0 ±
11.3
39.5-90.9
Aug 80
126
67.0 ±
12.2
32.2-91.6
Collected at Huayquique. N = number of specimens. S.D. = standard deviation.
564
M. BRETOS ET AL.
high or medium GSI values had ovaries of friable consistency and in which eggs
detached easily from the trabeculae at the time of dissection.
Assuming that an increase in GSI may be interpreted as a buildup of gametogenic
cells and gametes, while a decrease indicates spawning, GSI was used to estimate the
reproductive activity in both sexes in the present study.
After analyzing the data obtained, we concluded that the GSI was the most reliable
method for classifying F. maxima animals as "mature" or "immature".
Animals with spent and recovering gonads were grouped together as "immature"
specimens. Only fully mature animals, with high GSI were considered "mature".
Sexual maturity estimations are summarized in Table V. The highest number of
mature animals was detected in late July, 1980. Many mature specimens were also
found in October and November, 1979. Mature animals were present throughout
the year, although they were scarce in some months (Table V).
The onset of sexual maturity
The size at which F. maxima may first spawn is considered as the minimum size
at which estimated mature animals have been found.
The youngest mature female detected measured 49.8 mm and the youngest mature
male 47.7 mm in shell length (1.5-year-old animals). Nevertheless, numerous mature
specimens were usually observed in size classes over 65 mm (animals two or more
years old) (Bretos, 1982).
The highest GSI, meaning fully developed gonads, were detected in animals ranging
from 70 to 90 mm in shell length, with GSI values from 2 1 .3 (males) to 32.2 (females).
Spawning
Mean GSI were calculated separately per month for females and males over 60.0
mm shell length (Fig. 3). Mean GSI values exhibit the same tendencies in both sexes
although the highest values were observed in females in November.
TABLE V
Estimated maturity off. maxima over 60.0 mm in shell length, at Huayquique.
Date
Females
Males
Mature
Immature
Total
Mature
Immature
* Mature animals are equal or more numerous than immature animals.
Total
Jul 79
13
52
65
27
34
61
Aug 79
25
39
64
25
27
52
Sep 79
1
54
55
13
52
65
Oct 79
13*
13
26
23*
22
45
Nov 79
14*
8
22
19*
14
33
Dec 79
1
27
28
3
26
29
Jan 80
0
16
16
1
21
22
Feb80
6
16
22
8
15
23
Mar 80
10
24
34
5
27
32
Apr 80
21
32
53
28
33
61
May 80
0
13
13
2
9
11
Jun 80
6
29
35
21*
18
39
Jul 80
39*
12
51
40*
8
48
Aug 80
17
33
50
23*
16
39
REPRODUCTION OF FISSURELL4 MAXIMA 565
Two noticeable peaks appeared in November, 1979 and late July, 1980. The peak
of early July, 1979 is less conspicuous.
The lowest mean GSI value was observed in early January; only one male was
estimated as sexually mature in this sample (Table V). These facts strongly support
the idea that a massive spawning period is complete in December. Mean GSI also
decreased in August-September suggesting the occurrence of a winter spawning. Mean
GSI also decreased in May, but the sample obtained included few animals, indicating
that it may not be a representative sample of the population.
According to the GSI fluctuations, we assume that there are two spawning seasons
per year in the F. maxima population under study: a main reproductive period
occurring in late spring-early summer (November-December) and a secondary period
occurring in winter (July-August).
A close relationship appears to exist between the GSI cycle and the sea temperature
cycle (Fig. 2).
Surface sea water temperature at Huayquique exhibits two rises each year (see
Fig. 2, and Bretos, 1978). A little peak is found in winter; temperatures may reach
more than 17°C. The long and conspicuous rise of temperature begins in October-
November, i.e., in spring, and its highest values are observed in January-February
(summer). Sea water temperatures are particularly high, up to 25 °C, in years in which
El Nino current descends to northern Chile from Peru, as observed in the summer
of 1977 (Bretos, 1978).
DISCUSSION
According to Bacci (1947) there is "a certain degree of hermaphroditism" in F.
nubecula from the Gulf of Naples, detectable by statistical methods. This means that
sex reversal of the protandric type would occur in about 12% in this species of limpet.
On the contrary, other papers concerning the anatomy of reproductive organs in
Fissurella have described normal ovaries and testis and no signs of hermaphroditism
(Boutan, 1885; Ziegenhorn and Thiem, 1925). Ward (1966) reported that there was
no indication of hermaphroditism nor of change of sex at any shell length in F.
barbadensis. The results of the present study indicate that F. maxima is a dioecious
species in which sex reversal has not been detected. No significant differences were
found in size classes over 60 mm in shell length (Table III). Personal unpublished
observations on the gonads of the eight other Fissurella species from northern Chile
(Bretos, 1976) supports this.
LLJ
DC
3 17
tt
v 16
CL
I
UJ
15 -
J A S 0 N D
1979 1980
FIGURE 2. Mean sea surface temperatures at Huayquique.
566
M. BRETOS ET AL.
Not all of the individuals of the same species or population develop their gonads
at the same time, age, or shell length, since growth is variable from one animal to
another, and growth rate depends partly on endogenous factors (Wilbur and Owen,
1964; Bretos, 1978). Nevertheless, it is surprising to find sexually undetermined F.
maxima animals measuring as much as 72 mm in shell length (Table II). These are
juveniles in which development of the gonad has not yet, or only partially, begun.
Considering that it is possible to identify clearly ovaries and testis in small F. maxima
individuals (27.3 mm and 30.7 mm in shell length respectively), it may be assumed
that exogenous factors could be acting to retard gonad development. The main ex-
ogenous factor is the high incidence of trematod parasites in the gonad, attaining an
infection rate of up to 73.73% in sexually undetermined animals (Table II). Gastropods
30 -
25 -
Q 20
z
< '5
2:
o
o
Q
< 10
o
5
J
1979
— I — — I — — I —
A S 0 N
J F
1980
25 -
FIGURE 3. Monthly changes in gonadosomatic index in females and males. Mean values and standard
deviation.
REPRODUCTION OF FISSURELLA MAXIMA 567
are often part of parasitic life cycles, particularly those of digenetic trematodes. The
gametogenic activity of the mollusc is either curtailed or even completely suppressed
by parasitism (Webber, 1977), or the invaded gonad may be destroyed (McArthur
and Featherston, 1976), causing parasitic castration. It should be noted that the gonads
or other organs of molluscs are usually invaded by trematode cercaria, i.e., by trematode
larval forms. In F. maxima the digenetic trematodes that parasitize the gonads are
adult specimens (Bretos and Jiron, 1980), not larval stages which may cause consid-
erable damage. Gametogenetic activity does not seem to be suppressed in F. maxima
nor is the gonad destroyed by adult Proctoeces trematods.
F. maxima eggs are much larger (0.12-0.28 mm without envelopes) than those
of F. reticulata that measure 0. 1 mm (Boutan, 1885) or those of/7, barbadensis [0.08-
0.18 mm including the gelatinous coat (Ward, 1966)]. This may be related to the
size that each species attains: F. barbadensis can measure up to 3.3 cm in shell length
and F. reticulata is also small, but F. maxima can reach a shell length of 1 2 or more
centimeters.
The data obtained suggest that reproduction occurs rhythmically in the F. maxima
study population, showing a semiannual pattern of breeding. According to the GSI
calculated, two spawning periods per year were detected: one in winter and the other
in late spring-early summer (Fig. 3). A similar reproductive pattern is found in F.
barbadensis (Ward, 1966), which has two breeding seasons.
Semiannual breeding seasons occur in a number of tropical and temperate species.
One of the major environmental parameters affecting or influencing the reproductive
state of a population is temperature. Seasonally changing sea temperatures may in-
fluence reproductive activities and may serve to promote and synchronize spawning
(Webber, 1977). Mean sea surface temperature shows a bimodal cycle at Huayquique
(Fig. 2), presenting the main peak in summer and a little one in winter. There appears
to be a close correlation between sea temperature fluctuations and F. maxima gonado-
somatic index variations (Fig. 3), thus, its reproductive activity. F. maxima seems
to be one of the mollusc species whose spawning is influenced by sea temperature
changes.
F. gibba individuals at Banyuls, and F. reticulata individuals at Port-Vendres
seemed to be more numerous during the breeding season (Boutan, 1 885). The samples
of F. maxima obtained in winter and summer seem to be more numerous (Table I),
but this may be a coincidence. Nevertheless, more females than males were detected
in July-August (winter months). This period coincides with the secondary spawning
season of F. maxima at Huayquique. During the main spawning period (November-
December) however, males were more numerous than females (Table I), but there
was also a large number of sexually undetermined specimens which altered the real
sex ratio in these samples.
The youngest sexually mature specimens measured about 50 mm in shell length,
but the majority of the F. maxima population were mature at shell lengths of over
65 mm. On the other hand, a good number of sexually undetermined individuals
were observed up to 60 mm in shell length. It is therefore not advisable to catch F.
maxima animals smaller than 65 mm in shell length for commercial or industrial
purposes.
ACKNOWLEDGMENTS
These investigations were supported by SERPLAC I Region (Regional Devel-
opment Funds) and by the General Research Manag Jient of the Universidad del
Norte. We are also indebted to Mr. Jose Ignacio Moiaga for making the figures.
568 M. BRETOS ET AL.
LITERATURE CITED
BACCI, G. 1947. Osservazioni sulla sessualita degli Archaeogastropoda. Arch. Zool. ltd. 32: 329-341.
BOUTAN, L. 1885. Recherches sur Tanatomie et le developpement de la fissurelle. Arch. Zool. Exp. Gen.
3 Bis: 1-173.
BRETOS, M. 1976. Keyhole limpete of the genus Fissurella from northern Chile. Resumenenes de Communic.
y Simp., XIX Annual Meeting of the Biology Soc. of Chile, pp. 8.
BRETOS, M. 1978. Growth in the keyhole limpet Fissurella crassa Lamarck (Mollusca: Archaeogastropoda)
in Northern Chile. Veliger 21: 268-273.
BRETOS, M. 1980. Age determination in the keyhole limpet Fissurella crassa Lamarck (Archaeogastropoda:
Fissurellidae), based on shell growth rings. Biol. Bull. 159: 606-612.
BRETOS, M. 1982. Biologia de Fissurella maxima Sowerby (Mollusca: Archaeogastropoda) en el Norte de
Chile. 1 . — Caracteres generales, edad y crecimiento. Cahiers Biol. Mar. 23: 1 59- 1 70.
BRETOS, M., ANDC. JIRON. 1980. Trematods in Chilean Fissurellid molluscs. Veliger 22: 293.
McARTHUR, C. P., AND D. W. FEATHERSTON. 1976. Suppression of egg production in Potamopyrgus
antipodarum (Gastropoda: Hydrobiidae) by larval trematods. N. Z. J. Zool. 3: 35-38.
WARD, I. 1966. The breeding cycle of the keyhole limpet Fissurella barbadensis Gmelin. Bull. Mar. Sci.
GulfCaribb. 16: 685-695.
WEBBER, H. H. 1977. Gastropoda: Prosobranchia. Pp. 1-98 in Reproduction of Marine Invertebrates. Vol.
IV, A. C. Giese and J. S. Pearse, eds. Academic Press, New York.
WILBUR, K. M., AND G. OWEN. 1964. Growth. Pp. 211-242 in Physiology of Mollusca. Vol. I, K. M.
Wilbur and C. M. Younge, eds. Academic Press, New York.
ZlEGENHORN, A., AND H. THIEM. 1925. Beitrage zur Sistematik und Anatomic der Fissurellen. Jena Z.
Natumiss. 55: 1-78.
Reference: Biol. Bull. 165: 569-581. (December, 1983)
INDUCED DEVELOPMENT OF SWEEPER TENTACLES ON THE
REEF CORAL AGARICIA AGARICITES: A RESPONSE
TO DIRECT COMPETITION
ELIZABETH A. CHORNESKY
Division of Biological Sciences, The University of Texas, Austin, Texas 78712
ABSTRACT
The scleractinian coral Agaricia agaricites often has elongate sweeper tentacles
on colony margins close to other sessile animals. Sweeper tentacles can damage tissues
of opponents and are probably used in direct competition for substrate space. Fur-
thermore, contact with tissues or mesenterial filaments of other corals, or with tissues
of the gorgonian Erythropodium caribaeorum or the zooanthid Palythoa caribbea
can stimulate the development of sweeper tentacles by A. agaricites. Depending on
both the particular competitor species involved and the distance separating it from
A. agaricites, events leading to the development of sweeper tentacles may or may
not include tissue loss by A. agaricites. On average the development of sweeper
tentacles takes thirty days, and is localized exclusively on tissues close to the region
in contact with competitors. Sweeper tentacles do not develop in response to artificial
stimuli simulating tactile contact or damage such as occur in natural interactions
with other corals. Thus, recognition of competitor tissues appears to be a necessary
stimulus for sweeper formation.
INTRODUCTION
Sessile colonial animals, particularly scleractinian corals, crowd many tropical
reefs where space for growth often becomes limited (e.g.. Porter, 1972, 1974; Glynn,
1973; Connell, 1976, 1978; Sheppard, 1979, 1982). Although upright or branching
corals may partially escape this problem by growing up and over adjacent animals
(Porter, 1974; Connell, 1976; Glynn, 1976; Jackson, 1979; Wellington, 1980), many
corals growing along reef surfaces frequently encounter other sessile animals. Thus,
competition for substrate space is considered one of the processes structuring coral
reef communities and selecting for life history characteristics and other attributes of
sessile reef inhabitants (Connell, 1973, 1976; Glynn, 1973; Lang, 1973; Porter, 1974,
1976; Jackson, 1977, 1979; Potts, 1977; Bak and Engel, 1979; Sheppard, 1982).
When stony corals grow close together, they often directly damage one another
by using mesenterial filaments or sweeper tentacles (Lang, 1971, 1973; Richardson
etai, 19 79; Sheppard, 1979; Wellington, 1980; Bak et ai, 1982). If corals of different
species are placed in direct contact, many can extend their mesenterial filaments
within hours and use them to digest tissues on the opposing coral (Lang, 1971, 1973;
Glynn, 1976; Sheppard, 1979). The consequences of such interactions are generally
predictable; certain "digestively dominant" species, particularly of the suborder Faviina,
are consistently able to use mesenterial filaments to damage others (Lang, 1973;
Sheppard, 1979). The additional use of sweeper tentacles in natural interactions by
some corals, however, may alter the long term outcomes of these otherwise predictable
Received 21 March 1983; accepted 26 September 1983.
569
570 E. A. CHORNESKY
encounters (Richardson et al, 1979; Wellington, 1980; Bak et ai, 1982; Sheppard,
1982). Sweeper tentacles are longer than normal (Lewis and Price, 1975; Bak and
Elgershuizen, 1976) and armed with specialized cnidae (den Hartog, 1977; Wellington,
1980). When expanded, these tentacles increase the volume within reach of live coral
tissues and may deter other corals from growing too closely (Richardson et al, 1979)
or may actively damage competitor tissues (Wellington, 1980; Bak et al., 1982).
All coral polyps have mesenterial filaments. In contrast, the distribution of sweeper
tentacles is erratic, and the determinants of their presence are poorly understood.
For species which can form sweeper tentacles, neither every colony in a population,
nor every polyp on a colony necessarily possesses sweepers. On Montastraea cavernosa
(Linnaeus), these tentacles, which are present on most colonies, are thought to extend
in response to water currents and are most abundant around colony perimeters (Price,
1973 in den Hartog, 1977; den Hartog, 1977; Richardson et al., 1979). On Pocillopora
sp. (Wellington, 1980) and Madracis mirabilis (Duchassaing and Michelotti) (Bak et
al., 1982) sweepers develop on polyps next to wounds caused by the mesenterial
filaments of adjacent corals.
Since the stimulus for sweeper development determines their location on a colony,
it also determines whether they are used in competitive interactions. Thus, the con-
sequences of direct encounters involving corals depend not only on the relative effects
of mesenterial filaments and sweeper tentacles, but also on the factors which initiate
sweeper formation. Many of the responses of other cnidarians to direct competition
are thought to be stimulated by contact and recognition of opponent tissues (Theodor,
1970; Ivker, 1972; Francis, 1973; Purcell, 1977; Ottaway, 1978; Brace et ai, 1979;
Bigger, 1980; Watson and Mariscal, 1983).
This paper explores the conditions under which sweeper tentacles form on the
Caribbean reef coral Agaricia agaricites (Linnaeus). This species has short mesenterial
filaments which extend only a few mm away from the corallum (Bak et ai, 1982),
and can use them to digest only a few other species of coral (Lang, 1973). A. agaricites'
polyps are flat and normally have short tentacles (approximately two mm long, Lewis
and Price, 1975). In contrast, sweeper tentacles on A. agaricites may be over a cm
in length (Bak and Elgershuizen, 1976; pers. obs.; Fig. 1A). The occurrence of these
special tentacles only on portions of colonies of A. agaricites adjacent to other species
of sessile animals (pers. obs.) strongly suggests that they develop specifically in response
to direct competitive interactions.
Here I examine the potential function of sweeper tentacles on A. agaricites and
the stimulus for their development. The role of sweeper tentacles on A. agaricites in
determining the long term consequences of competitive encounters will be discussed
elsewhere.
GENERAL MATERIALS AND METHODS
This study has two components: I) determination of the potential function of
sweeper tentacles on A. agaricites and observation of their development under natural
and experimental conditions; and II) experimental determination of the nature of
stimuli which induce sweeper formation. The specific protocol and results for each
section follow this general discussion of information pertaining to the entire study.
All collections of animals, observations, and in situ experiments were accomplished
using SCUBA at a depth of- 10 meters on the west forereef at Discovery Bay, Jamaica.
Aquaria with running, unfiltered sea water for laboratory experiments concerning
sweeper function were provided by the Discovery Bay Marine Laboratory of the
University of the West Indies. Since the sweeper tentacles of A. agaricites expand
CORAL SWEEPER TENTACLE DEVELOPMENT
571
FIGURE 1. Coral polyps expanded at night. Corals are: A) Agaricia agaricites, B) Madracis decactis,
C) Montastraea cavernosa, and D) Montastraea annularis. Arrows indicate sweeper tentacles (sw) and
regular tentacles (t), and scale bars equal approximately 5 mm. In (A) compare length of regular and sweeper
tentacles on A. agaricites, here shown next to a damaged colony of Madracis decactis. Photographs were
taken in situ using a Nikonas camera, one to two framer, extension tube, and strobe.
maximally at night (Chornesky, unpub. data), all behavioral observations were made
after sunset between 2000 and 2400 hours. This study took place between March
1981 and September 1982.
Observations and experiments involved A. agaricites and various sessile animals
spanning a range of competitive strategies and including four corals, a zooanthid,
and a gorgonian (Table I). Among the stony corals used, A. agaricites can digest two
species (Madracis decactis [Lyman] and Porites astreoides Lesueur) and can be digested
by the two other species (Montastraea annularis [Ellis and Solander] and Montastraea
cavernosa} (Lang, 1973). Within each pair of digestively dominant or subordinate
corals, one can sometimes possess sweeper tentacles (Madracis decactis, pers. obs.;
M. cavernosa, Lewis and Price 1975) while the other doe not (P. astreoides and M.
annularis} (Table I). The zooanthid Palythoa caribbea D^chassaing and the gorgonian
Erythropodium caribaeorum Duchassaing and Michelotti sometimes overgrow A.
572 E. A. CHORNESKY
TABLE I
Characteristics of competitor species used in observations and experiments*
Order: Scleractinia Zooanthidae
Gorgonaceae
Species: P.a. M.d. M.a. M.c. P.c.
E.c.
A.a. can digest
Can digest A.a.
May have sweepers
Other cytotoxins overgrowth
overgrowth
Length polyps med. long short long med. long
Length tentacles short long short long short long
* Abbreviations and symbols used in table are: A.a. = Agaricia agaricites, P.a. = Porites astreoides,
M.d. = Madracis decactis, M.a. = Montastraea annularis, M.c. = Montastraea cavernosa, P.c. = Palythoa
caribbea, E.c. = Erythropodium caribaeorum, (+) = species has characteristic, (-) = species does not have
characteristic.
agaricites (Karlson, 1980). P. caribbea contains secondary chemicals which might be
used against competing animals (Cieresko and Karns, 1973).
Unless stated otherwise, all colonies of A. agaricites and Madracis decactis used
in these experiments lacked sweeper tentacles prior to treatment. Colonies of A.
agaricites used were of the formae A. agaricites f. purpurea or A. agaricites f. carinata
as described by Wells (1973). Where appropriate, data were analyzed using Chi-square
and Mann-Whitney tests for statistical significance.
/. FUNCTION AND DEVELOPMENT OF SWEEPER TENTACLES ON A. AGARICITES
MATERIALS AND METHODS
Function
Lewis and Price (1975) originally described the sweeper tentacles of A. agaricites
as appendages for feeding. Nevertheless, in hundreds of separate observations I have
never seen A. agaricites use sweepers to capture visible paniculate food, although
specifically searching for this behavior. This failure to observe feeding, combined with
my consistent observations that sweeper tentacles on A. agaricites only occur on
colony margins close to other animals, implied that on this species sweeper tentacles
might play a role in competitive interactions.
Experiments were conducted in sea water aquaria to determine the potential
function of sweeper tentacles on A. agaricites in spatial competition. Colonies of A.
agaricites already possessing sweeper tentacles, along with colonies of several competitor
species, were collected on the forereef and transferred to the aquaria. The A. agaricites
were then observed on several nights after expansion. After determining the location
of sweepers on these colonies, during the day, colonies of Madracis decactis (n = 5),
P. astreoides (n = 7), and M. annularis (n = 8) were placed within "sweeper length"
of the A. agaricites. These artificially arranged interactions were then observed on
several (3 to 5) nights for the behavior and condition of both corals in each pair.
Opponent species used in these experiments were selected because they all seem to
maintain normal behavior and health in a running sea water system.
CORAL SWEEPER TENTACLE DEVELOPMENT 573
Development: natural interactions
To determine the frequency with which sweeper tentacles occur on portions of
A. agaricites involved in competitive interactions, I labeled a series of natural en-
counters where colonies of A. agaricites were already within one cm of competitors.
These interactions were visited repeatedly at night and scored for the presence or
absence of sweeper tentacles on the A. agaricites. Interactions observed were with:
P. astreoides (n = 28), Madracis decactis (n == 17), M. annularis (n = 15), P. caribbea
(n = 14), and E. caribaeorum (n = 9).
To examine moreover, whether sweeper tentacles develop over time as competitive
interactions progress, the labeled encounters between A. agaricites and P. astreoides,
Madracis decactis, and M. annularis were subsequently scored for the presence or
absence of sweeper tentacles during four observation periods throughout the following
ten months.
Within each observation period, labeled interactions were visited on at least three
nights to minimize the chance that sweepers were contracted due to incidental activity
of other nocturnally active animals or other unpredictable events.
Development: experimentally induced
The following experiments tested whether sweeper tentacles form specifically as
a consequence of contact between A. agaricites and adjacent animals. Encounters
among reef corals generally result from gradual growth, and the first contact between
adjacent animals may often involve intermittently expanded tentacles and polyps.
Such intermittent contact may stimulate a different response from that of close tissue
and skeletal contact which presumably occur in natural encounters as the animals
grow closer and any interaction proceeds (see: Lang, 1973; Potts, 1977; Sheppard,
1979; Wellington, 1980; Bak et al, 1982). Two kinds of experiments were conducted
in situ in which: 1) animals were placed in very close tissue and skeletal contact; and
2) animals were fixed a small and consistent distance apart, simulating initial inter-
actions resulting from gradual growth.
1) Close contact. Colonies of A. agaricites were dislodged using a chisel and placed
in direct contact with the corals P. astreoides (n = 6), Madracis decactis (n = 5),
and M. annularis (n = : 11), the zooanthid P. caribbea (n = 5), and the encrusting
gorgonian E. caribaeorum (n = 6). Paired colonies touched even when both polyps
and tissues were contracted, ensuring constant contact independent of patterns of
tissue and tentacle expansion. Presence of sweeper tentacles on A. agaricites was
assessed nocturnally at weekly intervals for a period of up to fifty days.
2) Controlled distance. To more accurately simulate the initial contact between
competitors as it occurs in natural interactions, colonies of A. agaricites were cemented
by basal portions of bare skeleton onto cinder blocks (Fig. 2A) or onto stationary
asbestos tiles at a small distance from colonies of M. cavernosa (n = 10), Madracis
decactis (n = 18), and M. annularis (n == 34). The underwater epoxy-putty used to
fix corals in place was never in contact with live coral tissues and appears to be non-
toxic (Birkeland, 1976). Specimens of Madracis decactis and M. cavernosa (both of
which have long polyps and tentacles; Figs. IB, 1C, Table I) were positioned so that
contact occurred only between tentacle tips of these species and tissues of the A.
agaricites when corals were fully expanded at night. Approximate distances between
contracted corals were 3 mm with Madracis decactis and 2 cm with M. cavernosa.
Colonies of A. agaricites next to M. annularis (which ha • short polyps and tentacles;
Fig. ID, Table I) were positioned so that their polyps were separated by a 1-2 mm
gap even when both animals were fully expanded at night (distance between contracted
574
E. A. CHORNESKY
corals of 2-3 mm). In four of these interactions, tissues of A. agaricites and M.
annularis were in contact when expanded because colonies slipped into closer proximity
before the epoxy-putty hardened. Interactions were observed frequently during the
day and a minimum of once a week at night for a period of eighty days.
RESULTS
Function
Under laboratory conditions, without exception, tissues of competitor corals placed
close to the sweeper tentacles of A. agaricites were damaged. The behavior of sweeper
tentacles is similar to that of the catch tentacles of anemones (Purcell, 1977). Extended
sweepers brush against and sometimes adhere to opponents, creating patches of
sloughing necrotic tissues within their reach. Such lesions are easily distinguished
from the regions of clean bare skeleton resulting from digestion by mesenterial fil-
aments.
Development: natural interactions
In natural interactions sweeper tentacles were initially present on between forty-
seven and fifty-seven percent of A. agaricites colonies, depending on the competitor
species (Table IIA). In contrast to these initial frequencies, the cumulative frequency
of colonies having sweepers sometime during the ten months was between sixty-five
and eighty percent. This suggests that, over time, proximity to other corals stimulates
the development of sweeper tentacles on colonies of A. agaricites. Therefore, the
duration of observations may greatly influence the interpretation of the frequency of
sweeper occurrence in natural interactions.
Development: experimental induction
1) Close contact. Sweeper tentacles developed on A. agaricites in response to
close contact with all opponent species (Table IIB). Their development was restricted
to tissues within approximately 5 mm of the competitor. The sequence of events
TABLE II
Development of sweeper tentacles (s\v) on Agaricia agaricites in natural (A) and
experimental (B, C) interactions
A.
Natural
interactions
C.
Controlled
B.
Direct contact
distance
7o Lolonies witti Sw
% Develop
% Develop
Competitor
N
Initial
Cumulative
N
Sw
N
Sw
Porites astreoides
28
57
71
6
50
Madracis decactis
17
47
65
5
80
18
83
Montastraea annularis
15
47
80
11
64
34
56*
Montastraea cavernosa
—
—
—
—
—
10
90
Palvthoa caribbea
14
56
—
5
40
—
—
Erythropodium caribaeorum
9
57
—
6
50
—
—
* A total of 26 colonies of A. agaricites were digested by M. annularis. 76% of these colonies developed
sweeper tentacles.
CORAL SWEEPER TENTACLE DEVELOPMENT 575
varied with opponent species. For example, on colonies next to Madracis decactis
and P. astreoides, sweepers developed after the digestive filaments of A. agaricites
damaged tissues of the Madracis and the Porites. In contrast, on colonies adjacent
to M. anmdaris, sweepers developed around wounds caused by digestion of A. agaricites
by mesenterial filaments of M. annular is. Thus, direct contact with other animals
can stimulate development of sweeper tentacles on A. agaricites, and this response
is localized around the zone of contact.
2) Controlled distance. Although sweeper tentacles also developed on colonies of
A. agaricites at a fixed distance from opponents, the sequence of events differed in
perhaps important ways from that occurring when corals were in closer contact.
Most colonies of A. agaricites (90%) placed within reach of M. cavernosa tentacle
tips developed sweeper tentacles (Table IIC). No M. cavernosa ever digested tissues
of A. agaricites, nor did they develop sweeper tentacles in interactions prior to the
A. agaricites (although M. cavernosa can itself develop sweepers during competitive
interactions, Chornesky and Williams, 1983).
Fifteen of eighteen colonies (83%) of A. agaricites placed within tentacle reach
of Madracis decactis developed sweeper tentacles (Table IIC). There was no evidence
that the A. agaricites ever damaged tissues of Madracis decactis with mesenterial
filaments. Thirteen of the eighteen colonies of Madracis decactis also developed
sweeper tentacles. Interestingly, the three A. agaricites which did not form sweepers
during the experiment were adjacent to colonies of Madracis which had developed
sweeper tentacles first and then used them to create extensive wounds on the A.
agaricites. Sweepers developed around one of these wounds on A. agaricites at the
end of the study. In many of the interactions where sweeper tentacles did develop
first on the A. agaricites, nearby tissues of the Madracis decactis were damaged,
confirming laboratory predictions of sweeper function (test for association between
development of sweeper tentacles by A. agaricites and damage to Madracis decactis
tissues: x2 = 8.08, d.f. == 1, P < .005).
When paired with M. anmdaris sweepers developed on over half (56%) of the
colonies of A. agaricites placed adjacent to, but out of reach of opponents' polyps
and tentacles. Development occurred after the M. anmdaris digested A. agaricites
tissues (test for association between digestion and sweeper development: x2 = 7.99,
d.f. = 1 , P < .005) (Table IIC). Sweeper tentacles which developed in these interactions
seemed to function both to damage nearby M. anmdaris tissues and to prevent further
digestion by M. anmdaris. In forty-three percent of the interactions where sweeper
tentacles developed on the A. agaricites, wounds appeared on adjacent colonies of
M. anmdaris which could be attributed to the action of sweeper tentacles. In only a
total of five interactions (19%) were colonies of A. agaricites digested a second time
by the M. anmdaris. In four of these cases, sweeper tentacles had not yet developed
on the A. agaricites; the one colony redigested despite having developed sweeper
tentacles had previously been severely injured by the predaceous gastropod Coral-
liophila abbreviata (Lamarck).
The rate at which sweeper tentacles develop on A. agaricites is best reflected in
data from experiments with Madracis decactis and M. annularis. Mean development
time of sweepers on colonies of A. agancues adjacent to Madracis decactis was 30.2
days after corals were cemented close together (standard deviation of 16 days). For
colonies adjacent to M. anmdaris, the mean development time after digestion by M.
annularis was 31.6 days (standard deviation of 18.5 days). There is no significant
difference between rates of sweeper formation in experin jnts with Madracis decactis
and M. anmdaris (Mann- Whitney U == 132, nl - 18 i2 = 15, P > .1).
In a few cases, unexpected factors affected sweeper tentacle development. Shortly
576
E. A. CHORNESKY
after corals were cemented in place, seventeen of the thirty-four pairs of A. agaricites
and M. annularis were temporarily invaded by small crabs (identified tentatively as
Domecia acanthophora f. acanthophora [Desbonne and Schramm], Austin Williams
pers. comm.). A single crab was usually seen in the crevice formed between adjacent
corals (Fig. 2B). In hundreds of observations, during the day and at night, I have
never seen these crabs in natural interactions among corals. Comparison of interactions
with and without resident crabs shows that crabs decreased the likelihood that colonies
of A. agaricites already digested by M. annularis would develop sweeper tentacles
(test for association between presence of crabs and inability to develop sweepers: x2
= 3.87, d.f. = !,/*< .05). However, on those colonies which did develop sweepers
after crabs appeared, the crabs had no significant effect on the amount of time between
digestion by M. annularis and the appearance of sweeper tentacles (Mann-Whitney
U = 28, nl = 12 n2 == 6, P > .1). Other factors inhibiting sweeper formation after
digestion by M. annularis included enlargement of the wound by the predaceous
gastropod Coralliophila (1 of the 26 digested) and redigestion by M. annularis resulting
in destruction of tissues surrounding the initial wound (3 of the 26). Bak et al. (1982)
note the ability of Domecia and Coralliophila to damage coral tissues close to the
site of competitive interactions.
In summary, development of sweeper tentacles can occur prior to close tissue
and skeletal contact between adjacent corals. The distance at which the interaction
begins is a function of the length of competitor species' polyps, tentacles, and mes-
enterial filaments, and their readiness to evert mesenterial filaments. In addition,
development of sweeper tentacles or repeated use of mesenterial filaments by com-
petitors may delay sweeper formation by A. agaricites. Sweeper development may
also be inhibited by activity of epifauna such as crabs and gastropods. After devel-
opment, the sweeper tentacles of A. agaricites sometimes injure tissues of competitors
and may help prevent further damage by the mesenterial filaments of opponents.
II. STIMULUS FOR SWEEPER TENTACLE DEVELOPMENT
MATERIALS AND METHODS
The preceding experiments demonstrate that contact with various competitors
can stimulate development of sweeper tentacles on A. agaricites. Although differing
FIGURE 2. A) Corals cemented onto cinder blocks in controlled distance experiments. B) Photograph
taken at night of a small crab (cr) in the crevice formed between colonies of A. agaricites and M. annularis
in controlled distance experiments. Bar = ~5 mm.
CORAL SWEEPER TENTACLE DEVELOPMENT
577
in specific form among competitors, this contact generally involves three components
which occur simultaneously: 1) tactile contact; 2) damage, for example by tentacular
nematocysts or by the digestive enzymes or nematocysts of mesenterial filaments; or
3) chemical recognition of competitor tissues. The following experiments were designed
to separate the role of these factors in stimulating the formation of sweeper tentacles.
Corals were exposed to one of the following five stimuli: A) inanimate tactile contact;
B) inanimate damage; C) inanimate contact plus damage; D) animate damage; and
E) inanimate plus animate damage (see Fig. 3). After application of each treatment
in situ, corals were observed for development of sweeper tentacles approximately
once a week at night for a minimum of forty days.
To test whether inanimate tactile contact alone can induce sweepers, tufts of
artificial tentacles made of nylon monofilament line were nailed above colonies of
A. agaricites (n = 5; Fig. 3A). Tips of these artificial tentacles swayed slightly in the
surge and were constantly in contact with a portion of the A. agaricites.
To determine whether inanimate damage alone can induce sweeper tentacles,
portions of live tissues on colonies of A. agaricites were destroyed to mimic digestion
by mesenterial filaments (n = 5; Fig. 3B). Small amounts of concentrated hydrochloric
acid were applied to live tissues using a glass hypodermic syringe. Because the viscosity
and specific gravity of concentrated acid are greater than sea water, it remained where
applied and killed only a discrete patch of tissues. There was no apparent damage
to surrounding tissues. The acid left the skeleton denuded of coral tissues, resembling
lesions from digestion by other corals.
TREATMENT
METHODS
IAI
INANIMATE
CONTACT
ARTIFICIAL
TENTACLES
IBI
INANIMATE
DAMAGE
ICI
INANIMATE
DAMAGE +
CONTACT:
ACID +
ARTIF TENT
IDI
ANIMATE
DAMAGE
DIGESTION
DESIGN
IEI
ANIMATE*
INANIMATE
DAMAGE
DIGESTION*
ACID
RESULTS l%l
100
FIGURE 3. Experimental determination of stimulus for sweeper tentacle development. Corals were
treated with various combinations of artificial tentacles, artificial wounds created with HC1, and wounds
caused by mesenterial filaments of M. annularis. Results are presented 5 percent of colonies which developed
sweeper tentacles. In treatment E, sweeper development occurred onlj around wounds caused by mesenterial
filaments of M. annularis.
578 E- A- CHORNESKY
To simulate both the damage and tactile components of natural interactions, HC1
and artificial tentacles were applied to a series of A. agaricites (n = 5; Fig. 3C). Here,
artificial tentacles were positioned over live coral tissues next to the wound created
with HC1.
When coral tissues are destroyed by mesenterial filaments, damage is accompanied
by the potential for chemical recognition of competitors. In this treatment (animate
damage), corals were damaged by mesenterial filaments in a way which allows com-
parison with the inanimate damage treatment described above. Colonies of A agaricites
were allowed to be digested overnight by M. annularis (n = 6; Fig. 3D), and the
corals were separated the next day and then kept separate for the duration of the
experiment.
When a colony of A. agaricites is stimulated by a competitor, development of
sweeper tentacles is localized around the affected region. This final treatment was
designed to test the extent to which recognition of a competitor affects other tissues
within a colony by artificially creating a second inanimate wound on colonies already
digested by M. annularis. Colonies of A. agaricites were allowed to be digested by
M annularis, after which the corals were separated. One day later, a second wound
which overlapped the first wound slightly on one side was artificially created using
HC1 (n = 5; Fig. 3E). Development of sweepers around the artificial wound would
reflect the degree to which surrounding tissues were also affected by digestion of other
tissues within the colony. This treatment also controlled for whether the use of con-
centrated HC1 was appropriate to simulate damage in natural interactions, since
application of HC1 might conceivably disrupt normal physiological processes and
thereby inhibit sweeper tentacle formation. This would be apparent if sweepers did
not develop next to the M. annularis wound close to where the two wounds overlapped.
RESULTS
Sweeper tentacles did not form on colonies of A. agaricites in response to any of
the inanimate treatments — artificial tentacles, HC1 lesions, or a combination of the
two (Fig. 3A-C). Sweepers did form, however, on all A. agaricites with lesions from
M. annularis mesenterial filaments (Fig. 3D). These sweeper tentacles appeared within
eighteen days, were smaller than usual, and regressed within three weeks of devel-
opment. In nature, I have occasionally seen sweeper tentacles regress as the regenerating
edges of wounds caused by mesenterial filaments begin to advance. Sweeper tentacles
also formed on all colonies of A. agaricites with both M. annularis and HC1 lesions
(Fig. 3E), but only adjacent to the M. annularis wound. Similarly, these sweepers
were smaller than usual and regressed within three weeks. There was no evidence
that HC1 inhibited development of sweeper tentacles anywhere near the first wound.
There are two alternative explanations for why sweepers developed only adjacent to
the M. annularis wound on these colonies: 1) the response to recognition of another
animal within A. agaricites colonies may be quite localized, here occurring only
adjacent to the M. annularis wound; or 2) if recognition is colony-wide, tactile contact
may also be required to stimulate sweeper tentacle development.
DISCUSSION
Cnidarians display a notable array of responses to competitors, including: agonistic
behavior (Lang, 1971, 1973; Francis, 1973; Bigger, 1977, 1980; Ottaway, 1978; Brace
el al, 1979; Sheppard, 1979; Purcell and Kitting, 1982), development and use of
elongate tentacles (Purcell, 1977; Wellington, 1980; Bak et al., 1982; Watson and
Mariscal, 1983), directed growth (Ivker, 1972; Potts, 1977; Wahle, 1980), or an "im-
mune response" (Theodor, 1970; Hildeman et al., 1975; Rinkevich and Loya, 1983).
Most of these processes operate between animals within taxonomic orders, either
CORAL SWEEPER TENTACLE DEVELOPMENT 579
intraspecifically (Theodor, 1970; Ivker, 1972; Francis, 1973; Hildeman el al, 1975;
Potts, 1977; Purcell, 1977; Ottaway, 1978; Brace et al., 1979; Rinkevich and Loya,
1983; Watson and Mariscal, 1983) or interspecifically (Lang, 1971, 1973; Purcell,
1977; Sheppard, 1979; Bigger, 1980; Wellington, 1980), with a few exceptions (Bigger,
1977; Wahle, 1980; Sammarco et al., 1983). The scleractinian coral Agaricia agaricites
develops sweeper tentacles in response to encounters with a range of other animals,
including various corals, a gorgonian, and a zooanthid.
Interactions among sessile reef animals usually result from gradual growth. Par-
ticularly for a species like A. agaricites, having flat polyps, short tentacles, and short
mesenterial filaments, the nature of direct competitive encounters will vary with
characteristics of its opponents. The morphology of competitor polyps, tentacles, and
mesenterial filaments, as well as their readiness to evert mesenterial filaments, determine
how they first contact A. agaricites.
Regardless of the specific mode of contact between A. agaricites and various
anthozoan competitors, all such contact stimulates A. agaricites to develop sweeper
tentacles. For example, when A. agaricites grows close to corals having long tentacles,
the first contact will be with their tentacle tips. In experiments simulating such en-
counters, corals which are digestively dominant when in close contact (M. cavernosa
and A. agaricites) did not evert mesenterial filaments onto opposing corals (A. agaricites
and Madracis decactis, respectively). Contact with only tentacle tips of opponents
stimulated development of sweepers on nearby A. agaricites tissues. This differs from
controlled distance experiments with M. annularis, a digestively dominant coral having
short polyps and tentacles and long mesenterial filaments. Here, the first contact
between corals was digestion of A. agaricites, and sweeper tentacles developed around
the resulting wounds. The distance beween interacting corals did not affect the behavior
of M. annularis, as seen by Wellington (1980) for Pavona.
It is interesting that in these experiments the distance separating competitors
affected the readiness to evert mesenterial filaments of some corals (i.e., M. cavernosa)
and not of others (i.e., M. annularis). This, combined with evidence from controlled
distance experiments that contact was not necessarily required to stimulate eversion
of mesenterial filaments by M. annularis, suggests that controls over the use of mes-
enterial filaments in competitive interactions may be quite complex.
The development of sweeper tentacles on A. agaricites apparently occurs only
after recognition of competitor tissues. Tactile contact and tissue damage also may
be involved in initiating this process, although neither alone nor the two combined
is sufficient. In natural interactions however, contact, damage, and recognition are
probably inseparable. Corals being digested by mesenterial filaments surely have the
potential to recognize competitor tissues. Likewise, those in contact with tentacle tips
of adjacent corals may incur small scale damage from the tentacular nematocysts.
However, if chemical recognition is sufficient to induce sweeper tentacle formation,
the stimulus is probably not a diffusable substance (sensu Bigger, 1977). A. agaricites
colonies separated by only 1-2 mm from tissues of M annularis did not develop
sweeper tentacles until after digestion by M. annularis.
Sweeper tentacles develop only within a zone of approximately 5 mm surrounding
tissues stimulated by another animal. For example, on colonies with wounds resulting
from both digestion by M. annularis and HC1, sweepers formed only around the
wound inflicted by M. annularis. Moreover, disturbance by crabs or predaceous
gastropods to tissues immediately surrounding stimulated regions prevented a few A.
agaricites from developing sweeper tentacles at wounds caused by M. annularis.
Development of sweeper tentacles on A. agaricites w< •_, also inhibited or delayed by
the formation of sweeper tentacles on opposing colonies of Madracis decactis or by
redigestion by M. annularis. These various disturbances all damaged the small region
580 E. A. CHORNESKY
of "responsive" tissues where sweepers would have formed, thereby preventing their
development. This suggests that recognition of competitors stimulates a localized and
not a colony- wide response within colonies of A. agaricites.
In contrast to mesenterial filaments, sweeper tentacles are not generally present
on A. agaricites, but develop specifically as a response to competitive encounters with
other sessile animals. These sweepers have the potential to damage tissues of com-
petitors and may affect the long term outcome of competitive interactions (Chornesky,
in prep.). The exact sequence of events leading to formation of sweeper tentacles
depends upon characteristics of the opponent species and the distance at which en-
counters occur. These two factors will therefore determine the extent of damage to
A. agaricites before it develops sweeper tentacles. Understanding the dynamics of
such complex processes may be important for interpretation of the mechanism and
consequences of natural and experimental encounters among many reef corals.
ACKNOWLEDGMENTS
I am grateful to the many people who contributed to this work, especially Judith
Lang for her enthusiasm and for many generous and stimulating exchanges of ideas,
Charles Wahle for encouragement and assistance, and George and Mary Chornesky
for support throughout. I wish to thank: C. Brunet, J. C. Lang, C. M. Wahle, and
G. M. Wellington for useful comments on earlier drafts of this manuscript; J. Cripps
for assistance with photographic techniques; F. Chace and A. Williams for identification
of Domecia acanthophora; countless dive buddies including C. Brunet, G. Eberhart,
and R. Ramsay for assistance in the field; the West Indies Laboratory for supporting
preliminary stages of this research; and the Discovery Bay Marine Laboratory for
providing surface facilities. I gratefully acknowledge financial support from the Lerner-
Gray Fund of the American Museum of Natural History, the National Science Foun-
dation (DEB 8105172), and Sigma Xi. This paper is a partial fulfillment of the
requirements for a doctoral degree at The University of Texas at Austin. This is
contribution number 294 of the Discovery Bay Marine Laboratory of the University
of the West Indies.
LITERATURE CITED
BAK, R. P. M., AND B. W. ELGERSHUIZEN. 1976. Patterns of oil-sediment rejection in corals. Mar. Biol.
37: 105-113.
BAK, R. P. M., AND M. S. ENGEL. 1979. Distribution, abundance and survival of juvenile hermatypic
corals (Scleractinia) and the importance of life history strategies in the parent coral community.
Mar. Biol. 54: 341-352.
BAK, R. P. M., R. M. TERMAAT, AND R. DEKKER. 1982. Complexity of coral interactions: influence of
time, location of interaction and epifauna. Mar. Biol. 69: 2 1 5-222.
BIGGER, C. H. 1977. The acrorhagial response in Anthopleura krebsi: intraspecific and interspecific recognition.
Pp. 127-136 in Coelenterate Ecology and Behavior, G. O. Mackie, ed. Plenum Press, New York.
BIGGER, C. H. 1980. Interspecific and intraspecific acrorhagial aggressive behavior among sea anemones:
a recognition of self and not-self. Biol. Bull. 159: 117-134.
BIRKELAND, C. 1976. An experimental method of studying corals during early stages of growth. Micronesica
12: 319-322.
BRACE, R. C., J. PAVEY, AND D. L. J. QUICKE. 1979. Intraspecific aggression in the color morphs of the
anemone Actinia equina: the convention governing dominance ranking. Anim. Behav. 27: 553-
561.
CIERESZKO, L. S., ANDK. B. KARNS. 1973. Comparative biochemistry of coral reef coelenterates. Pp. 183-
203 in Biology and Geology of Coral Reefs, vol. II, O. A. Jones and R. Endean, eds. Academic
Press, New York.
CHORNESKY, E. A., AND S. L. WILLIAMS. 1983. Distribution of sweeper tentacles on Montastrea cavernosa.
In press in The Ecology of Deep and Shallow Coral Reefs. M. L. Reaka, ed. Symposium Series
on Undersea Research, Natl. Office of Undersea Research, NOAA, Rockville, Maryland.
CONNELL, J. H. 1973. Population ecology of reef-building corals. Pp. 205-245 in Biology and Geology of
Coral Reefs, vol. II, O. A. Jones and R. Endean, eds. Academic Press, New York.
CORAL SWEEPER TENTACLE DEVELOPMENT 581
CONNELL, J. H. 1976. Competitive interactions and the species diversity of corals. Pp. 5 1-68 in Coelenterate
Ecology and Behavior, G. O. Mackie, ed. Plenum Press, New York.
CONNELL, J. H. 1978. Diversity in tropical rain forests and coral reefs. Science 199: 1302-1310.
DEN HARTOG, J. C. 1977. The marginal tentacles of Rhodactis sanctilhomae (Corallimorpharia) and the
sweeper tentacles of Montastrea cavernosa (Scleractinia); their cnidom and possible function. Pp.
463-469 in Proc. Third Int. Coral Reef Sytnp, D. L. Taylor, ed. Univ. of Miami Press, Coral
Gables.
FRANCIS, L. 1973. Interspecific aggression and its effects on the distribution of Anthopleura elegantissima
and some related sea anemones. Biol. Bull. 144: 73-92.
GLYNN, P. W. 1973. Aspects of the ecology of coral reefs in the western Atlantic region. Pp. 271-324 in
Biology and Geology of Coral Reefs, vol. II, O. A. Jones and R. Endean, eds. Academic Press,
New York.
GLYNN, P. W. 1976. Some physical and biological determinants of coral community structure in the eastern
Pacific. Ecol. Monogr. 46: 431-456.
HILDEMANN, W. H., D. S. LlNTHlGUM, AND D. C. VANN. 1975. Transplantation and immunoincompatibility
reactions among reef-building corals. Immunogenetic 2: 269-284.
IVKER, F. B. 1972. A hierarchy of histo-incompatibility in Hydractinia echinata. Biol. Bull. 143: 162-174.
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. 1979. Morphological strategies of sessile animals. Pp. 499-555 in Biology and Systematics
of Colonial Organisms, G. Larwood and B. R. Rosen, eds. Academic Press, New York.
ICARLSON, R. H. 1980. Alternative competitive strategies in a periodically disturbed habitat. Bull. Mar.
Sd. 30: 894-900.
LANG, J. C. 1971. Interspecific aggression by scleractinian corals I. the rediscovery of Scolymia cubensis
(Milne Edwards and Haime). Bull. Mar. Sci. 21: 952-959.
LANG, J. C. 1973. Interspecific aggression by scleractinian corals II. why the race is not always to the swift.
Bull. Mar. Sci. 23: 260-279.
LEWIS, J. G., AND W. S. PRICE. 1975. Feeding mechanisms and feeding strategies of Atlantic reef corals.
J. Zool. Lond. 176: 527-544.
OTTAWAY, J. R. 1978. Population ecology of the intertidal anemone Actinia tenebrosa I. pedal locomotion
and intraspecific aggression. /. Mar. Freshwater Res. 29: 787-802.
PORTER, J. W. 1972. Patterns of species diversity in Caribbean reef corals. Ecology 53: 744-748.
PORTER, J. W. 1974. Community structure of coral reefs on opposite sides of the isthmus of Panama.
Science 186: 543-545.
PORTER, J. W. 1976. Autotrophy, heterotrophy, and resource partitioning in Caribbean reef-building corals.
Am. Nat. 110: 731-742.
POTTS, D. C. 1977. Growth interactions among morphological variants of the coral Acropora palifera. Pp.
79-88 in Coelenterate Ecology and Behavior, G. O. Mackie, ed. Plenum Press, New York.
PURCELL, J. E. 1977. Aggressive function and induced development of catch tentacles in the sea anemone
Metridium senile (Coelenterata, Actiniaria). Biol. Bull. 153: 355-368.
PURCELL, J. E., AND C. L. KITTING. 1982. Intraspecific aggression and population distribution of the sea
anemone Metridium senile. Biol. Bull. 162: 345-359.
RICHARDSON, C. A., P. DUSTAN AND J. C. LANG. 1979. Maintenance of living space by sweeper tentacles
of Montastrea cavernosa. Mar. Biol. 55: 181-186.
RINKEVICH, B., AND Y. LOYA. 1983. Intraspecific competitive networks in the Red Sea Coral Stylophora
pistallata. Coral Reefs 1: 161-172.
SAMMARCO, P. W., J. C. COLL, S. LA BARRE, AND B. WILLIS. 1983. Competitive strategies of soft corals
(Coelenterata: Octocorallia): allelopathic effects on selected scleractinian corals. Coral Reefs 1:
173-178.
SHEPPARD, C. R. C. 1979. Interspecific aggression beween reef corals with reference to their distribution.
Mar. Ecol. Prog. Ser. 1: 237-247.
SHEPPARD, C. R. C. 1982. Coral populations on reef slopes and their major controls. Mar. Ecol. Prog. Ser.
7: 83-115.
THEODOR, J. L. 1970. Distinction between "self and "not-self in lower invertebrates. Nature 221: 690-
692.
WAHLE, C. M. 1980. Detection, pursuit and overgrowth of tropical gorgonians by milleporid hydrocorals:
Perseus and Medusa revisited. Science 209: 689-69 1 .
WATSON, G. M., AND R. N. MARISCAL. 1983. The development of a sea anemone tentacle specialized for
aggression: morphogenesis and regression of the catch tentacle of Haliplanella luciae (Cnidaria,
Anthozoa). Biol. Bull. 164: 506-517.
WELLINGTON, G. M. 1980. Reversal of digestive interactions between Pacific reef corals: mediation by
sweeper tentacles. Oecologia 47: 340-343.
WELLS, J. W. 1973. New and old scleractinian corals from Jamaica. Bull. Mar. Sci. 23, 16-58.
Reference: Biol. Bull. 165: 582-590. (December, 1983)
ANALYSIS OF HEMOLYMPH OXYGEN LEVELS AND ACID-BASE
STATUS DURING EMERSION 'IN SITU' IN THE
RED ROCK CRAB, CANCER PRODUCTUS1
PETER L. DEFUR2, BRIAN R. MCMAHON, AND CHARLES E. BOOTH
Department of Biology, University of Calgary, Calgary, Alberta, Canada T2N 1N4
ABSTRACT
Hemolymph samples were taken from small (<100 g) individuals of Cancer prod-
uctus following ca. 3 h air exposure (emersion) on the beach, "in situ\ at Friday Harbor,
Washington. Compared with crabs of similar size in sea water in the laboratory, these
crabs emersed 'in situ had lower Pao2, and Pv02, but no significant change in pH
and a small, not significant, internal hypercapnia. Total CO2 (CCo2) content of the
hemolymph was elevated by 70% (15.2 versus 9.0 mM), possibly as compensation
for input of acid into the hemolymph. These responses are qualitatively similar to
those resulting from similar treatment in the laboratory, but differ in the reduced
magnitude of the internal hypercapnia and acidosis of the hemolymph. It is suggested
that the particular conditions of emersion 'in situ' permit some gas exchange with
interstitial sea water. Interstitial sea water was found to be hypoxic (P02 = 20-40
torr), which would limit oxygen supply yet permit CO2 excretion to continue, in
agreement with the data.
INTRODUCTION
Intertidal decapod crustaceans may face exposure to air and hence the transition
from aquatic to aerial respiration on a daily or more frequent basis, depending on
tide cycles and amplitudes. Reasonably complete patterns of respiratory responses
during short term air exposure (emersion) have been described for two marine crabs,
Carcinus maenas (Truchot, 1975; Taylor and Butler, 1978) and Cancer productus
(deFur and McMahon, 1984a, b) and also for freshwater crayfish, Austropotamobius
pallipes (Taylor and Wheatly, 1980). Less complete patterns of response to emersion
have also been described for several other marine crabs (McDonald, 1977; O'Mahoney,
1977; Batterton and Cameron, 1978). These laboratory studies indicate that short
term emersion is associated with an acidosis, which may be respiratory, as in Carcinus
maenas (Truchot, 1975; Taylor and Butler, 1978) or mixed respiratory and metabolic
as in Austropotambius pallipes (Taylor and Wheatly, 1980) and Cancer productus
(deFur and McMahon, 1984b). Compensation for the acidosis in all species studied
occurs largely via a rise of hemolymph bicarbonate, although the process is still
incomplete in 3-4 h. Gas exchange during emersion is probably diffusion limited in
all species studied and may also be perfusion limited in some species (deFur and
McMahon, 1984a). All of these studies, however, have been conducted in the laboratory
Received 17 March 1983; accepted 21 September 1983.
1 Supported by NSERC grant A5762.
2 Supported by Province of Alberta Graduate Fellowship.
Present address of P. L. deFur: Department of Biology, George Mason University, Fairfax, Vir-
ginia 22030.
582
IN SITU EMERSION OF C. PRODUCTUS 583
and it is not known how closely these 'laboratory' responses mimic those occurring
"in situ" under natural conditions.
The present study, therefore, describes hemolymph acid-base conditions and oxygen
levels in the crab. Cancer productus Randall during "in situ" emersion at low tide on
the beach. These hemolymph samples provide respiratory data obtained from crabs
air exposed in their natural habitat and unaffected by previous laboratory manipu-
lations.
MATERIALS AND METHODS
This study was undertaken at the Friday Harbor Laboratories of the University
of Washington on San Juan Island, Washington. Hemolymph was sampled while
animals were air exposed on the beach "in situ'' at low tide. Animals from the same
vicinity were also collected and maintained in flowing natural sea water at the Friday
Harbor Laboratories to provide comparative data from immersed crabs. All crabs
held in the laboratory were kept at ambient sea water temperature (9- 1 0°C) in darkened
aquaria provided with 5-10 cm of fine sand and gravel substrate. These crabs were
fed 2-3 times weekly except within 24 h of experiments.
Initially, an intertidal area where individuals were routinely emersed at low tide
was located, the time of exposure noted, and approximately 3 h later, hemolymph
samples were taken. Crabs were usually buried in the substrate beneath rocks or kelp;
hence it was necessary first to lift the kelp or a rock, locate a crab, and then rapidly
take the hemolymph samples. All hemolymph samples were withdrawn into iced, 1
ml glass syringes, which were immediately sealed and replaced on ice. Postbranchial
(arterial) samples were taken by carefully puncturing the dorsal carapace, anteriolateral
to the heart using the syringe needle, and then withdrawing 0.2-0.4 ml of hemolymph.
Prebranchial (venous) samples were taken from the base of the fifth walking leg by
gently restraining the crab and lifting the posterior end partially out of the substrate.
The iced samples were then returned to the laboratory for analysis. Burnett and
Bridges (1981) report that sealed, ice hemolymph samples may be kept for at least
1 h with no significant changes in acid-base or O2 variables. This conclusion was
tested and verified using three samples in the present study.
Postbranchial samples could be obtained quickly and with a minimum of dis-
turbance to the animals because the crabs remained motionless in the substrate.
However, partial removal from the substrate, as was necessary during prebranchial
sampling, or repeated prodding always provoked evasive behavior. Thus, if a pre-
branchial sample could not be obtained swiftly on the first attempt, the sample was
discarded. Both postbranchial and prebranchial samples were obtained sequentially
from 8 crabs fully emersed "in situ\ and these were treated statistically as paired
samples.
Hemolymph samples were analyzed for pH, total CO2 (CC02), CO2 tension
(Pco2X and O2 tension (P02), although small sample volume frequently prohibited
making all measurements on each sample. Hemolymph pH was measured with a
Radiometer capillary electrode (G299A) thermostatted to 9-10°C and connected to
an acid-base analyzer (Radiomenter PHM 71). CCo2 was determined on 40 ^1 of
hemolymph using the method of Cameron (1971) with each sample measurement
preceded and followed by 1 5 ^1 standard injections of 30 mM NaHCO3 . Hemolymph
PC02 was measured using a Radiomenter electrode (E 5036-0) thermostatted to 9-
10°C and the signal displayed on an acid-base analyzer (Radiometer PHM 71) set
to 10X sensitivity. The electrode was calibrated with humidified gases of known
PCo2 delivered via a Wosthoff pump. Measures of P02 were made with a Radiometer
584
P. L. DEFUR ET AL.
electrode (E 5047), thermostatted to 10°C, and an acid-base analyzer (Radiomenter
PHM 71).
Statistical analyses
Statistical analyses were performed using Student's Mest for either grouped or
paired variates and the 0.05 level was used as the criterion of significance. Regressions
were performed via the least squares estimation. Mean values (x) in the text are given
± one standard error (S.E.).
RESULTS
General observations
Individuals of Cancer productus were abundant in shallow water (<1.5 m) in the
vicinity of the Friday Harbor Laboratories during November, 1979, but large numbers
of crabs were found air exposed at low tide only in Beaverton Cove. This particular
area was protected, permitting the growth of a large kelp bed which covered the lower
intertidal zone during low tides. The substrate was predominantly coarse sand mixed
with fine gravel and restricted areas of loose, fine gravel. During low tide, C. productus
were most often found buried in the substrate beneath rocks or kelp with only the
most anterior-dorsal aspect of the shell protruding above the sand. Crabs were observed
emersed on top of the substrate only once and all but one of these were beneath
thick layer of kelp. On several occasions, crabs were found buried in fine substrate
which still held noticeable amounts of interstitial sea water. Postbranchial hemolymph
samples were obtained from 7 of these crabs and 4 samples of the interstitial water
were also obtained for measurement of P02 .
The mean weight of 30 crabs, which were not sampled but collected and returned
to the laboratory was 29.63 ± 2.40 g. Data from these animals were used to describe
the relationship between wet weight and carapace width (Fig. 1), from which the
mean wet weight of crabs sampled 'in situ' was estimated to be 21.02 ± 2.16 g. It is
interesting to note that there is a semilogarithmic relationship between carapace width
and wet weight. Tides which were sufficiently low to result in emersion of C productus
30
60
£ 40
0)
I/)
1/1
03 20
10L
4.0 5.0 6.0
Carapace Width
7.0
(cm)
FIGURE 1. Semilogarithmic relationship between body mass and carapace width in C. productus,
used to calculate wet weight of crabs sampled in situ, r = 0.99.
IN SITU EMERSION OF C. PRODUCTUS 585
occurred after dark, therefore air temperatures during sampling did not differ sig-
nificantly from the sea water temperatures of 9-10°C.
Crabs immersed in the laboratory
Mean values of P0:, pH, CC02, and PC02 in pre- and postbranchial hemolymph
of immersed crabs held in the laboratory for 3-10 days are given in Table I. Pao2
and Pv0, were typical of values reported previously for small C. productus exhibiting
primarily unilateral ventilation (deFur and McMahon, 1984a). The acid-base system
of immersed crabs in the laboratory was characterized by a high pH, low PCo2^ an<3
low CCo2 (Table I). These values and P02 levels of crabs immersed in the laboratory
were used as a baseline with which to compare emersed crabs in situ.
Hemolymph samples were obtained from 22 small C. productus which had been
emersed on the beach, in situ for approximately 3 h. Pao2 and Pv02 were significantly
lower in crabs emersed in situ than in immersed crabs, yet there remained a significant
difference between Pao2 and Pv02 of 6.85 torr (Table I). This difference was the same
regardless of whether the data were analyzed as grouped or as paired date, i.e., using
samples taken sequentially from the same crabs (Table I). Hemolymph O2 content
was not measured in the present study, yet the amount of oxygen delivered to the
tissues in crabs emersed in situ can be estimated using oxygen equilibrium curves
determined by deFur and McMahon (1984a) for hemolymph from crabs of similar
size at 10°C, 34%o (Fig. 2). In spite of the low in vivo P02's measured during in situ
emersion, hemocyanin was more than 70% oxygen saturated in transit through the
gills, and only 12% oxygen saturated in hemolymph returning from the tissues (Fig.
2). At a mean hemolymph oxygen carrying capacity of 0.466 mM (deFur and
McMahon, 1984a), this represents 0.141 mmol O2 per liter of hemolymph delivered
to the tissues. Unloading of oxygen from hemocyanin at the tissues was enhanced
by approximately 20% via a normal Bohr shift (see below).
TABLE I
Hemolymph oxygen tensions and acid base status ofC. productus during emersion
in situ and in interstitial water
Po2 (torr) PH CC02 (ml/) PC02 (torr)
Emersed in situ
postbranchial
12.38 ± 1.35
7.948 ± 0.023
15.23 ± 0.67
2.50 ± 0.22
(16)
(16)
(15)
(12)
prebranchial
5.85 ± 1.05
7.906 ± 0.031
15.96 ± 0.92
2.82 ± 0.29
(6)
(13)
(13)
(5)
pre-postbranchial (paired)' "
6.3 ± 1.9
0.072 ± 0.020
1.78 ± 0.47
0.267 ± 0.09
(6)
(8)
(8)
(5)
Interstitial
/// situ
16.86 ± 2.32
7.899 ± 0.055
15.89 ± 0.74
2.29 ±0.21
(7)
(7)
(7)
(6)
Immersed in laboratory
postbranchial
58.85 ± 10.7
7.960 ± 0.03
8.95 ± 0.75
1.97 ±0.31*
(13)
(12)
(7)
(7)
prebranchial
19.04 ± 3.0
7.921 ± 0.033
—
—
(14)
(13)
Mean ± S.E. (N).
1 ' Paired samples were taken sequentially and the data analyzed as paired variates.
* Calculated using the method of Wilkes et al. (1980).
586
P. L. DEFUR ET AL.
J /
20
40
PCX
60
(torn)
80 140
FIGURE 2. Oxygen binding curves for hemocyanin of small C productus at pHa = 7.98 and pHv
= 7.90 using data from deFur and McMahon (1984a) and deFur (1980).
Although hemolymph pH during in situ emersion was not significantly different
from corresponding values in immersed crabs (Table I) other variables in the acid-
base system of hemolymph in crabs emersed in situ were nonetheless dissimilar.
Paco2 was slightly (not significantly) higher in emersed crabs 'in situ\ but Caco2 was
significantly (70%) higher (P < .05) (Table I), indicating a large base excess. Sequential
samples of pre- and postbranchial hemolymph from these naturally emersed crabs
exhibited significant differences between the mean values of all three acid-base variables
(paired observations). Pvc02 and Cvc02 were higher and pHv lower than the corre-
sponding values for postbranchial hemolymph, indicating that branchial excretion
of CO2 continued during emersion (deFur and McMahon, 1984b).
Postbranchial hemolymph samples were also taken from seven crabs emersed in
situ but buried in substrate containing obvious interstitial sea water. These crabs were
clearly able to circulate some of this water through the branchial chambers since
water could often be seen flowing from the exhalant branchial apertures. Acid-base
conditions of hemolymph in these animals were, however, not significantly different
from crabs emersed in adjacent but drier areas, although Paco2 was slightly (P > .05)
lower. Mean Pao2 of the crabs obviously utilizing interstitial sea water was only 4.5
torr higher (P > ,05) than in those from drier areas, but was significantly reduced
from that of immersed crabs. This low mean Pao2 was likely a consequence of the
hypoxic nature of the interstitial water (P02 = 27 ± 4.5 torr; n = 4).
IN SITU EMERSION OF C. PRODUCTUS
587
The responses of small C. product us to emersion in substrate containing interstitial
water was further investigated in the laboratory. Ambient P02 fell from 1 50 torr to
5 1 torr in the first hour and decreased further to 3 1 torr by 4 h. Hemolymph Pao2
fell rapidly during initial exposure (Fig. 3), and continued to decline slowly; mean
Pao: over the 0.75-4.0 h period was only 14.3 ± 1.5 torr. Mean Pao2 of samples
taken from these crabs was not significantly different from mean Pao2 of either group
of crabs sampled in situ on the beach. Hemolymph pHa of crabs in interstitial water
in the laboratory was quite variable (Fig. 3) and the mean was not significantly
different from that of any of the groups of crabs sampled on the beach. Hemolymph
Caco2 of crabs exposed to interstitial water in the laboratory increased linearly during
4 h (Fig. 3), reaching levels similar to those in crabs emersed in situ.
DISCUSSION
The data obtained at Friday Harbor for crabs immersed in flowing natural sea
water, at sea level, in the laboratory at 9-10°C and 34%o salinity, compare well with
those obtained at similar temperature and salinity in a recirculating sea water system
at an altitude of 1050 m in Calgary (Table II). Pao,, Paco^ and Caco2 were slightly
?
'mmH
100r
8O-
60
9) 40
20
0
PH
81-
79-
77
75
0
Time
FIGURE 3. Postbranchial hemolymph P02, CC02, and pH in individual small C. produclus emersed
for 4 h in substrate containing interstitial sea water in the laboratory. Line fitted by eye for P02, by least
squares estimation for CCo2 (r = 0.99), and through x for pH. Symbols represent individual values.
588 P. L. DEFUR ET AL.
TABLE II
Hemolymph P0, and acid-base status of small C. productus immersed in sea water (10°C, 32-35%o
salinity) and the changes resulting from 3-4 h emersion in the laboratory and in situ
Immersed crabs
Location
Pao, (torr)
pHa
Caco2 (mA/)
Pacoj (torr)
Calgary'
50.7 ± 8.0
8.017 ±0.02
7.31 ±0.44
1.33 ± 0.05
(8)
(10)
(7)
(7)
Friday Harbor2
58.9 ± 1 1
7.960 ± 0.03
8.95 ± 0.75
1.97 ±0.31*
(11)
(12)
(7)
(7)
Changes during emersion
APao2
(torr)
APv02
(torr)
ApHa
APac02
(torr)
ACC02
(mA/)
pHa-pHv
In situ2 (3-4 h)
(Friday Harbor)
Laboratory1 (4 h)
(Calgary)
-46.5
-37.2
-13.2
-10.7
-0.012
-0.147
+0.53
+2.27
+6.28
+8.72
0.072
0.034
1 deFur and McMahon, 1984b.
2 Table I.
* Calculated using the method of Wilkes el al. (1980).
Data are x + 1 S.E. (n).
higher and pHa slightly lower in Friday Harbor than in Calgary as might be expected
from the change in altitude, but none of these differences was significant. deFur and
McMahon (1984a) also observed similar respiratory behavior patterns in immersed
C. productus regardless of location. These observations indicate that the respiratory
status of C. productus is affected little by the differences between experimental con-
ditions in Calgary and those more similar to the natural habitat.
The present data are the first hemolymph acid-base status or oxygen tensions
reported for decapods in situ during air exposure. A greater degree of variability than
usually occurs in laboratory studies was noted in some variables, perhaps because
factors such as nutritional state and molting stage are not controlled, as under laboratory
conditions. An important aspect of the present study is the qualitative similarity
between the responses of small C. productus to emersion on the beach in Friday
Harbor and in the laboratory in Calgary (Table II); under both experimental regimes
P02 and pH decreased, and CC02 and PCo2 increased. The decreases in both Pao2 and
Pv0;, were greater under natural conditions than in the laboratory, but these differences
between responses in situ and in the laboratory are not significantly different. Ad-
ditionally, under both conditions, hemocyanin is well oxygenated at the gill and
most of the O2 is removed in passage through the tissues (Fig. 2 and deFur and
McMahon, 1984a).
Crabs emersed under laboratory conditions (deFur and McMahon, 1984b) ex-
hibited a marked acidosis due in part to a significant increase in Pco2- In contrast,
crabs emersed "in situ' showed neither a significant acidosis nor increase in Pco2- The
small decrease in pH in these crabs (Table I) was less, however, than would be expected
on the basis of the in vitro buffering properties (deFur and McMahon, 1 984b), suggesting
that more effective compensation occurred "in situ\ The more than 6 mM increase
of Cco2 implies that there is some net input of acid which is compensated by elevation
of HCO3~. The relative contribution of other acids, especially metabolic ones such
IN SITU EMERSION OF C. PRODUCTUS
589
as lactic acid, to the acid-base status of crabs emersed 'in situ1 is not known. Thus,
the present study cannot identify with certainty the compensatory mechanisms in-
volved. However, the greater pHa-pHv difference and lower Pac02 measured in crabs
emersed "in situ' suggest that CO2 excretion may be more effective under these con-
ditions.
Maintenance of branchial CO2 excretion implies maintained ventilation and per-
fusion of the gills during emersion. deFur and McMahon ( 1984a) measured maintained
sub-ambient branchial pressures in small C. productus during emersion in the lab-
oratory, and reasoned that interstitial sea water could be aspirated into the branchial
chamber. This water could allow CO2 excretion to continue during emersion but
seems to have no effect on O2 uptake since Pao2 is depressed (Table I). This situation
is not paradoxical since a) CO2 diffuses more effectively in aqueous systems, and b)
interstitial sea water samples, though more highly oxygenated than finer sediments,
were still hypoxic. Thus, irrigation of the gills with interstitial sea water could allow
CO2 excretion with little effective oxygenation. Under the laboratory conditions used
by deFur and McMahon (1984a, b), care was taken to remove as much sea water,
including interstitial, as possible, precluding its use for branchial functions.
The observed acid-base changes during emersion in situ show a discrepancy between
measured and calculated PCo2 similar to that observed in the laboratory (deFur, Wilkes
and McMahon, 1 980). This discrepancy is clearly apparent on a "Davenport diagram"
(Fig. 4) and precludes use of such a diagram for analysis of the acid-base system. A
discrepancy occurs only during emersion and was associated with large, rapid elevations
of hemolymph CCo2i indicating dynamic rather than steady-state conditions. As noted
by deFur et al. (1 980), data from crabs immersed in sea water are described perfectly
on the Davenport diagram.
In a similar study, Toulmond (1973) described the responses of the intertidal
polychaete Arenicola marina during 4 h emersion "in situ\ Arenicola also experienced
a decrease of Pv02, nearly exhausting the otherwise substantial venous oxygen reserve.
CO.
(torn)
6-0 50
40
150
(mM)
120
9-0
60
3-0
2-0
10
790 800
PH
FIGURE 4 "Davenport diagram" relating CC02. PH, and PCOZ in the hemolymph of C. productus
according to the method of Wilkes et al. (1980). The diagonal line ( ) represents the in vitro buffer
capacity. Points depict mean in vivo values from Table I with measured PC02 given in (D) beneath the
symbol.
590 P. L. DEFUR ET AL.
Simultaneously, there was an internal hypercapnia with a subsequent acidosis (re-
spiratory) and rise of blood bicarbonate (Toulmond, 1973). This author concludes
that gas exchange is impaired under these conditions and anaerobiosis occurs, con-
tributing a metabolic component to the acidosis. The responses of small C. productus
under similar conditions (Table I) are qualitatively similar to those of Arenicola, but
are quantitatively quite different. The decrease in P02 and pH and the increase in
PCo2 are ^ss in small C. productus. These differences are likely due to some air
breathing capability of the crabs, and availability and utilization of sea water during
emersion. Arenicola marina ceases all ventilation, normally accomplished by body
movements forcing water through the burrow. Small C. productus, however, are able
to utilize the hypoxic interstitial sea water, permitting CO2 excretion but limiting
oxygen supply.
Small C. productus occupy a restricted habitat within the intertidal zone and
during air exposure remain buried in the substrate in locations where sea water drains
from the substrate relatively slowly. In this condition, the small crabs can maintain
acid-base balance for the few hours of emersion, yet must endure a reduction in
oxygen supply. Thus, these small crabs which have access to interstitial water may
not be able to maintain oxygen uptake in air, but do not have the problem of carbon
dioxide excretion which is the major respiratory problem of truly intertidal crabs and
true air breathers.
ACKNOWLEDGMENTS
The authors wish to thank the Director and staff of the Friday Harbor Laboratories
of the University of Washington for their cooperation and assistance. The fine technical
assistance of Alan W. Finder is gratefully acknowledged.
LITERATURE CITED
BATTERTON, C. V., AND J. N. CAMERON. 1978. Characteristics of resting ventilation and response to
hypoxia, hypercapnia, and emersion in the blue crab, Callinectes sapidus (Rathbun). /. Exp. Zool.
203:403-418.
BURNETT, L. E., ANDC. R. BRIDGES. 1981. The physiological properties and function of ventilatory pauses
in the crab. Cancer pagiirns. J. Comp. Physiol. 145B: 81-88.
CAMERON, J. N. 1971. Rapid method for determination of total carbon dioxide in small blood samples.
J. Appl. Physiol. 31: 632-634.
DEFuR, P. L., AND B. R. McMAHON. 1984a. Physiological compensation to short term air exposure in
Red Rock crabs, Cancer productus Randall, from littoral and sublittoral habitats. I. Oxygen uptake
and transport. Physiol, Zool. accepted.
DEFUR, P. L., AND B. R. McMAHON. 1984b. Physiological compensation to short term air exposure in
Red Rock crabs. Cancer productus Randall, from littoral and sublittoral habitats. II. Acid-base
balance. Physiol. Zool. accepted.
DEFUR, P. L., P. R. H. WILKES, AND B. R. McMAHON. 1980. Non-equilibrium acid-base status in C.
productus: role of exoskeletal carbonate buffers. Respir. Physiol. 42: 247-26 1 .
MCDONALD, D. G. 1977. Respiratory Physiology of the Crab, Cancer magister. Ph.D. Thesis, Department
of Biology, University of Calgary, Calgary, Alberta, T2N 1N4, Canada.
O'MAHONEY, P. M. 1977. Respiratory and Acid-base Balance in Brachyuran Decapod Crustaceans: the
Transition From Water to Land. Ph.D. Thesis, State University of New York, Buffalo, New York.
TAYLOR, E. W., AND M. G. WHEATLY. 1980. Ventilation, heart rate and respiratory gas exchange in the
crayfish Austropotamobius pallipes (Lereboullet) submerged in normoxic water and following 3
h exposure to air at 15°C. J. Comp. Physiol. 138B: 67-78.
TAYLOR, E. W., AND P. J. BUTLER. 1978. Aquatic and aerial respiration in the Carcinus maenas (L.),
acclimated to 15°C. J. Comp. Physiol. 127B; 315-323.
TRUCHOT, J. P. 1975. Blood acid-base changes during experimental emersion and reimmersion of the
intertidal crab Carcinus maenas (L.) Respir. Physiol. 23: 351-360.
TOULMOND, A. 1973. Tide-related changes of blood respiratory variables in the lugworm Arenicola marina
(L.). Respir. Physiol. 19: 130-144.
WILKES, P. R. H., P. L. DEFUR, AND B. R. McMAHON. 1980. A new operational approach to PC02
determination in crustacean hemolymph. Respir. Physiol. 12: 17-28.
Reference: Bio/. Bull. 165: 591-618. (December, 1983)
EXPERIMENTAL STUDIES ON EMBRYOGENESIS IN HYDROZOANS
(TRACHYLINA AND SIPHONOPHORA) WITH DIRECT DEVELOPMENT
GARY FREEMAN
Friday Harbor Laboratories, University of Washington and the Department of Zoology,
University of Texas, Austin, Texas 78712*
ABSTRACT
The normal embryology of the trachymedusa Aglantha digitale and the sipho-
nophores Nanomia cara and Muggiaea atlantica is described. Marking experiments
on these embryos indicate that the site of first cleavage initiation corresponds to the
oral pole of the oral-aboral axis. In Muggiaea the plane of the first cleavage corresponds
to the plane of bilateral symmetry. Experiments in which presumptive aboral and
oral regions are isolated from these embryos at different stages of development indicate
that there is an early determination of different regions along this axis. Only the oral
region of the Muggiaea embryo has the ability to regulate. These eggs have a pro-
nounced centrolecithal organization. As a consequence of cleavage, the outer ecto-
plasmic layer of the egg ends up in the cells that form the ectoderm, while the inner
or endoplasmic region of the egg ends up in the cells that form the endoderm.
Experimentally created fragments of fertilized eggs that contain only ectoplasm dif-
ferentiate to form an unorganized ectodermal cell mass, indicating that endoplasm
is necessary in order to differentiate endoderm.
The process of embryogenesis in these animals and the developmental mechanisms
they use are very different from those used by hydrozoans with indirect development.
These embryos use a suite of developmental mechanisms which are very similar to
those used by ctenophores. The significance of this similarity is discussed.
INTRODUCTION
From a developmental standpoint, the class Hydrozoa in the phylum Cnidaria
appears to be quite diverse (Tardent, 1978). All of the experimental studies that have
tried to define the mechanisms that underlie early embryogenesis in this group have
been done on one order, the Hydroida. (Table I presents the taxonomic classification
of the class Hydrozoa used in this paper.) On the basis of this experimental work a
list of the mechanisms that mediate early embryogenesis in this group has emerged
(see Discussion). One consequence of these mechanisms is that these embryos have
a remarkable ability to regulate (Teissier, 1931, Freeman, 1981).
Most of the species in the order Hydroida have a complex polymorphic life cycle.
In one major phase of the life cycle these animals are attached to a substrate and in
the other major phase they are free swimming animals which function in a pelagic
environment. The life cycle begins when an egg undergoes embryogenesis to generate
a planula larva which undergoes metamorphosis to form a sessile polyp. Typically
the polyp forms a colony that buds free swimming medusae. The medusae grow in
size and develop gonads which form the gametes that are the basis for the next
generation.
Received 25 April 1983; accepted 26 September 1983.
1 Author's permanent address.
591
592 G. FREEMAN
TABLE I
Classification of the class Hydrozoa to a sub order level
order Hydroida order Trachylina
Anthomedusae Trachymedusae
Leptomedusae Narcomedusae
Limnomedusae Pteromedusae
order Milleporina order Siphonophora
Cystonectae
order Stylasterina Physonectae
Calycophorae
order Actinulida
Not all orders of the class Hydrozoa have this kind of life cycle. In the order
Trachylina the egg develops via a planula directly into a medusa; the polyp stage is
absent or rudimentary. In the order Siphonophora the egg develops via a planula
into a larva with both medusoid and polypoid characteristics that forms a colony
composed of members of both types. The process of embryogenesis has been described
for a number of species from these two direct developing orders (see Metschnikoff,
1874, 1886; Brooks, 1886 for the Trachylina; Metschnikoff, 1874, Russell, 1938,
Carre, 1967, 1969 for the Siphonophora). Very little experimental work has been
done on embryogenesis in the Trachylina and Siphonophora (however, see Zoja,
1895; Maas, 1908; Carre, 1969). These two orders are normally found in deep water
in the open ocean; their life cycle occurs exclusively in a pelagic environment.
The present paper examines the process of embryogenesis in the trachymedusa
Aglantha digitals and the siphonophores Nanomia cam (Physonectae) and Muggiaea
atlantica (Calycophorae) from an experimental point of view. The results of these
experiments indicate that these animals share only a few developmental mechanisms
with the Hydroida. In each of these species there is a precocious segregation of
developmental potential and the embryos have only a limited ability to regulate.
The Trachylina and Siphonophora are not the only cnidarian orders that develop
directly. Some scyphozoan orders have a similar life cycle. Unfortunately no exper-
imental work has been done on early embryogenesis in these animals. However, in
the related radiate phylum Ctenophora, development is also direct. In these animals
the product of embryogenesis is a larva, but in most species no major polymorphic
change in the anatomy of the larva occurs as it grows into an adult. The embryology
of these animals has been studied in some detail from an experimental viewpoint
(see Reverberi, 197 1 for a review). In this group there is also a precocious segregation
of developmental potential. The life cycle of most ctenophores occurs exclusively in
a pelagic environment. In the discussion section of this paper the developmental
mechanisms used by the Hydroida and the Ctenophora will be compared. The sig-
nificance of the demonstration that Trachylina and Siphonophora embryos have a
suite of developmental mechanisms that are similar to those of the Ctenophora will
then be discussed.
MATERIALS AND METHODS
Animals
Aglantha and Nanomia were collected from the Friday Harbor laboratory dock
with a beaker attached to the end of a pole. These two species are not common in
HYDROZOAN EXPERIMENTAL EMBRYOLOGY 593
the surface waters at Friday Harbor during the spring and summer seasons. Their
distribution is patchy; during some years Aglantha appears to be much more abundant
than during other years. Aglantha is always more common than Nanomia. Muggiaea
eudoxids were collected by doing plankton tows 3-4 meters below the surface and
half way up East Sound on Orcas Island. Eudoxids are present there in the last half
of June, July, and August. Kozloff( 1974) was used for identifying the species employed
in this study. All of these species descriptions are based on animals found in the
Atlantic Ocean. While these animals resemble those found in the Atlantic, there are
some differences; it is not clear that these animals are identical to the Atlantic species.
Eggs were obtained through natural spawnings. It is difficult to predict the time
of spawning for these species. Bowls containing the animals were checked at 30 minute
to 2 hour intervals throughout the day for spawning. Aglantha tends to spawn between
0200-0400 and 1000-1200 hours. Nanomia tends to spawn one hour after it is
brought into the light. In the siphonophores the testes tend to become opaque 2-3
hours prior to spawning. In Aglantha and in Muggiaea eudoxids the sexes are separate.
In these species 2-4 sexually mature females were maintained in a bowl. When eggs
were found they were collected and a culture was set up by adding sea water containing
sperm from a bowl of males. A sexually mature Nanomia has both female and male
gonads. In this work eggs were fertilized by the sperm from the same individual as
they were spawned. The initiation of first cleavage was regarded as T0 for the purpose
of timing development.
The embryos were raised at 11-12°C. They were reared in millipore filtered
pasteurized sea water in wells (0.5-1.5 ml vol.) of spot plates. In many of the ex-
periments involving Aglantha blastomere isolates, 100 units/ml of penicillin was
added to the sea water; this significantly improved viability. The siphonophore embryos
frequently get caught at the air-water interface and are destroyed by surface tension
forces. In some experiments polyethylene oxide was added to the sea water (0.1 g/
10 ml) to increase its viscosity. The embryo develops normally in this medium, but
moves very little; as a consequence it is much less likely to get caught at the air-water
interface.
Experimental manipulations
Embryos were operated on in wells with a 2% agar bottom. Glass needles were
used as knives to cut the embryos into parts.
Early cleavage stage embryos were marked with chalk particles (chalk was used
because the vital dyes that were tried tended to diffuse throughout the early embryo).
A suspension of small chalk particles was produced by placing a drop of sea water
on a frosted glass slide and rubbing the tip of a stick of chalk in it. A small amount
of this suspension was placed in one corner on the agar surface of an operating dish
which was then filled with sea water. The part of the embryo surface to be marked
was placed in contact with one or more chalk particles and gently pressed against
the chalk with a blunt glass needle attached to a micromanipulator. This procedure
firmly attaches the chalk to the embryo's surface. Embryos at later stages were marked
with the stain nile blue; a 1% solution of the dye was prepared in distilled water. One
or more points on the surface of the embryo were marked by using a micromanipulator
to bring the open end of a fine capillary tube filled with 2% agar containing the dye
in contact with the surface of the embryo for a few minutes. Novikoff (1938) gives
directions for preparing these capillary tubes. The embryos tend to lose stain and
there is a diffusion of the stain into the endoplasmic region; however, the dyed spot
can usually be followed for 2-3 days. Since the embryos are translucent the dye spots
594 G. FREEMAN
on the surface can be observed even when they are very light by viewing the embryo
with a compound microscope under conditions of critical illumination. Frequently
embryos were first marked with chalk and subsequently remarked by staining. Too
much stain has a deleterious effect on the development of these embryos.
Eggs were centrifuged to create ectoplasmic and endoplasmic fragments. Cen-
trifugation stratifies the contents of the egg. In these species the yolky endoplasm
takes up a centripetal position and the ectoplasmic zone takes up a centrifugal position.
Following stratification the eggs elongate and may split into endoplasmic and ecto-
plasmic fragments. If they do not separate into fragments they can be easily cut into
fragments following centrifugation. Aglantha eggs were centrifuged in a mixture of
2 parts 1 molal sucrose and 1 part sea water for 15 minutes at 9500 rpm (10,800
X g). Nanomia eggs were centrifuged in a mixture of 1 part 1 molal sucrose and 1
part sea water for 10 minutes at 9500 rpm. The diameter of the fragments was
measured with a screw micrometer eye piece.
Histological procedures
Embryos and larvae were fixed in 1% osmium in cold sea water for one hour,
washed, dehydrated, and embedded in Epon. Sections were cut at 2 jum and stained
with methylene blue and azure II (Richardson et al., 1960).
RESULTS
Normal development oj Aglantha digitale
The normal development of Aglantha has not been described; however, Met-
schnikoff (1886) described the development of a related species, Aglaura hemistoma
from the Mediterranean. Figure 1 presents a series of photographs which outline the
development of Aglantha. The uncleaved egg has an average diameter of 1 39 /urn
(range 125-153 ^m, sample size 24). There is a membrane around the egg which is
closely applied to it; the embryo hatches out of the membrane during development
(Fig. 1 0- Polar bodies are not visible. Sections through fixed uncleaved eggs (Fig. 2a)
show a central zone containing large endoplasmic granules and a peripheral region
where these granules are absent. Cleavage is unipolar. The first two cleavages generate
four equal blastomeres. A number of embryos were marked at either the site of origin
of the first cleavage furrow (18 cases) or directly opposite this site (5 cases). These
marking studies showed that the second cleavage is always initiated at the same site
at the first cleavage. The third is unequal. Four micromeres are produced that are
largely devoid of endoplasmic granules (Figs. Ib, 2b). The embryos with the chalk
marks showed that the micromeres are given off opposite the site of first cleavage
initiation. During the fourth cleavage (Fig. Ic) the four macromeres divide equatorially
to form two tiers of macromeres along the axis specified by the first two cleavage
furrows. It is hard to follow cleavage beyond this point.
Gastrulation takes place during the next 3-4 hours. During this period the cells
that make up the micromere cap flatten and spread as a coherent layer over the
macromeres to create an ectodermal cell layer which surrounds the yolky macromeres
(Fig. 2c). The spreading movement can be followed by observing appropriately oriented
embryos at 10-15 minute intervals using Nomarski optics. It is not clear that epibolic
movement is the only mechanism of gastrulation; endoplasm free ectodermal cells
may also be generated by a cytokinesis which occurs tangentially to the external cell
membrane in some of the macromeres. As gastrulation takes place the embryo elongates
HYDROZOAN EXPERIMENTAL EMBRYOLOGY
595
F
H
•
FIGURE 1. Normal development ofAglanlha. A) Uncleaved egg. B) Eight cell stage. C) 16 cell stage
D) Five hour embryo. E) Eight hour embryo. F) 24 hour embryo. The arrow indicates the membrane that
surrounds the egg out of which the embryo is hatching. G) 40 hour embryo. Note the tentacle rudiments.
H) 54 hour embryo. I) Oral view of three day old embryo. The tentacles are contracted. The arrow indicates
a marginal sense organ. All photographs are at the same magnification. The bar indicates 50 ^m.
(Fig. Id). The marking experiments show that the direction in which the embryo
elongates corresponds to the axis of the first two planes of cleavage.
After gastrulation is completed, the ectodermal cells begin to form cilia. Over the
next 12 hour period the embryo hatches out of its membrane and begins to swim.
The planula rotates around its long axis as it moves forward; it does not reverse its
direction of movement. Experiments in which the chalk marks were replaced by dye
marks at 8-10 hours of development (8 cases) show that the site where cleavage is
initiated corresponds to the posterior or oral end of the planula.
Between 24 and 48 hours of development two tentacles begin to form opposite
each other in the oral region of the planula (Fig. Ig). At this point the planula begins
to transform into an actinula. Within a few hours after these two tentacles begin to
form, two more tentacles start to appear opposite each other between the first pair
of tentacles. During the next few days additional tentacles form. A tentacle is composed
of both ectodermal and endodermal cells; the ectoderm contains both nematoblasts
and nematocysts. The tentacles of an actinula are relatively rigid but they can contract
and change positions. Ciliary tracts that run the length of the tentacle serve as the
main locomotory organ of the actinula. The tentacles of the medusa have similar
ciliary tracts which beat in a coordinated manner. After four days of development
596
G. FREEMAN
c
FIGURE 2. Normal embryology of Aglantha. A) Section through an egg. Note the endoplasmic
granules. B) Section through an eight cell stage embryo. The arrows indicate the micromeres. Note the
relative paucity of endoplasmic granules in the micromeres. C) Section through seven hour embryo. Note
the ectoplasm containing ectodermal cells and the endoplasm containing endodermal cells. The ectodermal
nuclei have nucleoli. Note the change in the morphology of the endoplasmic granules between the onset
of cleavage and gastrulation. D) Section through a five day old larva. Note the gastrovascular cavity, C;
the gland cells associated with the mouth G; the digestive gland cells, D; and the endodermal cells which
line the gastrovascular cavity and make up the core of the tentacles, E. The bar indicates 50
marginal sense organs form between some of the tentacles. At the same time the
tentacles start to form, the manubrium begins to differentiate at the oral end of the
planula. At this site a heavily ciliated mouth forms between days two and three of
development. Between 3 and 4 days of development manubrial gland cells form
adjacent to the mouth. The endoderm, which is composed of highly vacuolated cells,
organizes itself into an epithelium and a space which is continuous with the manubrium
forms in the center of the larva. As these events occur a basement membrane forms
between the ectodermal and endodermal cell layers. By five days of development
gland cells with small vacuoles begin to appear between the large vacuolated endo-
dermal cells in the oral region of the larva (Fig. 2d).
HYDROZOAN EXPERIMENTAL EMBRYOLOGY
597
Experimental work on Aglantha
The ability of parts of embryos that normally form different germ layers or different
regions of the actinula larva, to differentiate these structures when isolated was studied
by doing the following experiments.
1 ) At the eight cell stage the micromeres and macromeres were isolated (Fig. 3a).
Since the micromeres form ectoderm and the macromeres form both ectoderm and
endoderm in the intact embryo, this experiment asks whether or not there is a seg-
regation of germ layer specific developmental potential at the eight cell stage. Nineteen
micromere and 25 macromere halves were raised for six days; four cases of each type
were sectioned. All of the micromere isolates formed a compact sphere (Figs. 4a, 5a).
The surface cells were ciliated but there was no indication of swimming polarity. In
every case nematocysts were present. There was no indication of a basement membrane
separating the external from the internal cells of the sphere. The development of the
macromere isolates was more variable (Figs. 4b, 5b). The isolates were spherical to
oblong in shape. All of them formed both ectoderm and endoderm; however, en-
dodermal gland cells did not differentiate. In all of the cases the ectodermal cells
appeared to be very thin in places or did not completely cover the endoderm. The
ectodermal cells were ciliated; in 13 cases the isolates showed swimming polarity.
Nematocysts were present in the ectoderm in 1 5 cases. Two of the isolates formed
a stubby tentacle and one of these cases formed a manubrium.
2) At the 16 cell stage the embryo was cut into two parts in such a way that one
isolate consisted of four oral macromeres and the other consisted of the micromeres
and their adjacent macromeres (Fig. 3b). This experiment was done to find out if an
embryo which lacks the endodermal cells that are normally present at the oral end
of the embryo still has the ability to form tentacles and a manubrium. Twenty-two
micromere and aboral macromere isolates were raised for six days; six of these cases
were sectioned. Fifteen of these isolates formed normal actinula larvae (Fig. 4c); three
cases formed a manubrium but no tentacles and one case formed tentacles but no
manubrium. The three remaining isolates formed ectoderm containing nematocysts
and endoderm, all of them showed swimming polarity. Sixteen oral macromere isolates
were raised for six days; three of these cases were sectioned. These cases resembled
the eight cell stage macromere isolates (Fig. 4d); however, they did not form nematocysts
and no case showed swimming polarity.
3) At eight hours of development (Fig. le) (5 hours after the 16 cell stage), the
embryo was cut into an oral and an aboral half (Fig. 3c). In order to identify each
half, these cases were marked at the site of origin of the first cleavage furrow. At eight
FIGURE 3. Operations performed to isolate parts of the Aglantha embryo at different stages of de-
velopment. A) Eight cell stage; isolation of micromeres and macromeres. B) 16 cell stage; isolation of
micromeres with aboral macromeres and oral macromeres. C) Eight hour embryo; isolation of oral and
aboral halves. D) Two cell stage; isolation of individual blastomeres. E) Eight hour embryo; isolation of
lateral halves, x, chalk mark placed at the site of origin of the first cleavage indicating the oral end of the
embryo. The dashed line indicates how the embryo was cut.
598
G. FREEMAN
A
D
F
G
H
Y
FIGURE 4. The development of isolates from Aglantha embryos. A) Five day old micromere isolate
from eight cell stage embryo. Note the lack of endoderm. B) Five day old macromere isolate from eight
cell stage embryo. The arrow indicates the boundary between the ectoderm and endoderm. C) Five day
old micromere and middle macromere isolate from a 16 cell stage embryo. D) Five day old oral macromere
isolate from 16 cell stage embryo. E) Five day old aboral half isolated from eight hour old embryo. F)
Three day old isolate from two cell stage embryo. The arrow indicates the ectodermal cap. G) Y, yolky
fragment and C, clear cytoplasmic fragment from a centrifuged egg. Note the nucleus in the clear cytoplasmic
fragment. H) Four day old embryo from clear cytoplasmic fragment. Note the lack of endoderm. All
photographs are at the same magnification. The bar indicates 50 ^m.
hours of development both regions of the embryo have ectodermal and endodermal
cell layers; a number of hours will elapse before there is an obvious indication of
organogenesis. This experiment was done to find out if the aboral half of the embryo
can regulate to form the tentacles and manubrium which are normally formed in
the oral half of the embryo. Twenty-one aboral halves were raised for six days; eight
of these cases were sectioned. All of the aboral halves developed swimming polarity.
Eighteen cases showed no indication of tentacle or manubrium formation (Fig. 4e).
Fourteen of these cases formed nematocysts at their oral end; the six embryos in this
category that were sectioned had small vacuole endodermal gland cells. Three cases
formed one or two stubby tentacles at their oral end, in all three cases nematocysts
were present. None of these cases showed any indication of forming a manubrium,
this point was checked by sectioning two of these cases. Sixteen oral halves were
raised for six days. All of these cases formed a normal larva with tentacles and a
manubrium.
4) Two kinds of control operations were performed. At the two cell stages each
blastomere was isolated (Fig. 3d) and at the eight hour stage the embryo was cut into
two lateral halves (Fig. 3e). These isolates contain both oral and aboral regions and
ectodermal and endodermal germ layers or have the potential to form these germ
layers. Twenty-eight two cell stage isolates were raised for six days; three of these
cases were sectioned. All of these cases formed both ectodermal and endodermal germ
layers. Eighteen of the isolates formed a ball of endodermal cells with a cap of
ectoderm covering primarily one end of the endoderm (Figs. 4f, 5c). Nematocysts
were frequently present in the ectodermal cell layer. Many of these isolates ( 1 4)
showed swimming polarity. The ectodermal cap was always at the aboral end of the
swimming isolate. (Eight isolates had a mark indicating where the first cleavage was
HYDROZOAN EXPERIMENTAL EMBRYOLOGY 599
A .. A , . C
,
B «~ D
mfr i
%
\
•«
FIGURE 5. The histology of sectioned isolates from Aglantha embryos. A) Two five day micromere
isolate from eight cell stage embryo. Note the lack of endoderm. B) Five day macromere isolate from eight
cell stage embryo. Note the places where the ectoderm does not completely surround the endoderm. C)
Five day isolate from two cell stage embryo. The endoderm is surrounded by ectoderm and there is an
ectodermal cap at one end of the embryo. D) Four day embryo from clear cytoplasmic fragment. Note
the lack of endoderm. All photographs are at the same magnification. The bar indicates 50
initiated; the cap was opposite the mark.) These cases probably failed to gastrulate
properly. The remaining cases (10) formed more or less normal actinula larvae with
one or more tentacles and a manubrium. Fifteen, eight hour lateral half isolates were
raised for six days. Twelve of these cases formed actinula larvae with one or more
tentacles and manubrium. These experiments show that the patterns of development
seen in experiments 1-3 cannot be ascribed to the operative procedures used, but
must reflect a program of differentiation inherent in the various regions of the embryo
at the time these regions were isolated.
5) The last experiment investigated the effect of the yolky endoplasm on devel-
opment. Nucleated egg fragments that lacked endoplasm were produced by centrifuging
fertilized eggs and examining the ability of the resulting fragments to develop. Figure
4g shows the ectoplasmic and endoplasmic fragments produced by centrifugation.
The average diameter of the ectoplasmic fragments was 90 yum (range 81-95 ^m,
sample size 35), while the average diameter of the endoplasmic fragments was 126
^m (range 109-156 /urn, sample size 28). The ectoplasmic fragments contained ap-
proximately 27% of the egg volume. Five ectoplasmic fragments were sectioned; they
600
G. FREEMAN
contained the nucleus and all of the visible egg constituents except the larger en-
doplasmic yolk granules. None of the endoplasmic fragments that were produced
cleaved (115 cases). Sixty-two (38%) of the ectoplasmic fragments cleaved compared
with approximately 70% for the untreated control eggs from the same batches. The
first two cleavages of the ectoplasmic fragments are normal; however, at the third
cleavage micromeres are not formed. There was no indication of gastrulation. After
24 hours of development a solid ciliated ball of small cells formed. Most of these
isolates disintegrated between the second and third day of development. It was possible
to raise 13 cases for five days (Figs. 4h, 5d); four of these cases were sectioned. They
resembled eight cell stage micromere isolates, but they lacked nematocysts. They
showed no swimming polarity; they lacked endoderm and showed no indication of
manubrium or tentacle formation. This experiment shows that in the absence of
endoplasm, endoderm will not differentiate.
Normal development of Nanomia cara and Muggiaea atlantica
The normal development of a Mediterranean species of Nanomia (Metschnikoff,
1874; Carre, 1969) and M. atlantica and a related Mediteranean species of Muggiaea
(Metschnikoff, 1874; Russell, 1938) have been described. Figures 6 and 7 present a
D
F
G.
H
<
i\
• *
FIGURE 6. Normal development of Nanomia. A) First cleavage. Note the unipolar furrow. B) Eight
cell stage. C) Six hour embryo. D) 1 2 hour embryo. E) 1 8 hour embryo. F) 44 hour embryo. Vacuolated
cells have begun to form in the anterior region of the planula (arrow). Anterior and lateral endodermal
thickenings are present. G) 57 hour embryo. The arrow indicates the tentacle rudiment. H) Four-and-a-
half day old siphonula. Note the pneumatophore rudiment P, the tentacle with cnidobands at its base T,
the oral pigment O, and the first indications of gastric cavity formation. I) Seven day old functional
siphonula. Note the cnidobands on the tentacle. All photographs are at the same magnification. The bar
indicates 50 j/m.
HYDROZOAN EXPERIMENTAL EMBRYOLOGY 601
series of photographs that outline the development of Nanomia and Muggiaea, re-
spectively. The uncleaved Nanomia egg has an average diameter of 274 ^rn (range
252-292 ^m, sample size 13), while the Muggiaea egg has an average diameter of
319 nm (range 307-331 ^m, sample size 17). The Muggiaea egg is less dense than
sea water. There are no membranes around these eggs. In Muggiaea the polar bodies
are associated with an extracellular structure, the cupule (Carre and Sardet, 1981)
that tends to fall off the egg shortly after fertilization. Sections through uncleaved
eggs show a central zone containing large endoplasmic granules and a peripheral
region where these granules are absent (Fig. 8a). The distinction between the ecto-
plasmic zone and the endoplasm is much sharper in these eggs than it is in the
Aglantha egg. Cleavage is unipolar; the first two cleavages generate four equal blas-
tomeres. In both species the site of origin of the second cleavage furrow was established
by marking the site of origin of the first cleavage furrow or the point directly opposite
this site. The second cleavage furrow occurred at the site of origin of the first cleavage
furrow in 64% of the cases (sample size 22) for Nanomia and in 84% of the cases
(sample size 18) for Muggiaea. In the remaining cases the second cleavage was initiated
at the equator. The variable origin of the second cleavage furrow has been noted in
Nanomia by Carre (1969), it has also been observed in other hydrozoans (Teissier,
1931; Freeman, 1981). The third cleavage is always perpendicular to the preceding
cleavage and gives rise to two tiers of blastomeres with four equal sized cells in each
tier (Figs. 6b, 7b).
Gastrulation begins following the 64 cell stage. Prior to gastrulation each blastomere
contains part of the initial surface of the egg. The cell nucleus and the ectoplasmic
region of the egg are found here. Gastrulation occurs when a tangential cell division
gives an external daughter cell that contains the ectoplasmic region and an internal
daughter cell that contains the endoplasm (Fig. 8b-d). Gastrulation is completed by
six to seven hours of development in both species.
The Nanomia embryo begins to elongate between 12 and 18 hours of development;
cilia develop and the embryo begins to show swimming polarity. The embryo has
now transformed into a planula. Experiments in which the site of origin of the first
cleavage furrow was marked show that this region corresponds to the posterior end
of the planula (16 cases). During the next six hour period ( 1 8-24 hours of development)
the embryo continues to elongate and an endodermal thickening begins to develop
at the anterior end and along one side of the planula. At the same time the anterior
ectodermal cells begin to enlarge (Fig. 60- Between 42 and 54 hours of development
the ectoderm at the anterior end of the planula invaginates and begins to form the
pneumatophore, while a tentacle begins to grow out from the side of the planula
where the endoderm has thickened. At this point the planula begins to transform
into a siphonula larva. By seven days a feeding siphonula has developed. During this
period the pneumatophore begins to secrete gas. Red pigment cells form at the posterior
end of most larvae and a mouth with associated muscle and gland cells forms at this
site. The large endodermal cells that filled the interior of the larva disappear and a
gastric cavity forms in their place. Cnidobands form at the base of the tentacle and
take up positions on the tentacle.
The Muggiaea embryo begins to form cilia between six and eight hours of de-
velopment. By 12-1 5 hours the embryo shows swimming polarity and an endodermal
thickening forms along one side of the planula; during this period the embryo also
begins to elongate. At about 24 hours of development an invagination begins to form
in the anterior part of the lateral endodermal thickening, this is the first indication
of nectophore development; in the lateral band just below the invagination a bulge
forms, this is the first indication of tentacle formation. Experiments in which the site
602
G. FREEMAN
o
B
C
D
E
F
G
H
N
M
FIGURE 7. Normal development of Muggiaea. A) Uncleaved egg. B) Eight cell stage. C) Six hour
embryo. D) 1 5 hour embryo. The brackets indicate the lateral endodermal thickening. E) 28 hour embryo.
The nectophore and tentacle rudiments are beginning to form. F) 43 hour embryo. G) Two-and-a-half day
old larvae. The nectophore is functional. H) Three-and-a-half day old larvae. The mouth is forming. I)
Four-and-a-half day old larvae. Cnidobands are present on the tentacle. N, nectophore or rudiment;
T, tentacle or rudiment; M, mouth or rudiment. All photographs are the same magnification. The bar indi-
cates 50
HYDROZOAN EXPERIMENTAL EMBRYOLOGY
603
C
D
* *
* .
FIGURE 8. Normal embryology of Nanomia and Muggiaea. A) Section through part of a Nanomia
egg. Note the large endoplasmic granules and the sharp transition between the ectoplasm and the endoplasm.
B) Section through the peripheral region of a six hour Nanomia embryo. C) Section through the peripheral
region of an 18 hour Nanomia embryo. D) Section through the peripheral region of an 18 hour Muggiaea
embryo. C) and D) are similar. Note the ectoplasm containing ectoderm cells. Muggiaea has cortical granules
under the outer ectodermal cell membrane; these were initially just under the egg membrane. The nuclei
of these cells have nucleoli. The primary endoderm is made up of large endoplasm containing cells. The
bar indicates 50
of origin of the first cleavage furrow was marked show that this region corresponds
to the posterior end of these larvae ( 1 7 cases).
In both Muggiaea and Nanomia the formation of the lateral endodermal thickening
and organogenesis in this region give the larva a bilateral character. In Muggiaea the
relationship between the plane of the first cleavage and the plane of bilateral symmetry
was studied by placing a series of chalk marks around the egg on the first cleavage
furrow and examining the positions of the chalk marks in the 24 hour larva. Over
30 cases were marked, but only six cases were suitable for analysis because in most
cases the chalk tends to get displaced from the surface of the planula when it begins
to swim. In each of these six cases one side of the circle of chalk marks was coincident
with the lateral endodermal thickening. An attempt to do this experiment in Nanomia
failed because a much longer period elapses before an unambiguous lateral thickening
develops and too many of the chalk granules were lost.
Between days one and four of Muggiaea development the nectophore rudiment
grows rapidly and transforms into a functional locomotory organ. During this same
period the somatocyst forms, the tentacle rudiment transforms into a functional
tentacle and cnidobands form and take up positions on the tentacle. A mouth forms
at the posterior end of the larva and the large endodermal cells that filled the interior
of the planula disappear and a gastric cavity forms in their place.
604
G. FREEMAN
Experimental work on Nanomia and Muggiaea
1 ) The first set of experiments on the siphonophore embryos were done to de-
termine when the various regions along the oral-aboral axis, which will become the
mouth, tentacle, and pneumatophore or nectophore, are specified to form these struc-
tures. The embryos were cut into oral and aboral halves at various times from the
eight cell stage on and these halves were raised to see how they differentiated. Figure
9 indicates how these operations were done. When these operations were done at
early stages of development (prior to 20 hours), the site of origin of the first cleavage
furrow was marked so that the oral end of the embryo could be unambiguously
identified. After an operation at the eight cell stage, gastrulation appeared to take
place in both pieces at the normal time. When the operation was done after gastrulation,
the ectodermal covering spread over the yolky endoderm cells within an hour. During
this period the isolate sometimes lost one or two large endodermal cells. If cell loss
was excessive the case was discarded. The results of these experiments are summarized
in Tables II and III.
The results indicate that in Nanomia the specification of the mouth, tentacle, and
pneumatophore forming regions along the oral-aboral axis has already occurred by
the eight-cell stage of development. Aboral halves produced at this stage and later
stages correspond to the anterior third of the siphonula (Fig. lOa). Most of the surface
of these isolates is covered with large vacuolated cells. There is frequently a pneu-
matophore at the anterior end of these isolates and a rudimentary tentacle at the
posterior end; frequently pigment cells are found at the posterior end but a mouth
does not form. This point was checked by sectioning three of these cases. The muscle
and gland cells that are characteristic of the mouth were not present. Oral isolates
produced at the eight cell stage and later corresponded to the posterior two thirds of
the siphonula (Fig. lOb). In most cases they have a tentacle at their anterior end
(there may also be a few large vacuolated cells in this region) and a mouth and
B
B'
FIGURE 9. Operations performed to isolate oral and aboral halves of Muggiaea and Nanomia embryos
at different stages of development. A-E, Muggiaea. A'-E', Nanomia. A) Eight cell stage. B) Six to seven
hour embryo. C) 15-17 hour embryo. D) 22-24 hour embryo. E) 30-36 hour embryo. A') Eight cell stage.
B') Six to seven hour embryo. C) 12-13 hour embryo. D') 18-19 hour embryo. E') 31-41 hour embryo,
x, chalk mark placed at the site of origin of the first cleavage indicating the oral end of the embryo. The
dashed line indicates how the embryo was cut.
HYDROZOAN EXPERIMENTAL EMBRYOLOGY
605
TABLE II
The differentiation o/'Muggiaea embryo halves isolated at different times during development
Kind
of differentiation
Time of
Number
Corresponding
Isolate type
isolation
of cases
member of pair
Nectophore1
Tentacle2
Mouth
Oral
8-cell st.
4
2
4(2)
4(2)
4
6-7 h
6
4
6(3)
5(3)
6
15-17 h
4
2
4(2)
4(3)
4
22-24 h
4
4
1 (0)
4(3)
4
30-36 h
3
2
0
3(3)
3
Aboral
8-cell st.
2
2(2)
0
0
6-7 h
4
4(3)
0
0
15-17 h
3
3(3)
0
0
22-24 h
5
5(5)
0
0
30-36 h
4
4(4)
0
0
Lateral
2-cell st.
22
9
22 (22)
21 (20)
22
8-cell st.
1
0
1 (1)
1 (1)
1
6-7 h
4
1
4(3)
3(2)
4
22-24 h
2
1
2(1)
2(1)
2
30-36 h
5
2
5 (4)
5 (4)
5
Dorso-ventral
6-7 h
10
4
10(8)
10(8)
10
Dorsal
15-17 h
5
3
4(2)
4(1)
5
22-24 h
5
3
3(1)
4(0)
5
30-36 h
1
1
0
1 (0)
1
Ventral
15-17 h
7
7(6)
7(6)
7
22-24 h
7
7(6)
6(4)
7
30-36 h
1
1 (1)
1 (1)
1
1 The parenthesis indicates the number of cases that formed functional nectophores.
2 The parenthesis indicates the number of cases that formed tentacles with nematoband brackets.
pigment cells at their posterior end. The Nanomia data can also be analyzed by
examining the 24 examples of pairs of aboral and oral isolates from the same embryo.
The tentacle forming region is found in the zone between the oral and aboral halves.
Usually the tentacle is better developed in the oral isolate than it is in the aboral
isolate. In those cases (3) where the tentacle is well developed in the aboral isolate it
is rudimentary in the oral isolate. There is no indication that both halves develop
more complete tentacles when they are isolated at an early stage versus a later stage.
However, there is a tendency for oral isolates to differentiate large vacuolated cells
more frequently when they are isolated at early stages rather than later stages. This
suggests that the region which will differentiate large vacuolated cells may not have
been definitively positioned along the oral-aboral axis of the embryo by the eight cell
stage. The only feature which regulates its position along the oral-aboral axis is the
pigment cells. These regularly form at the most posterior end of aboral halves regardless
of the time at which these halves were isolated.
The results of the isolation experiments involving oral and aboral halves of the
Muggiaea embryo are more complex. Aboral halves produced at the eight cell stage
and later differentiated only the nectophore (Fig. lOc); these nectophores attain the
size of nectophores from an intact embryo. This result suggests that the aboral half
606
G. FREEMAN
A
\
C
D
F
\
FIGURE 10. The development of aboral and oral isolates from Nanomia (A-B) and Muggiaea
(C-F) embryos. All isolates are seven days old. A) Aboral half from eight cell stage embryo. Note the
anterior pneumatophore rudiment (arrow). B) Oral half from same eight cell stage embryo as (A). Note
the anterior tentacle with a cnidoband and the posterior mouth and pigment cells. C) Aboral half from
eight cell stage embryo. Note the lack of a mouth and tentacle. D) Oral half from 1 5 hour embryo. The
embryo has a mouth and a tentacle rudiment. The arrow points to an abnormal nectophore rudiment. E)
Oral half from the same eight cell stage embryo as (C). Note the mouth and tentacle. F) Oral half from
22 hour embryo. The embryo has an anterior protrusion, a tentacle with a position along the body which
is more anterior than normal and a posterior mouth. All photographs are at the same magnification. The
bar indicates 50
of the embryo is specified to form the nectophore some time prior to the eight cell
stage. Oral isolates produced at early developmental stages can form a normal larva
with a nectophore, tentacle, and mouth. While the larva is smaller than normal, the
HYDROZOAN EXPERIMENTAL EMBRYOLOGY 607
parts show the correct proportions (Fig. lOe). However, in half of these cases the
nectophore was smaller than normal and nonfunctional or rudimentary (Fig. lOd).
Since many of these oral isolates with a nectophore have a corresponding aboral
isolate from the same embryo that has also formed a nectophore, the oral isolate
must have formed a nectophore as a consequence of a regulatory adjustment that
occurred as a result of the operation. Between 15 and 22 hours of development there
is a marked decrease in the ability of the oral halves to differentiate a nectophore.
These cases form a tentacle at their anterior end and a posterior mouth (Fig. 100-
These experiments show that in both the Muggiaea and Nanomia embryos there
is an early specification of the ways in which the different regions will differentiate
along the oral-aboral axis of the future larva. However, in Muggiaea differentiation
of part of the oral region of the embryo can be respecified until just before the first
indications of organogenesis.
2) Since the different regions along the oral-aboral axis of the Nanomia embryo
appear to be specified some time before the eight cell stage, an attempt was made to
bracket the time period when specification occurs. Eggs which were undergoing their
first cleavage ( 1 4 cases) and two cell stage embryos that were marked at the site of
first cleavage initiation (5 cases) were cut into oral and aboral halves (Fig. 11). In
each case only the oral half contained the nuclei, and thus it was the only half that
developed. When an operation was done the diameter of each fragment was measured
to calculate the relative volume of the oral isolate.
The results of these experiments are presented in Table III. Unlike the oral isolates
produced at the eight-cell stage and at later stages, a substantial proportion of these
cases formed large vacuolated cells and a pneumatophore. These cases can be further
categorized by examining the kinds of structures that differentiate and the proportions
of the larvae. Six cases formed miniature larvae of normal proportions (Fig. 12a).
Ten cases formed the components of a normal larva but the anterior region was
abnormally small (Fig. 12b). About half of these cases looked like the eight cell stage
oral isolate that had a few large vacuolated cells at its anterior end. The anterior end
of the other isolates was better developed. There were also three cases that did not
form large vacuolated cells or a pneumatophore (Fig. 12c). There did not appear to
be a correlation between the time of the operation or where the cut was placed with
reference to the cleavage furrow and the kind of larva that differentiated. However,
larger isolates tended to form more normal larvae than smaller oral isolates (Fig. 1 3).
This experiment suggests that the specification of different regions along the oral-
aboral axis of the Nanomia embryo is either taking place during the first cleavage
and the two cell stage, or that it has occurred prior to this time, but the oral region
FIGURE 1 1 . Operations performed to isolate oral portions of first cleavage and two cell stage Nanomia
embryos. A) First cleavage. B) Two cell stage. In both operations the plane of the cut was perpendicular
to the oral-aboral axis of the embryo; however, the position of the plane along the oral-aboral axis varied
from case to case giving oral isolates of varying size. When operations were done on embryos that were
undergoing their first cleavage in some cases the cut was made through the first cleavage furrow, in other
cases the cut was made before the furrow reached that point, x, chalk mark placed at the site of the origin
of the first cleavage indicating the oral end of the embryo. The dashed line indicates how the embryo
was cut.
608
G. FREEMAN
TABLE III
The differentiation o/~ Nanomia embryo halves isolated at different times during development
Cor-
Kind of differentiation
responding
Isolate
Time of
Number
member
Pneumato-
Vacuolated
Posterior
type
isolation
of cases
of pair
phore
cells
Tentacle1
Mouth
pigment
Oral
8-cell st.
9
8
0
3
4(1)
9
7
6-7 h
5
5
0
1
4(1)
4
2
12-13 h
7
7
0
1
7(2)
7
6
18-19 h
7
2
0
1
7(2)
7
5
31-41 h
2
2
0
0
2(2)
2
2
Aboral
8-cell st.
9
7
9
4(0)
0
4
6-7 h
7
3
7
3(0)
0
0
12-13 h
7
6
7
5 (0)
0
1
18-19 h
3
1
3
3(0)
0
3
31-41 h
6
5
6
6(1)
0
5
Lateral
2-cell st.
16
4
7
16
16(4)
16
12
8-cell st.
8
4
5
7
7(0)
7
4
6-7 h
6
3
0
5
4(0)
5
2
12-13 h
6
3
4
6
6(2)
6
4
Oral
1st cleavage-2
19
9
16
18(4)
19
17
cell
' The parenthesis indicates the number of cases that formed tentacles with nematoband brackets.
of the Nanomia embryo can regulate in much the same way that the oral region of
the Muggiaea embryo regulates at later developmental stages.
3) As a control experiment embryos were cut into lateral halves at the two cell
FIGURE 12. The development of oral isolates from Nanomia eggs undergoing their first cleavage. All
isolates are seven days old. A) Normal larvae. The pneumatophore has secreted a gas bubble. This case
developed from an egg fragment with 53% of the volume of a normal egg. B) Larvae with a reduced aboral
end; a pneumatophore rudiment is present. This case developed from an egg fragment with 49% of the
volume of a normal egg. C) Larva which lacks an aboral end. This case developed from an egg fragment
with 50% of the volume of a normal egg. All of these cases developed from eggs that were cut through a
non-furrow region when the furrow was a third of the way across the egg. All photographs are at the same
magnification. The bar indicates 50
HYDROZOAN EXPERIMENTAL EMBRYOLOGY
609
Anterior region
missing
Anterior region
abnormally small
Larvae with
normal proportions
4/10
Pneumatophore
5/6
Pneumatophore
30 40 50 60 70
% Egg Volume
FIGURE 13. Graph relating the development of oral isolates from first cleavage and two cell stage
Nanomia embryos to the size of the isolate.
stage and later developmental stages (Fig. 14). All of the operations on Nanomia
embryos were performed before the development of the lateral thickening. Since the
point where the lateral thickening will develop in these embryos is not known, it is
more accurate to say that these embryos were cut along their oral-aboral axis. The
Muggiaea embryos were cut into lateral halves. Since the first cleavage furrow defines
the plane of bilateral symmetry the blastomere isolation experiments at the two cell
stage produces lateral halves. The experiments at the eight cell stage and at six to
seven hours of development were performed on embryos in which the first cleavage
furrow was marked. At later stages the lateral thickening was obvious. Virtually all
of these cases developed into normal larvae regardless of the stage when the operation
was performed (Table II). Each pair of lateral halves from the same embryo always
form the same structures. This experiment shows that the results obtained when these
embryos are cut into oral and aboral halves cannot be ascribed to the operation
per se.
FIGURE 14. Operations performed to isolate lateral halves of Nanomia and Muggiaea embryos at
different stages of development. A-D, Nanomia. A'-E', Muggiaea. A) Two cell stage. B) Eight cell stage.
C) Six to seven hour embryo. D) 12-13 hour embryo. A') Two cell stage. B') Eight cell stage. C) Six to
seven hour embryo. D') 22-24 hour embryo. E') 30-36 hour embryo, x, chalk mark placed at the site of
origin of the first cleavage indicating the oral end of the embryo. The dashed line indicates how the embryo
was cut. Embryos B' and C were cut along a set of chalk marks that indicate the plane of the first cleavage.
The ventral thickening was used to orient embryos D' and E'.
610
G. FREEMAN
4) The region along the oral-aboral axis where the lateral thickening forms is
referred to as the ventral side of the embryo. The differentiation of the dorsal and
ventral sides of the Muggiaea embryo was studied by cutting these embryos into
halves along their frontal plane at various time periods during development (Fig. 15).
The earliest stage when this operation was done was at six to seven hours of development
on embryos in which the first cleavage furrow was marked. At this stage it is not
possible to distinguish between a dorsal and a ventral side. All of these cases developed
into normal larvae (Table II). In two pairs of isolates from the same embryo the chalk
marks stayed on until the ventral thickening had formed. In both cases both members
of each pair formed their ventral thickening under the same chalk mark, indicating
that the outer surface of either side of the embryo is capable of becoming the ventral
side. The other operations were performed after the ventral thickening had formed
(15-36 h) (Table II). At all time periods when the operation was done the ventral
halves formed normal larvae (Fig. 16a). The behavior of the dorsal halves depended
upon when the operation was performed. When dorsal halves were produced at
15-17 hours of development a ventral thickening quickly formed opposite the cut
and in most cases a nectophore and tentacle formed. When dorsal halves were produced
at 22-24 and at 30-36 hours of development the ventral thickening took much longer
to form and there was a marked decline in the ability of these halves to form a
nectophore even though they formed a rudimentary tentacle (Fig. 16b). These ex-
periments show that both of the regions defined by the plane of bilateral symmetry
as potential dorsal or ventral sides of the embryo have the capacity to become the
ventral side of the embryo. Even after the ventral side of the embryo has begun to
differentiate, the dorsal side which is morphogenetically quiescent can differentiate
as a ventral side. One of the embryos that was to be used for these operations at 22
hours of development illustrates this point in a different way. The glass needle that
was used broke during the operation and only the aboral end of this embryo was cut
along the frontal plane. The cut was rather jagged, however it healed over. Subsequently
an endodermal thickening developed on the dorsal side of the embryo opposite the
region where the nectophore would form on the ventral side. The embryo went on
to form two nectophores (Fig. 16c).
Experiments have not been done that address the issue of dorsal ventral specification
in Nanomia, however a few of the embryos operated on in experiment 3 must have
been cut along or close to the presumptive frontal plane. Since both halves developed
normally in all cases one can tentatively conclude that at the time the operations
were performed the presumptive dorsal side (if it exists) can still regulate.
5) The last experiment investigated the effect of the yolky endoplasm of the
FIGURE 15. Operations performed to isolate dorsal and ventral halves of Muggiaea embryos at
different stages of development. A) Six to seven hour embryo. This embryo was cut along the oral-aboral
axis in a plane perpendicular to a set of chalk marks that indicate the plane of the first cleavage. B) 1 5-
1 7 hours of development. C) 22-24 hours of development. D) 30-36 hours of development. The ventral
thickening was used to orient embryos B-D. The dashed line indicates how the embryo was cut.
HYDROZOAN EXPERIMENTAL EMBRYOLOGY
611
A
C
\
FIGURE 16. Muggiaea larvae from operated embryos. A) Six day old ventral isolate from 22 hour
embryo. B) Six day old dorsal isolate from 22 hour embryo. Note the tentacle rudiment. The dorsal half
is from the same embryo as (A). C) Five day old larva with two nectophores. All photographs are at the
same magnification. The bar indicates 50 ^m.
siphonophore egg in development. Eggs were centrifuged to produce ectoplasmic and
endoplasmic fragments. These experiments were only done on Nanornia. The en-
doplasmic fragments that were produced moved to the air water interface either during
centrifugation or shortly after centrifugation and were destroyed. Figure 17a shows
A
FIGURE 17. The development of ectoplasmic fragments and older Nanomia embryos that have lost
their endoplasm. A) Ectoplasmic fragment from centrifuged egg. B) Five day ciliate sphere from ectoplasmic
fragment. C) Seven day larva from embryo which lost its endoplasm at 16 hours of development. The
embryo has a pneumatophore rudiment, vacuolated anterior cells, a tentacle rudiment and a mouth. It is
much smaller than a normal larva. Compare this figure with 61. All photographs are at the same magnification.
The bar indicates 50 nm.
612 G. FREEMAN
an ectoplasmic fragment. The average diameter of these fragments was 1 3 1 yum (range
107-142 ^m, sample size 12). An ectoplasmic fragment contains about 10% of the
egg volume. Only about a fourth of the eggs that were centrifuged produced ectoplasmic
fragments. Most (88%) of the ectoplasmic fragments cleaved. The early cleavages were
normal. Cilia developed by 15 hours, however, the embryos did not elongate and
there was no indication of swimming polarity. Twenty embryos were raised for five
days. Ten of these cases were sectioned. There was no indication of organogenesis
(Fig. 17b). This experiment suggests that the endodermal plasm is necessary for
normal development.
This conclusion is supported by observations on post gastrula Nanomia embryos
that get caught on the air-water interface. When this happens most or all of the
endodermal cells are lost and one is left with an ectodermal hull. Unfortunately this
procedure for removing the endoderm of the embryo is not exactly well controlled.
When an embryo is de-endodermized between 6 and 12 hours of development (5
cases) the ectoderm that remains forms a ciliated ball and no organogenesis occurs.
If even a small amount of endoderm remains the embryo will show swimming polarity
and a mouth and/or a few large vacuolated cells will form (12 cases). If an embryo
is de-endodermized after it has begun to elongate it will differentiate most structures
even though the larva will be very small (8 cases) (Fig. 17c). This suggests that the
yolky endoderm is only necessary for the early stages of development.
DISCUSSION
The generality of the findings
Trachylina. There are a number of descriptive studies on early development of
other species in the order Trachylina. Most of this literature dates from the last
century; a great deal of it is summarized in Metschnikoffs (1886) monograph on the
embryology of medusae. This monograph describes the egg and/or early developmental
stages of seven species in the order Trachylina; it also provides comparative data on
a number of species in the order Hydroida. All of the species in the order Trachylina
appear to have eggs with large endoplasmic granules; these granules are much smaller
in the eggs and early developmental stages of species in the order Hydroida. Within
the order Trachylina there appears to be some variation in the size of these granules
and their packing in different species.
In all of the species in the order Trachylina there appears to be an early estab-
lishment of ectodermal and endodermal cell layers. In Aglantha this process begins
at the eight cell stage: I suspect that this may also occur at this stage in two of the
species Metschnikoff studied, Aglaura and Polyxenia. Prior to the work on Aglantha
described here, gastrulation was considered to be the time when ectodermal and
endodermal cell layers formed. Gastrulation can occur in several ways in cnidarians;
several schemes describe the ways in which this process can occur (Tardent, 1978).
Different species in the Trachylina have been placed in different slots in these schemes.
However in every case gastrulation involves a delamination in which a cell division
takes place in such a way that an inner larger cell inherits primarily the granular
endoplasm and a smaller outer daughter cell inherits primarily the cortical cytoplasm.
This type of gastrulation is not too different from the formation of an endoplasm-
poor micromere at the eight cell stage. In every case these gastrulation events occur
at an early stage of development before a great deal of cell division has taken place.
The kinds of cnidarian gastrulation that are associated with later cleavage stages, such
as ingression and secondary delamination, do not occur in these embryos.
The process of embryogenesis in Aglantha is very similar to the process of em-
HYDROZOAN EXPERIMENTAL EMBRYOLOGY 613
bryogenesis in Aglaura which Metschnikoff ( 1 886) has studied. Metschnikoff considered
the possibility that epiboly might occur in Aglaura but rejected it because he had no
evidence that smaller blastomeres were moving over the larger blastomeres; however,
I have seen epiboly occur in Aglantha. The fact that two cell stage blastomere isolates
sometimes have an ectodermal cap at their aboral end supports this view — these cases
would be generated when epiboly does not occur.
At this point no experiments have been done on the embryos of other species in
the order Trachylina that elucidate the issues considered here. Both Maas (1908) and
Zoja (1895) separated and reared blastomeres from early cleavage stage embryos of
the narcomedusae Liriope and Geryonia. Single blastomeres isolated from two and
four cell stage embryos form medusae or medusa larvae.
Siphonophora. The early development of only a few species of siphonophores has
been studied. Carre's papers (1967, 1969) contain the best histological descriptions
of early development. All species of siphonophores appear to have relatively large
eggs. In every case where the egg has been examined, it contains relatively large
endoplasmic granules and there is a sharp boundary between the endoplasm and the
cortical layer of the egg. In every case where gastrulation has been described, it appears
to take place early in development by delamination.
The only experimental work on early embryogenesis in siphonophores has been
done by Carre (1969) on Nanomia bijuga. This work addresses the issue of whether
regulation can occur along the oral-aboral axis of the embryo and the effect of de-
velopmental age on regulatory ability. The results she obtained contradict the results
presented here. Because her work is only briefly described, many crucial details that
would aid in interpreting the experiments are not given. One set of experiments
involved the isolation of blastomeres at the 2, 4, 8, and 16 cell stages. In a crucial
experiment an eight cell stage embryo was separated into eight blastomeres; seven of
these isolates formed a pneumatophore bud and a tentacle. Another set of experiments
divided gastrulae, young planulae, and planulae with a pneumatophore bud into two
halves. In the experiment on the planulae with a pneumatophore bud, and presumably,
in the young planulae, the cut created oral and aboral halves, but when gastrulae
were cut into halves the cut was not oriented because marked embryos would have
to be used. Carre reports that when gastrulae or young planulae were cut in half
regulation occurred in all cases. The only developmental stage where regulation did
not occur was the planula with the pneumatophore bud. At this stage the aboral half
developed into a small siphonula without a gastrozoid and the oral half formed a
gastrozoid but did not differentiate a pneumatophore. Carre concluded that regulation
is total in young planulae and disappears when organogenesis begins.
It is hard to believe that two species belonging to the same genus should behave
in such different ways. At present on the basis of my experiments on Nanomia cara
and Muggiaea I would argue that there is probably an early specification of different
regions along the oral-aboral axis of all siphonophore embryos. However Carre's
report suggests that the situation may be more complex. It is conceivable that some
species in this Order may show an early specification of different regions along the
oral-aboral axis ot the embryo and that these same species may differ in their ability
to regulate. In other species the ability of different regions along the oral-aboral axis
to regulate may be so extensive that it may be difficult to define when a particular
region along this axis is specified.
The comparative embryology of the Hydroida and Ctenophora
Hydrozoans with both direct and indirect development and ctenophores share a
number of developmental traits. Both of these groups have a centrolecithal egg with
614 G. FREEMAN
a central yolky endoplasmic region that is surrounded by a peripheral layer of cortical
cytoplasm. In both groups cleavage is unipolar. The oral-aboral axes of the embryos
are established at the time of first cleavage under conditions where the oral pole of
the axis corresponds to the site of first cleavage initiation (Freeman, 1977, 1980). In
the order Hydroida this region corresponds to the posterior end of the planula which
becomes the mouth of the polyp after metamorphosis.
When the basic features of development in the order Hydroida are compared
with a similar set of features in the Ctenophora, several major differences between
these two groups that involve the structure of the egg, the process of embryogenesis
and the mechanisms that underlie this process become apparent. Each of these dif-
ferences will now be examined.
Egg organization. While the Hydroida and Ctenophora have centrolecithal eggs,
these two groups differ in the way this organization is expressed. The endoplasmic
granules of ctenophore eggs are larger and more closely packed than those of Hydroida
eggs, as a consequence the transition between the ectoplasmic and endoplasmic regions
is much sharper in ctenophore eggs (see Fig. 30 in Freeman and Reynolds, 1973 for
a section through a typical ctenophore egg and Fig. 1 in Freeman and Miller, 1982,
for sections through Hydroida eggs.). In the ctenophore egg both of these cytoplasmic
layers behave to a large extent like immiscible fluids (Spek, 1926). This kind of
cytoplasmic behavior appears to be absent or much less pronounced in the Hydroida.
The Aglantha egg is similar to a ctenophore egg in that it has large endoplasmic
granules; however these granules are not closely packed. Nevertheless the ectoplasmic
region of the Aglantha egg appears to be more distinct than it is in Hydroida eggs.
Both siphonophore eggs have large closely packed endoplasmic granules (see Fig. 1
in Carre and Sardet, 1981, for sections through the egg of a related species ofMuggiaea)
and a distinct ectoplasmic region. These eggs closely resemble ctenophore eggs.
Cleavage pattern. In the Hydroida it is difficult to talk about cleavage patterns
during early embryogenesis. After the first cleavage there is generally not a set ori-
entation for subsequent cleavage furrows, even though certain cleavage planes are
more probable than others. There is no evidence that ectoplasm and endoplasm are
differentially distributed to different blastomeres during early cleavage (Tardent, 1978).
In ctenophores early cleavage occurs according to a stereotypic pattern. The first three
cleavages take place along the oral-aboral axis of the embryo generating eight ma-
cromeres. Ctenophores are biradially symmetrical; there is a one-to-one relationship
between the planes of the first cleavages and the sagittal and tentacular planes of
these embryos. At the fourth cleavage each macromere gives off a micromere at the
aboral pole of the embryo. During this division there is a differential distribution of
cytoplasm so that the micromeres inherit very little endoplasm. During the next few
divisions additional yolk-free micromeres are given off at the aboral pole of the
embryo. These micromeres will become the ectodermal covering of the embryo;
gastrulation occurs by epiboly (Reverberi, 1971).
In Aglantha the initial cleavage divisions also generate a stereotypic pattern. This
embryo closely resembles the ctenophore embryo in that micromeres which are largely
yolk-free are generated at the aboral end of the embryo. These micromeres will also
form at least part of the ectodermal covering of the embryo. Gastrulation takes place
in the same way in both forms. The two siphonophores do not generate a stereotypic
cleavage pattern, in this sense they are Hydroida-like. However, in Muggiaea the
plane of the first cleavage corresponds to the plane of bilateral symmetry of the
embryo. Thus there is a relationship between the plane of cleavage and a symmetry
property of the embryo as there is in ctenophores. Both siphonophore embryos undergo
differential divisions at early stages of development that generate endoplasm and
HYDROZOAN EXPERIMENTAL EMBRYOLOGY 615
ectoplasm free cells. These divisions which constitute gastrulation produce the external
ectoplasm containing cells that form the ectoderm of the embryo; although this
differential division does not occur in the same way it does in the ctenophore embryo,
it has the same effect.
The establishment of embryonic regions with different developmental potentials.
During the early cleavage stages of embryogenesis in ctenophores several cell divisions
have been identified that give rise to daughter cells with different developmental
potentials (Reverberi, 197 1 ; Freeman and Reynolds, 1973). The first of these divisions
occurs at the third cleavage. If a blastomere is isolated at the four cell stage it will
continue to cleave and subsequently differentiate comb plate cilia cells and light
producing cells. When the four cell stage blastomere divides, it produces E and M
daughter cells. If the E blastomere is isolated it will subsequently differentiate comb
plate cilia cells, but not light producing cells, while the isolated M macromere will
differentiate light producing cells but not comb plate cilia cells. In these embryos
cleavage does not passively divide up special cytoplasmic regions of the egg that have
been in place for some time. The factors that specify these two cell types are gradually
localized in the future E and M macromere forming regions of the embryo during
the two cleavages which precede this division (Freeman, 1976). These embryos behave
like a mosaic of parts which have been largely specified during early cleavage stages.
In the Hydroida that gastrulate by unipolar ingression it is possible to map the
position of the presumptive ectodermal and endodermal cells prior to gastrulation.
The ectodermal cells are found at the presumptive anterior end while the endodermal
cells are found at the presumptive posterior end of the embryo. At any time prior
to gastrulation it is possible to isolate each of these presumptive regions and both
kinds of isolates will form a normal planula larva with both ectodermal and endodermal
cell layers (Freeman, 1981). When an isolated region regulates to form a normal
planula it always conserves its polarity properties (Teissier, 1931). During gastrulation
the presumptive ectodermal cells lose their capacity to form endodermal cells; this
is the first point during development where there is a restriction of developmental
potential (Freeman, 1981). Following gastrulation the embryo differentiates into a
planula larva. If a post gastrula embryo or planula is cut up into regions with different
presumptive fates, each part will regulate to form a normal planula, as long as both
ectodermal and endodermal cell layers are present (Miiller et al, 1977; Freeman,
1981, however, see Lesh-Laurie, 1976). These embryos behave like developmental
fields (Wolpert, 1969). The way a given cell differentiates in these embryos ultimately
depends upon its position with respect to its neighbors.
In Aglantha and the two siphonophores there is an early specification of different
embryonic regions. In Aglantha the micromeres that are produced at the eight cell
stage differentiate only ectoderm while the macromeres differentiate both ectoderm
and endoderm. In ctenophores the micromeres and macromeres produced at the 16
cell stage differentiate in the same way. This is quite different from the Hydroida
where ectoderm and endoderm are not specified until gastrulation. Gastrulation in
the hydroida is not an early event as it is in Aglantha, but a relatively late event, at
least in those forms which gastrulate by ingression. After gastrulation the aboral half
of the Aglantha embryo cannot regulate to form a mouth and tentacles; it behaves
differently from the aboral half of the Hydroida embryo which can regulate. Unfor-
tunately this experiment has not been done on ctenophore embryos.
In Nanomia there is a specification of different regions along the oral-aboral axis
of the embryo by the eight cell stage; this specification occurs before ectodermal cells
have formed as it does in the ctenophore embryo. In Muggiaea the situation is more
complicated, while the aboral region of the embryo is specified by the eight cell stage.
616 G. FREEMAN
the oral part of the embryo is capable of regulation until just before organogenesis
begins; the same is true of the presumptive dorsal half of the embryo. In this embryo
the timing of determinative events appears to be a mix which has some of the elements
of the ctenophore situation and some of the elements of the Hydroida situation.
The role of ectoplasm and endoplasm in cell specification. In both the Ctenophora
and the Hydroida, experiments have been done to create egg fragments that lack
endoplasm (see Beckwith, 1914; Freeman and Miller, 1982, for the Hydroida, and
LaSpina, 1963; Freeman and Reynolds, 1973 for the Ctenophora). This experiment
is done by centrifuging fertilized uncleaved eggs to stratify the egg contents and then
increasing the centrifugal force or cutting the egg to give a nucleated ectoplasmic
fragment. When this experiment is done on ctenophores the initial cleavages are
normal. However there is not a normal segregation of developmental potential, both
the E and M macromeres differentiate comb plate cilia. These embryos fail to dif-
ferentiate certain cell types such as light producing cells and they develop into a
poorly organized ectodermal mass (see Fig. 35 in Freeman and Reynolds, 1973 for
a cross section through one of these "embryos"). In the Hydroida ectoplasmic fragments
form normal planulae. This comparison indicates that endoplasm is necessary for
normal embryogenesis in the Ctenophora, but not in the Hydroida.
Ectoplasmic fragments of both Aglantha and Nanomia differentiate ectodermal
masses that are similar to the ectodermal mass produced under similar conditions
by the ectoplasmic fragments of Ctenophore eggs. The behavior of the ectoplasmic
fragments reflects the marked distinction between the ectoplasm and the endoplasm
in the eggs and embryos and the inheritance of the ectoplasm by the ectodermal cells
in these three groups of animals. The lack of morphogenesis in these ectodermal
masses probably reflects the lack of endoderm. Hydroida embryos which lack endoderm
are capable of undergoing metamorphosis but cannot form a polyp (Freeman, 1981).
This comparison of development in the Ctenophora, the Hydroida, the Trachylina,
and the Siphonophora shows that the Trachylina and the Siphonophora each have
an egg organization, a mode of early development, and a set of mechanisms for
specifying embryonic regions that is very similar to those found in Ctenophores.
The bases for developmental parallelism
The Cnidaria and the Ctenophora are thought to be closely related (Hyman, 1 940).
It is possible that the development parallelism between the Trachylina, the Siphon-
ophora, and the Ctenophores could be explained on the basis of common descent.
At present there is no agreement about how the classes and orders in the phylum
Cnidaria are related. It is not even clear what the most primitive members of the
phylum looked like. Some students of this group have argued that the first Cnidarians
were polyps (Werner, 1973) while others have argued that the first Cnidarians were
medusae (Brooks, 1886, Rees, 1966). It is also not clear how the phylum Ctenophora
is related to the Cnidaria. However, a number of speculative phylogenies have been
developed that have the status of educated guesses. Hyman ( 1 940) has argued that
the Trachylina and the Ctenophora are closely related. No one has suggested the
Siphonophora are closely related to either the Ctenophora or Trachylina. The spec-
ulations concerning the origin of the Siphonophora derive this order from the Hydroida
(Totton, 1965).
This parallelism may also reflect the fact that these embryos develop directly.
During embryogenesis a set of structures are going to develop which are more elaborate
than those of a planula larva. This reflects the fact that these animals have to function
in a pelagic environment. It will take a certain amount of time to generate these
HYDROZOAN EXPERIMENTAL EMBRYOLOGY 617
structures. Because the egg is a closed system, only so much time is available for
building these structures before the embryo's nutrient reserves are depleted. These
two considerations could place a premium on the way time is allocated during em-
bryogenesis.
Before a structure develops a decision has to be made about its placement. An
embryonic field is one mechanism for specifying structure placement and is used by
the Hydroida. This mechanism relies on physiological machinery which assigns each
cell an address with respect to its neighbors. In order for this mechanism to function,
its physiological machinery has to be created and it must function for a period of
time. This means that this could be a relatively costly mechanism in terms of time
utilization. However, if the differentiation of a structure depends on the inheritance
of localized cytoplasmic regions, as it appears to be in the direct developers, the time
needed to decide where a given structure will be placed is reduced substantially.
The process of embryogenesis in Cnidarians also depends upon interactions between
ectodermal and endodermal cell layers. This means that these cell layers have to exist
before structure formation can begin. When the specification of these cell layers
depends upon the position of a given cell with respect to its neighbors, as it does in
the Hydroida with indirect development, this process is going to take much longer
than it will in direct developing embryos where the parcelling out of ectoplasm and
endoplasm at cleavage accomplishes the same end.
Embryonic field mechanisms and cytoplasmic localization mechanisms are fre-
quently regarded as separate and distinct ways of specifying the developmental potential
of different parts of embryos. The experiments described here suggest that during the
course of evolutionary divirsification within a group of animals, a transition from
one mechanism to the other can occur relatively easily (see Freeman, 1982, for a
general discussion of this mode of evolutionary change).
The developmental similarities that the Trachylina, the Siphonophora, and the
Ctenophora share is impressive. If one assumes that all three groups evolved inde-
pendently from a Hydroida like stock, one would have to argue that while there are
no constraints which prevent the transition from a field to a cytoplasmic localization
mechanism of embryonic determination, the way one undergoes the transition is
highly constrained. For example, in all three groups the axial relationships are similar,
and when cleavage is related to symmetry, the same relationship holds in different
groups. This kind of constraint provides a basis for explaining the developmental
similarities within these groups.
ACKNOWLEDGMENTS
The impetus for this work was a Dahlem workshop on Evolution and Development
held in Berlin in May, 1981. I want to thank Professor A. O. D. Willows, the director
of the Friday Harbor Laboratories for facilitating my work there. I am especially
grateful to R. Emlet, G. Mackie, R. Miller, C. Mills, S. Smiley, R. Satterlie, and A.
Spencer for collecting siphonophores. This work was supported by grant GM 20024
from the National Institute of Health.
LITERATURE CITED
BECKWITH, C. 1914. The genesis of the plasma-structure in the egg of Hydractinia echinata. J. Morphol.
25: 189-251.
BROOKS, W. 1886. The life history of the hydromedusae: a discussion of the origin of the medusae and
of the significance of metagenesis. Mem. Boston Soc. Nat. Hist. 3: 359-430.
CARRE, D. 1967. Etude du developpement larvaire de deux Siphonophc.es: Lensia conoidea (Calycophore)
et Forskalia edwardsi (Physonecte). Cah. Biol. Mar. 8: 233-2^1.
618 G. FREEMAN
CARRE, D. 1969. Etude histologique du developpement de Nanomia bijuga (Chiaja, 1841) Siphonophore,
physonecte, aglamidae. Cah. Biol. Mar. 10: 325-341.
CARRE, D., ANDC. SARDET. 1981. Sperm chemotaxis in siphonophores. Biol. Cell. 40: 1 19-128.
FREEMAN, G. 1976. The role of cleavage in the localization of developmental potential in the Ctenophore
Mnemiopsis leidyi. Dev. Biol. 49: 142-177.
FREEMAN, G. 1977. The establishment of the oral-aboral axis in the Ctenophore embryo. J. Emb. Exp.
Morphol. 42: 237-260.
FREEMAN, G. 1980. The role of cleavage in the establishment of the anterior-posterior axis of the hydrozoan
embryo. Pp. 97-108 in Developmental and Cellular Biology ofCoelenterates. P. and R. Tardent,
eds. Elsevier/North Holland, Amsterdam.
FREEMAN, G. 1981. The role of polarity in the development of the hydrozoan planula larva. Wilhelm
Roux'sArch. 190: 168-184.
FREEMAN, G. 1982. What does the comparative study of development tell us about evolution? Pp. 155-
167 in Evolution and Development. J. Bonner, ed. Springer- Verlag, Berlin.
FREEMAN, G., AND R. MILLER. 1982. Hydrozoan eggs can only be fertilized at the site of polar body
formation. Dev. Biol. 94: 142-152.
FREEMAN, G., ANDG. REYNOLDS. 1973. The development of bioluminescence in the Ctenophore Mnemiopsis
leidyi. Dev. Biol. 31: 61-100.
HYMAN, L. 1940. The Invertebrates: Protozoa through Ctenophora. McGraw-Hill, New York. 726 pp.
KOZLOFF, E. 1 974. Keys to the Marine Invertebrates ofPuget Sound, the San Juan Archipelago and Adjacent
Regions. University of Washington Press, Seattle. 226 pp.
LASPINA, R. 1963. Development of fragments of the fertilized egg of ctenophores and their ability to form
ciliated plates. Acta. Embryol. Morphol. Exp. 6: 204-2 1 1 .
LESH-LAURIE, G. 1976. Stolon vs. hydranth determination in Pennaria tiarella planulae: a role for DNA
synthesis. Pp. 365-375 in Coelenterate Ecology and Behavior, G. Mackie, ed. Plenum, New York.
MAAS, O. 1908. Uber der Bau des Meduseneis. Verh. Deutsch. Zoo/. Ges. Leipzig. 1 14-129.
METSCHNIKOFF, E. 1874. Studien iiber die Entwicklung der Medusen and Siphonophoren. Z. Wiss. Zoo/.
24: 15-80.
METSCHNIKOFF, E. 1886. Embryologische Studien an Medusen: Ein Beitrag zur Genealogie der Primiti-
vorgane. and Atlas, Alfred Holder, Vienna. 159 pp.
MULLER, W., A. MITZ, J. WiCKHORST, ANDH. MEIER-MENGE. 1977. Polar morphogenesis in early hydroid
development: action of Caesium, of neurotransmitters and of an intrinsic head activator on pattern
formation. Wilhelm Roux'sArch. 182: 311-328.
NOVIK.OFF, A. 1938. Embryonic determination in the annelid Sabellaria vulgaris. II. Transplantation of
polar lobes and blastomeres as a test of their inducing capacities. Biol. Bull. 74: 21 1-234.
REES, W. 1966. The evolution of the hydrozoa. Pp. 199-221 in The Cnidaria and their Evolution, W.
Rees, ed. Academic Press, London.
RICHARDSON, K., L. JARRET, AND E. FINKE. 1960. Embedding in epoxy resins for ultrathin sectioning in
electron microscopy. Stain Technol. 35: 313-325.
REVERBERI, G. 1971. Ctenophores. Pp. 85-103 in Experimental Embryology of Marine and Freshwater
Invertebrates, G. Reverberi, ed. North Holland, Amsterdam.
RUSSELL, F. 1938. On the development of Muggiaea atlantica Cunningham. J. Mar. Biol. Assoc. U. K.
22: 440-446.
SPEK, J. 1 926. Uber gesetzmassige substanzverteilung bei der Furchung des Ctenophoreneis und ihre Beziehung
du den Determinations-problem. Wilhelm Roux's Arch. 107: 54-73.
TARDENT, P. 1978. Pp. 67-415 in Morphogenese der Tiere Coelenterata, Cnidaria 1, Al F. Seidel, ed.
Gustave Fischer, Jena.
TEISSIER, G. 1931. Etude experimentale du developpement de quelques hydraires. Ann. Sci. Nat. Ser. Bot.
Zoo/. 14: 5-60.
TOTTON, A. 1965. A Synopsis of the Siphonophora. British Museum (Natural History) London. 230 pp.
WERNER, B. 1973. New investigations on systematics and evolution of the class scyphozoa and the phylum
cnidaria. Publ. Seto Mar. Biol. Lab. 20: 35-6 1 .
WOLPERT, L. 1969. Positional information and the spatial pattern of cellular differentiation. / Theoret.
Biol. 25: 1-47.
ZOJA, R. 1895. Sullo svillupo dei blastomeri isolati di alcuni medusae (e di altr organismi). Wilhelm Roux's
Arch. 1: 578-595, 2: 1-37.
Reference: Biol. Bull. 165: 619-636. (December, 1983)
CIRCULATION OF FLUIDS IN THE GASTROVASCULAR SYSTEM OF
THE REEF CORAL ACROPORA CERVICORNIS
ELIZABETH H. GLADFELTER
West Indies Laboratory, Teague Bay, Christ iansted, St. Croix. USVI 00820, and Department of Biology,
University of California, Los Angeles, California 90024
ABSTRACT
Circulation of fluids in the gastrovascular system of A. cervicornis was determined
by observing the movement of fluorescein dye injected via a lateral polyp and viewed
in the dark under ultra-violet light. Scanning electron microscopy and petrographic
thin sections were used to describe the general morphology of the gastrovascular
system. This consists of two functional units: an axial unit composed of the coelenteron
of the axial polyps and a peripheral unit composed of tubes oriented axially ramifying
through the skeleton lying just beneath the outer ectoderm. These units are connected
by radially oriented tubes including the coelenterons of the lateral polyps. The entire
gastrovascular system is lined by flagellated endoderm cells.
Flow in the axial unit is always proximal. Row in the peripheral unit is both
distal and proximal and the velocity is always less than the flow in the axial unit.
Light does not appear to change the rate of flow. Rates of flow in the peripheral unit
show a diel cycle, with increased flow rates occurring between 2100 and 0600.
INTRODUCTION
Reef corals are symbioses between colonial cnidarians (Anthozoa: Scleractinia)
and intracellular dinoflagellates (=zooxanthellae). The animal colony consists of polyps
connected by coenosarc through which extensions of the gastrovascular system ramify
(Wells, 1956). Thus, there exists the potential for transport of materials (e.g., dissolved
or particulate organic matter) from one site in the colony to another.
Gastrovascular transport systems in Cnidaria have been investigated in hydro-
medusae (Roosen-Runge, 1967); hydroids (Rees et ai, 1970); pennatulids (Musgrave,
1909; Parker, 1920; Brafield, 1969); and gorgonians (Murdock, 1978a, b). To date,
work on transport in scleractinian corals is limited to a few studies in which materials
introduced at one site in the colony have been detected at another site (Pearse and
Muscatine, 1971; Taylor, 1977).
The reef coral Acropora cervicornis is a branching form consisting of a relatively
large axial corallite and polyp at the terminus of each branch, and many smaller
lateral calices with polyps along the length of the branch. The distal portion of the
axial corallite is a site of rapid skeletal development (e.g., Goreau and Goreau, 1959;
Pearse and Muscatine, 1971; Gladfelter, 1982, 1983) and cell division (Gladfelter,
1983); both processes occur in a characteristic diel pattern (Gladfelter, 1983). When
soluble organic molecules and 45Ca++ have been introduced at a distance from the
tip of a branch, they have been detected later in the tissues of the axial polyp and
its skeleton at the extreme distal portion of the branch; it has been inferred that these
molecules and ions have been transported in some way to the tip (Pearse and Muscatine,
Received 9 May 1983; accepted 26 September 1983.
619
620
E. H. GLADFELTER
FIGURE 1 . Several SEMs of decalcified axial polyps, showing the axial unit of the fluid transport
system, a. Cross-section of an axial polyp ca. 10 mm from the tip, exposing the axial unit (a) of the fluid
transport system, the canal formed by the coelenteron of the axial polyp lying within the calyx of the
skeleton. A number of lateral polyps (Ip) can be seen. Scale bar = 500 ^m. b. Cross-section of the axial
polyp ca. 20 mm from the tip, showing the now partly occluded axial canal (a). Scale bar = 250 ^m. c.
View of the membrane surface of the endodermal cells lining the axial canal. Note that each flagellum is
surrounded by a circlet of raised projections of the cell membrane. Scale bar = 5
197 1; Taylor, 1977). To date, however, neither the morphological basis of this transport
nor the patterns of flow in this hypothesized transport system have been described.
The present study was undertaken with these goals in mind.
CIRCULATION IN ACROPO&4 CER11CORNIS
621
FIGURE 1. (Continued)
MATERIALS AND METHODS
Morphology of the gastrovascular system
Scanning electron microscopy (SEM) was used to describe the general morphology
of the gastrovascular system of A. cervicornis. Specimens examined by SEM were
prepared as described by Gladfelter (1982, 1983). Measurements of the volume oc-
cupied by certain parts of the gastrovascular system were made from petrographic
thin cross-sections of the skeleton (made along the branch length) as described by
Gladfelter (1982); spaces void of skeleton are occupied by the coelenteron of the
gastrovascular system (Gladfelter, 1982). The total cross-sectional area of the axial
corallite (calyx and theca) and the area of secondary radial growth was determined
by direct measurement of 10 colonies (with 2-3 branches per colony).
Patterns and rates of fluid transport
Collection and maintenance of specimens. Acropora cervicornis was collected from
a depth of 10-12 m in Buck Island Channel, adjacent to Teague Bay forereef, St.
Croix, U. S. Virgin Islands. Straight branches, ca. 20 cm long, with a single axial
corallite were removed from the colonies. Within 20 min of collection, the branches
were transported submerged in a plastic tub filled with sea water to the West Indies
Laboratory. The coral branches were placed in shaded outdoor aquaria supplied with
fresh continuously flowing sea water. Corals were routinely allowed to acclimate for
24 h before measuring rates of fluid transport.
Detection of fluid transport. The fluorescent dye, fluorescein, was used to detect
transport of fluids along the axis of a branch of A. cervicornis. For each experiment,
10-12 branches were brought into a darkened laboratory and allowed to acclimate
in running sea water for 1 h prior to measurement. The temperature during all the
measurements of rates of fluid movement was the same as that of the natural en-
622
E. H. GLADFELTER
CIRCULATION IN ACROPOR.A CERVICORNIS 623
vironment, 27° ± 1°C. A single branch was placed horizontally in a plastic tub
(90 cm X 45 cm X 23 cm) filled with sea water. A hypodermic syringe with a #26
needle was used to inject 0.05 ml of a saturated solution of fluorescein dye in sea
water into a lateral polyp. The distance traveled by the moving dye front was measured
each minute after the initial injection by observing the branch in the dark with an
ultraviolet light. As a control, 3 coral branches were fixed in 10% buffered formalin,
injected with fluorescein, and observed as described above. On some coral branches
either the distal portion of the axial polyp or the proximal portion of the axial polyp
was injected with fluorescein dye, and observed as described above.
RESULTS
Morphology of the gastrovascular system
Canals. The gastrovascular system of a branch of Acropora cervicornis is a series
of interconnected large (ca. 1000 /urn in cross-section) and small (ca. 100 ^m in cross-
section) canals. The largest canal in each branch is the portion of the coelenteron of
the axial polyp within the calyx of the axial corallite (Fig. la, b); this is referred to
as the axial canal. Slightly smaller are the somewhat radially oriented canals formed
by the coelenterons of the lateral polyps within the lateral corallites. The smallest
canals ramify through the porous skeleton (Figs. 2, 3a, b); the canals oriented axially
and lying just beneath the outer ectoderm are referred to as peripheral canals. In the
distal 5 mm of the branch, the canals within the wall of the axial corallite are the
peripheral canals, but as the branch increases in diameter the canals just below the
outer ectoderm, between the pseudocostae of the skeleton, serve in this capacity (Fig.
3). Petrographic thin cross-sections of the skeleton were used to determine the cross-
sectional areas of component parts of the gastrovascular system. As the branch increases
in girth by radial accretion of skeleton, the resulting secondary growth of skeleton
contains both the coelenterons of the lateral polyps as well as small canals connecting
the axially oriented canals (both peripheral and axial). The total cross-sectional area
of the calyx (containing the axial canal) does not decrease significantly until ca. 30
cm from the tip, while the cross-sectional area of the combined peripheral canals
increases several fold (Table I). The cross-sectional areas of the canals oriented radially,
in the secondary radial growth of the skeleton increases from 0 cm2 at the tip of the
branch (where there is no radial growth) to a large cross-sectional area 30 cm from
the tip (Table I).
Endodermal cells. The entire gastrovascular system is lined by flagellated endo-
dermal cells (Figs. Ic, 2c, 4, 5). The flagella are ca. 200 nm wide and ca. 10-15 /nm
long; they are surrounded by a circlet of ca. 10 membrane ridges (Fig. 4a, c) about
1 ^m long and up to 200 nm above the surface of the membrane. Each endodermal
cell appears to have 1 flagellum (Fig. 5). Zooxanthellae are located primarily in those
endodermal cells which lie beneath the outer ectoderm (Figs. Ib, 2c, 5b) although
FIGURE 2. SEMs of a decalcified axial polyp, exposing a cross-section ca. 1 mm from the tip of a
branch, a. Low magnification showing the entire axial polyp. The tentacles (t) are withdrawn into space
left in the calyx of the corallite. The porous wall (i.e.. theca) of the corallite contains the ramifying canals
of the peripheral unit of the fluid transport system. Lower edge is magnified in 2b, c. Scale bar = 250 jum.
b. The ramifying canals of the peripheral unit (p) are located within the porous skeleton (sk), seen in this
view as empty space after the removal of the mineral. Scale bar = 50 ^m. c. Enlargement of 2b showing
the tissue layers at the edge of the axial polyp: outer ectoderm (ec), calicoblastic ectoderm (cec) and
endoderm (en) which contains zooxanthellae (zx). Scale bar = 12.5
624
E. H. GLADFELTER
CIRCULATION IN ACROPOR.4 CERVICORNIS 625
TABLE I
Cross-sectional area (cm2) of the components of the fluid transport system at different distances
from the branch tip
Distance Axial Peripheral unit Radial unit
from tip (cm) unit-axial canal peripheral canals lateral canals
0 .020 .028 —
10 .016 .074 .30
20 .015 .105 .50
30 .008 .135 .90
occasionally they are found deeper within the colony. There appears to be one zooxan-
thella per cell (Fig. 5b).
The surface of the endodermal cells facing the coelenteron may have relatively
few membrane projections (Fig. 4a) between flagella. However, there may be numerous
folds projecting above the surface of the membrane, particularly in those cells con-
taining zooxanthellae and lying adjacent to the outer ectoderm (Figs. Ic, 4c) or
endodermal cells in the gastrovascular pockets of the distal portion of the axial polyp
in the specimens fixed at night (i.e., either 2400 or 0500; Fig. 4b). The surfaces of
cell membranes in specimens fixed during the day (i.e., 1 100 and 1800) have fewer
projections (Fig. 4a).
Endodermal cells at different sites in the gastrovascular system have different
shapes. Cells at the distal tip of the axial corallite, at the distal end of the peripheral
canals are columnar, ca. 12 ^im tall and ca. 1 ^m in diameter (Fig. 6a). Proximal to
the tip, the shape of the endodermal cells becomes squamous (Fig. 6b), only a few
^m tall, and ca. 10 nm in diameter; by 200 pm below the distal tip, most of the
endodermal cells lining the canals of the gastrovascular system have this shape. The
exceptions are cells containing zooxanthellae; these cells, lying beneath the outer
ectoderm and lining the peripheral canals (Figs. 2c, 3b) are tall (ca. 12 ^m) and broad
(ca. 10 /um). There is a large subepidermal space in the endoderm (Fig. 5a, b); it is
more noticeable where cells are not occupied by zooxanthellae.
The distal end of a peripheral canal has a high density of flagella. This is due to
the columnar shape of the cells, with a correspondingly small membrane surface
(containing one flagellum per cell) facing the coelenteron (Fig. 7).
In this study the digestive role of the gastrovascular system was not investigated;
nevertheless, certain observations can be made from SEMs. As noted above, numerous
projections, microvilli, often vastly increase the surface area of the endodermal cell
membrane. Foreign particles were found in contact with these microvilli (Fig. 4b, c).
In several SEM preparations, particulate matter was present in the canals of the
gastrovascular system. In one case the particulate matter was a mass of unidentifiable
smaller particles (perhaps partially decomposed food) entangled by flagella (Fig. 7a,
FIGURE 3. SEMs of the outer edge of a decalcified branch, exposing a tangential section located 45
cm from the tip. a. View of peripheral canals (p) between pockets left after the dissolution of the pseudocostae
of the skeleton (sk). Scale bar = 50 ^m. b. Higher magnification of 3a, showing the configuration of the
tissue layers at the edge of the branch, labeled as in Figure 2. The label, en, is located in the approximate
region magnified in 3c. Scale bar = 12.5 /an. c. View of the membrane surface of endodermal cells lining
the coelenteron. Note the flagella (f) and numerous small projections of the membrane surface. Small
particles, possibly bacteria (b) are also seen on the surface. Scale bar = 1.3 //m.
626
E. H. GLADFELTER
FIGURE 4. SEMs of endodermal cells, showing the surfaces of the membranes which face the coelenteron.
Scale bar = 1.3 jum. a. Endoderm located in a gastrovascular pocket at the tip of the axial polyp; specimen
fixed at 1 500. Note circlet (c) of projections surrounding each flagellum and foreign particles, possibly
bacteria (b) on the surface, b. Endoderm located in a gastrovascular pocket at the tip of the axial polyp;
specimen fixed at 0500. Note the possible phagocytic event (pe). c. Endoderm located beneath outer
ectoderm, 30 mm from branch tip. These cells contain zooxanthellae. Note the possible phagocytic event (pe).
b). In another case a mass of zooxanthellae plus some smaller objects (perhaps bacteria)
were attached to the wall of a canal. In freshly collected coral tips, viewed with a
50X dissecting microscope, free zooxanthellae were observed in the peripheral canals
at the tip of the branch.
CIRCULATION IN ACROPORA CERVICORNIS
627
FIGURE 4. (Continued)
Patterns and rates of fluid transport
To detect the pattern and rate of flow in the gastrovascular system, fluorescein
dye was injected into the system via a lateral polyp. Initially, just after the injection
(t = 0), dye extended 0.7 cm proximally and 0.3 cm distally from the injected polyp.
Usually the dye moved in both directions from the point of injection; distance traveled
was measured each minute. The rate of movement was determined from the slope
of a linear regression, plotting distance versus time; the coordinates at t = 0 were 0.7
cm for the proximal rate and 0.3 cm for the distal rate. In 15% of the trials the initial
dye movement was in a right hand helical direction. The dye front moving proximally
appeared fainter than that moving distally.
To determine if rate or pattern of flow was affected by distance from the branch
tip, corals were injected at either 3 cm, 7 cm, or 10 cm proximal to the tip. To
determine the effect of light on transport of fluids, some corals were maintained under
daylight fluorescent light (750 ft candles), except during the 10 s-min ' when the
room was darkened to observe the position of the fluorescein visible under ultraviolet
light. All determinations of the rate of transport as affected by distance from tip or
by light were made between 1000 and 1500. The results of these experiments are
shown in Tables II and III. The rate of flow was greater in the proximal direction
than in the distal direction in 83% of the branches measured (Table II). This proportion
is significantly different than expected if there were no difference in the rates in the
two directions (P < 0.005, x2 =92.1). Table II also shows that neither light
nor distance from tip affected pattern of flow, i.e., proximal was greater than the
distal rate.
Table III shows that rate of both proximal and distal flow was not significantly
altered (as determined by Mests between the means) by either distance from the tip
or by the presence of light; i.e., for each direction (e.g., proximal) and each distance
from the tip (e.g., 3 cm) the dark value for rate (2. 1 1 cm • min"1) is virtually the same
628
E. H. GLADFELTER
B
">^ Mf
FIGURE 5. SEMs of endoderm: cross-section through the tissue layer exposing flagellated outer mem-
brane surface (om) facing the coelenteron, body of the endodermal cells (cb), subepithelial space (ses), and
inner membrane (im) adjacent to the mesoglea. Scale bar = 2.5 pm. a. Endoderm located in gastrovascular
pockets near the tip of the axial polyp. The cells do not contain zooxanthellae. There is one flagellum per
cell. b. Endoderm located adjacent to outer ectoderm 10 mm from the tip of the axial polyp. These cells
do contain zooxanthellae (zx).
as the light value (2. 10 cm • min ') and the range of values in the proximal direction
(1.72-2.14) found in the three distances from the tip are not statistically different.
However, the mean rate of flow is significantly greater in the proximal than in the
CIRCULATION IN ACROPOH4 CERVICORNIS
629
'W.
i
FIGURE 6. SEMs of longitudinal sections through a decalcified axial polyp, showing differences in
morphology of endodermal cells from different locations, a. Gastrovascular pocket is located near the tip
of the axial polyp with an opening to an adjacent pocket. Outer ectoderm (ec) covers the distal tip and is
separated by mesoglea (m) from the calicoblastic ectoderm (cec) and endoderm (en), b. View ca. 300 pm
from the tip showing the change in the shape of cells of the endoderm (en) and the calicoblastic ectoderm
(cec) from columnar (see 7a) to squamous.
distal direction (P < 0.001, F == 43.51, ANOVA) under all conditions tested (Ta-
ble III).
About 1 h after the termination of a series of measurements, the coral branches
had expelled the fluorescein dye from all portions of the colony. The dye remained
630
E. H. GLADFELTER
FIGURE 7. SEM of a decalcified axial polyp, a. A gastrovascular pocket located at the tip of the
polyp. Note higher density of flagella towards the tip of the pocket. Also note the bolus of foreign material
in the canal of the peripheral unit. Scale bar = 25 ^m. b. Higher magnification of bolus in 7 a. Scale bar
= 6 fj.m.
as a "cocoon" in the coral mucus surrounding each branch until the branches were
rinsed and replaced in an aquarium with fresh flowing sea water.
To determine if rates or pattern of flow varied with time of day, one set of coral
branches were monitored every 3 h, from 1200 on one day up to and including 1200
on the following day. A second set of corals were monitored ca. every 3 h from 1000
to the following 0130. The results are presented in Figure 8. The rate of fluid flow
CIRCULATION IN ACROPOR.4 CERVICORNIS 631
TABLE II
Summary of the direction of fluid movement in the peripheral unit of individual branches^
Pr>
Di
Pr -
Di
Pr <
Di
Distance
Dark (D)
No.
from tip (cm)
or Light (L)
of trials
#
%
#
%
#
%
D
31
25
81
2
7
4
13
3
L
13
10
77
2
15
1
8
D
21
17
82
1
5
3
14
7
L
5
5
100
0
0
0
0
D
5
4
80
1
20
0
0
L
26
22
85
3
12
1
4
Total no. trials
101
83
9
9
% of total trials
83
9
9
' Data presented are the number of trials in which the rate of movement of the dye front was greater
in a preferred direction (i.e., distal, Di or proximal Pr) or equal in both directions (within 0.1 cm • min~').
in the proximal direction was always greater than that in the distal direction, confirming
the results presented above. There was, however, a diel pattern in the rate of fluid
transport. Highest rates occurred between 2400 and 0900, with a rapid decline in the
rate of flow at mid-morning to the low between 1000-1200 until about 1800 when
the flow rates began a gradual increase to the early morning peak.
No movement of fluid occurred in the control branches which had been fixed in
formalin.
All measurements of fluid transport described above refer to observations of dye
moving in an axial orientation (either proximal or distal) just beneath the surface of
the outer ectoderm of the colony. To determine the rate and direction of fluid move-
ment in the axial core of a branch, several approaches were taken. The first was to
observe the time at which dye injected into a lateral polyp was first observed in the
proximal portion of the axial polyp exposed on the open portion of a branch. The
second approach involved directly injecting the axial polyp and noting the time at
which the dye reached the proximal end of the axial polyp. Finally the distal tip (ca.
2 cm) of a coral branch injected either via a lateral polyp or via the proximal end
TABLE III
Rate of fluid movement (cm/min) in peripheral canals in the proximal (Pr) and distal (Di) direction as
affected by distance from branch tip and illumination (L, light; D, dark)
Proximal
Distal
Distance
Dark (D)
from tip (cm)
or Light (L)
X
S.D.
n
X
S.D.
n
D
2.11
0.64
30
1.35
0.54
31
3
L
2.10
0.78
14
1.58
0.74
13
D
1.72
0.82
21
1.02
0.44
20
7
L
1.90
0.80
5
1.05
0.21
5
D
2.14
0.56
5
1.28
0.40
5
10
L
2.08
0.71
26
1.37
0.33
26
632
E. H. GLADFELTER
t 5
o
Q,
1 E 3
!T i> o
O ^
W
_L
_L
1200 1500 1800 2100 2400 0300 0600 0900 1200 1500 1800
Time of day
2200 0130
FIGURE 8. Diel patterns of the rate of fluid movement in the peripheral unit of the gastrovascular
system. Each point represents data from 12 coral branches; closed points are from the second set. Values
plotted are the mean of 12 measurements and the 95% confidence limits. Proximal flow is indicated by
circles; distal flow is indicated by triangles.
of the axial polyp was broken so that dye moving distally in the axial polyp could
be detected. The results of these investigations are summarized in Table IV. Dye
injected into either the axial polyp or a lateral polyp was first observed at the center
of the broken proximal end of the branch (in the axis) and later at the edges (i.e.,
circumference) of the proximal end of the branch. Fluid transport in the distal direction
in the axial canal was never observed. Rate of fluid conduction in the axial canal
was always 2-3 times greater than the rate of fluid transport just beneath the outer
ectoderm. Dye was never seen in the area between the periphery and the axis, indicating
that the canals in this area served mainly for radial conduction of fluids. Some of
the dye injected into a lateral polyp is transported to the axial polyp, and some of
the dye injected into the axial polyp is transported to the periphery.
TABLE IV
Summary of rate and direction of fluid movement in the axial canal
Position of injection
Position of dye after transport
Rate (cm- min ')
S.D.
17 tips*
axial polyp (distal)
axial canal (proximal)
lateral polyp, 5 cm
from tip
axial canal-proximal end
peripheral canal-proximal end
axial canal-distal end
peripheral canals
axial canal-distal end
axial canal-proximal end
peripheral canals
11.4
10.7
rate not measured,
but dye present
NEVER observed
1.9 0.6
NEVER observed
6.5 2.2
range (3.4-9.9)
see Figure 8
range (1-3)
* 74 trials.
CIRCULATION IN ACROPOR.4 CERVICORNIS 633
DISCUSSION
The fluid transport system of Acropora cervicornis consists of the canals of the
gastrovascular system. These canals conduct fluids axially along a branch and radially
between the periphery and axis of the branch. Two units, peripheral (P) and axial
(A) are responsible for conduction along the axis of a branch. The peripheral unit
consists of small peripheral canals. It conducts fluid proximally and distally. The
axial unit consists of the large axial canal. It differs in two respects from the peripheral
unit: 1 ) flow is always in the proximal direction and 2) the rate of flow is 2-3 times
greater. The peripheral and axial units are connected by a radially conducting unit
consisting of the canals of the lateral polyps and smaller short canals within the radial
secondary growth of the skeleton. This radial unit can conduct fluids both towards
and away from the branch axis. Fluid moving from the outside medium into the
gastrovascular system was never directly observed. The "cocoon" of mucus with
expelled dye which surrounds the entire coral branch ca. 1 h after the injection of
the dye suggests that exchange of gastrovascular fluid with the outside medium occurs
via all the lateral polyps and perhaps the axial polyp as well.
The mechanism of fluid propulsion is probably flagellar action. Musgrave (1909)
described a ciliated canal system in a pennatulid, and suggested that it functioned in
intracolonial transport of fluids. Parker ( 1 920) observed that circulation in Renilla
followed a specific route. Thus, the idea of a circulatory system in colonial cnidarians,
with fluid propelled by cilia has been in the literature for a number of years. In A.
cervicornis the flagella are short (10-15 ^m) like cilia, but since there is only one per
cell, the conventional terminology employed by Robson (1957) will be used in this
discussion. In references to past literature, when the term "ciliated canals" is used,
I will refer to the tubules in that form.
The fluid transport system operates under low Reynolds numbers; the Reynolds
number of the axial transport unit (A) and the peripheral unit (P; and p, for one
canal in the unit) can be calculated (Alexander, 1968):
Re = pua/rj
where
p = density of fluid
u = velocity of fluid
a = radius of canal
77 = viscosity of fluid; and
ij/p = v = kinematic viscosity = 10~2 crrr-s '
UA == 10 ' cm-s~'; UP = 3 X 10~2 cm-s~'
aA = 5 X 10~2 cm; ap = 5 X 10~3 cm
so that
ReA = 5 X KT1 and Rep = 1.5 X 10~2.
Since a Reynolds number of >2000 is necessary to produce turbulent flow (Vogel,
1981), the fluids in the gastrovascular system of A. cervicornis have a laminar flow.
Roosen-Runge (1967) described a very similar system in the circulation of fluids
in the canals of a small hydromedusa, Phialideum sp. He observed rates of flow ca.
100 ^ni'S"1 in canals with radii of 25 /mi. Using these values and the Poiseulle
equation he concluded that the circulatory system of the medusa was operating at
pressures of .12 mm-Hg. Unfortunately, Poiseulle's equation cannot be applied in a
634 E. H. GLADFELTER
situation in which fluid is propelled by flagella because Poiseulle's equation depends
on a pressure differential and assumes that "the fluid velocity at the edge of the tube
is zero" (Feigl, 1974). In ciliated tubes the "pump" is located all along the length of
the tube and the flow velocity profile is reversed from that seen in Poiseulle flow
(Vogel, 198 1). A fluid flow profile normal to a ciliated wall shows a maximum velocity
ca. 2 cilia lengths from the ciliated wall with a decrease in velocity to zero at 10 cilia
lengths from the ciliated surface (Cheung and Winet, 1975). In a tube lined with
cilia, each of whose length is 20% of the radius of the tube, the flow velocity profile
is almost flat due to the combined effect on water particles from cilia located on
opposite sides of the tube (Gray, 1928). A peripheral canal of Acropora cervicornis
presents such a situation. The length of a flagellum is ca. 25% of the radius of the
canal. In fluid flow along this type of canal, the most important force is tangential,
the wall shear stress (Brennan and Winet, 1977). Descriptions of fluid flow in ciliated
tubes have largely been confined to mucociliary systems. Even in these accounts there
are not enough sufficient observations or quantitative information to adequately
describe the hydrodynamics of flow (Brennan and Winet, 1977). Perhaps the two
dimensional model, Couette flow (R. Kelly, pers. comm.), describing a wall moving
in relation to a fluid in which tangential force is the most important component
affecting the velocity profile, is most applicable.
In the axial canal of A. cervicornis, flow induced by flagellar beating would produce
a flow velocity profile decreasing from a maximum velocity 2 flagellar lengths from
the wall to a velocity of zero at 10 flagellar lengths from the wall. If flagella are the
only propulsive force, then fluid in the center of the axial canal would be stationary,
since 10 flagellar lengths is equal to about 130 ^m from the wall while the radius of
the axial canal is 500 nm. Whether the central fluid is stationary or whether some
other force moves this fluid cannot be ascertained from this study.
In the canals of the medusa, Phialidium sp. (Roosen-Runge, 1967), muscular
action could affect the direction, and sometimes the rate of flow, but the flagella were
the main driving force. Brafield (1969) concluded that in the pennatulid, Pteroides
sp., the peristaltic muscular contractions were the most important driving force in
the circulation of fluids throughout that colony. In Acropora cervicornis, muscular
action probably plays no role because the canals of the gastrovascular system are set
at a fixed size due to their position, embedded in a rigid skeleton.
In Roosen-Runge's study (1967) he observed the actual movements of particles
within the canals and he was able to discern that flow could proceed in opposite
directions in the same canal. This might provide an explanation for the observation
that the peripheral unit of A. cervicornis can carry fluids in two directions at the same
time. In an analysis of a stationary protozoan, Cheung and Winet (1975) found flow
velocity profiles showing a backflow of fluid between a ciliated wall and up to 0.5
cilia lengths from the wall, with the maximum forward velocity occurring at 2 cilia
lengths from the wall. If this pattern occurs in a peripheral canal of A. cervicornis,
it could be the mechanism by which flow could proceed in the opposite direction in
the same tube. In systems operating under low Reynolds numbers, such as cnidarian
circulatory systems, the fluids act very viscous. Consequently, very little mixing of
adjacent streams need take place (Vogel, 1981). In these peripheral canals, a large
surface area relative to that of the axial canal, presents a site for exchange of dissolved
and paniculate matter. The cell membranes of the endodermal cells lining the pe-
ripheral canals are often highly folded, and phagocytic events can be observed
in SEMs.
Pearse and Muscatine (1971) and Taylor (1977) demonstrated that soluble organic
molecules and inorganic ions are transported distally to the growing tip of Acropora
CIRCULATION IN ACROPOH4 CERVICORNIS 635
cervicornis. Other investigators found that only after a short time (30 min) can ra-
dioactive food fed to a colonial cnidarian polyp be detected in adjacent polyps (Rees
et al, 1970; Murdock, 1978a, b). Furthermore, Rees et al. state that in a growing
hydroid colony, "radioactive food fed to the terminal hydranth seemed to be pref-
erentially utilized by the growing regions." Thus, it is not surprising that the growing
tip of A. cervicornis (ca. 300 /*m • day"1; Gladfelter, 1982) serves as a "sink" for soluble
organic material and Ca+\ required for the development of the axial polyp and the
skeleton (Pearse and Muscatine, 1971; Taylor, 1977). The role of the gastrovascular
system in removing materials from the growing tip has not been investigated. Several
hypotheses to explain light enhancement of calcification depend on the removal of
substances from the sites of crystal deposition (e.g., Goreau, 1959; Simkiss, 1964;
Gladfelter, 1983); the role of the fluid transport system in effecting removal of materials
is unknown.
To resolve some of the questions concerning calcification it would be useful to
know the chemical properties of the fluid in the gastrovascular system. Obviously,
most useful would be data on the chemical composition of fluids outside the cali-
coblastic ectoderm, i.e., just adjacent to the developing skeleton, but these data are
extremely difficult to obtain, even in relatively large volume reservoirs such as the
extrapallial fluids of molluscs (Simkiss, 1982). Additionally, it would be useful to
know the rate of exchange of fluids between the gastrovascular system and the external
medium. On one hand, if the gastrovascular system serves to distribute soluble organic
matter throughout the colony to sites which can use it as an energy source or as
precursor molecules, it would be disadvantageous to rapidly exchange the gastrovascular
fluid for sea water. The zooxanthellae can serve to clean the system of metabolic
wastes, as they effectively remove ammonia (Muscatine, 1980) and probably other
materials as well. However, at some point, the fluid in the gastrovascular system
would become depleted of such things as calcium ions, which the colony needs for
extension of its skeleton as well as increasing the strength of the skeleton by subsequent
infilling of pores (Gladfelter, 1982).
To summarize, the gastrovascular system of Acropora cervicornis serves as a cir-
culatory system, characterized by: (1) two units (axial and peripheral) conducting
fluids by means of flagella along the axis of the branch; (2) a low Reynolds number,
leading to laminar flow; (3) a predictable diel pattern in the rate of flow in the
peripheral unit; and (4) no change in the rate of flow due to light.
ACKNOWLEDGMENTS
I would like to thank W. B. Gladfelter for helpful field assistance and suggestions
during the course of this study, L. Muscatine for encouragement and editorial com-
ments, and N. Merrell for initially developing the technique used to observe fluid
flow. R. Trench, P. Nobel, and R. Kelly provided valuable suggestions to improve
the discussion.
This study was partially funded by a Biomedical Research Support Grant (NIH)
to UCLA and the Chancellor's Patent Fund of UCLA. I wish to thank the West
Indies Laboratory of Fairleigh Dickinson University, particularly the past and present
directors, R. F. Dill and J. C. Ogden, for the use of its facilities. I also wish to thank
S. Bedlyar and J. Berliner for the use of the SEM facility in the School of Medicine
at UCLA. Support during the final year of this study was provided through the Meta
McBride Haupt Dissertation Fellowship from the American Association of Univer-
sity Women.
This paper is West Indies Laboratory Contribution #99.
636 E. H. GLADFELTER
LITERATURE CITED
ALEXANDER, R. M. 1968. Animal Mechanics. University of Washington Press, Seattle. 346 pp.
BRAFIELD, A. E. 1969. Water movements in the pennatulid coelenterate P leroides grise urn. J. Zoo/., Land.
158: 317-325.
BRENNEN, C., AND H. WINET. 1977. Fluid mechanics of propulsion by cilia and flagella. Ann. Rev. Fluid
Mech. 9: 339-398.
CHEUNG, A. T. W., AND H. WINET. 1975. Flow velocity profile over a ciliated surface. Pp. 223-234 in
Swimming and Flying in Nature. T. Y-T. Wu, C. J. Brokaw, and C. Brennen, eds. Plenum Press,
New York.
FEIGL, E. O. 1974. Physics of the cardiovascular system. Pp. 10-22 in Physiology and Biophysics II,
T. C. Ruch and H. D. Patton, eds. W. B. Saunders, Co., Philadelphia.
GLADFELTER, E. H. 1982. Skeletal development in Acropora cervicornis: I. Patterns of calcium carbonate
accretion in the axial corallite. Coral Reefs 1: 45-51.
GLADFELTER, E. H. 1983. Skeletal development and related aspects of the reef coral Acropora cervicornis.
Ph.D. dissertation. University of California, Los Angeles.
GOREAU, T. F. 1959. The physiology of skeleton formation in corals. I. A method for measuring the rate
of calcium deposition by corals under different conditions. Biol. Bull. 116: 59-75.
GOREAU, T. F., AND N. I. GOREAU. 1959. The physiology of skeleton formation in corals. II. Calcium
deposition by hermatypic corals under various conditions in the reef. Biol. Bull. 117: 239-250.
GRAY, J. 1928. Ciliary Movement. The University Press, Cambridge. 162 pp.
MURDOCK, G. R. 1978a. Digestion, assimilation, and transport of food in the gastrovascular cavity of a
gorgonian octocoral (Cnidaria: Anthozoa). Bull. Mar. Sci. 28(2): 354-362.
MURDOCK, G. R. 1978b. Circulation and digestion of food in the gastrovascular system of gorgonian
octocorals (Cnidaria: Anthozoa). Bull. Mar. Sci. 28(2): 363-370.
MUSCATINE, L. 1980. Uptake, retention and release of dissolved inorganic nutrients by marine alga-in-
vertebrate associations. Pp. 229-244 in Cellular Interactions in Symbiosis and Parasitism, C. B.
Cook, P. W. Pappas, and E. D. Randolph, eds. Ohio State University Press, Columbus.
MUSGRAVE, E. M. 1909. Experimental observations on the organs of circulation and powers of locomotion
in pennatulids. Q. J. Microsc. Sci. 54: 443-482.
PARKER, G. H. 1920. Activities of colonial animals. I. Circulation of water in Renilla. J. Exp. Zool. 31:
342-367.
PEARSE, V. B., AND L. MUSCATINE. 1971. Role of symbiotic algae (zooxanthellae) in coral calcification.
Biol. Bull. 141: 350-363.
REES, J., L. V. DAVIS, AND H. M. LENHOFF. 1970. Paths and rates of food distribution in the colonial
hydroid Pennaria. Comp. Biochem. Physiol. 34: 309-316.
ROBSON, E. A. 1957. The structure and hydrodynamics of the musculo-epithelium in Metridium. Q. J.
Microsc. Sci. 98(2): 265-278.
ROOSEN-RUNGE, E. C. 1967. Gastrovascular system of small hydromedusae: Mechanisms of circulation.
Science 156(3771): 74-76.
SIMKISS, K. 1964. Phosphates as crystal poisons. Biol. Rev. 39: 487-505.
SIMKISS, K. 1982. Mechanisms of mineralization (normal). Pp. 351-366 in Biological Mineralization and
Demineralizatiofi, G. H. Nancollas, ed. Springer- Verlag, New York.
TAYLOR, D. L. 1977. Intra-colonial transport of organic compounds and calcium in some Atlantic reef
corals. Proc. 3rd Int'l Coral ReefSymp. I: 431-436.
VOGEL, S. 1981. Life in Moving Fluids. Willard Grant Press, Boston. 352 pp.
WELLS, J. W. 1956. Scleractinia. Pp. 328-444 in Treatise in Invertebrate Paleontology, R. C. Moore, ed.
Geol. Soc. of Am. and University of Kansas Press.
Reference: Biol. Bull. 165: 637-685. (December, 1983)
SURVIVAL, GROWTH, AND BEHAVIOR OF THE LOLIGINID SQUIDS
LOLIGO PLEI, LOLIGO PEALEI, AND LOLLIGUNCULA BREVIS
(MOLLUSCA: CEPHALOPODA) IN CLOSED SEA WATER SYSTEMS
ROGER T. HANLON, RAYMOND F. HIXON, AND WILLIAM H. HULET
The Marine Biomedical Institute, The University of Texas Medical Branch,
200 University Boulevard, Galveston, Texas 77550-2772
ABSTRACT
Over 1200 squids were captured by night lighting, trawling, or seining in the
northern Gulf of Mexico for laboratory maintenance. Two types of recirculating sea
water systems were designed and evaluated: a 2 m circular tank ( 1 500 liter capacity)
and a 10 m long raceway (10,000 liters). Mean laboratory survival was: Loligo plei
(12 to 252 mm mantle length, ML) 1 1 days, maximum 84 days; Loligo pealei (109
to 285 mm ML) 28 days, maximum 71 days; Lolliguncula brevis (27 to 99 mm ML)
19 days, maximum 125 days. Smaller squids showed significantly poorer survival
than larger ones. All squids fed well on a variety of live estuarine fishes and shrimps.
Growth rates depended upon stage of maturity. The highest rates were Loligo plei
59 mm/month (23.8 g/mo), Loligo pealei 77 mm/mo (67.3 g/mo), and Lolliguncula
brevis 31 mm/mo (17.2 g/mo). General aspects of behavior and body patterning were
species-specific and were useful indices of the squids' condition. Key factors for lab-
oratory survival were (1) prevention of skin damage, (2) tank systems with sufficiently
large horizontal dimensions, (3) high quality water, (4) ample food supply, (5) no
crowding, (6) maintaining squids of similar size to reduce aggression and cannibalism,
and (7) segregating sexes to reduce aggression associated with courtship, mat'-ig, and
egg laying.
INTRODUCTION
Pelagic, schooling squids of the Order Teuthoidea are powerful swimmers that
forage over great distances in coastal and open-ocean waters. These dynamic predators,
with their highly developed organ systems, are of great interest and use to the scientific
community, mainly because they have a network of giant axons that mediates a
simultaneous contraction of the mantle for jet-propulsed swimming from predators.
Historically, researchers have experienced difficulty in collecting and maintaining
these animals alive in captivity, due primarily to damage of the delicate squid skin
during capture, transport, and maintenance. Over the past ten years, considerable
progress has been made in identifying and resolving problems associated with keeping
squids alive under laboratory conditions. Since 1975 we have reviewed, tested, and
refined many techniques for the capture and maintenance of squids, with the ultimate
goal of supplying neuroscience investigators at The University of Texas Medical
Branch with live squids. We present here our capture, transport, and maintenance
methodology, the design of our closed sea water systems, and we describe the survival,
growth, and general aspects of behavior of squids maintained in these systems.
Received 9 May 1983; accepted 29 August 1983.
637
638 R. T. HANLON ET AL.
Three loliginid squid species (Suborder Myopsida, Family Loliginidae) are com-
monly found on the continental shelf in the northern Gulf of Mexico near Galveston:
the tropical arrow squid Loligo (Doryteuthis) plei Blainville, 1823, the common long-
finned squid Loligo pealei Lesueur, 1821, and the bay or brief squid Lolliguncula
brevis (Blainville, 1823). Aspects of the areal and bathymetric distribution of these
species are described by Rathjen et al (1979), Hixon (1980a) and Hixon et al. (1980).
Loligo plei and L. pealei in the Gulf of Mexico attain maximal reported sizes of 297
mm and 285 mm mantle length (ML), respectively (Rathjen et al., 1979; Hixon,
1980a; Hixon et al., 1980), and they are well-established experimental models, primarily
for studies of the giant fiber system (cf., Rosenberg, 1973; Arnold et al., 1974; DiPolo,
1976; Tasaki, 1982). Lolliguncula brevis is a smaller species, maximal 107 mm ML,
that has potential for a variety of scientific applications (Hulet et al., 1980; Hendrix
et al.. 1981).
MATERIALS AND METHODS
Capture
Field collections were made from two University of Texas research vessels, the
16 m stern trawler R/V ERIN LEDDY-JONES and the 12 m R/V VIRGINIA
BLOCKER. The R/V ERIN LEDDY-JONES was equipped for bottom trawling and
for night lighting with three quartz iodide lamps controlled by rheostats (Fig. 1 ). One
1000-watt lamp was mounted on the stern A-frame, and two 500-watt lamps were
located on either side of the rigging amidships. The R/V VIRGINIA BLOCKER was
used for night lighting only. It deployed two portable 500-watt lamps astern or a 500-
watt underwater mercury vapor lamp.
Great emphasis was placed on obtaining squids by methods that imparted little
or no skin damage, particularly to the fins (Hulet et al., 1979). Both species of Loligo
were captured alive by attracting them to bright lights at night and dipnetting them
onboard. Squid jigs were often used at night-light stations to lure squids to the surface
where they were more easily dipnetted. The dipnets were 3 or 5 m-long aluminum
poles attached to a 46 cm-diameter stainless steel hoop with a shallow net made of
soft 1.3 cm ('/2 inch) knotless nylon mesh. Every effort was made to handle the squids
briefly and gently. After dipnetting, squids were immediately immersed into a shipboard
sea water transport tank so that their water-to-air-to-water transfer lasted only several
seconds.
Lolliguncula brevis was captured by bottom trawling and beach seining. Trawl
durations were very short (5 to 15 minutes) and in shallow water (3 to 10 m) in and
around Galveston Bay, so that residence time in the net was short and squids were
not tightly compressed in the codend for long periods. Forward speed of the vessel
was reduced during trawl retrieval and only the codend was swung onboard, placed
in water, and the squids quickly placed by hand into transport tanks. Several trawl
nets were used, including a 9.1 m-wide (length of foot rope) semi-balloon trawl, a
3.0 m-wide shrimp try net, and 3.0 m-, 6.4 m- or 9.1 m-wide box trawls constructed
by Marinovich Trawl Co. (Biloxi, Mississippi). The semi-balloon trawl and the try
net were made of 3.8 cm stretch mesh nylon netting with a codend inner liner of
1.3 cm mesh knotless nylon netting. The box trawls were constructed entirely with
knotless nylon netting ( 1 .9 and 1 .3 cm mesh) and were fitted with stainless steel hoops
in the codend. Beach seining for Lolliguncula brevis took place at night in summer
on the bay side of Galveston Island. Short tows (5 minutes) were made with a 30.5
m-long by 2 m-wide bag seine constructed of 1.3 cm knotless nylon mesh. Squids
were transported to the laboratory within one hour of capture.
SQUID MAINTENANCE, GROWTH, AND BEHAVIOR 639
Shipboard transport and laboratory transfer
Squids were always immediately segregated from other captured organisms (e.g.,
fishes and other invertebrates) and transported in one of three types of shipboard
tanks (Fig. 1 ). The first type consisted of simple, vertically oriented Nalgene cylinders
of 200 or 380 1 capacity ( VT, vertical tank), with no flow-through of sea water. The
second consisted of fiberglass cylinders of 380 or 550 1 capacity mounted horizontally
on skids (HCT, horizontal cylindrical tank) and fitted with a rectangular, hinged
opening along the length of the upper surface. These horizontal cylinders substantially
reduced sloshing, thereby decreasing haphazard contact between the squids and tank
wall. The third type was a 1020 1 fiberglass rectangular horizontal tank (RHT) with
rounded corners. This tank was separated into two layers that doubled the transport
capacity over other designs. In the latter two systems, fresh running sea water was
pumped into the top and flowed out the bottom. These systems were switched to a
recirculating mode and 100 percent oxygen was bubbled into the water when Loligo
spp. were being transported from offshore and water quality deteriorated nearshore.
Most tanks were covered with polystyrene on the outside for insulation.
For transfer from shipboard to the laboratory, the squids were caught with small
dipnets and placed by hand into clear plastic bags (38 cm X 80 cm), with care being
taken not to startle the squids and cause inking. One to three squids and approximately
4 1 of sea water were put in each bag, which was then pumped full with 100 percent
oxygen and tied off. Several bags were placed horizontally in an insulated container,
the top was closed, and they were taken by truck several hundred meters to laboratory
tank systems with similar temperature and salinity. The bags were floated in the tanks
for 1 5 to 30 minutes until temperatures equilibrated. Each bag was then opened and
the squids were released directly into the tank without handling.
Throughout this paper we report our results as mean and median values, but only
median values were compared statistically because we used non-parametric tests (see
Gibbons, 1976). The shipboard transport, laboratory transfer, and 24 hour acclimation
mortality data were analyzed statistically to test for ( 1 ) differences in mortality among
the three species, (2) differences in the performance of the three tank designs, and
(3) differences in mortality associated with squid size. The first comparison (Kruskal-
Wallis test) was carried out among all three species, using only the HCT data. The
performance of the tank systems was evaluated (Kruskal-Wallis test) using the data
of Loligo plei because it was the only species transported in all three tanks. Finally,
the third comparison (Mann- Whitney U test) was made between the sizes of L. plei
that died versus those that survived in the VT and HCT transport tanks; similar data
were not available for the other two species.
Closed sea water systems
A major objective was to develop a large-volume, inexpensive, easily reproducible
sea water system that could be modified to test different techniques for maintaining
and growing squids. Two basic systems were developed, both being closed systems
that recirculated and filtered their own set volume of sea water.
The 2 m circular tank (CT) system (Fig. 2) is a simple and readily modified design
that we developed in 1975 and continues to be our standard system for maintenance
and experimentation (Hanlon et al, 1978). Its capacity is approximately 1500 1 of
sea water. Biological filtration, which includes mineralization, nitrification, and dis-
similation of nitrogenous compounds (cf., Spotte, 1979a, b), is carried out principally
in the filter bed. This layer is 6 cm deep and consists exclusively of crushed oyster
shell (approximate particle size 10 X 5 X 2 mm; total weight approximately 160 kg)
640
R. T. HANLON ET AL.
FIGURE 1 . Capture and transport. A. R/V ERIN LEDDY-JONES nightlighting for Loligo plei off
the coast of Galveston, Texas in 17 m of water. Note the 1000-watt quartz-iodide lamp on the A-frame
and two 500-watt quartz-iodide lamps amidships that are used to attract squids. Squids are dipnetted on
board (left) and placed in a transport container (arrow). B. Three types of transport containers: VT is the
vertical tank; HCT is the horizontal cylindrical tank; RHT is the rectangular horizontal tank. The tanks
and squids are all drawn to the same scale. The squids equal the approximate size of 200 mm mantle
length. Water flow is indicated by arrows. In the RHT, (A) is the removable partition that is replaced when
approximately 15 adult Loligo spp. are put in the tank. The tank top (B) is then secured with stainless
steel bolts (C) that force a rubber gasket (D) against the top edge of the tank, producing a water-tight seal.
Another 15 squids are placed in the upper compartment through the chimney (E). When water quality is
good, sea water is continually pumped into the base of the tank through (F) and allowed to overflow from
the chimney. When water quality deteriorates near shore, the tank water is circulated by a submersible
pump (G) that pushes the water through an exterior filter (H) and back into the tank. Pure oxygen or air
may be added through a valve (I).
on which bacteria attach and grow. Newly constructed systems are "conditioned"
for several weeks to allow bacterial populations to equilibrate. Toxic ammonia, directly
excreted by tank animals or produced indirectly through mineralization of organic
products, is oxidized by nitrifying bacteria in the filter bed to nitrite and then to less
toxic nitrate. Nitrate is either assimilated by green algae growing in the algal tank
under continuous illumination, removed through partial water changes, or removed
through dissimilation by bacteria into a completely reduced state in which inorganic
nitrogen is released from the water into the atmosphere (Painter, 1970).
Mechanical filtration reduces water turbidity by separating and concentrating
particulate organic carbon (i.e., particles, aggregates, detritus, free floating algae, and
bacteria) in the filter bed and in two layers of polyester fiber within an auxiliary filter
(Fig. 2). Physical adsorption of dissolved organic carbon is accomplished with granular
activated carbon in the auxiliary filter or with the periodic use of a foam fractionator
SQUID MAINTENANCE, GROWTH, AND BEHAVIOR
641
HCT
550 liters
Scale = 0.5 m
VT
380 liters
X
RHT
1020 liters
FIGURE 1. (Continued)
or "protein skimmer" that physically binds surface-active organic material to the air-
water interface of bubbles and chemically binds non-surface-active compounds with
surface-active material (Rubin et al, 1963). This is necessary when the tank is loaded
to high capacity and partially eaten food accumulates in the system. Flow rate through
the system is approximately 16 1 per minute.
All fabrication materials are fiberglass, polyvinyl chloride (PVC), or some other
inert synthetic product. The only metal components are in the pumps and they do
not come in contact with sea water. Tank walls are painted with various patterns
made with an inert black paint (Thixochlor, Napko Paint Co., Houston, Texas) to
increase contrast and make the walls more visible to squids. Partitions that divided
the tank into two or four segments were used occasionally and were constructed of
642
R. T. HANLON ET AL.
Side View
Top View
*" 2m
SQUID MAINTENANCE, GROWTH, AND BEHAVIOR 643
a PVC frame with soft knotless nylon netting. Several CT systems constructed since
1978 have been used without the algal tank.
Our second design is a 10,000 1 raceway (RW) system based upon a design for
the intensive closed-system culture of penaeid shrimps (Mock el al, 1977). Our
original raceway (Fig. 3) was 10 m long, 2 m wide, and had rounded ends. The
raceway framework consisted of aluminum struts that supported walls made of plywood
panels, and rounded ends made of curved aluminum sheeting. A watertight rubberized
liner was placed inside the framework, and a corrugated fiberglass partition was
suspended lengthwise inside the raceway. Air-lift pumps attached to the central partition
provided aeration and water circulation. A CT system (similar in design to Fig. 2)
was connected to the raceway to provide water filtration. The algal tank consisted of
eight rotating polystyrene "biodiscs" that provided a larger surface area for algal and
bacterial growth (Antonie, 1976).
Newer raceways now in operation consist of a single long fiberglass sheet that is
curved upward on the sides by supporting struts and closed at each end by a rounded
fiberglass half-circle. They may be ordered in a variety of lengths and widths (Ewald
Mfg., Karnes City, Texas). The central partition can be omitted to provide greater
horizontal space. In this case, water is pumped through auxiliary filters (similar to
those shown in Fig. 2, part B) and re-enters the raceway to provide aeration and
directional water flow.
When necessary, water is chilled by cooling units (Model Dl-100, Frigid Units,
Inc., Toledo, Ohio). A deionized water unit provides water for mixing artificial sea
water and for replacing water lost through evaporation. Polystyrene panels are fitted
over the tops of the raceway and the CT system. These covers reduce evaporation,
provide some temperature insulation, and prevent outside activity from disturbing
experimental animals.
Both natural and artificial sea water (Instant Ocean Brand, Aquarium Systems
Inc., Eastlake, Ohio) have been used in our tanks. Water quality was monitored
frequently. Temperature, salinity, and pH were recorded every one to three days.
Estimates of inorganic nitrogen buildup were made biweekly with field test kits (Hach
Chemical Co., Ames, Iowa) and precise measurements were made periodically for
ammonia (Solorzano, 1969), nitrite (Strickland and Parsons, 1972) and nitrate (Rand
el al., 1976). No tolerance levels for these ions have been established for cephalopods,
but a partial water change was made when the concentrations exceeded those rec-
ommended for most marine animals (Spotte, 1973, 1979a, b). At approximately
monthly intervals, a trace element mix (Wimex Trace Elements, Hawaiian Marine
Imports, Houston, Texas) was added to each system to replenish those trace elements
lost through algal metabolism. Dissolved oxygen measurements were made infre-
quently, but were always near saturation. Activated carbon in the auxiliary filters
was changed every four to ten weeks, depending upon the animal load in the system.
The foam fractionators and UV sterilizers were used continuously. Lighting was from
indirect natural sunlight and from overhead fluorescent lights regulated to provide a
natural light/dark photoperiod.
FIGURE 2. The 2 m circular tank (CT). This closed sea water system is shown with 1 1 female Loligo
plei. A pump (A) pushes water to an auxiliary filter (B), where it then flows by gravity through two layers
of polyester fiber (C) and granular activated carbon (D) into an algal tank (E) that is under continuous
illumination (F) and back into the squid holding tank (G). Water circulation in G moves in a clockwise
direction that is caused primarily by the flow from air-lift pumps (H). Water is drawn through the filter
bed (I) into the perforated subsurface pipes of the air-lift pumps (H). Water is also drawn into another set
of subsurface pipes (J) by the pump (A). Various painted patterns (K) make the wall more visible to the
squids. Viewing ports (L, and arrows in photograph) are used for underwater observations.
644
R. T. HANLON ET AL.
SQUID MAINTENANCE, GROWTH, AND BEHAVIOR 645
Recapturing, handling, and anaesthetizing live squids
Squids could be dipnetted from both tank systems due to the narrow dimensions
of each. It was usually possible to slowly herd the schools into particular sections of
the tank and then isolate individuals for netting (Fig. 3). Dipnetted animals were
handheld and gently immersed into a solution of 1.0 to 1.5 percent ethanol in sea
water for one to three minutes until respiratory movements slowed or stopped. Squids
occasionally inked in the ethanol solution, and the ink was immediately dipnetted
out with a fine-mesh net. The anaesthetized animals could be examined, weighed,
or measured for a period of five to ten minutes. Thereafter, each squid was placed
by hand into sea water and rocked to and fro for 30 to 180 seconds until it regained
alertness and body control and swam off. Squids released directly into the tank before
full recovery were often attacked by other squids.
Survival, growth, and mortality analyses
When squids were maintained, daily records were kept for each laboratory tank
system. When each squid died, the date of death, number of days since capture, sex,
mantle length, stage of sexual development, and probable cause of mortality were
recorded. Data on mantle length, sex, and sexual development were unavailable when
squids died from cannibalism or when squid remains were eaten by food organisms
in the tank. Small-sized, usually immature, squids less than 40 mm ML in Lolliguncula
brevis and less than 50 mm ML in Loligo plei were termed "juveniles." The analysis
of variance procedure by ranks (Conover and Iman, 1 976) was used to detect differences
in laboratory survival time observed among (1) the three species and (2) males,
females, and juveniles within each species.
For determinations of growth of laboratory animals, squids were maintained in
the CT systems. Individuals that were in good condition after one week in captivity
were measured at intervals of 7 to 2 1 days. After being anaesthetized, each squid was
(1) held head-down for several seconds to allow the mantle cavity to drain, (2) gently
blotted on absorbent paper towels, (3) sexed, (4) weighed to the nearest gram, and
(5) measured (dorsal mantle length) to the nearest mm. No attempts were made to
mark squids for identification, but notes were taken of recognizable differences in
individuals (e.g., scars, damaged chromatophores) and this was sufficient to identify
squids in subsequent examinations. Throughout the growth observations, palaemonid
shrimps and various small cyprinodont fishes were fed to the squids at least twice
daily. There was an excess of live food in the tanks at all times. Male and female
squids were segregated in the Loligo spp. observations but not in those of Lolliguncula
brevis.
Increases or decreases in mantle length or wet weight over the duration of the
growth observation were expressed in two ways: (1) as the change in mantle length
or wet weight per month (30 days; abbreviated mo), and (2) as an instantaneous
FIGURE 3. The 10,000 1 raceway (RW), a closed sea water system. Seventeen Loligo plei, mostly
males, are seen swimming over a white fiberglass sheet (put in for the photograph only). The air-lift pumps
are turned off for the photograph. Water leaves the raceway (A) via a siphon to a pump (B) that pushes
the water to a rotating biodisc tank (C) that is under continuous illumination (D). The water flows by
gravity first into the auxiliary filter (E) then through the main filter (F, a CT system) and then back to the
raceway. Water within the raceway is circulated in a clockwise direction by the discharge from the main
filter and by the air-lift pumps. Two air blowers (I) drive the air-lift pumps. A foam fractionator (J) is
mounted in the raceway. Note the accessibility of the squids for recapture.
646 R. T. HANLON ET AL.
relative growth rate expressed as the percent gain in length or weight per day ( Winberg,
1960). Statistical analyses were conducted only with Lolliguncula brevis, and using
only monthly changes in mantle length and wet weight; no statistical comparisons
were made with Loligo spp. due to the small sample size. Tests were made on Lol-
liguncula brevis to compare the monthly growth rates of males and females (Mann-
Whitney U test), and to detect size-dependent differences in growth rate within each
sex (Kruskal-Wallis test).
Two six-day observations were conducted with six Loligo plei to collect preliminary
data on gross growth efficiency (or food conversion efficiency). This was measured
as the ratio G/I X 100, where G was wet weight increase of the squid and I was total
wet weight of food ingested. Only fishes were used as food. All fishes were weighed
before entry into the tank (twice per day) and their remains removed and weighed
daily. The ingested wet weight of fish was calculated simply by subtracting the total
weight of food remains from the total weight of fishes.
Feeding
Daily feeding consisted of small live estuarine fishes and shrimps. Principal food
organisms included the sheepshead minnow Cyprinidon variegatus, the longnose
killifish Fundulus similis, the diamond killifish Adinia xenica, the sandtrout Leiostomus
xanthurus, the sailfin molly Poecilia latipinna, the tidewater silverside Menidia ber-
yllina, juvenile mullet Mugil spp., juvenile menhaden Brevoortia spp., juvenile and
adult penaeid shrimp Penaeus spp., and the palaemonid shrimp Palaemonetes pugio.
These species were all readily seined throughout the year in nearby salt marsh areas,
estuarine bay waters, and low-energy beachfronts. These organisms are part of the
natural diet of Lolliguncula brevis, but not of Loligo plei or L. pealei, which come
from offshore. In most cases food was dropped into the tank two or three times per
day in quantities that allowed feeding ad libitum throughout the day and night. Most
prey organisms were equal to or slightly greater than the length of the squids' arms,
but on some occasions mid-sized squids attacked and ate prey organisms nearly their
own length. Food remains were netted or siphoned out of the tanks daily.
Behavioral observations
In the laboratory, squids were observed carefully and often from above the tanks
or, more commonly, through the windows in the tank walls. In order to determine
possible direct and indirect causes of mortality, particular attention was paid to chro-
matophore patterning, postures, and general aspects of behavior associated with tem-
perature or salinity shock, fin damage, feeding, and intraspecific interactions.
In the field, Loligo spp. were observed occasionally by skin or SCUBA diving
near the boat during night lighting stations offshore from Galveston and throughout
the western Gulf of Mexico. More extensive night diving observations were made on
Loligo plei in St. Croix in 1978 (Hanlon et al, 1980) and at Grand Cayman Island
in 1980 (Hanlon and Hixon, 1981).
RESULTS
Capture
For the analyses of shipboard transport and laboratory maintenance, a total of
700 Loligo plei and 89 Loligo pealei were captured by night lighting and dipnetting,
SQUID MAINTENANCE. GROWTH, AND BEHAVIOR 647
and approximately 425 Lolliguncula brevis were captured by bottom trawling. Many
more squids were obtained during these collections between 1976 and 1982, but they
were preserved for other studies. A wide size range of animals was collected in the
northern Gulf of Mexico for these analyses: Loligo plei 12 to 252 mm ML; Loligo
pealei 109 to 285 mm ML; and Lolliguncula brevis 21 to 99 mm ML.
Capture results for each species varied with season, year, and collection site. The
areal and bathymetric distributions of the three species near Galveston have been
outlined previously (Rathjen el al., 1979; Hixon el al., 1980), and Hixon (1980a, b)
has described aspects of the seasonal movements and abundance of each species. At
present, we estimate the seasonal availability of each species as follows. Loligo pealei
is generally present on the edge of the continental shelf (40 to 250 m deep), but this
species is more abundant in fall, winter, and spring and less abundant in summer.
Loligo plei is present closer to shore in depths between 20 and 75 m. Large adults
are most abundant in spring and summer. Small and mid-sized L. plei are found
farther offshore during fall and winter, but large adults disappear from the northwestern
Gulf in early fall and do not reappear until spring. Lolliguncula brevis is present year-
round. It is usually abundant in Galveston Bay ( 1 to 20 m deep) between early spring
and late fall. When bay temperatures drop in winter, this species moves out of the
bay to nearshore waters less than 40 m deep along the Texas coast.
Both species of Loligo came to night lights, but Loligo plei did so more readily
and consistently than Loligo pealei. We conducted 164 night light stations for these
species totaling 30 1 hours of observations. Capture rates by dipnet were low for both
species: 5.0 squids/h for L. plei and 0.9 squids/h for L. pealei. Occasionally there
were highly productive nights in which hundreds of squids could be easily dipnetted;
on these nights capture was terminated quickly when onboard tanks were filled, so
the capture rates are conservative. Other contributing factors to the low numerical
catch rate were that: (1) squids were often present, but out of dipnet range, (2) very
small squids were often not collected, and (3) there were seasonal and yearly decreases
in squid abundance and many of these observations were taken during year-round
exploratory fishing.
The quartz-iodide lamps deployed above water generally attracted more squids
than the underwater mercury vapor lamp, but our attempts to quantify this observation
have failed (e.g., Hanlon el al., 1980; Hanlon and Hixon, 1981). Changing the light
intensity to draw in squids seen on the periphery of the lighted area did not work
consistently. Some squids were caught with squid jigs, but usually jigs attracted squids
near the boat for dipnetting. Thus far, no particular style of jig has been effective for
consistently capturing these species of Loligo, although a wide variety of jigs from
Japan and South America has been used (cf, Rathjen el al., 1979, Fig. 4).
Trawling and seining have been reasonably successful capture methods for Lol-
liguncula brevis. The slow-moving nets were effective because this species lives in
nearshore waters of high turbidity, thus reducing net avoidance. We believe that the
large trawls (6.4 and 9.1 m) caught higher numbers of Lolliguncula brevis than the
small trawls (3.0 m), but comparisons could not be made because of the wide variability
in the seasonal and yearly use of the nets and differences in the abundance of the
animals.
Various injuries were sustained by the squids during capture, and these affected
their subsequent survival during transport and later in the laboratory maintenance
tanks. Loligo plei and Loligo pealei caught with dipnets were practically unharmed
when placed in the onboard transport tanks. Squids caught with jigs had small puncture
wounds in the arms, tentacles, and funnel, but no permanent damage was done to
the skin on the mantle and fins. In comparison, most of the squids caught by the
648 R. T. HANLON ET AL.
trawls sustained skin abrasion caused by the net or other captured animals. The use
of nets such as box trawls or beach seines, which are constructed entirely of knotless
nylon netting, may reduce skin abrasion caused by the knots in conventional nets.
Survival in the trawls was poor when squids were caught with stinging jellyfishes or
organisms with hard or pointed exoskeletons such as crabs. Squids generally survived
capture better when caught with moderately large numbers of small schooling fishes
such as anchovies or menhaden.
Shipboard transport and laboratory transfer
Success in shipboard transport varied greatly depending upon the species caught,
the squids' size, and physical condition after capture, time in transport, sea and
weather conditions, and type of shipboard transport tank (Tables I, II, and III).
Mortality associated with shipboard transport and laboratory transfer included squids
that died any time from capture through their first 24 hours of acclimation in the
laboratory tank systems. Average mortality was 35 percent for Loligo plei during a
mean transport time of 7 hours (standard error of the mean, Sx, 1.1 hours). Average
mortality was 48 percent for Loligo pealei during a mean transport time of 1 5 hours
(Sx = 3.2 hours). Average mortality was 27 percent for 324 Lolliguncula brevis during
a mean transport time of 1 hour (Sx = 0.4 hours). However, no statistically significant
differences were found in median mortality (L. plei, 17.5 percent; L. pealei, 33 percent;
Lolliguncula brevis, 29 percent) among the three species when transported in the
HCT. Most mortality in Loligo plei occurred in small squids less than 50 mm ML,
some of it due to cannibalism by larger squids. High mortality in Loligo pealei was
attributable to the long transport times and the relatively small horizontal tanks (380
and 550 1 HCT) in which this large species was transported. In contrast, Lolliguncula
brevis had the shortest transport time and low mortality; a contributing factor was
that mortality rates associated with beach seining (Table III, Observations 9, 10, 11,
and 12) were between only 0 and 13 percent.
Mortality in the vertically oriented cylinders (VT) was high compared to the
horizontal cylindrical tank (HCT) or the rectangular horizontal tank (RHT). When
mortality of all squids of all three species was compared by type of transport tank,
overall pooled mortality in the vertical tanks was 47 percent versus 28 and 24 percent
in the other tank designs. For Loligo plei, transport in the vertical tanks resulted in
53 percent overall pooled mortality versus 20 and 24 percent in the HCT and RHT
tanks, respectively (Table I). However, for this species no statistically significant dif-
ferences in median mortality (VT, 33 percent; HCT, 17.5 percent; RHT, 16 percent)
were found among transport containers (Kruskal-Wallis test, .05 < P < .10). Nev-
ertheless, we found the vertical tanks unacceptable because of the lack of flowing sea
water and because their narrow horizontal dimensions led to crowding, uncontrolled
water sloshing, and fin and skin damage due to collisions with the tank wall.
The horizontally oriented cylinders and the rectangular tank functioned better
than the vertical tanks. The closed tops in both designs substantially reduced sloshing,
thereby decreasing haphazard contact between the squids and the tank walls. When
sea conditions were good, squids swam in the middle of the water column or slightly
nearer the bottom; in general, the upper half of the water column was unused by the
squids. The 1020 1 rectangular horizontal tank successfully utilized this upper part
of the water column by insertion of a horizontal divider after a number of squids
had already distributed themselves across the bottom of the tank. The next batch of
squids was then collected and placed in the upper level.
Small-sized squids did not withstand capture and transport as well as larger con-
SQUID MAINTENANCE, GROWTH, AND BEHAVIOR 649
specifics. For example, Loligo plei that survived transport in the HCT were significantly
(Mann- Whitney U test, P < .001) larger (median ML 1 13 mm) than squids that died
during transport (median ML 53.5 mm). Similar, but not statistically significant,
results were found in L. plei transported in the VT. Smaller squids were generally
more damaged during capture, they incurred more skin damage from wall contact
during transport, and adult Loligo often cannibalized smaller squids in the same
transport tank.
Transferring squids in plastic bags to the laboratory was successful. Although
somewhat time consuming, it insured that each squid had adequate sea water, oxygen,
and space during this critical period. There was also little sloshing, and when there
was sloshing the rounded sides of the horizontally oriented bags reduced fin abrasion.
The squids transferred well in the darkness of the closed container. This served to
cut off their view of all external commotion, to which they reacted poorly. It was
important not to jar the squids during this process. All movements were gentle including
driving, closing truck doors, and carrying the squids into the laboratory; otherwise
the squids were startled and would ink in the bag.
Healthy, undamaged squids of all three species were able to survive substantial
temperature and salinity changes between capture and release into the laboratory
tanks. During transport, temperature and salinity usually changed slightly from con-
ditions at the capture sites. When the squids were transferred to the laboratory mainte-
nance tanks in plastic bags, temperature equilibration usually took place within 30
minutes, whereas salinity changes occurred abruptly when the squids were released
into the tanks. These changes in temperature and salinity are listed for each species
in Tables I, II, and III.
Lolliguncula brevis (Table III) was subjected to the largest temperature and salinity
changes. The largest temperature changes were +1 1 or — 9°C (average change was
approximately ±6°C), and the largest salinity shocks were +12 or —8 ppt (average
shock was slightly less than ±5 ppt). The combination of most extreme change was
in Observation 1, with a salinity decrease of 8 ppt combined with a temperature
decrease of 9°C. As expected, this estuarine nearshore species tolerated salinity and
temperature changes quite well. Hendrix el al. (1981) have recently analyzed salinity
tolerance in this squid and shown that this species is an osmoconformer that readily
moves within salinities between 17 and 36 ppt.
Loligo plei and Loligo pealei tolerated surprisingly large changes with little or no
apparent harm to their subsequent laboratory survival. In L. plei (Table I), the largest
temperature changes were +8 or — 1 1°C (average change approximately ±5°C), and
the largest salinity shocks were +9 or —8 ppt (average shock was about ±3.5 ppt).
The combination of most extreme change was in Observation 17, with a salinity
increase of 7 ppt and a concurrent temperature decrease of 1 1 °C. For L. pealei (Table
II), the largest temperature changes were +3 or — 8°C (average change approximately
±5°C), and the largest salinity shocks were +4 or —5 ppt (average shock was about
+2 ppt). The combination of most extreme change was in Observation 2, with a
salinity increase of 4 ppt and a temperature decrease of 8°C. Presumably the salinity
changes were dealt with by equilibrating blood osmolality through volume regulation,
as found in Lolliguncula brevis (Hendrix et al., 1981).
It was very difficult to detect any deleterious effects of these physiological stresses.
Even in the extreme cases cited above, most of the undamaged animals survived well
in captivity. Squids that had sustained skin trauma during capture and transport were
probably most affected by the additional physiological stress of salinity and temperature
shock. We believe that these squids probably accounted for most of the deaths within
one to five days in captivity.
650
R. T. HANLON ET AL.
TABLE I
Loligo plei: summary oj capture, transport and transfer, and laboratory maintenance
CAPTURE
SHIPBOARD TRANSPORT (TP), LABORATORY TRANSFER
(TF), AND 1-DAY ACCLIMATION (AC,)
Obs.
No.
Date
No.
Squids
Collected
Transport
Container
Trans-
port
Time
(h)
Salinity
Change
(ppt)
Temp.
Change
No. Dead
in TP,
TF, AC,
Percent
Mor-
tality
1
7
JUL 76
17
200 1
VT
18
28-36
32-29
10
59%
2
8
AUG
76
21
200 1
380 1
VT
VT
5
33-32
33-25
30-27
30-24
2
10%
3
22
AUG
76
18
2001
VT
4
33-33
33-28
30-24
30-24
6
33%
4
23
SEP 76
27
2001
VT
4
30-35
30-30
27-23
27-24
2
7%
5
2
OCT
76
15
2001
VT
24
29-32
27-24
11
73%
6
11
OCT
76
29
2001
VT
3
30-32
24-22
6
21%
7
1
NOV
76
80
2001
VT
3
35-32
17-22
77
96%
8
2
NOV
76
14
200 1
380 1
VT
VT
3
36-32
18-22
7
50%
9
9
NOV
76
35
380 1
VT
3
36-32
21-22
15
43%
10
4
DEC
76
28
380 1
VT
24
35-32
18-22
25
89%
11
20
JAN 77
3
3801
HCT
11
35-37
17-21
1
33%
12
17
MAR
77
1
200 1
VT
4
26-35
16-23
0
0%
13
6
APR
77
12
3801
VT
8
27-36
19-22
4
33%
14
18
APR
77
14
3801
HCT
13
34-35
22-24
0
0%
15
25
MAY
77
11
3801
HCT
6
31-34
26-21
2
18%
16
25
MAY
77
22
3801
HCT
6
31-34
26-21
2
9%
17
25
MAY
77
58
3801
HCT
1-6
30-37
32-21
10
17%
18
16
JUL 77
11
3801
HCT
3
34-34
30-21
0
0%
19
16
AUG
77
10
2001
VT
3
30-34
32-22
0
0%
SQUID MAINTENANCE, GROWTH, AND BEHAVIOR
651
TABLE I (Continued)
LABORATORY MAINTENANCE
AND SURVIVAL
Main-
tenance
Tank
Salinity
Range
(ppt)
Temp.
Range
(°C)
No. of
Squids
Main-
tained
Size (mm ML) at
Death
Laboratory
Survival (days)
Sex
(x)
(Range)
(x)
(Range)
CT
34-37
24-30
3
M
182
(165-205)
17
(15-18)
4
F
92
(82-98)
16
(14-19)
CT
25-26
24-26
10
M
129
(105-153)
6
(2-11)
CT
32-33
23-27
9
F
98
(83-118)
9
(3-18)
CT
33-35
23-25
6
M
155
(140-175)
17
(5-28)
CT
28-30
23-25
6
F
102
(100-107)
19
(5-29)
CT
30-35
22-24
4
M
103
(71-138)
4
(2-6)
5
F
87
(74-102)
21
(15-29)
15
J
41
(31-50)
3
(2-6)
CT
30-32
22-24
4
J
35
(30-41)
3
(2-3)
CT
31-32
22-24
11
M
63
(52-74)
7
(3-17)
8
F
59
(52-67)
3
(3-6)
3
J
45
(39-50)
3
(2-3)
CT
32-33
20-22
3
J
41
—
16
(2-43)
CT
33-34
21-23
3
M
59
(55-63)
4
(3-6)
CT
32-33
20-22
4
J
42
(34-48)
3
(2-6)
CT
33-34
21-23
1
M
53
—
3
CT
32-33
20-22
19
J
30
(21-45)
3
(2-6)
CT
35-36
20-21
3
J
37
(32-43)
4
(3-6)
CT
35-37
19-21
1
M
85
—
84
—
1
J
43
—
22
—
CT
35-36
21-23
1
M
69
—
4
—
CT
35-35
23-23
2
M
95
—
3
(2-4)
CT
36-36
22-22
3
F
—
—
2
(2-3)
CT
35-36
22-25
7
M
226
(204-243)
33
(16-54)
RW
34-37
16-23
5
F
—
—
14
(3-21)
CT
34-37
21-22
2
M
140
—
16
(8-24)
5
F
79
(54-95)
54
(14-52)
2
J
43
—
4
(2-6)
CT
34-37
20-21
3
M
135
(105-160)
56
(55-57)
17
J
40
(38-44)
7
(2-16)
RW
31-37
20-21
23
M
118
(64-223)
17
(2-49)
21
F
76
(51-123)
18
(3-45)
2
J
48
(47-50)
25
(22-28)
CT
34-36
20-21
4
M
139
(110-164)
10
(4-16)
7
F
83
(65-101)
10
(5-16)
CT
34-35
21-22
10
J
19
(12-22)
3
(2-4)
652
R. T. HANLON ET AL.
TABLE I (Continued)
CAPTURE
SHIPBOARD TRANSPORT (TP), LABORATORY TRANSFER
(TF), AND 1-DAY ACCLIMATION (AC,)
Trans-
Obs.
No.
Date
No.
Squids
Collected
Transport
Container
port
Time
(h)
Salinity
Change
(ppt)
Temp.
Change
(°C)
No. Dead
in TP,
TF, AC,
Percent
Mor-
tality
20
16
AUG 77
4
2001
VT
3
30-34
32-22
0
0%
21
18
AUG 77
7
3801
HCT
3-7
28-32
29-21
4
57%
22
15
OCT 77
14
380 1
HCT
12-36
33-35
27-21
12
86%
23
30
OCT 77
13
380 1
HCT
7
36-34
26-21
5
38%
24
10
MAR 78
3
3801
HCT
4
35-38
14-20
0
0%
25
26
APR 78
15
5501
HCT
3
35-35
20-28
3
20%
26
1
MAY 78
1
5501
HCT
3
34-36
22-24
0
0%
27
16
MAY 78
17
5501
HCT
6
38-32
24-23
0
0%
28
5 JUN 78
550 1 HCT
15 34-36 27-22
100%
29 12 JUN 78 4
30 10 AUG 78 54
31 20 MAY 82 75
550 1 HCT
25-31 28-22
550 1 HCT 3-5 25-32 29-21
30-32 29-22
1020 1 RHT 6 — —
1
7
26
25%
13%
35%
32
33
8 JUL 82
1 1 AUG 82
37
27
1020 1 RHT
1020 1 RHT
6
6
6
1
16%
4%
2 = 700
x = 7 Max. A = Max. A =
+9, -8 +8, -11
Abbreviations: VT, vertical tank; CT, 2 m circular tank system; HCT, horizontal cylindrical tank;
RHT, rectangular horizontal tank; RW, raceway tank; J, juvenile; * artificial sea water.
In all cases it was imperative not to overload the transport tanks or transfer bags,
since this promoted wall contact, general excitement among the squids, and occa-
sionally cannibalism. Long transport times and hot summer temperatures also increased
mortality. Determination of the proper number of squids to be transported per tank
is a behavioral consideration, not a physiological one, because water quality is good
throughout the trip. The important considerations are the relative positioning of the
squids to one another (this depends on the sizes of the squids) and to the tank
SQUID MAINTENANCE, GROWTH, AND BEHAVIOR
TABLE I (Continued)
653
LABORATORY MAINTENANCE AND SURVIVAL
No. of
Size
(mm ML) at
Laboratory
Main-
Salinity
Temp.
Squids
X It
Death
Survival (days)
tenance
Range
Range
Main-
Tank
(ppt)
(°C)
tained
Sex
(x)
(Range)
(x)
(Range)
CT
34-35
20-21
2
M
106
(99-113)
3
2
F
91
(82-100)
3
—
RW
32-35
20-22
1
M
—
23
—
2
F
—
3
—
RW
36-39
18-22
1
M
70
2
1
F
67
—
2
—
CT
34-36
20-22
1
M
53
2
6
J
48
—
3
(2-7)
CT
34-38
20-24
3
F
64
(62-65)
20
(5-31)
CT
35-37
18-24
3
M
123
(113-133)
12
(7-19)
8
F
102
(83-110)
14
(6-20)
CT
35-36
24-24
1
M
145
—
7
—
CT
35-37
22-23
7
M
210
(155-252)
13
(8-22)
CT*
30-32
20-22
10
F
119
(110-133)
14
(6-27)
CT*
30-32
21-24
1
M
—
—
6
—
1
F
73
—
23
—
1
J
—
—
3
—
RW*
32-35
21-24
26
M
173
(112-232)
10
(2-23)
CT
35-36
21-23
19
F
109
(107-139)
13
(2-29)
CT
34-34
22-23
12
M
145
(88-180)
5
(3-18)
CT*
31-32
22-23
4
F
112
(92-131)
5
(4-6)
RW*
26
M
162
(115-200)
7
(2-16)
23
F
106
(86-124)
6
(2-12)
RW*
—
—
30
M+F+J
—
—
14
(3-36)
RW*
—
—
25
M+F+J
—
—
13
(3-30)
(25-39)
(16-30)
2 = 453
M
145
(52-252)
12
(2-84)
F
95
(51-139)
13
(2-52)
J
35
(12-50)
5
(2-43)
M+F+J
107
(12-252)
11
(2-84)
configuration (especially the size of the horizontal dimensions of the tank) because
the squid schools are generally dispersed horizontally, not vertically, in tanks. Based
upon our experience in observing squids during transport and analyzing the reasons
for mortality, we recommend ten full-sized Loligo spp. (150 to 250 mm ML) or 25
Lolliguncula brevis (40 to 80 mm ML) per 550 1 HCT tank. For the RHT tank we
recommend 15 full-sized Loligo spp. per level (30 total). These are conservative
estimates; under ideal conditions we have successfully transported greater numbers.
654
R. T. HANLON ET AL.
TABLE II
Loligo pealei: summary of capture, transport and transfer, and laboratory maintenance
CAPTURE
SHIPBOARD TRANSPORT (TP), LABORATORY TRANSFER
(TF), AND 1-DAY ACCLIMATION (AC,)
Obs.
No.
Date
No.
Squids
Collected
Transport
Container
Trans-
port
Time
(h)
Salinity
Change
(PPt)
Temp. No. Dead
Change in TP,
(°C) TF, AC,
Percent
Mor-
tality
1
20
FEE 77
7
CT
35-36
16-16
2
29%
2
18
AUG
77
1
380 1
HCT
7
28-32
29-21
0
0%
3
15
OCT
77
8
3801
HCT
24
35-35
27-21
2
25%
4
19
OCT
77
2
380 1
HCT
28
33-35
27-21
1
50%
5
30
OCT
77
1
380 1
HCT
7
36-34
26-21
0
0%
6
23
MAY
78
1
5501
HCT
10
34-36
25-22
0
0%
7
2
JUN
78
7
550 1
HCT
15
34-30
27-22
6
86%
8
23
JUN
78
10
550 1
HCT
10
34-32
27-22
7
70%
9
25
OCT
78
6
5501
HCT
10
35-30
26-21
4
67%
10
27
APR
79
15
5501
HCT
48
36-36
23-17
8
53%
11
10
JUN
80
22
550 1
HCT
10
33-32
26-22
—
—
12
23
JUL80
3
550 1
HCT
10
35-33
28-20
1
33%
13
13
AUG
80
4
550 1
HCT
10
36-36
28-22
—
—
14
1
MAR
82
2
1020 1
RHT
10
36-35
18-21
—
—
2 = 89
x == 15
Max.A =
+4, -5
Max. A =
+3, -8
Abbreviations: CT, 2 m circular tank system; HCT, horizontal cylindrical tank; RHT, rectangular
horizontal tank; RW, raceway tank; * artificial sea water.
Sea water systems and water quality
Both systems provided adequate filtration capability as well as space for squids.
As a rule of thumb, we determined that the 2 m circular tank system could maintain
the following numbers of adult squids in a healthy state for several weeks: ten to 15
Loligo spp. ( 1 50 to 250 mm ML) or 25 Lolliguncula brevis (40 to 80 mm ML).
Estimates for the 10,000 1 raceway were determined to be: 50 Loligo spp. or 100
Lolliguncula brevis.
The tank systems were usually kept at the same approximate temperature and
salinity as each species encountered in the wild at that month of the year, although
fluctuations occurred. The reported ranges that squids are found in the northern Gulf
of Mexico and were subjected to during our transport and maintenance work were:
Loligo plei 13 to 32°C and 25 to 39 ppt; Loligo pealei 13 to 30°C and 28 to 39 ppt;
Lolliguncula brevis 11 to 34°C and 18 to 39 ppt (Tables I, II, III; Rathjen et al,
1979; Hixon, 1980a; Hixon et al, 1980).
SQUID MAINTENANCE, GROWTH, AND BEHAVIOR
655
TABLE II (Continued)
LABORATORY MAINTENANCE AND SURVIVAL
No. of
Size
(mm ML) at
Laboratory
Main-
Salinity
Temp.
Squids
Death
Survival
(days)
tenance
Range
Range
Main-
Tank
(ppt)
(°C)
tained
Sex
(x)
(Range)
(x)
(Range)
CT
34-36
16-23
1
M
213
_
71
_
4
F
172
(154-200)
23
(2-67)
RW
32-35
21-22
1
F
167
—
21
—
RW
36-39
18-21
6
M
202
(158-285)
30
(15-41)
RW
36-39
18-21
1
M
183
—
5
—
CT
34-36
20-22
1
F
121
—
30
—
CT
32-37
21-23
1
M
109
—
53
—
CT*
31-32
21-22
1
F
174
—
25
—
CT*
31-32
21-22
2
M
152
(139-166)
3
(3-3)
1
F
137
—
2
—
RW*
30-34
15-22
4
F
163
—
36
(21-60)
CT
36-36
17-18
6
M+F
—
—
17
(3-35)
RW*
32-34
21-22
4
M
190
(140-153)
44
(25-54)
CT
33-36
20-22
2
M
—
—
27
(27-27)
CT
36-36
21-22
1
F
155
—
32
—
CT*
34-36
20-22
1
F
154
—
47
—
(30-39)
(15-23)
2 = 37
M
184
(109-285)
31
(3-71)
F
160
(121-200)
28
(2-71)
M+F
174
(109-285)
28
(2-71)
Water quality remained high except in rare cases when high densities of animals
(substantially greater than those mentioned above) were maintained for long periods
of time. The Hach field test kits were useful only for gross estimates of nitrogenous
buildup and for indicating increases, at which time detailed chemical tests were
performed. The pH of each separate system was different, but the mean value for all
experiments was 7.9, with a range of 7.7 to 8.5. Thirty-nine detailed water chemistry
tests were performed among five CT systems during 1 977 and 1978; these tests covered
six different maintenance trials and all three squid species. The mean recorded level
of total ammonia-nitrogen (NH4-N) from detailed chemical tests was .103 mg/1, with
a range of .020 to .161 mg/1 (n = 12). Mean level of total nitrite-nitrogen (NO2-N)
was .003 mg/1, with a range of .002 to .007 mg/1 (n =: 12). Mean level of total nitrate-
nitrogen (NO3-N) was 14.65 mg/1, with a range of 9.98 mg/1 to 20.73 mg/1 (n = 15).
In one separate observation, a male Loligo plei (124 mm ML) survived alone for 10
days in a 150 1 aquarium that had approximate levels (from Hach test kits) of .185
mg/1 nitrite-nitrogen and 32.50 mg/1 nitrate-nitrogen on Day 7. Even assuming that
656
R. T. HANLON ET AL.
TABLE III
Lolliguncula brevis: summary oj capture, transport and transfer, and laboratory maintenance
CAPTURE
SHIPBOARD TRANSPORT (TP), LABORATORY TRANSFER
(TF), AND 1-DAY ACCLIMATION (AC,)
Obs.
No.
Date
Trans-
No, port Salinity
Squids Transport Time Change
Collected Container (h) (ppt)
Temp. No. Dead
Change in TP,
(°C) TF, AC,
Percent
Mor-
tality
1
17 JUN 77
3801HCT 1 29-21
30-21
—
2
14 SEP 77
30 380 1 HCT 1 23-30
28-21 23
77%
3
30 SEP 77
32 380 1 HCT 1 27-32
27-26
28-21 5
28-21
16%
4
24 OCT 77
7 380 1 HCT 1 24-36
24-2 1 2
29%
5
1 DEC 77
20 380 1 HCT 1 24-36
17-21 7
35%
6
14 DEC 77
15 380 1 HCT 1 25-26
16-18 6
40%
7
26 JAN 78
9 380 1 HCT 8 34-35
13-18 1
11%
8
7 MAR 78
63 380 1 HCT 1 24-26
13-20 23
37%
9
20 JUN 78
•J 550 1 HCT 1 25-25
28-21 0
0%
10
29 JUN 78
38 200 1 VT 1 24-32
29-22 5
13%
11
7 JUL 78
200 1 VT 1 24-30
29-22
—
12
24 JUL 78
13 200 1 VT 1 26-30
28-21 1
8%
13
14 AUG 78
36 550 1 HCT 1 32-24
31-23
—
14
24 OCT 78
550 1 HCT 1 27-30
23-21
—
15
22 JAN 79
12 550 1 HCT 1 30-32
11-22 4
33%
16
5 JUL 79
42 550 1 HCT 1 18-18
29-21 12
29%
17
27 FEE 80
550 1 HCT 1 33-27
19-20
—
18
24 OCT 80
— — — —
— —
—
v ~_ 4">5 x = 1 max. A =
+ 12, -8
max. A =
+ 11, -9
Abbreviations: VT, vertical tank; CT, 2 m circular tank system; HCT, horizontal cylindrical tank;
RW, raceway tank; J, juvenile; * artificial sea water.
SQUID MAINTENANCE, GROWTH, AND BEHAVIOR
657
TABLE III (Continued)
LABORATORY MAINTENANCE AND SURVIVAL
Main-
Salinity
Temp.
No. of
Squids
Size (mm ML) at
Death
Laboratory
Survival (days)
tenance
Range
Range
Main-
Tank
(ppt)
(°C)
tained
Sex
(x)
(Range)
(x)
(Range)
CT
22-23
21-22
8
M
56
(51-62)
13
(2-58)
23
F
62
(44-85)
10
(2-59)
10
J
34
(27-40)
4
(2-6)
CT
23-25
20-22
3
M
57
(41-67)
45
(8-64)
CT
29-33
21-22
2
F
82
(79-85)
40
(25-54)
2
J
35
—
5
—
RW
34-39
18-22
15
M
51
(45-57)
23
(2-68)
CT*
26-36
20-23
11
F
73
(63-80)
33
(12-55)
CT
34-36
20-21
1
J
34
—
39
—
CT
35-38
20-23
2
M
55
(53-56)
21
(5-37)
3
F
70
(67-72)
38
(33-44)
CT
34-36
15-21
7
M+F+J
—
—
29
(3-106)
CT*
24-27
15-24
7
M
57
(53-60)
76
(33-123)
CT
30-36
15-24
2
F
65
(59-72)
64
(58-70)
CT*
28-34
18-19
3
M
56
—
22
(15-33)
5
F
83
—
24
(16-35)
CT
25-26
19-24
40
M+F+J
—
—
20
(2-49)
CT
22-24
21-22
3
M
59
(58-60)
46
(40-49)
4
F
71
(62-76)
38
(33-40)
CT*
32-34
21-22
5
M
43
(41-50)
6
(4-11)
7
F
49
(42-69)
5
(3-12)
21
J
36
(29-40)
3
(2-5)
CT*
28-30
20-21
18
M+F+J
—
—
4
(2-8)
CT*
30-30
21-23
12
M+F+J
—
—
8
(2-12)
CT
22-24
22-24
9
M
55
(50-59)
6
(2-12)
8
F
71
(52-99)
7
(2-17)
CT
30-32
20-21
28
M+F+J
—
—
13
(3-29)
CT
32-34
20-22
8
M+F+J
—
—
20
(3-27)
CT
18-18
20-21
30
M+F+J
—
—
12
(2-17)
CT
26-27
19-22
3
M
59
(56-60)
72
(67-74)
3
F
68
(66-69)
67
(54-73)
CT
34-36
16-25
6
M
52
(48-56)
68
(51-115)
2
F
63
(60-66)
120
(115-125)
2
J
35
(32-38)
41
(28-53)
(18-39)
(15-25)
2 = 313
M
54
(41-67)
32
(2-123)
F
67
(42-99)
25
(2-125)
J
36
(27-40)
6
(2-53)
M+F+J
55
(27-99)
19
(2-125)
658 R. T. HANLON ET AL.
there is a large source of error in the Hach test, these levels indicate that squids can
tolerate concentrations at least somewhat higher than those recommended for marine
animals by Spotte (1979a, b): ammonia 0.1 mg/1 NH4-N, nitrite 0.1 mg/1 NO2-N,
and nitrate 20.0 mg/1 NO3-N.
Circular tank systems used without algal tanks since 1978 have not shown sub-
stantially increased levels of inorganic nitrogen, nor has our recent raceway tank,
which does not have a biodisc but depends mostly on the bacterial population in the
filter bed of the adjoining CT system for biological filtration. The biodisc filter in our
early raceway system (Fig. 3) increased the capacity for biological filtration. However,
a drawback of the biodisc was the lack of control over the types of organisms that
grew on it, some of them undesirable in a closed system. In our new raceway without
the biodisc, it is likely that this extra filtration capacity is not needed with our currently
used animal loads. Certainly our attention to cleanliness contributes to this result,
since food remains are carefully removed daily and nearly all maintenance procedures
recommended by Spotte ( 1979a, b) are followed. Slight shifts in pH and corresponding
increases in nitrogen levels are dealt with quickly, usually by replacing a small per-
centage of the water volume with fresh, clean sea water.
Some other problems are noteworthy. In uncovered tanks in bright illumination,
various algae and other unknown organisms grew on the tank walls and raceway
bottoms (note the black growth on the raceway bottom in Fig. 3). Growth of these
types of organisms is uncontrollable and some forms can be deleterious (e.g., some
blue-green algae). Bacterial buildup, especially of potentially toxic Vibrio spp., can
also occur on these substrates as well as on biodisc filters. Therefore, we occasionally
clean the bottoms or keep tops on the tanks to reduce illumination.
No conspicuous differences were noted between the performance or longevity of
natural sea water and artificial sea water. Some CT systems have been in continuous
use for as long as two years with no major alterations, aside from periodic ten percent
water changes, occasional addition of trace metals, and occasional gentle stirring of
the filter bed to siphon off excess detritus buildup that can clog the filter bed and
reduce denitrification by bacteria.
General aspects of behavior
Healthy, calm squids of these three species do not bang haphazardly into aquarium
walls. Squid vision is keen and they can quickly and deftly maneuver without hitting
walls or other objects. Their behavior changes, however, if: (1) they are placed in
small tanks, (2) they have incurred significant skin or fin damage, (3) they are engaged
in intense intraspecific aggression, or (4) they are not fed. It is important to recognize
normal versus altered behavior because it is possible to preclude or reduce circumstances
that promote altered behavior, which leads to decreased survival in captivity.
Loliginid squids are social, schooling, inquisitive creatures that actively react to
everything in their environment. Nearly all aspects of squid behavior are mediated
through expression of the chromatophore system, as well as particular postures and
movements; collectively these are referred to as body patterns (Hanlon, 1982).
Loligo plei (Fig. 4) has the widest range of body patterns and the most complex
behavior. To date, 16 chromatic and six postural components of body patterning
have been described and associated with specific behavior (Hanlon, 1982, and in
prep.). Males grow larger than females, they are far more aggressive, and they possess
seven male-only chromatic components that are used in an intraspecific aggressive
context and are inextricably connected with courting and mating behavior (Hanlon,
1981, 1982). Males establish and maintain a rank order based upon size and ag-
SQUID MAINTENANCE, GROWTH, AND BEHAVIOR
659
' "*
y"*" .
, -' *
FIGURE 4. Loligo plei. A. Five females (74 to 102 mm ML) schooling loosely during Observation 4
(see Table I). The Clear pattern indicates calmness. Note the well-developed ovaries (white arrow) and
nidamental glands (black arrow) characteristic of very mature females. In color, the red accessory nidamental
glands are also visible. B. Three males (165 to 205 mm ML) and four females (82 to 98 mm ML) from
Observation 1, schooling tightly in a CT system. The All Dark pattern indicates that the squids are alarmed.
C. Intraspecific aggression and mate pairing during Observation 1. The large dominant male (center, 205
mm ML) is performing a "lateral display" towards the male on its right (bottom, 175 mm ML) in order
to keep his female mate (93 mm ML in Ring pattern) segregated from the school. Mating and egg laying
occurred the same day. D. A small live fish is seized by the extended tentacles of a male squid, 174 mm
ML. Note the buckling of the tentacles (see Kier, 1982). E. A female (1 10 mm ML) from Observation 27
eating a small fish. Note that the fish is held vertically and that the viscera are being eaten first. The digestive
gland is swollen and reddish (white arrow) and the stomach is approximately '/? full (black arrow). F. A
male (left, 1 13 mm ML) and female ( 1 10 mm ML) from Observation 25 in a tug-of-war over a fairly large
fish. G. Cannibalism. Six males (approx. 220 mm ML) jointly eating another male that had been moribund
prior to cannibalization.
660
R. T. HANLON ET AL.
gressiveness. They accomplish this mainly through visual signalling, in particular a
"lateral display" in which the males position themselves in parallel and then unilaterally
flash flame-like streaks of chromatophores on the lateral mantle towards one another
(Fig. 4C). Up to five additional chromatic components may be expressed in this
display, depending upon its intensity. In some cases the squids may also engage in
"fin beating" while parallel to one another, and in extremely rare cases the dominant
(and usually larger) squid may execute a forward attack and grasp or bite the other
squid. In contrast, females are generally passive and docile in the laboratory (Fig.
4A, B, C) and they seldom engage in aggressive behavior except occasionally during
the pursuit of prey (Fig. 4F). Mating and egg laying are common in captivity and
can be artificially stimulated by placing egg strands or a facsimile in the tank in the
manner described for Loligopealei by Arnold (1962). Feeding and growth in captivity
are good, with cannibalism (Fig. 4G) occurring rarely. L. plei has delicate skin and
is more vulnerable to skin abrasion than the other two species (Fig. 7).
Loligopealei (Fig. 5) has the second widest range of body patterns and its behavior
I
i f
FIGURE 5. Loligo pealei. A. Intraspecific compatibility is obvious in this school of three Loligo pealei
(two males, one female at far left) and two Loligo plei males (arrows). All squids are approximately 220
mm ML. Note the wall pattern and how squids stay near the middle of the tank. B. Female (174 mm ML)
in a Ring pattern while bottom sitting. This is a normal posture for this species. C. Female (180 mm ML)
actively securing an egg strand into the substrate. Note the egg strand on the left; also the bold stripes on
the wall.
SQUID MAINTENANCE, GROWTH, AND BEHAVIOR 661
is similar in complexity to L. plei. Approximately 12 chromatic and four postural
components and their associated behavior are recognizable in this species. Males and
females are similar in size (Hixon et al, 1981) and grow larger than L. plei. Males
and females both display intraspecific aggression. Males are slightly more aggressive
and they also establish a rank order based upon size and aggressiveness (similar to
that reported by Arnold, 1962), but they do not show any obvious male-only or
female-only chromatic components. Mating and egg laying are common in captivity
(Fig. 5C) and can be easily stimulated (Arnold, 1962). Feeding and growth are good,
as in L. plei, but cannibalism by large males is more common. The skin is nearly as
subject to injury as in L. plei (Fig. 7).
Loligo pealei commonly sits on the bottom (Fig. 5B). This is a normal posture,
exclusive to this species, that is assumed for long periods of time on sand or gravel
substrates. Bottom sitting is conducive to laboratory survival because it conserves
energy (compared to constant swimming), it maintains calmness among the tank
animals, and it minimizes contact with the tank walls.
Loligo plei and Loligo pealei are very similar morphometrically and they are
difficult to distinguish visually, especially when they are smaller than 100 mm ML.
They can, however, be distinguished by their specific chromatic components (par-
ticularly L. plei males) and their behavior, and this is useful for the identification
and segregation of animals in the laboratory.
Lolliguncula brevis (Fig. 6) is distinctly different from the two species of Loligo.
Aside from being smaller, its behavior is less complex and only seven chromatic and
four postural components of body patterning have been noted thus far. A common
threat posture is illustrated in Figure 6D. Females grow markedly larger than males
(Dragovich and Kelly, 1962; Hixon, 1980a). Little intraspecific aggression has been
observed and there has been no evidence of rank ordering among males. Mating has
been seen fairly often, and large females are often seen with conspicuous white patches
of spermatophores attached to a pad on the inside of the mantle on the left side (Fig.
6A); however, egg laying in captivity is rare. Efforts to stimulate egg laying with egg
strands were negative, but occasionally a temperature increase resulted in egg laying.
Feeding and growth in captivity are very good (Fig. 6B, C). This species is less vulnerable
to fin and skin damage than Loligo spp. For these reasons, males and females may
be kept in the same tank at higher densities (Fig. 6A) than Loligo spp. and for longer
periods of time.
There is some interspecific compatibility among the three species. On several
occasions mid- to large-sized Loligo pealei and Loligo plei have been kept in CT
systems for up to 15 days with no noticeable problems. It was important that these
animals were all of a similar size (about 200 mm ML) and were put into this tank
at the same time. They schooled together in a seemingly random arrangement, i.e.,
individuals were found in different parts of the school at different times (Fig. 5A).
During another observation, nine Loligo plei were put into a tank that held a mating
pair and eggs of Loligo pealei. The male Loligo pealei continuously displayed towards
and attacked the L. plei, which had to be removed within one day. It was difficult
to tell whether this was territorial defense of the eggs, the female, or the tank, or
simply the usual aggression shown by large males to define the rank order. Lolliguncula
brevis is compatible with both species of Loligo if all animals are of the same size;
they even school together with little interaction. But if the Loligo are larger they will
display towards the Lolliguncula brevis, which in turn will often display and attack
as well. It is characteristic of Loligo plei and Loligo pealei to cannibalize smaller
squids or weakened squids such as those with impaired swimming due to skin damage
or the effects of anaesthetic agents. Cannibalism by Lolliguncula brevis has been
662
R. T. HANLON ET AL.
.
*h^ * • * .
Mil*
ft ~jm
f
* -v *•",-
v
FIGURE 6. Lolliguncula brevis. A. Twenty-three squids in a CT system. Note the white patch of
spermatophores inside the mantle on the female in the left foreground. B. Small squid feeding on a penaeid
SQUID MAINTENANCE, GROWTH, AND BEHAVIOR
663
'
T»V •'.
**T* *
^%> '
**£*+- \*
. ,-m&. 4 ~* %
FIGURE 7. Fin damage. A. Loligo pealei with fairly severe fin damage incurred initially from transport.
B. The same squid as in (A) showing the amount of damage on the posterior fin and the ventral mantle
that resulted from hitting the transport tank walls. C. Loligo plei female (82 mm ML) from Observation
1 with moderate fin damage. Note the expanded chromatophores that are usually present around the
periphery of damage. D. Loligo plei female (87 mm ML) with several round patches of damage; these are
less lethal than damage to the periphery of the fin.
observed only on one rare occasion. In summary, under ideal circumstances there is
interspecific compatibility among the three species, but when a size difference exists
the larger individual usually dominates.
As the three species were exposed to salinity shock when first brought to the
laboratory, their reactions were immediate. The first manifestation of stress was the
curling of the extreme distal portions of the eight arms. The squids also showed
sluggish hovering and swimming movements and usually did not school or feed
immediately. In extreme cases, some squids would go to the bottom and sit, a posture
very uncommon to Lolliguncula brevis and Loligo plei, but not unusual for Loligo
pealei. Normal behavior usually resumed within one hour or less.
Field behavior was also species specific. Loligo plei was nearly always found in
moderate- to large-sized schools both during the day and night. Around night light
stations, the schools usually stayed deep and would characteristically rise en masse
to the surface under the light, then quickly dive. They preferred the periphery of
shrimp nearly as long as the squids' mantle length. C. Small squid from Observation 16 (Table III) eating
a very large silverside, Menidia beryllina. D. A female swimming in a typical threat posture three days
after brain surgery, in which the vertical lobe was cut. E. Narcotized squid with the characteristic chro-
matophore pattern that is usually produced while the squid is under anaesthesia.
664 R. T. HANLON ET AL.
light. On occasion, one or several squids would leave the school to feed. Only rarely
would the schools swarm for any period of time right at the surface under the lights
in the manner described for the California market squid Loligo opalescens (Kato and
Hardwick, 1975). Individual schools would seldom stay around the night light station
very long, but other schools would appear later, indicating that squids were moving
and actively foraging. Schools became closer knit and usually left the night light
stations upon the appearance of schools of scad or jacks (Family Caranjidae), mackeral,
or sharks. Schools always were comprised of squids of similar size, an observation
corroborated by laboratory results that showed squid schools being incompatible when
size disparities were present. In some cases it appeared that schools were not only
size specific but sex specific as well [the white testis of mature males is highly con-
spicuous in live squids and is even used in signalling (Hanlon, 1982)]. At Grand
Cayman Island, B.W.I., large schools could be attracted to night lights set on sandy
patches between coral reefs at 10m. These schools were never seen near reefs during
the day, at which time they presumably moved to deeper water. Off Eleuthera Island,
Bahamas, we have observed one small Loligo (probably plei) swimming on three
consecutive days with a school of 1 2 Sepioteuthis sepioidea. Moynihan and Rodaniche
(1982) observed this association frequently in Panama.
Loligo pealei behaved quite differently from Loligo plei in the field. Our only
observations were at deep-water night light stations, and in most cases large adults
were seen singly or in pairs. No tight, well-formed adult schools were ever observed
at night light stations, although on some nights enough individuals would arrive at
the lights over a period of time to form a loose aggregate of squids. In a few cases
we could identify the pairs as male-female, presumably a mating pair. In contrast to
adults, young juveniles were often seen in large schools under the lights, indicating
that L. pealei becomes more solitary at night as it becomes larger. Lolliguncula brevis
observations are scarce due to the turbid waters in which it lives.
Feeding
All three squid species fed readily on a variety of live fishes and shrimps (Figs.
4, 5, 6). The feeding response of adult squids in good physical condition has been
excellent, with detection, pursuit, and capture of prey usually taking five seconds or
less. Some squids fed within ten minutes of their release into the tank and nearly all
fed within the first day of confinement. The feeding behavior of healthy squids is
predictable and provides an indication of the animals' overall condition. Damage to
the delicate fins seriously impaired the squids' ability to deftly maneuver for prey
capture. Squids actively participating in intraspecific aggressive behavior often did
not feed well; conversely, feeding occasionally led to aggressive behavior when two
or three squids would capture the same prey organism and a vigorous tug-of-war
would result (Fig. 4F).
The response of juvenile squids to the presence of food was usually slower, with
feeding sometimes not beginning for a day or two and remaining sporadic thereafter.
One probable cause for this was their greater susceptibility to fin damage during
capture and transport to the laboratory. Another cause, in some cases, was the presence
of large conspecifics that were aggressive.
Movement on the part of the prey provided an essential visual stimulus to the
squids. Fishes or shrimps that made it to the bottom of the tank without detection
would go uneaten for hours if they remained motionless. In the CT systems, the
oyster shell substrate and the painted walls provided partial concealment because of
the similarity in coloration between the substrate and prey organisms. Palaemonid
SQUID MAINTENANCE, GROWTH, AND BEHAVIOR 665
shrimps blended in especially well, and even when there were hundreds of them in
the tank, only those that moved quickly or swam into the water column were attacked
and eaten. In the raceway system, squids had little difficulty in sighting and capturing
prey against the pale interior of new or cleaned systems, but on algae-covered bottoms
(e.g., Fig. 3) small shrimps were difficult to detect. Normally, all squids ate daily at
each of the two or three times food was dropped into the tanks, even though food
was usually in the tanks at all times. During growth experiments, Loligo plei consumed
10 to 18 percent of its body weight in food per day (see Growth section).
Small fishes (less than about 25 mm long) were usually eaten entirely. Larger
fishes were captured with the two long tentacles (Fig. 4D) and were bitten several
times through the vertebrae just behind the head, after which the viscera were eaten
(Fig. 4E) and all the meat on either side of the skeleton was cleanly stripped away.
Shrimps were eaten completely except for some of the head and the thin exoskeleton.
Hungry squids sometimes took prey nearly as long as their own mantle length (Fig.
6B, C).
It was not possible to detect any clear-cut diet preferences for different species or
different growth stages. The younger stages of all three species seemed to prefer
crustaceans, and the larger animals generally preferred fishes, but many individual
and collective exceptions to this statement occurred. Cannibalism occurred rarely
(see Behavior section). Growth rates were equally high on shrimp-only, fish-only, and
mixed diets, and our conclusion is that estuarine food organisms are suitable for
maintaining and growing loliginid squids.
Field and laboratory observations both confirmed that squids of all sizes eat prey
organisms of a wide size range. Underwater observations during night lighting stations
off Texas and Grand Cayman verified that adult Loligo plei commonly fed on very
small plankton in the vicinity of the night light. The squids always seemed to be very
selective about these planktonic organisms, for they would carefully orient towards,
follow, and seize specific organisms even when great masses of plankton were present.
Conversely, squids at the same station would inspect and sometimes attack squid jigs
up to 70 mm long, objects that were many orders of magnitude larger than the
planktonic organisms they had seized minutes before. Laboratory observations cor-
roborated this behavior. We commonly saw adult squids follow and inspect small
bubbles that were only several millimeters in diameter.
Survival
Loligo plei ranging in size from 12 to 252 mm ML (mean ML 107, Sx == 3.0)
were maintained in 33 laboratory observations (Table I). The mean survival time for
455 squids was 1 1 days (Sx == 0.5, median =: 7 days, Fig. 8). The longest-lived male
(85 mm ML) survived 84 days, and two females (89 and 95 mm ML) survived for
a maximum of 52 days. There were no significant differences in survival time between
males (n = 149, median survival time 7 days) and females (n = 132, median survival
time 10 days). In contrast, the survival times of 81 juveniles (less than 50 mm ML)
were low. Median survival was three days, which was significantly (P < .001) lower
than both males and females.
Loligo pealei showed best overall survival in our tank systems (Table II). Squids
(n = 37) ranging in size from 109 to 285 mm ML (mean 173 mm ML, Sx = 7.7)
had a mean survival time of 28 days (Sx :: 3.1, median = 27 days, Fig. 8). The
maximum survival time was 7 1 days for a male measuring 2 1 3 mm ML. The longest-
lived female survived 67 days and measured 200 mm ML. There were no statistically
significant differences in survival by sex; 1 7 males had a median survival time of 28
666
R. T. HANLON ET AL.
50 -
40-
30-
20 -
10 -
>•
Ih..
Loligo plei
= 1 1 days
: 7 days
mean survival (x)
(Sx = 0.5)
median survival =
n = 453
— \J i i i i i i i i i i i
(0
S 20-,
i i i i i i i i
Loligo
i i i i
pealei
10-
o u
o3 50 -i
0.
40 -
30 -
20-
10-
lliilllii.l. ..
x = 28(Sx = 3.
median = 27
n = 37
i i i i i
I I •, • •-_••
Lolliguncula brevis
x = 1 9 (S x = 1~2F
median = 9
i i r i i i r i i i i i i i i i i i i " i i i i
10 20 30 40 50 60 70 80 90 100 120
Survival (days)
FIGURE 8. Survival summary of all squids in recirculating circular tanks and raceways.
days, and 14 females had a median survival time of 27.5 days. No small-sized juveniles
of this species were maintained during the 14 laboratory observations.
A total of 3 1 3 Lolliguncula brevis ranging in size from 27 to 99 mm ML (mean
ML 55 mm, Sx ~ 1 .2) were maintained during the course of 1 8 laboratory observations
(Table III). The mean survival time for all squids was 19 days (Sx = 1.3, median = 9
days, Fig. 8). Some squids commonly survived much longer; the longest-lived male
(60 mm ML) survived 123 days, and the longest-lived female (also 60 mm ML)
survived for 125 days. There were no statistically significant differences in survival
times between 63 males (median survival 19 days) and 74 females (median survival
14 days). The median survival time of 36 juveniles (less than 40 mm ML) was only
3.5 days; this was significantly lower (P < .001) than that of males and females.
There were statistically significant differences in laboratory survival among the
three species. Loligo pealei survived in the laboratory tanks significantly longer (P
< .001) than both Loligo plei or Lolliguncula brevis, and Lolliguncula brevis survived
significantly longer (.001 < P < .01) than L. plei.
It should be pointed out that the mean survival times for all three species shown
above are conservative figures. In these estimates all squids alive in the laboratory
after one day of acclimation were included, regardless of their size or physical condition
after capture and transport. The inclusion in the calculations of the short-lived juveniles
and sexually-mature adults near the end of their life cycle also reduced the overall
mean and median values. If the effects of these factors are reduced by computing
laboratory survival using only squids that lived beyond five days, instead of one, then
the mean survival time for each species increases substantially. The mean survival
SQUID MAINTENANCE, GROWTH, AND BEHAVIOR
667
of Loligo plei becomes 15 days (Sx := 0.7) compared to 11 days. Likewise, mean
survival for Loligo pealei increases to 3 1 days (Sx = 3. 1 ) from 28 days, and Lolliguncula
brevis increases to 29 days (Sx = 1.7) from only 19 days.
These higher figures probably represent a more realistic approximation of how
long squids survive in captivity, because they do not include squids that incurred
extensive skin damage during capture, transport, and transfer. In effect, one can then
begin to define the limitations to survival among these squids in terms of aspects of
behavior and maturation that take place in the laboratory tank system. These are
explained below (Principal causes of mortality).
Growth
Laboratory observations on growth were obtained from three male and three
female Loligo plei and from seven male and one female Loligo pealei; temperatures
ranged from 18 to 23°C (Table IV). The results suggest that adult males of both
species of Loligo are capable of growing at high rates in both length and weight in
the laboratory. Males of L. plei grew at a mean rate of 47 mm/mo (Sx = 7.6) and
13.3 g/mo (Sx = 5.30), while L. pealei males grew at a mean rate of 44 mm/mo (Sx
= 10.7) and 37.3 g/mo (Sx = 10.94). In contrast, adult females grew little or not at
TABLE IV
Laboratory growth of male and female Loligo plei (top) and Loligo pealei (bottom)
GROWTH RATE
Obs.
No.
Temp.
(°C)
Dura- ML
tion Changes
Sex (days) (mm)
WW
Changes ML % Length
(g) (mm/mo) gain/day
WW % Weight
(g/mo) gain/day
16
20-21
M
33
68-104
10.0-17.6
33
1.
3
6.
9
1.7
16
20-21
M
33
75-140
13.0-39.2
59
1
9
23,
8
3.3
16
20-21
M
22
124-160
37.6-44.3
49
1
2
9.
1
0.7
x:
29
47
1
,5
13
3
1.9
(Si):
(3.7)
(7.6)
(0
.22)
(5.
30)
(0.76)
15
21-22
F
10
56-54
8.0-7.5
-6
-0
,4
-1.
,5
-0.6
15
21-22
F
33
88-93
20.5-28.2
5
0
2
7.0
1.0
15
21-22
F
12
93-87
23.4-26.0
-15
-0
.6
6
.5
0.9
x:
18
-5
-0
,3
4
.0
0.4
(Sx):
(7.4)
(5.8)
(0
.24)
(2
.75)
(0.52)
6
11
11
11
11
3
21-23
21-22
21-22
21-22
21-22
18-21
M
M
M
M
M
M
x:
(Sx):
30
14
43
44
28
7
28
(6.1)
LOLIGO PEALEI
86-112 27.6-49.1
130-140 54.2-60.5
132-153 70.4-78.9
135-248 55.5-133.5
149-209 72.2-130.5
262-276 216.3-232.0
26
21
15
77
64
60
44
(10.7)
0.9
0.5
0.3
1.4
1.2
0.7
0.8
(0.17)
21.5
13.5
5.9
53.2
62.5
67.3
37.3
(10.94)
1.9
0.8
0.3
2.0
2.1
1.0
1.4
(0.31)
20-22
124-127 64.6-65.0
13
0.3
1.7
0.1
668
R. T. HANLON ET AL.
UJ
H
X
H
o
o
LU
_J
oa
o
c
00
_^
~5
00
(SO
I
i
c
o -^
2 w
Q ^
x
o
C/3
a —
E^
O
- o
eg
oo
so
O
SO
2 ^
c oo
* o.-§
o
E
o. S
C 3 ~
•- O K
OOJ2 rM — iy~i l — ^r -rr ^o r — • Os ro • — i -^r — ^ri/~>sOOsOv~»sOOsOsOfN — « fN
....„.,.,., -. ,,-,-, -.O — fNroOO — ' O O O O O
f
O i/"i C> ro *O ^t OO fN ^^ ^" ON ~~ 'O so t — ^J° <O ^T fN ^^ OO r— OO ro sD ro
' — — fN — fNOOO
00^ o^f^J^TC^OO— sOOOOOOOOt — ro ro ^ >n r— — ' fN fN ro fN • — ' ro ^- —
> do— 'od — d — o'o'ood — doo — o'ddoo'ddo
j .E I
a
"o
^^ ^O oo c*~\ oo f*~) ^"^4 */"~i r^i CNJ i/^ r*^ "^
g fN I
E
ro ro O O^ O SO fNfNO'sf'^fN — ' Tj-roroOvr—
<L> P ^. fN ^. Tj- — • d °°. fN ^ fN u~> r-i ro d — od O ^t ro d fN r- — O O
op^, Tj-rvj — oo — — — oo — r ^ — ' — — rN oo ' ' — —
Cnf! IIIIIIIIIIIIIIIIIIIJ^IIIIII
U— — ^ ^— fxi ro r*^ ro ^ ^ ^~ ^~ ^ *D r~~~ i/~i */^ \O ^O OO I — OO GO — • ^" ^^
C/5
I I t I I I I I I I I I I I I I I I I I I I I I I I
oK •— — ^ ^ */*i \o r** o^ ^^ ^™ ^^ fN fNi fM ^~ ^~ "/^ ^- oo ^^ ^o r^J ^J f"^ *^
-•^ r^rornfnf^fOrnrorn^^TjTtT^Tt^^^-^-Tj-iy^*r>»/^i/^t/^»o
or-—
*O O^ I — *"O 1O ^v ro
OO CJ ^O <N ro »^> m
I t I I I I I I 1 I I I I I t I I
r^l ro ^J1 CO ^f ^^ ro ro ^f
rjr^^r^fNrNir^rNirNirs
I I t I I I I I I
OOO*/^OO"/^*/~>OOOO1/^
o
\O ^O co ro "O
SQUID MAINTENANCE, GROWTH, AND BEHAVIOR
669
r7
O
^
ON*
oo
ci2
ON^
ON
c!^
ON "^
ON
ON
fN
•HHq
§3
§3
cU
in r- fN oo oo so
— oo r*~i — so O^
^J" r^i O^ ^^ ^O ^^
OOfNOO — fNW-^SO ON r^i
— — O fN ro fN
fN fN r*1"! -*vj" TJ- c^j
fN f> fN — O fN
— — d d d d d — d
E
E
ON
I I
E
u
I
4J
U
c
£>
o
sO so '^ NO ^^
— W m d <N
o «/i
o d
'T v~>
_ ; 06
r^» f^i OO OO '^C1 ^^ fN ^^ ^^ \^^ i/^ O^ s^^
d d wi ^t fN fN r-' fN wS fN d fl 06
PH fN SO
d — •*'
10 O
d d
OO T}-
d —
oo r-~
d d
oo o r-_ oq — o ON_ —
d— '• — '• — ;d— ;d —
II
<^>-~ w~)sO<^im — — fN fN
dddddddddd
I I I
O^ fN C
r~ — OQ
fN
xq —
d d
o
«
<u
.E E
so
J=
— ON
I
— O
— fNv>O — fNOr--— • r~
fN O fN
r~ so fN — so
I I I
2
<u *
— * ^^
C '^~ '
1 E
o S
s -
— oo
I I
>n o
fi fN
^^ r*^ i/^ ^ ^^ Vi "^ OO ^O 1O V~> 0s */~i fN ^" f^4 OO *O r*"i \O */~* ™"
^O f^J r^ ^^ OO OO f^i OO "^ SiO «^D N^^ i/*i ^^ **O f^"l ^^ ^^ "^ fl '^ V^
"~* C^4 f^-J ^^ fN ^^ fN "^ "^ C^J r^i oj fN ^^ ^1 ^^i ^^ ""^ f**l ^J f*^ r*"l
i i i i i i i i i i i i i i i i i i i i i i
fNi^iONsOfN — O'^O^l;r--'^)fNir)OOOOOr*^sDfv)O'^>
fNf^<^ir^Tj-»y~isDmr — ONOOONOON^-ON( — OO — < — • OO OO
_ ___,_._,fN|fNfNfN
••52
O
S s
flj C
SON
— <^1
— so
SO «/-)
O ro
SO so
— O^ \O *O OO fN *O ^" rN O^ <"*"! ^" */^ <^
*O ^O **O ^O ^O ^"0 {*"*• ^~~~ *^~* f"^ t^ t*^ t — - */^
I I I I I I I I I I I I I I
*3" *S~) I — <^ ^O f^J *^ SO ON — -* fN ^ */"} *^)
•^1-OOt^fNfNfN —
t^-ooooso(~-r^-oooo
l I I 1 I l l I
&\ oo O^ ^^ ^f ^f W) ^O
i/^ \^^ \^ f~-~. ^* f^. [^«. (•••».
o £
ob ^
^ C^
E §
ON r«1
>^> fN
' — - O ON — ON ON
r — V~i SO I — • SO <"*">
r-
fN
OO OO fN O
f*^ fN ON 1^-
ON
fN
ON — OO SO
"" f^ *n f*^
ob <«
u 'c
'ob oo
U.U.U.U_U.U.U.U-U-UHU-U.U.U_U-U_U.U.lJ_lJ-U.U.
I I
wi O
— fN
so
^fNfN'VifNfNfNfN— 'fNfN r*^ — ^ T}- r*^ f*^ rn m m Tf rn r^i
fNfNfNfNfNfNfNfNfNfNfNr--lfNfNfNfNfNfslfNfNfNfN
I I I I I I I I I I I I I I I I I I I I I I
sOONONsOON — — — V>O— - OO'n'/^ooooooOOv-ioooo
_ _ — _- ,»- .fNfNfN— -fNfN — — — — — — 'fNfN — — ' — •
oo r- r- oo
ON ON ON
fN ON
so ro
sO
il* s:
cw .2
^*^ "s3
E
E
670 R. T. HANLON ET AL.
all in these observations; negative mantle length values resulted from damage to the
posterior mantle during confinement. All females were mature when captured and
their oviducts were full of eggs when death occurred. Unfortunately, no growth ob-
servations on juvenile Loligo were made.
Growth observations were obtained on 28 male and 22 female Lolligimcula brevis;
temperatures ranged from 15 to 25 °C (Table V). Males and females survived equally
well during these observations; overall mean survival was 50 days (Sx = 4.4) for males
and 48 days (Sx = 4.1) for females. Males grew in length at an overall mean rate of
8 mm/mo (Sx = 1.2) and 3.5 g/mo (Sx = 0.62), while equivalent rates for females
were 1 1 mm/mo (Sx = 2.1) and 7.9 g/mo (Sx = 1.20). There were no statistically
significant differences in median monthly growth rates in length between the sexes
(males 8.5 mm/mo, females 1 1 mm/mo), but males (2.95 g/mo) differed significantly
from females (8.9 g/mo) in median monthly growth in weight (.01 > P > .001).
These weight differences reflect the maturation of reproductive organs and the pro-
duction of eggs in adult females.
In both sexes of Lolligimcula brevis there were size-dependent differences in growth
rate. Small young squids grew faster than larger (and presumably older) adults. Males
were divided into three categories (<39 mm ML, 40 to 49 mm ML, >50 mm ML)
based on their mantle length at the beginning of the growth observation (Table V).
Males >50 mm ML grew in length at a mean rate of only 0.9 mm/mo (Sx = 1.6)
compared to 10 mm/mo (Sx = 1.7) for the 40 to 49 mm ML group and 1 1 mm/
mo (Sx = 1.5) for the <39 mm ML group. Similar differences in monthly growth
rates in wet weight were measured (Table V). The median monthly growth rate in
length of the >50 mm ML group (2 mm/mo) differed significantly (P < .05) from
the median growth rates of the other two groups (both 1 1 mm/mo). The same statistical
results among the three groups were obtained using the monthly growth in weight
measurements. The reason for this reduced growth rate is that males >50 mm ML
are nearing maximal size and the end of their life cycle.
Female Lolligimcula brevis were grouped into four categories using the same
criterion: <30 mm ML, 40 to 49 mm ML, 50 to 59 mm ML and >60 mm ML
(Table V). The mean monthly growth rate in length of the >60 mm ML group was
only 3 mm/mo (Sx = 2.3) compared to over 13 mm/mo for the other three groups.
However, there was only a statistically significant difference (.10 > P > .05) between
the median monthly growth in length of the >60 mm ML group (4 mm/mo) and
the 40 to 49 mm ML group (13.5 mm/mo). Similar results were obtained using the
monthly wet weight data from the four groups. The mean monthly increase in wet
weight of the >60 mm ML group was low (mean 3.9 g/mo, Sx = 2.39) compared
to the other three groups which were all above 8.6 g/mo. However, the median
monthly growth rates in weight among the four groups were not statistically different.
Females showed reduced growth rates beyond 60 mm ML because they, like males,
were reaching maximal size.
The growth measurements suggest that the three species generally grow in the
laboratory at similar instantaneous relative growth rates (Tables IV and V); comparisons
among species of differing sizes are best done using instantaneous relative growth
rates (percent gain per day). Female Loligo plei and female Loligo pealei are not
included due to the reasons mentioned earlier. The mean instantaneous relative
growth rates in weight ranged from 1.4 %/day for L. pealei males to 1.9 %/day for
L. plei males and 1.9 %/day for Lolligimcula brevis females; the highest measured
rate was 4.6 %/day for a Lolligimcula brevis female. The mean instantaneous relative
growth rates in length were 0.8, 0.5, and 0.6%/day for L. pealei males, and Lolligimcula
brevis males and females, respectively. The mean instantaneous relative growth rate
SQUID MAINTENANCE, GROWTH, AND BEHAVIOR 671
for L. plei males was 1.5 %/day, but since this is based on only three squids it is
difficult to make comparisons. The highest measured rate was 1.9 %/day for a
L. plei male.
Gross growth efficiency (GGE) was estimated separately for three males and three
females of Loligo plei. All squids had been maintained previously for 42 days in CT
systems. At 21°C, the three males (107, 136, 136 mm ML) collectively gained 12.5
g over six days while ingesting 56.9 g of fish, for an estimated 22 percent GGE.
However, one male was dominant and very aggressive, and he was taking the vast
majority of food and accounted for 12.3 g of the weight gain. With a conservative
estimate that he obtained 80 percent of the fishes, his GGE was 27 percent, and his
daily food intake was 18 percent of his body weight per day. The three females (56,
89, 95 mm ML) were sexually mature and full of eggs. Collectively they gained only
1.4 g in six days while ingesting 36. 1 g offish, for an estimated 4 percent GGE. These
females were eating five to 17 fishes each day, and their collective mean daily food
intake was 10 percent. Apparently, either 10 percent daily food intake represents the
females1 required maintenance ration, or egg production utilized most of the energy
that otherwise may have contributed to somatic growth.
Principal causes of mortality
The majority of deaths have been related to (1) fin damage, (2) intraspecific
aggression, (3) sexual maturation, mating, and subsequent egg laying by females, and
(4) crowding.
Fin damage (Fig. 7) was very critical because it impaired normal swimming and
hovering and it eliminated stabilization during jet-propulsed movements, which were
necessary for deftly pursuing and attacking prey and avoiding aggressive conspecifics.
Details of the effects of fin damage were reported elsewhere by Leibovitz et al. (1977)
and Hulet et al. (1979). Although survival during shipboard transport and laboratory
transfer was fairly good, injuries incurred during shipboard transport of all species
and during trawl capture of Lolliguncula brevis often accounted for many deaths
during the first few days in captivity. Shipboard movement during heavy weather
and long transports caused increased wall contact that resulted in skin abrasion to
the squids, especially smaller ones.
The cumulative effects of fin damage from sporadic wall contact during long
maintenance periods also contributed to mortality in all species. There were rare
cases in which minor fin damage healed in some squids. Usually, however, the damage
remained in a steady state or slowly spread from bacterial infection. Subsequent wall
contact exacerbated existing wounds until eventually the fins became useless. The
patterns painted on the walls apparently helped reduce wall contact, but they did not
eliminate it.
Intraspecific aggression was one primary cause of mortality once the squids were
in the laboratory. It was characteristic among Loligo plei males and, to a slightly
lesser degree, Loligo pealei males; Lolliguncula brevis did not show obvious signs of
aggression. During establishment of their rank order and during mate selection, the
males vigorously made lateral displays and frontal attacks on subordinate males and
sometimes females. This disrupted feeding and led to increased fin damage from wall
contact when subordinate squids escaped. If Loligo spp. squids of a large size difference
were put in the same tank, the smaller squids were nearly always badly harassed and
died from fin damage and/or starvation within days, and on occasion they were
cannibalized.
Sexual maturation and its manifestations were another primary cause of mortality.
From the standpoint of laboratory survival, mating in Loligo was a fatal event because
672 R. T. HANLON ET AL.
females usually laid eggs and died within a few days. After repeated matings, males
of Loligo plei occasionally underwent an apparent catabolic change in which the arms
and fins deteriorated until the squids could not swim or capture food. Females of
Lolliguncula brevis (42 to 99 mm ML) and Loligo plei (51 to 139 mm ML) often
showed very rapid sexual maturation and egg development within two to three weeks
in captivity. Lolliguncula brevis and Loligo plei females that were segregated from
males often produced so many eggs that the mantle bulged and the internal organs
were pushed forward, probably affecting digestion; they would often die without
laying eggs.
Crowding caused increased intraspecific aggression, fin damage from more frequent
contact with the wall, and disruption of feeding. Had crowding been allowed over
long periods, it would have resulted in deterioration of water quality if the biological
carrying capacity of the tank system were exceeded.
There are other factors that contributed to mortality. Loligo plei that inked during
transfer in plastic bags died quickly in the inky water. Another similar event, which
we called the "shock syndrome," occurred when L. plei squids were startled and
began to ink. However, the ink was only ejected into the mantle and over the gills,
but not forcibly enough to get it out of the mantle. Ventilatory movements ceased
immediately and the squids invariably died. On rare occasions Loligo spp. would
leap completely out of the tank during the night. As previously mentioned, cannibalism
by Loligo spp. accounted for some mortality. A certain number of deaths were inex-
plicable, i.e., there was no skin damage, no aggression, etc. It is possible that an
inconspicuous pathologic condition existed, that parasites weakened the squids, or
that there was a nutritional deficiency. Although these were not obvious, they deserve
future attention.
A typical scenario of how fin damage, aggression, and sexual maturation affected
survival in a typical summer experiment on Loligo plei is as follows. Out of 20 adult
squids (ten male, ten female) caught at a night lighting station, 1 7 would survive to
dockside during a seven-hour transport. Within the first five days in a CT system,
four squids would die as a result of fin damage incurred during capture and transport.
The other 13 squids would school together and feed well for the next two weeks
except for isolated and mild aggression by the largest male as he established and
maintained rank order. During this time the size of the females' ovaries and nidamental
glands would swell noticeably and the accessory nidamental gland would become
bright red. Pair formation would begin, with the large male herding two to three
females from the school and laterally displaying towards subordinate males, who
would begin to accrue fin damage from hitting the walls during escape. Two males
would die within two days of this (about Day 18). Mating by several pairs would
take place over several days; five females would lay eggs within one day and then die
(about Day 21). Intraspecific aggression would increase, two males would die from
repeated matings or fin damage, rank order would change, and several more matings
with egg laying would occur by Day 25. Conditions would briefly stabilize for the
remaining four squids. Then another three would die within one week — one female
from having too many eggs but not laying them, and two males from aggression and
fin damage — until only one large male remained alive for several more weeks (Day
50). Mean survival would be about 20 days.
DISCUSSION
Our results clearly demonstrate that successful transport and long-term mainte-
nance of live loliginid squids are strongly dependent upon avoiding damage to the
SQUID MAINTENANCE, GROWTH, AND BEHAVIOR 673
skin and fins during capture, and upon using sufficiently large tanks during laboratory
maintenance to sustain high quality sea water. These points cannot be overstressed.
Key factors for laboratory survival may be summarized as follows: (1) prevention of
skin abrasion during capture, transport aboard ship, and transfer to the laboratory,
(2) the tank system must be sufficiently large, with opaque walls and preferably no
corners, (3) water quality must be high, (4) squids must have an ample food supply,
(5) they must not be crowded, (6) only squids of similar size should be in the same
tank to reduce aggression and cannibalism, and (7) sexes should be segregated to
reduce aggression associated with courtship, mating, and egg laying.
Capture and transport
From the outset we recognized that capturing a live, undamaged squid is difficult.
Over the past five years we experimented with several capture strategies: trawls, dipnets
and squid jigs with night lights, and encirclement nets such as lampara nets and purse
seines. Trawling is the least satisfactory capture method because of the high percentage
of dead and damaged squids due to prolonged contact with the net or other animals,
and to dropping of the catch on deck, which is a common practice of fishermen.
Trawling is the capture method presently used to capture squids for physiological
work at Woods Hole (Summers, 1968, 1969; Summers and McMahon, 1970, 1973;
Summers et ai, 1974), at Plymouth, England (Holme, 1974) and in the past off
Ocean City, Maryland (Brinley and Mullins, 1964). Few of the squids reach shore
alive because of skin damage, and those that do live stay alive briefly or for only a
few days (Holme, 1974). We have tested five trawl nets, but during 226 trawl stations
we had very little success in capturing live undamaged Loligo spp. in depths between
20 and 200 m. These nets have increased our catch of live Lolliguncula brevis, and
for this species trawling is our primary collection method. Success with Lolliguncula
brevis is mostly attributable to the short-duration tows in very shallow water, less
than 1 0 m deep. From our experience and that of many others, it appears that trawl
capture of large Loligo spp. from deep water may not ever by a satisfactory collection
technique if squids are to be kept alive more than a few days.
Less traumatic capture methods include squid jigging (day or night) or attracting
squids to lights and either dipnetting them, jigging them, or encircling them with a
lampara net or purse seine. Unfortunately, the mean catch rate has been low, primarily
because of the inconsistent attraction of squids to lights. It is likely that there is a
species-specific response to light and that a host of other factors such as hydrographic
conditions, moon phase, food availability, and sexual condition can influence squid
behavior in relation to artificial light. These parameters are not well defined for our
species. In some other geographic areas, squids may be caught alive with these methods
or with pound nets or floating fish traps, and these are certainly the preferred methods
if long-term maintenance is a key objective (Tardent, 1962; Summers and McMahon,
1970; Flores et ai. 1976, 1977; Matsumoto, 1976; OT3or et ai, 1977; Hurley, 1978;
Matsumoto and Shimada, 1980). Without doubt, future work on improving light
attraction and atraumatic capture methods that impart little or no skin damage should
receive high priority because it affects all aspects of squid maintenance.
Factors affecting survival during shipboard transport have been discussed in Results.
We believe that the configurations of the HCT and RHT tanks and their recommended
stocking densities provide adequate transport survival if the squids are in good condition
and water quality is not allowed to deteriorate. The larger the volume of the tanks,
the better, but vessel size will limit this in most cases. For comparison, Flores et ai
(1976, 1977) reported that fishermen transported 1000 Todarodes pacijicus in shipboard
674 R. T. HANLON ET AL.
live wells of 4000 1 capacity for about 1 2 hours, but they noted that the extreme
crowding (1 squid/4 1) resulted in extensive fin damage to most squids. O'Dor et al.
(1977) transported 20 Illex illecebrosus per container (60 X 90 X 30 cm deep, or 1
squid/8 1), but because of the short transport time of one hour, no mortalities occurred.
Matsumoto (1976) transported 15 Doryteuthis bleekeri in a 1 X 1 X 1 m tank (1
squid/66 1) for 3 to 5 hours with no mortalities; this is more space per squid than
our recommendation often Loligo spp. per 580 1 HCT tank (1 squid/58 1). These
results verify that squids cannot be crowded during long transport.
In many operations, a major breakdown in the successful handling of live squids
takes place at dockside. Our method of placing squids in plastic bags eliminates many
of the problems encountered at this stage, especially sloshing water that led to skin
damage and external commotion that startled the squids. Flores et al. (1976) used a
similar method that worked equally well. It would be desirable to reduce as much
as possible the large salinity and temperature shocks that squids encounter during
laboratory transfer, but this is often impractical.
Sea water systems
The performances of the 2 m circular tank systems and the 10,000 1 raceways
were satisfactory. Both designs provided two essential criteria: the capability to sustain
high quality water, and the physical dimensions to accommodate the movements
and habits of the squids. The advantages of our closed sea water systems are ( 1 )
independence from a natural sea water supply and hence, reproducibility at inland
laboratories, (2) efficient filtration of recirculated water, (3) large volume and wide
horizontal space for distribution of squids, (4) accessibility to and observation of live
animals, (5) simple construction, and (6) low cost.
We chose a closed (recirculating) system over an open (flow through) system for
several reasons. Water quality adjacent to Galveston Island is variable and often
unsatisfactory. Closed systems offer better control over temperature and salinity fluc-
tuations, disease organisms, turbidity, pollutants, and undesirable animals that compete
with cultured organisms for space and nutrients (Spotte, 1979a). Moreover, once the
requisite conditions for each species are identified, they can be carefully and contin-
uously regulated. It is clear that appropriately designed closed systems are suitable
for squid maintenance, since a comparison of squid maintenance work done in open
versus closed systems showed that maintenance success with closed systems equaled
or surpassed that in open systems (Boletzky and Hanlon, 1983).
Water quality is of great importance. Artificial sea water is a satisfactory substitute
for natural sea water, as evidenced from our present results and our success in rearing
Loligo opalescens from hatching to adult size over an 8-month period (Hanlon et
al., 1979; Yang et al., 1983). Aside from its biological usefulness, we found it to be
as cost effective as natural sea water because of the ship and personnel time required
to obtain high salinity offshore water, and the time and space needed to filter and
store it. Buildups of inorganic nitrogen (ammonia, nitrite, nitrate) were not particularly
high in our systems and were not a probable cause of mortality. However, our detailed
chemistry tests were few, and the subject of nitrogen tolerance is critical to closed
system maintenance and culture. Since 1982, we have had detailed chemistry tests
performed weekly on all systems. Preliminary results from transport experiments of
Lolliguncula brevis in plastic bags (one squid per 4 1 of sea water) indicate strongly
that they die primarily from decreased pH (increased hydrogen ion concentration)
and secondarily from ammonia buildup. For example, if pH is maintained within
±0.2 of its original level (e.g., 8.0), squids can survive up to 30 hours even when
SQUID MAINTENANCE, GROWTH, AND BEHAVIOR 675
levels of ammonia gradually increase to 10 mg/1 NH4-N (or 100 times the recommended
levels of Spotte, 1979a). In contrast, squids usually die if the pH is allowed to drop
below about 7.0. Therefore, it seems that pH is probably the most important barometer
of water quality for squids. Obviously, a great deal more work must be done to
understand aspects of water quality that most affect squid survival.
Matsumoto (1976) and Matsumoto and Shimada (1980) are the only authors that
give any nitration information on closed systems for squids. Matsumoto's first system
(1976) utilized sand filtration. In an improved system (Matsumoto and Shimada,
1980) they added 20 kg of zeolite and 10 kg of crushed oyster shell to the filtration
system. They attributed longer survival of squids to the zeolite, but the reasons are
unclear. Zeolite is a naturally occurring porous material that removes selective ions
by a combination of ion exchange and adsorption, but its use in marine systems is
limited because of competition from other ions in sea water that quickly reduce the
number of exchange sites available for binding contaminant ions such as ammonium,
nitrate, and phosphate (Spotte, 1979a). Johnson and Sieburth (1974) examined the
efficacy of zeolite in removing ammonium ions in salinities ranging from zero to 25
ppt. They found that, although initially it removed ammonium ions very efficiently,
it lost its effectiveness after only two or three liters of sea water (25 ppt.) had passed
through the ion exchange column. Furthermore, they found that the ideal size for
granules was 1.00 by 0.35 mm; Matsumoto and Shimada (1980) used an average
diameter of 3.00 mm. Based upon this scant information, it appears as though the
use of zeolite in marine systems is limited to occasional, brief use to complement
existing biological filters, but it does not seem likely that its continued use enhances
filtration.
Matsumoto and Shimada (1980) did not give values for pH or nitrogenous buildup.
However, it seems likely that improved survival of their squids was due partly to the
buffering capacity of the added oyster shell rather than to zeolite. Our CT systems
resulted in comparable survival using 360 kg of oyster shell as the only biological
filtration substrate. Reports by Hirayama (1970) and Bower et al. (1981) show that
sand filters (predominantly silica) have poor buffering capacities and that some cal-
careous filtrant (e.g., oyster shell or coral with calcium carbonate, or dolomite with
calcium carbonate and magnesium carbonate) is necessary to buffer closed sea water
systems. In view of these data, it is possible that the use of zeolite is unnecessary.
Rather, it is more important to have a large filtering bed area of calcareous material
and a small animal load, and to monitor pH and inorganic nitrogen buildup closely
to insure high quality water.
Tank size and configuration are also important to squid maintenance. Survival
is generally better in tanks with wide horizontal dimensions and no corners, all other
factors being equal. The narrow rectangular tanks used by Summers and McMahon
(1970, 1974) and Summers et al. (1974) ranged in size from 0.92 m wide X 1.83 m
long X 0.31 m deep to 1.37 m wide X 3.66 m length X 0.31 m deep, and mean
survival was two weeks or less. Larger round tanks from 1.5 to 2.0 m wide were used
by a variety of investigators to improve mean survival up to two to four weeks (e.g.,
Neill, 1971; Matsumoto, 1976;Soichi, 1977; Hurley, 1978; Matsumoto and Shimada,
1980; the CT system in this report). Large rectangular tanks (e.g., LaRoe, 1971;
Mikulich and Kozak, 1971; Flores et al., 1976, 1977; the raceway system in this
report) produced similar mean survival of several weeks. Finally, the very large 1 5
m-diameter circular tank used by O'Dor et al. (1977) resulted in survival between
26 and 82 days.
The painted wall patterns probably reduced wall contact by the squids. However,
we believe their effect was minimal on healthy, undamaged squids because they easily
676 R. T. HANLON ET AL.
avoided the walls in white walled tanks as well. Although damaged squids or those
engaged in intraspecific aggressive behavior hit walls regardless of their pattern, the
painted walls seemed to result in fewer collisions. These situations argue in favor of
bumper systems to lessen impact, but our impression is that this is not usually worth
the logistical difficulties involved. Rather, it is more practical to keep fewer squids
in larger tanks, so that the decrease in wall-to-volume ratio compensates for the
bumper. We used a polyethylene bumper sloping at 45° in Observation 1 for Loligo
plei (Table I), but it did not noticeably enhance survival.
Several worthwhile comparisons may be made between our closed system 10,000
1 raceway and the 580,000 1 open system Aquatron used by O'Dor et al. (1977) to
study Illex illecebrosus. In one sense, the Aquatron may represent the ultimate squid
holding tank because its great size provides a more natural environment for aspects
of normal behavior such as schooling, foraging, and reproduction. Two major draw-
backs are its cost and the difficulties of recapturing squids. We believe raceway systems
similar to that described herein offer a reasonable compromise. The raceways are
simple in design, inexpensive, and manufactured in a variety of lengths and widths.
Furthermore, squids survive well in them and are easily observed and recaptured (see
Fig. 3). O'Dor et al. (1977) kept a maximum of 50 squids in the Aquatron at one
time. By comparison, we kept 46 Loligo plei in a raceway for a mean survival of
~20 days and a maximum of 57 days (Observation 17, Table I). O'Dor et al. (1977)
speculated that the tank diameter required to allow "relaxed" behavior in Illex il-
lecebrosus was between 3.7 and 15 m. Our observations of loliginid squids in our
2 m diameter CT system indicate this distance is less for loliginid squids. Certainly
the 10 m long X 2 m wide raceway provided sufficient room for relaxed behavior
for small numbers of all three of our species, especially in later versions of the raceway
in which the central partition was removed.
Behavior, survival, and growth
In the course of initially testing the prototype sea water systems, it became apparent
that the behavior of the squids provided the best evaluation of the systems. This
observation led to more detailed analyses of behavior that provided feedback on how
to refine the methods and systems in order to accommodate the needs of the squids
for long-term maintenance. The significance of this seemingly simple philosophy for
providing the basic requirements of squids tends to be overlooked by many. Clearly,
aspects of behavior are the true limiting factors to survival and growth of wild-caught
loliginid squids in a laboratory environment. The fin and skin damage that were
often cited in this report as causes of mortality were merely manifestations of either
aspects of behavior, transport in small tanks, or, in Lolliguncula brevis, trawl capture.
By carefully observing the squids it was possible to correlate body patterns of
chromatophores and postures with specific aspects of behavior such as stress, calmness,
aggressiveness, and precopulatory behavior (Hanlon, 1978, 1981, 1982, and in prep.).
Accordingly, these clues are now used to avoid some problems before they develop.
For example, it is often difficult to segregate newly-caught animals by sex. When
Loligo plei males begin to show lateral displays, the squids are segregated by size and
by sex, with the usual result of restoring calmness and normal feeding, which in turn
promotes increased survival. However, in cases such as this, the effects are relatively
short-lived, on the order of several days or weeks only.
For long-term survival and growth in laboratory tanks, intraspecific aggression
and sexual maturation in Loligo spp. are the two most restrictive factors. In Lolliguncula
brevis, the factor most responsible appears to be sexual maturation. Feeding in all
SQUID MAINTENANCE, GROWTH, AND BEHAVIOR 677
species is clearly not the problem. In Loligo spp., the size relationships among squids
exert a strong influence on survival. One reason is that larger squids dominate prey
capture. In one 16-day growth observation period (part of Observation 16, Table IV),
three males (136, 136, 107 mm ML) were kept together in a CT system with a diet
of only fishes (Cyprinidontidae). One of the 136 mm ML males quickly became
dominant, harassed the other two squids, and ate nearly all the fishes. During this
period he grew at a rate of 51 mm ML/mo, while the other two squids grew the
equivalent of 9 and -4 mm ML/mo. In contrast, the latter two males had grown at
rates of 73 and 48 mm ML/mo during the 20 days previous to this observation when
they were in a tank with squids of initial sizes of 68, 75, and 88 mm ML. It is
noteworthy that the squids were less aggressive when they were smaller and new in
the laboratory. A similar effect of intraspecific aggression on feeding was reported for
fishes by Peter (1979).
A more dramatic intraspecific aggressive effect of size disparity was cannibalism
by Loligo spp. Cannibalism was not solely a result of food deprivation because in
some cases it occurred in tanks that were stocked with food organisms. Cannibalized
squids were either smaller or injured. The field observation that schools of Loligo
plei usually contained squids of similar size suggests that cannibalism is a means by
which size specificity is maintained and by which weakened squids are eliminated.
However, when schools of mating pairs are formed, as seen by Waller and Wicklund
(1968) in the Bahamas, the smaller females are readily accepted as mates. Neither
ourselves nor Waller and Wicklund (1968) observed cannibalism among mates.
Sexual maturation seemed to progress at an accelerated rate in the laboratory.
Our evidence is twofold: the gonads of most squids usually grew rapidly within 1
to 4 weeks in the tanks, and wild-caught females of Loligo plei generally had less
well-developed gonads than females of similar size that had been captured in the
same geographic area but kept in the laboratory for three weeks or so. The effects of
extrinsic regulators of sexual maturation such as light (intensity and cycle), temperature,
and food are not understood. It is possible that the general stress of capture and
maintenance, combined with constant food availability and a different light regime,
was enough to accelerate sexual maturation. In any event, even the longest-lived
squids of each species were always sexually mature when they died. Our recent ob-
servation that Loligo opalescens reared through the life cycle in the laboratory all
attained sexual maturation and died within eight months indicates that the effects of
maturation are, at least in part, intrinsically regulated and may be difficult to control
in the laboratory.
Interspecific compatibility among the three species in the laboratory had an in-
teresting correlate in the field. Although the three species were never observed together
in situ, different combinations of all three species were captured together in 1 5-minute
trawls, including all three species in the same trawl on 1 1 occasions (Hixon, 1980a).
The results of an interspecific association analysis (Cox, 1980 based on Cole, 1949)
based on 150 trawl stations showed that there was a positive coefficient of association
between Lolliguncula brevis and Loligo plei, indicating that these species are found
frequently in close proximity to one another. Although the other two combinations
showed negative coefficients, this was a reflection of the species' areal and depth
distribution as well as temperature and salinity preferences (Hixon, 1980a). These
findings do not mean that these three species co-occur in the same schools, but it
does indicate that species-specific schools may co-occur in the same habitat.
In Table VI are comparisons of survival among squids maintained in the laboratory
by various researchers. Direct comparisons are impossible because of the highly varying
conditions surrounding each worker's geographic area, the species, time of year,
678
R. T. HANLON ET AL.
1
0 •* OOO
r-~ t~- oo oo oo
ON ON ON ON ON
sO C C
ON o o ~Q ~Q "Q
0^
U
u
—~- r— .— CQ ^ C^
•S -S -*. c c c
f ' • C C C
Oj, .C S S ON oo 00 '-C £ IS
r^ r~- r-
^2
oi) 3 22: ob ob^'OTD'O
ON ON ON
^ ON ON ON
uT
O
_ E S^-o-o'a E E -§S§ 2
(^ ' »^ so 55^" ^^ ' ' ri O O O ON
T3 "" ""
%
o\t c^ °ot 5 •stit;^ ~~I^!
- — iO ctf ~~ t/) c« t/^ ON O O OOOO
a a a
w a a a g;
"*•
• & >•> T-T <" ^ <D — g- g- EESE c —
^J ^5 ^5
"g 12 «; 'S ~
o1* ^~EEE>>" " 5S33 5,
OOO
3 <u <u u JS
3£ S < 1 1 1 1 S H S S S S ^Z
bob
•^000-5
D.
a a a.
•o
u °o
222
O
£ ^ o. ~ -E .M"7
^ ^ ^
4-1
qj
.%?.%£ -^ g « — S4 ' "^ £ "> i«i«on« inin
OOO
C W5 V3 ^ 00
£
Z< ^^ ^^^ ^ •• VD QJ) ^* 0^ Q^ O^ O^ QJQ QJ5
-C -C -C
f|lQ Q^ Q^ Q^
— ' • — • — — • —
"§)"§) *J'«<:^-t:$|*- <-
OO OO 00
c c c
<2 •—>•—>•—.•—.
s
3
C C «^ ^ ^j oJj
4-* *-• '4—1
c^ cfl c^
O.
03 C
OOO
U
tS)
•g
t/~j rn — ^^ OO 'O ^^ r^- r**} <N r*"^ fN) ^5
in >r> >/•>
r*1 r~ ON O TJ-
§r ~
^^ S fQ^^lS^ ^ — »^ir*-i— -
— m <^
so — • -<t
>
<^ 3
) )
u
o 13
CQ
d oj
Z
H
13
'> ^
1— C/3
oo ^f *^i r^~ r^* o^ O1^ fi ^^ */~^ i ^^5 ^^ ^^
5 00 JN
£ - ;qo ^
3 >>
•^
. T3
^
X ' — '
S-
%
03
•5
a
'§
>
^.__ fN«-)Tt-<J-r~-t--00 ON r-TTu^rn TJ-
O O m
ON t- 0 0 0
3 >,
/ i 7 7 T ' 7
77'
r^t j — • rn CM
? Ill
§5
c 3
K
^j
(U
C
S
3
£
2
r\
- -5
3
mparisons oflabi
u
'1
o.
•3 1 1
•1 ^ -Si -a
Q. -^ -G J^
O ^* ^ t^ ^
£3 1 1 « i5
p« ft. CJ, ?S >
^ .|) .§3 J? |, .|)
•o 5
'in K
II
6 1
u. r^
OJ ~
•o K
J-
Todarodes pacific
r*J
3
3 *""'
O
C/^
t/3
SQUID MAINTENANCE, GROWTH, AND BEHAVIOR 679
number of animals evaluated, capture and transport methods, maintenance tank size,
etc. In many cases, the parameters upon which survival was described were not stated
or denned clearly; more significantly, information regarding selection criteria of squids
that were included or deleted from survival analyses was not always provided. Nev-
ertheless, the table provides an overview and forms a basis for discussion. Overall,
the results of our work compare favorably with other research efforts.
Survival of Loligo pealei in the laboratory (Table VI) has been very low historically.
Undoubtedly, a major reason for this is that most evaluations were made on trawl-
caught squids that had substantial skin damage. All of these earlier evaluations have
been on L. pealei from New England waters. Our results of 28 days mean survival
are based on few animals (n == 37), but they are a considerable improvement upon
past efforts. The main reason for improvement is that the squids were caught in
nearly perfect condition with dipnets. The very long transport times (mean 15 hours)
resulted in some degree of skin damage that affected long-term survival. Seasonality
had no obvious effect on survival because squids collected throughout the year survived
equally well (Table II). Survival was enhanced by the fact that no juveniles were
maintained, but only mid- to full-sized adults which generally do better in captivity.
Survival was strongly enhanced by the bottom sitting behavior and general calmness
of this species in captivity. The fallacy that bottom sitting in L. pealei is abnormal
behavior must be dispelled once and for all. Williams (1909), Stevenson (1934), and
Macy (1982) have all reported this behavior as normal, and our observations confirm
their findings. From a maintenance standpoint, it may be important to provide a
substrate that is suitable to the squids for bottom sitting; the crushed oyster shell in
our systems was acceptable to them.
Survival of Loligo plei was fairly low overall (Table VI). This resulted partly
because we analyzed every squid we caught at those stations regardless of size, sex,
or condition. When conditions were good (notably Observations 11, 14, 15, 16, 17,
in Table I) mean survival of squids (excluding juveniles) ranged from 14 to 84 days.
This maximal survival of 84 days is the longest that any squid of the genus Loligo
has been maintained. The steep mortality slope in Figure 8 is attributable initially
to skin damage during transport and generally poor survival by juveniles, and later
to intraspecific aggression and sexual maturation which limited long-term survival.
We expect that long-term survival would improve by selecting only mid- or large-
sized squids in the best condition at capture, transporting fewer squids per tank, and
segregating sexes in the laboratory.
Lolliguncula brevis survival was good. This is the only species we know of that
withstands trawl capture well. As mentioned, part of the reason is the short towing
period in shallow water, but this species also is apparently less susceptible to skin
damage than other loliginid squids. If the high early mortality (Fig. 8) attributable
to capture trauma is eliminated, then mean survival for 197 squids becomes 29 days
for all sizes of squids. This compares favorably with any species studied thus far. The
maximal survival of 125 days is the longest that any wild-caught squid has ever been
maintained in captivity. The long survival and high growth rates of this species in
captivity make it a potentially useful species for long-term in vivo experimentation.
Loligo vulgaris (from the Mediterranean) and Doryteuthis bleekeri (from Japan)
may be compared best to the three species mentioned above because they are all in
the Family Loliginidae. Tardent (1962) and Neill (1971) demonstrated that jig-caught
Loligo vulgaris could be kept for about 14 days in large tanks. Matsumoto (1976)
and Matsumoto and Shimada (1980) showed that jig-caught Doryteuthis bleekeri
could be maintained consistently for about 14 days (Table VI). They also reported
one run in which ten squids had a mean survival of approximately 43 days. This is
680 R. T. HANLON ET AL.
an excellent result that, although not quantified, shows survival is high in jig-caught
adult squids that are transported carefully and not crowded in laboratory tanks.
Unfortunately, these authors give no details of squid size or sex, nor of the details
of selection at the capture site. Therefore, it is difficult to make other comparisons
between Doryteuthis bleekeri and Loligo plei (considered by some to belong to the
genus Doryteuthis), which is similar in size and appearance.
Survival in captivity of the oceanic, oegopsid squids Todarodes pacificus and Illex
illecebrosus has been good (Table VI). These high survival times are a result of capture
and transport methods that are atraumatic, as well as the use of very large maintenance
tanks and good feeding. OTJor et al. (1977) found that mid- to large-sized adult Illex
illecebrosus survived a mean of about 30 days or more, although in one group of
squids mean survival was only 13 days. Mikulich and Kozak (1971), Flores et al.
(1976, 1977), and Soichi (1977) reported mean survival up to 30 days for mid- to
large-sized Todarodes pacificus. All of the results above are excellent examples of
how squids can be kept alive for weeks if certain principals are adhered to.
To summarize the criteria necessary for good survival of squids in captivity, we
once again reference the first paragraph in this discussion but also the statements by
other successful researchers that reached similar conclusions (e.g.. Summers et al.,
1974: pg. 300; O'Dor et al., 1977: pg. 334; Flores et al., 1977). Since the squids of
greatest immediate interest to neurobiologists are mainly of the genus Loligo, we
believe that future researchers can expect mean survival of two weeks or more for
loliginid squids captured, transported, and maintained by the methods outlined in
this communication. Our demonstration that artificial sea water is a suitable substitute
for natural sea water, and that a relatively simple, inexpensive closed system maintains
squids well, will also provide alternate ways for others to keep squids alive for ex-
perimentation.
Growth comparisons may be made between our laboratory results and those of
other researchers, and between our laboratory results and field estimates of growth.
In general, all of our laboratory growth rates are higher than estimates from size-
frequency analyses of field data. Our Loligo pealei mean growth rate of 44 mm/mo
for males (Table IV) was higher than the 23 mm/mo reported from the laboratory
studies of Macy (1980) as well as the calculated field growth rate of 15.7 mm/mo
(range 6.5 to 24.5 mm/mo) based upon 618 males caught over a two-year period off
the Texas coast (Hixon et al., 1981). Hixon et al. (1981) also provided a historical
comparison of field growth rate estimates, nearly all of which are under 20 mm/mo.
Our single observation of 13 mm/mo in one female compares closely with the 1 1.7
mm/mo (range 8.6 to 14.2 mm/mo) calculated rate of 733 females caught off the
Texas coast (Hixon et al., 1 98 1 ). The high growth rates in males are partly a reflection
of ideal laboratory conditions, but they indicate that males are probably capable of
very rapid growth in the field when conditions are favorable.
Loligo plei males grew in our laboratory at a mean rate of 47 mm/mo (Table
IV), substantially greater than the only other laboratory estimate of 15 to 25 mm/
mo given by LaRoe (1971) for comparable temperatures. Field estimates are also
lower. Whitaker (1978) estimated growth rates of 5.0 to 14.3 mm/mo for 1065 squids
caught off the southeastern U. S. during 1974 and 1975; the 14.3 mm/mo rate was
for a period of 1 32 days during spring and summer, when temperatures were similar
to the laboratory temperatures in our tank systems. Hixon ( 1 980a) calculated a growth
rate of 1 1.5 mm/mo (range 2.0 to 20.0 mm/mo) for 1819 male squids caught over
a two-year period off the Texas coast. His estimate for 1887 females was 6.8 mm/
mo (range 2.7 to 9.5 mm/mo); in comparison, our laboratory females did not grow
(Table IV) due to sexual maturation. As in Loligo pealei, the high growth rates
SQUID MAINTENANCE, GROWTH, AND BEHAVIOR 681
attained by males in the laboratory indicate that this species is capable of very rapid
growth during brief, ideal periods.
Lolliguncula brevis males grew at a mean rate of 8 mm/mo, and females at 1 1
mm/mo in our laboratory observations (Table V). No other laboratory data are
available for comparison, but Hixon (1980a) estimated field growth of 1141 males
at 8.6 mm/mo (range 5.7 to 1 1.4 mm/mo), and 1045 females at 7.9 mm/mo (range
4.3 to 12.5 mm/mo) off the Texas coast. Although the mean growth rates of males
agree well, the maximal laboratory rate of 20 mm/mo is much higher than the
maximal field estimate of 11.4 mm/mo. Among females, both the mean ( 1 1 mm/
mo) and maximal (31 mm/mo) laboratory rates are much higher than those from
field estimates (7.9 mm/mo and 12.5 mm/mo, respectively). In all cases, Lolliguncula
brevis shows the capability of growing at rates higher than previously thought when
conditions are particularly good.
Some comparisons of growth in body weight may be made also. Among the three
species in this study, the instantaneous relative growth rates in weight were on the
same order of magnitude: males of all three species and female Lolliguncula brevis
grew at mean rates of 1.4, 1.7, and 1.9%/day at temperatures of 18 to 23°C (Tables
IV and V). In comparison, Hirtle et al. (1981) reported that Illex illecebrosus grew
at rates of 1.1 to 1.9%/day at 7 to 10°C. In the cuttlefish Sepia officinalis, Richard
(1971) and Pascual (1978) reported growth rates of approximately 1.0 to 4.0%/day
in mid-sized to adult animals at temperatures of 14 to 26 °C. The only other growth
rate reported in the literature is by Choe ( 1 966), who calculated a very fast rate of
7.1%/day in mid-sized Sepioteuthis lessoniana at 23 to 31°C. Growth this fast is
usually only attained by very young animals during their exponential growth phase,
but apparently Sepioteuthis lessoniana is capable of continuing fast growth for a long
period under ideal laboratory conditions.
The gross growth efficiency (GGE) estimate of 27 percent in a male Loligo plei
and the estimated feeding rates of 18 and 10 percent for male and female Loligo plei
are comparable to other squids. LaRoe (1971) reported that Sepioteuthis sepioidea
(10 weeks old) showed GGEs of 20 to 40 percent and daily food intakes of 10 to 30
percent. Macy (1980) reported a mean daily food intake of 1 1 percent for adult Loligo
pealei in the laboratory. Yang et al. (1983) found that laboratory-cultured Loligo
opalescens had a mean daily food intake of 14.9 percent between Days 108 and 232
(adult size). Hirtle et al. (1983) reported that captive Illex illecebrosus showed a mean
GGE of 40 percent and an average daily food intake of 10 percent. Soichi (1977)
calculated that Todarodes pacificus had a mean daily food intake of 24.3 percent
(range 10.6 to 38.9 percent).
The effects of specific diets on growth seem small. Laboratory and field studies
show consistently that squids feed predominantly on crustaceans and fishes (e.g.,
Fields, 1965; Vovk, 1974; Ennis and Collins, 1978; Vinogradov and Noskov, 1979;
Macy, 1982). In the present study, squids grew equally well on fish-only, shrimp-
only, or mixed diets. Hirtle et al. (1981) noted similar results with Illex illecebrosus.
Previous research has indicated that smaller squids generally appear to have a slight
preference for crustaceans, while fishes are preferred when the squids are larger (Hirtle
et al., 198 1; this report). These differences are so small that, for laboratory maintenance
or growth, either diet is acceptable.
The growth results given above indicate that it is feasible to grow mid-sized squids
to adult size in a reasonably short time. This may be useful for in vivo experimentation,
both short- and long-term. It might also be useful as an alternate way of providing
larger axons. We have already demonstrated this on a small scale in Lolliguncula
brevis (Table V). With growth rates of 10 mm/mo for mid-sized males and 13 mm/
682 R. T. HANLON ET AL.
mo for mid-sized females (Table V), squids were grown another 1 7 to 2 1 mm over
50 days to bring them to full adult size, with axons as large as 200 p.m in the largest
females (Hulet et al, 1980). There are possibilities with Loligo spp. as well. For
example, a mid-sized Loligo plei male 100 mm ML could possibly be grown to 160
mm ML in about 45 days, assuming that only the best animals were selected and
that they had a sustained growth rate of 40 mm/mo. At 160 mm ML, the giant axon
measures approximately 325 nm in this species, sizeable enough for many types of
axon experiments. The same type of operation could apply to Loligo pealei which,
from our experience, is a better candidate because (1) it is less aggressive than L.
plei, (2) it sits on the bottom, (3) it is calmer, and (4) it grows larger. Since the
majority of Loligo spp. caught by night lighting off Galveston are around 100 mm
ML, testing of this concept deserves future attention.
ACKNOWLEDGMENTS
We thank John W. Forsythe, Joseph P. Hendrix, Jr., and Deirdre A. McConathy
for excellent technical assistance. We also gratefully acknowledge continued support
from DHHS Grant No. RR 01024, Division of Research Resources, National Institutes
of Health, and the Marine Medicine General Budget of the Marine Biomedical Institute,
University of Texas Medical Branch at Galveston. Portions of this work were submitted
by R.T.H. and R.F.H. in partial fulfillment of the Ph.D. requirements at the Rosenstiel
School of Marine and Atmospheric Science, University of Miami, Florida.
LITERATURE CITED
ANTONIE, R. L. 1976. Fixed Biological Surfaces — Wastewater Treatment. CRC Press, Cleveland, Ohio.
200 pp.
ARNOLD, J. M. 1962. Mating behavior and social structure in Loligo pealei. Biol. Bull. 123: 53-57.
ARNOLD, J. M., W. C. SUMMERS, D. L. GILBERT, R. S. MANALIS, N. W. DAW, AND R. J. LASER. 1974.
A Guide To The Laboratory Use Of The Squid, Loligo pealei. Marine Biological Laboratory,
Woods Hole, Massachusetts. 74 pp.
BOLETZKY, S.v., AND R. T. HANLON. 1983. A review of the laboratory maintenance, rearing and culture
of cephalopod molluscs. Mem. Nat. Museum Victoria 44: 147-187.
BOWER, C. E., D. T. TURNER, AND S. SPOTTE. 1981. pH maintenance in closed seawater culture systems:
limitations of calcareous filtrants. Aquacullure 23: 21 1-217.
BRINLEY, F. J., JR., AND L. J. MULLINS. 1964. The collection of squid for use at an inland laboratory.
Unpublished Report to NSF (6B-1765). 6 pp.
CHOE, S. 1966. On the growth, feeding rates and the efficiency of food conversion for cuttlefishes and
squids. Korean J. Zoo/. 9(2): 12-19. (In Korean).
COLE, L. C. 1949. The measurement of interspecific association. Ecology 30: 41 1-424.
CONOVER, W. J., AND R. L. IMAN. 1976. On some alternative procedures using ranks for the analysis of
experimental designs. Communications In Statistics Series A5: 1349-1368.
Cox, G. W. 1980. Laboratory Manual Of General Ecology. William C. Brown, Dubuque, Iowa. 237 pp.
DiPoLO, R. 1976. The influence of nucleotides on calcium fluxes. Fed. Proc. 35(14): 2579-2582.
DRAGOVICH, A., AND J. A. KELLY, JR. 1962. A biological study and some economic aspects of squid in
Tampa Bay, Florida. Proc. Gulj and Caribbean Fish. Inst. 15: 87-103.
ENNIS, G. P., AND P. W. COLLINS. 1978. Food and feeding of the short-finned squid (/. illecebrosus) during
its seasonal occurrence inshore at Newfoundland and a brief review of the tropic relationships of
the species. ICNAF Res. Doc. 78/II/7.
FIELDS, W. G. 1965. The structure, development, food relations, reproduction, and life history of the squid
Loligo opalescens Berry. Calif. Fish & Game. Fish Bull. 131: 1-108.
FLORES, E. E. C., S. IGARASHI, R. MIKAMI, AND K. KOBAYASHI. 1976. Studies on squid behavior in
relation to fishing. I. On the handling of squid, Todarodes pacificus Steenstrup, for behavioral
study. Bull. Fac. Fish. 27(3-4): 145-151.
FLORES, E. E. C., S. IGARASHI, AND T. MIKAMI. 1977. Studies on squid behavior in relation to fishing.
II. On the survival of squid, Todarodes pacificus Steenstrup, in experimental aquarium. Bull. Fac.
Fish. 28(3): 137-147.
SQUID MAINTENANCE, GROWTH, AND BEHAVIOR 683
GIBBONS, J. D. 1976. Nonpammetric Methods For Quantitative Analysis. Holt, Rinehart and Winston,
New York.
HANLON, R. T. 1978. Aspects of the Biology of the Squid Loligo (Doryteuthis) plei in Captivity. Ph.D.
Dissertation, University of Miami, Coral Gables, Florida.
HANLON, R. T. 1981. Courting and mating behavior in the squid Loligo (Doryteuthis) plei. P. 325 (Abstract)
in Advances in Invertebrate Reproduction, W. H. Clark, Jr., and T. S. Adams, eds. Elsevier/North-
Holland, New York. 400 pp.
HANLON, R. T. 1982. The functional organization of chromatophores and iridescent cells in the body
patterning of Loligo plei (Cephalopoda: Myopsida). Malacologia 23(1): 89-1 19.
HANLON, R. T., AND R. F. HIXON. 1981. In situ behavioral observations of cephalopods at Grand Cayman,
B.W.I. Bull. Am. Malacol. Union, 1980: 12. (Abstract)
HANLON, R. T., R. F. HIXON, AND W. H. HULET. 1978. Laboratory maintenance of wild-caught loliginid
squids. Pp. 20.1-20.14 in Proceedings of the Workshop on the squid Illex illecebrosus, Fisheries
and Marine Service Technical Report, no. 833. N. Balch, T. Amaratunga, and R. K. O'Dor, eds.
HANLON, R. T., R. F. HIXON, J. W. FORSYTHE, AND J. P. HENDRIX, JR. 1980. Cephalopods attracted to
experimental night lights during a saturation dive at St. Croix, U. S. Virgin Islands. Proc. Am.
Malacol. Union, 1979: 53-58.
HANLON, R. T., R. F. HIXON, W. H. HULET, AND W. T. YANG. 1 979. Rearing experiments on the California
market squid Loligo opalescens Berry, 1911. Veliger 21(4): 428-431.
HENDRIX, J. P., JR., W. H. HULET, AND J. M. GREENBERG. 1981. Salinity tolerance and the responses to
hypoosmotic stress of the bay squid Lolliguncula brevis, a euryhaline cephalopod mollusc. Comp.
Biochem. Physiol. 69 A: 641-648.
HIRAYAMA, K. 1970. Studies on water control by filtration through sand bed in a marine aquarium with
closed circulating system — VI. Acidification of aquarium water. Bull. Jpn. Soc. Sci. Fish. 36(1):
26-34.
HIRTLE, R. W. M., M. E. DEMONT, AND R. K. O'DOR. 1981. Feeding, growth, and metabolic rates in
captive short-finned squid, Illex illecebrosus, in relation to the natural population. / Shell Res.
1(2): 187-192.
HIXON, R. F. 1980a. Growth, Reproductive Biology, Distribution and Abundance of Three Species of
Loliginid Squid (Myopsida, Cephalopoda) in the Northwest Gulf of Mexico. Ph.D. Dissertation,
University of Miami, Coral Gables, Florida.
HIXON, R. F. 1980b. Potential commercial squid resources of the Gulf of Mexico — an updated review.
Pp. 54-73 in Proceedings of a Workshop for Potential Fishery Resources of the Northern Gulf of
Mexico, Mississippi-Alabama Sea Grant Consortium, M. Flandorfer, and L. Skuplen, eds. Gulf
Coast Research Laboratory, Ocean Springs, Mississippi. 92 pp.
HIXON, R. F., R. T. HANLON, S. M. GILLESPIE, AND W. L. GRIFFIN. 1980. Squid fishery in Texas; biological,
economic and market considerations. Mar. Fish. Rev. 42(7-8): 44-50.
HIXON, R. F., R. T. HANLON, AND W. H. HULET. 1981. Growth and maximal size of the long-finned
squid Loligo pealei in the northwestern Gulf of Mexico. J. Shell. Res. 1(2): 181-185.
HOLME, N. A. 1974. The biology of Loligo forbesi Steenstrup (Mollusca: Cephalopoda) in the Plymouth
area. J. Mar. Biol. Assoc. U. K. 54: 481-503.
HULET, W. H., R. T. HANLON, AND R. F. HIXON. 1980. Lolliguncula brevis — a new squid species for the
neuroscience laboratory. Trends in Neuroscience 3(4): 4-5.
HULET, W. H., M. R. VILLOCH, R. F. HIXON, AND R. T. HANLON. 1979. Fin damage in captured and
reared squids. Lab. Anim. Sci. 2">(4): 528-533.
HURLEY, A. C. 1978. School structure of the squid Loligo opalescens. Fish. Bull. 76(2): 433-442.
JOHNSON, P. W., AND J. McN. SIEBURTH. 1974. Ammonia removal by selective ion exchange: a backup
system for microbiological filters in closed-system aquaculture. Aquaculture 4: 61-68.
KATO, S., AND J. E. HARDWICK. 1975. The California squid fishery. FAO Fish. Rep. 170. Suppl. 1: 107-
127.
KIER, W. M. 1982. The functional morphology of the musculature of squid (Loliginidae) arms and tentacles.
J. Morphol. 172: 179-192.
LARGE, E. T. 1971. The culture and maintenance of loliginid squids Sepioteuthis sepioidea and Doryteuthis
plei. Mar. Biol. 9(1): 9-25.
LEIBOVITZ, L., T. R. MEYERS, R. ELSTON, AND P. CHANLEY. 1977. Necrotic exfoliative dermatitis of
captive squid (Loligo pealei). J. Invert. Path. 30: 369-376.
MACY, W. K., III. 1980. The Ecology of the Common Squid Loligo pealei Lesueur, 1821 in Rhode Island
Waters. Ph.D. Dissertation, University of Rhode Island.
MACY, W. K., III. 1982. Feeding patterns of the long-finned squid, Loligo pealei, in New England waters.
Biol. Bull. 162(1): 28-38.
MATSUMOTO, G. 1976. Transportation and maintenance of adult squid (Doryteuthis bleekeri) for physiological
studies. Biol. Bull. 150: 279-285.
684 R. T. HANLON ET AL.
MATSUMOTO, G., AND J. SHIMADA. 1980. Further improvement upon maintenance of adult squid (Dor-
ytenthis bleekeri) in a small circular and closed-system aquarium tank. Biol. Bull. 159(2): 319-
324.
MIKULICH, L. V., AND L. P. KOZAK. 1971. Experimental rearing of Pacific Ocean squid under artificial
conditions. Ekologiya No. 3: 94-96. (In Russian)
MOCK, C. R., L. R. Ross, AND B. R. SALSER. 1977. Design and preliminary evaluation of a closed system
for shrimp culture. Proc. of the 8th Annual Meeting of the World Mariculture Society. Pp. 335-
372.
MOYNIHAN, M., AND A. F. RODANICHE. 1982. The behavior and natural history of the Caribbean reef
squid Sepioteuthis sepioidea. Advances in Ethology No. 25, 151 pp.
NEILL, S. ST. J. 1971. Notes on squid and cuttlefish; keeping, handling and colour-patterns. Pubbl. Staz.
Zool. Napoli 39: 64-69.
O'DoR, R. K., R. D. DURWARD, AND N. BALCH. 1977. Maintenance and maturation of squid (lllex
illecebrosus) in a 15 meter circular pool. Biol. Bull. 153: 322-335.
PAINTER, H. A. 1970. A review of the literature on inorganic nitrogen metabolism in microorganisms.
Water Res. 4: 393-450.
PASCUAL, E. 1978. Crecimiento y alimentacion de tres generaciones de Sepia officinalis en cultivo. Inv.
Pesq. 42(2): 421-442.
PETER, R. E. 1979. The brain and feeding behavior. Pp. 121-159 in Fish Physiology, Vol. 8. W. S. Hoar,
D. J. Randall, and J. R. Brett, eds. Academic Press, New York.
RAND, M. C., A. E. GREENBERG, AND M. J. TARAS, eds. 1976. Standard Methods for jhe Examination
of Water and Wastewater, 14th ed. American Public Health Association, Washington.
RATHJEN, W. F., R. T. HANLON, AND R. F. HIXON. 1979. Squid fishery resources and development in
the Northwest Atlantic and Gulf of Mexico. Proc. GulfCarribb. Fish. Inst. 31: 145-157.
RICHARD, A. 1971. Contribution a I 'Etude Experimental de la Croissance el la Striation de la Coquille
de Sepia officinalis L. (Mollusque, Cephalopode). These de Doctoral d'Etat, Univ. Lille (No. 243).
ROSENBERG, P. 1973. The giant axon of the squid: A useful preparation for neurochemical and phar-
macological studies. Pp. 97-160 in Methods of Neurochemistry, Vol. 4, R. Fried, ed. Marcel
Dekker, Inc., New York, 322 pp.
RUBIN, E., R. EVERETT, JR., J. J. WEINSTOCK, AND H. M. SCHOEN. 1963. Contaminant removal from
sewage plant effluents by foaming. PHS Publ. no. 999-WF-5, Public Health Service, Cincinnati,
Ohio. 56 pp.
SOICHI, M., 1977. On the growth and food quantity of young common squid, Todarodes pacificus, in
captivity. Dosuishi 13(4): 79-82. (In Japanese)
SOLORZANO, L. 1969. Determination of ammonia in natural waters by the phenolhypochlorite method.
Limnol. Oceanogr. 14: 799-801.
SPOTTE, S. 1973. Marine Aquarium Keeping: The Science, Animals, and Art. John Wiley & Sons, Inc.,
New York. 173 pp.
SPOTTE, S. 1979a. Fish and Invertebrate Culture: Water Management in Closed Systems. 2nd Ed. John
Wiley & Sons, Inc., New York. 179 pp.
SPOTTE, S. 1979b. Seawater Aquariums: The Captive Environment. John Wiley & Sons, Inc., New York.
413 pp.
STEVENSON, J. A. 1934. On the behaviour of the long-finned squid [Loligo pealii, (Lesueur).]. Canadian
Field-Naturalist XL VIII: 4-7.
STRICKLAND, J. D. H., AND T. R. PARSONS. 1972. A Practical Handbook of Seawater Analysis, 2nd Ed.
Bull. 167, Fish. Res. Board Can., Ottawa, 310 pp.
SUMMERS, W. C. 1968. The growth and size distribution of current year class Loligo pealei. Biol. Bull.
135: 366-377.
SUMMERS, W. C. 1969. Winter population of Loligo pealei in the Mid-Atlantic Bight. Biol. Bull. 137: 202-
216.
SUMMERS, W. C., AND J. J. McMAHON. 1970. Survival of unfed squid, Loligo pealei, in an aquarium.
Biol. Bull. 138: 389-396.
SUMMERS, W. C., AND J. J. MCMAHON. 1974. Studies on the maintenance of adult squid (Loligo pealei).
I. Factorial survey. Biol. Bull. 146(2): 279-290.
SUMMERS, W. C., J. J. MCMAHON, AND G. N. P. A. RUPPERT. 1974. Studies on the maintenance of adult
squid (Loligo pealei). II. Empirical extensions. Biol. Bull. 146: 291-301.
TARDENT, P. 1962. Keeping Loligo vulgaris L. in the Naples Aquarium. 1st International Congress on
Aquarology. Bull. Inst. Oceanographique, Monaco, Special no. 1A: 41-46.
TASAKJ, I. 1982. Physiology and Electrochemistry of Nerve Fibers. Academic Press, New York. 343 pp.
VINOGRADOV, V. E., AND A. S. NOSKOV. 1979. Feeding of the short-finned squid, lllex illecebrosus, and
long-finned squid, Loligo pealei, off Nova Scotia and New England, 1974-75. ICNAF Sel. Papers
5: 31-36.
SQUID MAINTENANCE, GROWTH, AND BEHAVIOR 685
VOVK, A. N. 1974. Feeding habits of the North American squid Loligo pealei Les. Fisheries and Marine
Service, Translation Series No. 3304. 14 pp.
WALLER, R. A., AND R. I. WICKLUND. 1968. Observations from a research submersible — mating and
spawning of the squid, Doryteuthis plei. Bioscience 18(2): 110-111.
WHITAKER, J. D. 1978. A Contribution to the Biology of Loligo pealei and Loligo plei (Cephalopoda,
Myopsida) off the Southeastern Coast of the United Slates. Master's thesis. The College of Charleston,
Charleston, North Carolina.
WILLIAMS, L. W. 1909. The Anatomy of the Common Squid. The American Museum of Natural History,
New York. 9 1 pp.
WINBERG, G. C. 1960. Rate of Metabolism and Food Requirements of Fishes. Fish. Res. Board Can.,
Trans. Ser. No. 194, 202 pp.
YANG, W. T., R. T. HANLON, M. E. KREJCI, R. F. HIXON, AND W. H. HULET. 1983. Laboratory rearing
of Loligo opalescens, the market squid of California. Aquaculture 31: 77-88.
YANG, W. T., R. F. HIXON, P. E. TURK, M. E. KREJCI, R. T. HANLON, AND W. H. HULET. 1983. Culture
of California market squid from hatching — completion of the rearing cycle to second generation
hatchlings. Aquabiology. 5(5): 328-339. (In Japanese)
Reference: Bio/. Bull. 165: 686-698. (December, 1983)
THE LATITUDINAL COMPENSATION HYPOTHESIS: GROWTH DATA
AND A MODEL OF LATITUDINAL GROWTH DIFFERENTIATION
BASED UPON ENERGY BUDGETS. I. INTERSPECIFIC COMPARISON
OF OPHRYOTROCHA (POLYCHAETA: DORVILLEIDAE)
JEFFREY S. LEVINTON
Department of Ecology and Evolution, State University of New York, Stony Brook. New York 11794
ABSTRACT
A northern (North Carolina) sibling species of Ophryotrocha grew more rapidly
than a southern sibling species (Florida); this presumed advantage, however, diminished
to zero as temperature increased from 15 to 30°C. Survival of the northern sibling
species was low at 30°C. The differential response probably had a genetic basis since
both species had been reared for 2-3 generations under the same conditions. The
effect lasted in laboratory populations reared for a year in the laboratory at 25 °C (ca.
10 generations).
My results are consistent with a graphical model that suggests an evolutionary
shift of metabolism-temperature curves and feeding efficiency curves for the two
sibling species. These shifts predict a changing advantage of growth of one species
relative to the other as temperature increases.
INTRODUCTION
Many studies have demonstrated physiological differences among latitudinally
separated or otherwise thermally disparate populations of the same species or among
closely related species. Differentiation has been recorded for metabolic rate (Mangum,
1963), temperature tolerance (Zhirmunsky, 1959), egg development time (McLaren
el ai, 1969) and spawning temperature (Loosanoff and Nomejko, 195 1 ). Krogh (1916)
predicted that such differences should be consistent with a compensatory adaptation
to maximize growth rates in a given temperature regime. Animals living in low
temperature (high latitude) locales would thus be expected to "compensate" by in-
creasing metabolic and growth rates at a given temperature, relative to animals from
high temperature (low latitude) locales. This difference would be analogous to the
seasonal adjustment of Q10 found in many marine invertebrate species. Winter-ac-
climated animals can sustain more activity than summer animals maintained at the
same low temperature (see Kinne, 1964; Newell, 1973 for literature summaries and
general discussion).
An appropriate physiological compensation for latitudinal position occurs for
many, but not all, examined species. Scholander et al. (1953) found compensatory
metabolism-temperature (M-T) adaptation in a comparison of arctic and tropical
poikilotherms. Compensatory differences occur for latitudinally separated populations
of a single species. Heart-beat rate, water propulsion speed, somatic growth rate, and
oxygen consumption rate differ among populations of the mussel Mytilus californianus
on the west coast of North America (e.g., Rao, 1953; Dehnel, 1956; Pickens, 1965).
In cases of compensation, high latitude populations show an upwards displacement
Received 27 May 1983; accepted 29 August 1983.
686
GROWTH AND LATITUDE 687
of the M-T curves relative to low latitude populations. This form of compensation
is not universally observed, however (e.g., Fox, 1936; Vernberg and Vernberg, 1966).
There are two generalizations from studies on latitudinal variation in growth rates.
Individuals of high latitude populations of poikilotherms often obtain larger maximum
body size than conspecifics or closely related species living at low latitudes (e.g.,
Weymouth and McMillan, 1931; Ray, 1960). Secondly, although cold temperatures
often reduce activity and constrain individuals to grow more slowly, they compensate
by accelerating growth rate or larval development rate, relative to low latitude-derived
individuals, when both are reared at the same temperature (Schneider, 1967; Ament,
1979; Bervan et ai, 1979).
Although adaptation to low temperature would probably entail a form of com-
pensation involving relative acceleration of growth of the high latitude form at low
temperature, one might expect that this shift in metabolism would result in an increased
cost at higher temperature, leaving these forms at an energetic disadvantage in higher
temperature environments. In other words, "latitudinal compensation" may not be
compensation at all. Rather, local populations may shift their metabolic properties
to maximize growth under local temperature conditions. A manifestation of this shift
is a presumed acceleration of forms living in low temperatures, relative to high tem-
perature forms reared at the same temperature. This shift would be either in the form
of local evolution, or non-genetic response such as acclimation.
It is the purpose of this paper to present evidence for genetically based differences
in somatic growth rates among latitudinally separated sibling species of Ophryotrocha
(Polychaeta; Dorvilleidae). A companion paper will demonstrate differences between
subspecies. I will propose a model based upon energy budgets to explain latitudinal
clines in growth rate and body size in marine poikilotherms. The model assumes
that all populations evolve to maximize growth rate; observations of acceleration are
merely a manifestation of this selection pressure.
Evidence for genetically-based physiological compensation
Latitudinal differences in allozyme variants occur in a wide variety of invertebrates
and fishes (e.g., O'Gower and Nicol, 1968; Johnson and Utter, 1973; Williams et ai,
1973; Koehn et al, 1976). Although this clinal variation is obviously correlated with
temperature change, it is not clear that the genetic differences account for the phys-
iological differences observed in the studies cited above. In the mussel, Mytilus edulis,
extensive latitudinal differentiation in allele frequencies occurs on the east coast of
North America. By contrast, little differentiation is found in M. californianus along
the west coast. This correlates well with the steep latitudinal thermal gradient on the
east coast as opposed to the gentle gradient on the west coast (Levinton and Suchanek,
1978). Heat-stable variants of the enzyme phosphoglucomutase are more common
in more southern relative to northern populations of the east coast ribbed mussel
Geukensia demissus (Gosling, 1979). Thus physiological differentiation may be ge-
netically based. Adaptation at the molecular level has been shown between species
living in different environments (e.g., Somero and DeVries, 1967; Hochachka and
Somero, 1973; and references therein) and some evidence exists for adaptive enzyme
variants within a marine invertebrate species (Burton and Feldman, 1983; Hall, 1983;
Koehn et al., 1980).
Latitudinal differences in whole-animal physiological parameters may or may not
have a genetic basis. There are three possible components of physiological response
(Kinne, 1962). (1) Labile Compensation: individuals differ physiologically due to
local acclimatization. These differences are reversible after a period of laboratory
688 J. S. LEVINTON
acclimation (e.g., Pickens, 1965). (2) Irreversible Non-genetic Compensation: field
conditions induce irreversible physiological changes that cannot be eliminated through
laboratory acclimation (e.g., Gibson, 1954; Zamer and Mangum, 1979). These dif-
ferences are acquired independently of genotype; irreversible effects are fixed because
of the environment only. An obvious example is temperature-induced sex in fishes
(e.g., Conover and Kynard, 1981). (3) Genetically-based Physiological Differences:
this is difficult to distinguish in the field from irreversible non-genetic compensation.
An irreversible physiological response could result from irreversible non-genetic, or
genetically-based responses (e.g., Levinton and Lassen, 1978). Unfortunately, it is
not possible to distinguish between these two alternatives in most studies demonstrating
differences among field-collected adults. A useful approach would be the examination
of progeny of populations that have been reared in the laboratory under identical
conditions, preferably for several generations (e.g., Battaglia, 1957; Schneider, 1967;
Ament, 1979). The examination of progeny of laboratory-reared stocks, however,
may involve unrealistic laboratory conditions and ignoring field parameters that might
affect gene expression. Despite these potential problems, I take the latter approach
in this study.
MATERIALS AND METHODS
Three populations of two sibling species ofOphryotrocha (Polychaeta; Dorvilleidae)
were collected. Ophryotrocha species are found commonly in microhabitats of fouling
communities, and are readily cultured in the laboratory (e.g., see Akesson, 1976,
1978; Sella, 1978). O. costlowi Akesson was collected from Morehead City, North
Carolina, and O. macrovifera Akesson was collected from Tampa Bay, Florida and
at St. Lucie's Inlet, Florida. Identifications were confirmed via crossing tests performed
by Prof. Bertil Akesson, Goteborg University. Initial populations of approximately
fifty individuals were reared on ground, par-boiled spinach, at room temperature
(20 °C) and 30%o salinity. Sea water used for culture was twice glass-fiber filtered,
sterilized for 24 hours at 80°C, and cooled to room temperature. Worms were cultured
in glass evaporating dishes, whose curved sides permit easy examination of worms
and egg cases under the dissecting microscope at 12-50X magnification.
Both species belong to the "labronica" group of Ophryotrocha (Akesson, 1978).
Sexes are separate and the male follows the female for some time (hours to over a
day) before sperm is transferred. Females construct a tubular egg mass and reside in
the tube until juveniles develop directly and emerge from the egg case. The near
simultaneous emergence permits experiments with large numbers of siblings of identical
age to be initiated at the same time.
Somatic growth rates were measured in two ways. First, animals were collected
in the field and held at ca. 20°C (approximately 2 degrees variation) for 2-3 generations.
Approximately twenty newly emerged juveniles from the same family were placed
in a dish with spinach. Individuals were transferred to constant temperature envi-
ronmental chambers held at 15, 20, 25, or 30°C (variation was less than 0.5°C).
Reciprocal transplants from different pairs of temperatures demonstrated no significant
effect of starting the experiment from conditioned populations at 20°C. Therefore,
it is unlikely that laboratory acclimation to 20°C significantly influenced the results
of the growth experiments. Each day, five to ten randomly selected individuals were
isolated, and the number of setigerous segments was counted. They were then returned
to the bowl. I found no difference in worm length versus number of setigerous segments
for the three populations (Fig. 1). I therefore assume that my measure is an homo-
geneous indicator of growth over all populations. From these data, I determined a
GROWTH AND LATITUDE
689
20-
Q
O
Q.
<
<r
<
Q.
U-
O
a:
UJ
m
15-
10-
5-
OO D
O 0°0
• O
O
Qi
••O
•
•O
•O
O
O
O
con
•D
(ED
• Moreheod City
O st Lucie
Q Tampa
20
40 60
BODY LENGTH
80
100
120
FIGURE 1. Body length (mm) versus number of setigerous segments for the two Florida populations
of O. macrovifera (Tampa and St. Lucie Inlet) and the North Carolina population of O. costlowi.
somatic growth rate by regression analysis of setigers on time. Difference in growth
rates within and between populations was evaluated using standard analyses of variance
related to regression (Sokal and Rohlf, 1981).
Second, newly emerged juveniles were placed separately (and arranged randomly)
in 1 ml wells of glass spot plates, and provided with sea water and spinach. After
seven days, the number of setigerous segments was counted for each individual, which
permitted comparisons between populations via nested analysis of variance. This
experiment was performed at 25 °C only. These experiments were designed such that
several families from each sibling species were employed; this permitted an estimate
of variation in growth rate among families, that is, a full-sib analysis. All experiments
were done after the field-collected populations had been maintained in the laboratory
at room temperature (usually ca. 20°C) for one year (ca. 10 generations), in order
to eliminate physiological characteristics that may have been fixed in the field. For
this comparison, I employed North Carolina O. costlowi and St. Lucie's Inlet, Florida
O. macrovifera.
Egg diameter was measured with an ocular micrometer fitted to a Wild dissecting
microscope (at 50X). I also recorded the time and number of setigers corresponding
to the acquisition of adult jaws. Finally, the number of eggs per case was counted.
Life history characteristics of two populations of O. macrovifera were compared.
Although geographically separated, these two populations were completely interfertile.
An energy budget model of growth rate
A simple model based upon energy budgets may be used to predict genetic dif-
ferences between North Carolina and Florida sibling species. Imagine the presence
of cold- and warm-adapted genotypes. Figure 2 shows a hypothetical difference between
690
J. S. LEVINTON
I
M
cool
warm
\
TEMPERATURE
FIGURE 2. Graphical model explaining differential adaptation among latitudinally separated populations.
Top diagram: Rate of ingestion, I, and metabolic cost rate, M, is illustrated for two hypothetical populations
living in different thermal regimes. The regimes have similar maximum but differing minimum temperatures.
Lower diagram: Difference between I and M curves yields reserves available for growth, G. The maximum
growth rate of the warm-adapted form is displaced, relative to the cold-adapted form, towards higher
temperature.
the two genotypes in energy acquired versus metabolic expenditure as a function of
increasing temperature. The two hypothetical genotypes differ in that the energy
intake and metabolic expenditure curves are displaced from each other such that the
cold-adapted genotype enjoys a growth advantage at lower temperature (Fig. 2b).
This model predicts that at lower temperature, the cold-adapted form should grow
more rapidly than the warm-adapted form. As temperature increases, this difference
should diminish to a zero point, beyond which the warm-adapted genotype should
enjoy the advantage. This advantage may simply involve relatively rapid growth. If
the warm-adapted genotype lives in temperatures never experienced by the cold-
adapted form, then the cold-evolved form might die at higher temperature, due to
an excess of metabolic cost relative to rate of gained energy. Thus, the differential
growth among individuals adapted to different temperatures would be seen only in
the lower part of the temperature scale.
RESULTS
Temperature at the sites
Seasonal differences in temperature at the three sampling localities differ more
in the distribution of temperature and winter minima than in summer maxima (Fig.
GROWTH AND LATITUDE
691
30"
l±J
20"
rr
UJ
CL
^
LU
10"
0
FS
M
M
0 N D
MONTH
FIGURE 3. Mean monthly temperature of surface waters of the three localities: NC = Morehead
City, North Carolina (actually nearby Beaufort, North Carolina); FSL = St. Lucie Inlet of Indian River,
Florida; FT = Tampa Bay, Florida. Data derived from NOAA records.
3). Neither Florida locales experience mean monthly winter temperatures lower than
15°C. The North Carolina locale temperature surpasses 25°C for only two months,
while both Florida sites are above 25 °C for 5-6 months.
Whole family analyses
Figure 4 demonstrates the nature of the data obtained for somatic growth estimated
for a given family (the complete data set is available upon request from the author).
Growth rate was relatively uniform at 20, 25, and 30°C, but quite variable among
individuals at 15°C.
Table I summarizes the variance analysis of the family growth regression on time,
when comparing combined Florida locales with the North Carolina locale. At 15,
20, and 25 °C, somatic growth rate of North Carolina animals surpasses that of Florida
animals (P < .001). At 30°C, growth rates are approximately equal. The ratio of
growth rate of North Carolina: Florida animals diminishes progressively from 15°C
(1.58) to 30°C (1.02). The absolute difference in growth rate, however, is greatest at
the two intermediate temperatures.
Significant differences between Florida populations were found at 15°C (P
< .001) and 25°C (P < .05) (Table II). The magnitude of difference, however, is
great only at 15°C, as demonstrated by the ratio of somatic growth rates at the four
temperatures. At 15°C, growth differed between the two populations by a surprising
factor of ca. 2. The variance analysis (Table III) also demonstrates that, for a given
site and temperature, among-family growth rates were significantly different within
sample populations from all three source localities. A more appropriate experimental
design would spread members of the same family among different temperature treat-
ments to estimate the family variance component. This approach is taken in the
companion paper on intraspecific latitudinal differences (Levinton and Monahan,
1983).
692
J. S. LEVINTON
LU
S
CD
UJ
CO
CO
ID
o
rr
UJ
o
ui
CO
10 -
5-
15"
10
5-
o ooooooo o o ooc
ooo oooq_^D — • — """So ooo ooooo
~oo oo o oo o
20
10
20
30 0
10
20
— i
30
10 -
25
10-
10
20
30 0
I I I
10
20
30
TIME (DAYS)
FIGURE 4. Examples of growth regressions of families of O. costlowi run at 15, 20, 25, and 30°C.
Survivorship of families
Eleven families each from North Carolina and the combined Florida locales were
run at 30°C. The survival (defined as greater than 20 percent survival of individuals)
of families from the combined Florida locales was much greater (10 out of 1 1) than
that of the North Carolina site ( 1 out of 1 1 ). This difference in survival is significant
= .01; 2 X 2 contingency table; Fisher's Exact Test). At lower temperatures survival
was very high and similar between the two areas. The high mortality at 30°C in the
North Carolina samples was surprising, in light of the rapid growth observed up to
the day that death was observed, usually near the time of sexual maturity. Referring
GROWTH AND LATITUDE
693
TABLE I
Growth differences and variance analysis of regression statistics relating setigers to time
Growth rate:
Experimental
Degrees
North
temperature
Fs
of freedom
Florida
Carolina
P
R
15
12.85
917
0.041
0.065
<.001
1.58
20
154.37
1732
0.364
0.520
<.001
1.43
25
41.99
962
0.853
0.956
<.001
1.12
30
1.58
828
1.215
1.243
NS
1.02
NS = not significant.
F statistic measures among-family versus between-locality variance of growth rate. Florida localities
(O. macrovifera) are pooled for comparison with North Carolina (O. costlowi) locality. R = ratio of North
Carolina to Florida growth rate (in setigerous segments/day).
to Figure 3, it is apparent that 30°C is greater than the mean monthly temperature
experienced by all three populations. The data, however, probably underestimate the
temperature achieved in the shallow microhabitats occupied by the worms. The Tampa
site experiences the greatest summer maximum mean temperature.
Growth of individuals
Growth of North Carolina O. costlowi was found to be greater than that of Florida
(St. Lucie) O. macrovifera (Table IV). Additionally, growth rate differed significantly
among families. It is not known whether this difference can be attributed to genetic
differences or to maternal effects (now under investigation). In both source populations,
among-family differences were strongly significant (P < .001). These data indicate
that among-family differences must be accounted for in physiological experimentation.
Life-history differences
Significant differences were found between the two Florida populations of O.
macrovifera, despite complete interfertility between adults in crosses. Both egg size
and number of setigerous segments at release (Table V) were larger in individuals
from the Tampa population, relative to the St. Lucie population. Both eggs and larvae
TABLE II
Comparison of growth rates ofO. macrovifera populations from Tampa (Gulf Coast, Florida)
and St. Lucie Inlet (Atlantic Coast, Florida)
Growth rate:
Experimental
temperature
F5
DF
Tampa
St. Lucie
P
R
15
4.87
527
0.028
0.061
<.001
2.18
20
0.54
866
0.363
0.372
NS
1.02
25
6.31
702
0.842
0.799
<.05
0.95
30
0.19
538
1.149
1.136
NS
0.99
NS = Not Significant.
F statistic is a measure of among-family versus between locality variance. R = St. Lucie/Tampa
growth rate.
694 J. S. LEVINTON
TABLE III
Variance among families within the populations from each locality, at four temperatures
15°C
20°C
25°C
30°C
Temperature
population
N
F
P
N
F
P
N
F
P
N
F
P
OCNC
5
5.61
<.001
11
22.32
<.001
5
6.68
<.001
7
17.86
<.001
OMT
5
32.76
<.001
5
16.76
<.001
5
9.56
<.001
5
8.71
<.001
OMSL
5
5.19
<.001
5
3.60
<.01
5
21.85
<.001
6
19.41
<.001
/"statistic measures difference in variance within and among families from a given locality. (N = number
of families; F = value of F statistic, P = significance level for among-family heterogeneity in somatic growth
rate (setigerous segments/day)). OCNC = O. costlowi. North Carolina; OMT = O. macrovifera, Tampa;
OMSL = O. macrovifera, St. Lucie Inlet.
of North Carolina O. costlowi were smaller than the eggs and larvae of Florida pop-
ulations of O. macrovifera. The size at which the adult jaw developed was greater in
O. macrovifera (Table V). Akesson (1978) presents similar data, except for noting
intraspecific variation in O. macrovifera.
DISCUSSION
These results are consistent with an energy budget model that postulates a difference
in adaptation of high and low latitude (i.e., thermally differing) populations. The
northern O. costlowi shows greater somatic growth rate, but this advantage decreases
with increasing temperature as predicted by the model. At 30°C, the difference is
nonexistent and North Carolina families show very high mortality. The rich food
(spinach) used in the experiments may have tended to shift the growth differences,
favoring growth of the northern populations at higher temperatures than found under
field conditions with a food supply of lower nutritional content. Such an effect was
documented by Bayne et al. (1973) in the energy budget of the mussel, Mytilus edulis.
The rich food provided in the present experiments might explain the rapid growth
observed at 30°C for all three populations. A poorer food that is available in nature
might not permit a favorable energetic balance at this extreme temperature.
As illustrated by Figure 3, the thermal regimes at the three localities do not differ
TABLE IV
Analysis of variance for growth rates of individuals distributed among nine families each for O. costlowi
from Morehead City, North Carolina and O. macrovifera from St. Lucie 's Inlet, Florida
O. costlowi, North Carolina: Total growth = 7.48 + .37 (95% CL)
Source of Variation df MS Fs
Among Families 8 15.08 9.40***
Within Families 72 1.60
O. macrovifera. Florida: Total growth = 5.52 + .23 (95% CL)
Source of Variation
df
MS
Fs
Among Families
8
4.42
5.79***
Within Families
72
0.76
* P < .001.
Growth is measured as number of setigerous segments added in seven days after hatching.
GROWTH AND LATITUDE 695
TABLE V
Some differences in life histories between O. costlowi for Morehead City, North Carolina, and O.
macrovifera from St. Lucie Inlet (Atlantic Coast of Florida) and Tampa (Gulf Coast of Florida)
O. macrovifera
Characteristic
O costlowi
Tampa
St. Lucie
Egg Diameter (^m)
104.4 + 9.85
(225)
134.2 + 17.42
(128)
145.5 ± 16.84
(80)
Setigerous Segments When
Hatching
0 ± 0
(100)
1.5 ± 0.74
(170)
2.6 ± 0.68
Acquisition of Adult Jaws
(Males)
11.4 ± 0.91
(39)
14.8 ± 0.75
(12)
—
Acquisition of Adult Jaws
(Females)
13.5 ± 0.93
(44)
16.1 ± 0.64
(18)
—
Sample size (in parentheses) and standard deviations are given.
very much in maximum summer temperature. The most important difference lies
in the seasonal distribution of temperature and the winter minimum. The Florida
locales have winter minima near 15°C, which is substantially greater than for North
Carolina. Increasing the experimental temperature from 15 to 30°C provides an
opportunity for increasing growth rate (Fig. 3). In the lower range, the high latitude
population enjoys the advantage in growth efficiency. At 30°C, however, the two
sibling species do not differ in growth, although North Carolina O. costlowi suffer
high mortality. This may stem from an inability to acclimate and a predetermined
pattern of investment of energy for somatic growth, despite the cost in maintenance.
At present, there are no data on reproductive output as a function of temperature.
It is likely that reproductive investment will follow the patterns found for somatic
growth. Akesson (1976) investigated the effect of temperature on the life cycle of O.
labronica and demonstrated optimum intermediate temperatures for eggs per egg
mass and egg output per female per day. In the sequential hermaphrodite, Ophryotrocha
puerilis, the size at which sex change from male to female occurs is greater in a
northern (Atlantic) subspecies, relative to a southern (Mediterranean) subspecies (Sella,
1978). Sella (1980) has presented evidence that the size at sex change in O. puerilis
is genetically regulated and maintained by stabilizing selection. This suggests that
thermal limitations may influence the life history patterns of sequential hermaphrodites
much as spatially varying mortality patterns can (e.g., Charnov, 1978, 1979, 1981).
Although the northern subspecies of O. puerilis switches sex at a larger number of
segments, individuals grow to this size in the same number of days as individuals of
the southern subspecies, implying accelerated growth for the northern form. This
result is consistent with the present findings. Further work on these subspecies dem-
onstrates intraspecific differentiation in growth rates comparable to the interspecific
data presented here (Levinton and Monahan, 1983).
These results suggest that, given the strong differences in temperature along the
eastern coast of North America, some compensation is possible for life at low tem-
perature. The model and results also suggest that populations evolve locally to maximize
growth rate. It is therefore incorrect to state, for example, that southern (high tem-
perature) populations evolve slower growth rates, relative to northern populations.
Rather, all populations are adapted for maximum growth rate, and they sacrifice
696 J. S. LEVINTON
efficiency at temperatures rarely experienced to maximize growth efficiency at tem-
peratures that are experienced commonly. Thus southern individuals grow more
slowly at lower temperature because evolution has shifted the metabolism-temperature
relationship to minimize metabolic cost at high temperature; this shift, however,
restricts activity and growth at low temperature, relative to higher latitude forms.
Bervan et al. (1978) developed a temperature-related explanation for growth differences
in salamanders living at different altitudes. They described the necessary compensation
of living in cold climates as "countergradient selection," implying that selection for
increased growth rate works against the limiting effects of cold temperatures on poiki-
lotherms. But animals in warm climates have their problems as well; genetic variance
for a favorable metabolism-temperature response would therefore be of great advantage,
given a geographic/altitudinal gradient of temperature.
Newell and Kofoed (1977) demonstrate that thermal constraints can be met with
compensation via physiological acclimation. Thus the presumed genetic component
we observe here must be placed aside this non-genetic response in any accounting
of response to the thermal regime. The ability to acclimate is just as much an evo-
lutionary response as the differences in growth we have discussed.
Growth experiments performed in this study were done on individuals from
different populations, reared under the same laboratory conditions. Assuming lab-
oratory conditions are reasonably related to field conditions, these results suggest that
the growth differences are genetic, and confirm the expectation that populations of
coastal invertebrates show strong regional differentiation in physiologically important
characters (Battaglia, 1959; Gooch and Schopf, 1971; Levinton and Fundiller, 1975).
These results suggest that an integrated study of energy budgets, genetics, and somatic
growth rates will be useful in understanding regional differentiation within marine
species. As the differences found in this study lasted (apparently) for ca. 1 0 generations
of laboratory rearing (as in Battaglia, 1959), one can be reasonably certain that en-
vironmental effects can be ruled out.
ACKNOWLEDGMENTS
The initiation of this project is due to the kind provision of facilities and support
by Prof. John D. Costlow, Duke University Marine Laboratories. I am also grateful
to Prof. Bertil Akesson of Gothenburg University, Sweden, who performed necessary
identifications and provided much-needed advice. During the course of this work I
was lucky to have the assistance in the laboratory of Susan Hutchison, Reed Johnson,
and Lois Mingalone. Linda Weinland provided the specimens of Ophryotrocha ma-
crovifera. Rosemary Monahan edited and criticized the manuscript. This work was
partially supported by U. S. National Science Foundation grants OCE-78-09057 and
OCE-80- 18743 (to J.S.L.), and by a grant from the Department of Energy (Contract
DE-AS05-76EV04377) awarded to Duke University Marine Laboratory under the
direction of J. D. Costlow. Contribution number 474 from the Graduate Studies in
Ecology and Evolution, State University of New York at Stony Brook.
LITERATURE CITED
AKESSON, B. 1976. Temperature and life cycle in Ophryotrocha labronica (Polychaeta, Dorvilleidae). Ophelia
15: 37-47.
AKESSON, B. 1978. A new Ophryotrocha species of the Labronica group revealed in crossbreeding experiments.
Pp. 573-590 in Marine Organisms: Genetics, Ecology, and Evolution, B. Battaglia and J. L.
Beardmore, eds. Plenum Publ. Co., New York.
AMENT, A. 1979. Geographic variation in relation to life history in three species of the marine gastropod
GROWTH AND LATITUDE 697
genus Crepidula: growth rates of newly hatched larvae and juveniles. Pp. 61-76 in Reproductive
Ecology of Marine Invertebrates, S. E. Stancyk, ed. Univ. of South Carolina Press, Columbia, SC.
BATTAGLIA, B. 1959. Ecological differentiation and incipient intraspecific isolation in marine copepods.
Ann. Biol. 33: 259-268.
BAYNE, B. L., R. J. THOMPSON, AND J. WIDDOWS. 1973. Some effects of temperature and food on the
rate of oxygen consumption by Mytilus edulis L. Pp. 181-193 in Effects of Temperature on
Ectothermic Organisms, W. Wieser, ed. Springer- Verlag, New York.
BERVAN, K. A., D. E. GILL, AND S. J. SMITH-GILL. 1979. Countergradient selection in the green frog,
Rana clamitans. Evolution 33: 609-623.
BURTON, R. S., ANDM. W. FELDMAN. 1983. Physiological effects of an allozyme polymorphism: glutamate-
pyruvate transaminase and response to hyperosmotic stress in the copepod Tigriopus californicus.
Biochem. Genet. 21: 239-251.
CHARNOV, E. L. 1978. Sex ratio: adaptive response to population fluctuations in pandalid shrimp. Science
200: 204-206.
CHARNOV, E. L. 1979. Natural selection and sex change in pandalid shrimp: test of a life-history theory.
Am. Nat. 113: 715-734.
CHARNOV, E. L. 1981. Sex reversal in Pandalus borealis: effect of a shrimp fishery. Mar. Biol. Lett. 2: 53-
57.
CONOVER, D. O., AND B. E. K.YNARD. 1981. Evolution of sex determination: interaction of temperature
and genotype in a fish. Science 213: 577-579.
DEHNEL, P. A. 1 956. Growth rates in latitudinally and vertically separated populations of Mytilus californianus.
Biol. Bull. 110: 43-53.
Fox, H. M. 1936. The activity and metabolism of poikilothermal animals in different latitudes. Proc. Zool.
Soc. Land. 106: 945-955.
GIBSON, M. B. 1954. Upper lethal temperature relations of the guppy, Lebistes reticulatus. Can. J. Zool.
32: 393-407.
GOSLING, E. 1979. Hidden genetic variability in two populations of a marine mussel. Nature 279: 713-
715.
GOOCH, J. L., AND T. J. M. SCHOPF. 1971. Genetic variation in the marine ectoproct Schizoporella errata.
Biol. Bull. 141: 235-246.
HALL, J. G. 1983. Temperature-related Catalytic Differentiation Among Homologues of Glucosephosphate
Isomerase (EC 5.3.1.9) Isolated From the Bivalves Mytilus edulis and Isognomon alatus. PhD.
Dissertation, State University of New York at Stony Brook.
HOCHACHKA, P. W., AND G. N. SOMERO. 1973. Strategies of Biochemical Adaptation. W. B. Saunders,
Philadelphia.
JOHNSON, A. G., AND F. M. UTTER. 1973. Electrophoretic variants of aspartate aminotransferase of the
bay mussel, Mytilus edulis (Linnaeus, 1758). Comp. Biochem. Physiol. 44B: 317-323.
KJNNE, O. 1962. Irreversible non-genetic adaptation. Comp. Biochem. Physiol. 5: 265-282.
K.INNE, O. 1964. Non-genetic adaptation to temperature and salinity. Helgol. Wiss. Meeresunters. 10: 433-
458.
KOEHN, R. K., R. MILKMAN, AND J. B. MITTON. 1976. Population genetics of marine pelecypods. IV.
Selection, migration and genetic differentiation in the blue mussel, Mvtilus edulis. Evolution 30:
2-32.
KOEHN, R. K., R. I. E. NEWELL, AND F. IMMERMAN. 1980. Maintenance of an aminopeptidase allele
frequency cline by natural selection. Proc. Nat. Acad. Sci. USA 9: 5385-5389.
KROGH, A. 1916. Respiratory Exchange of Animals and Man. Longmans Green, London.
LEVINTON, J. S., AND D. L. FUNDILLER. 1975. An ecological and physiological approach to the study of
biochemical polymorphisms. Pp. 165-178 in Proc. 9th Europ. Mar. Biol. Symp.. H. Barnes, ed.
Aberdeen University Press, Aberdeen, Scotland.
LEVINTON, J. S., ANDH. H. LASSEN. 1978. Selection, ecology, and evolutionary adjustment within bivalve
mollusc populations. Phil. Trans. Roy. Soc. Land. B 284: 403-415.
LEVINTON, J. S., AND R. K. MONAHAN. 1983. The latitudinal compensation hypothesis: growth data and
a model based upon energy budgets. II. Intraspecific comparisons between subspecies of Ophry-
otrocha puerilis (Polychaeta, Dorvilleidae). Biol. Bull. 165: 699-707.
LEVINTON, J. S., ANDT. H. SUCHANEK. 1978. Geographic variation, niche breadth, and genetic differentiation
at different geographic scales in the mussels Mvtilus californianus and Mytilus edulis. Mar. Biol.
49: 363-375.
LOOSANOFF, V. L., AND C. A. NOMEJKO. 1951. Existence of physiologically different races of oysters,
Crassostrea virginica. Biol. Bull. 101: 151-156.
MANGUM, C. P. 1963. Studies on speciation in maldanid polychaetes of the North American Atlantic
coast. III. Intra- and interspecific divergence in oxygen consumption. Comp. Biochem. Physiol.
10: 335-349.
698 J. S. LEVINTON
MCLAREN, I. A., C. J. CORK.ETT, AND E. J. ZILLIOUX. 1969. Temperature adaptations of copepod eggs
from the arctic to the tropics. Biol. Bull. 137: 486-493.
NEWELL, R. C. 1973. Environmental factors affecting the acclimatory responses of ectotherms. Pp. 151-
164 in Effects of Temperature on Ectothermic Organisms, W. Wieser, ed. Springer- Verlag, New
York.
NEWELL, R. C., ANDL. H. KOFOED. 1977. Adjustment of the components of energy balance in the gastropod
Crepidula fornicata in response to thermal acclimation. Mar. Biol. 44: 275-286.
O'GOWER, A. K., AND P. I. NICOL. 1968. A latitudinal cline of haemoglobins in a bivalve mollusc. Hereditv
23:485-491.
PICK.ENS, P. E. 1965. Heart rate of mussels as a function of latitude, intertidal height, and acclimation
temperature. Physiol. Zool. 38: 390-405.
RAO, K.. P. 1953. Rate of water propulsion in Mvtilus californianus as a function of latitude. Biol. Bull.
104: 171-181.
RAY, C. 1960. The application of Bergmann's and Allen's rules to the poikilotherms. J. Morphol. 106: 85-
108.
SCHNEIDER, D. E. 1967. An Evaluation of Temperature Adaptations in Lalitudinally Separated Populations
ofthe.XanthidCrab, Rhithropanopeus harrisii (Gould), by Laboratory Rearing Experiments. PhD.
Dissertation, Duke University.
SCHOLANDER, P. F., W. FLAGG, V. WALTERS, AND L. IRVING. 1953. Climatic adaptation in arctic and
tropical poikilotherms. Physiol. Zool. 26: 67-93.
SELLA, G. 1978. Responses adaptatives a differentes temperatures chez les deux sousespeces d'Ophryotrocha
puerilis, pendant la phase male. Cah. Biol. Mar. 18: 1 17-125.
SELLA, G. 1980. The genetic system controlling the body length at which sex change occurs in Ophryotrocha
puerilis (Annelida, Polichaeta). Atti Soc. Toscana Sci. Nat. P. V. Mem. Ser. B 87: 197-203.
SOKAL, R. R., AND F. J. ROHLF. 1981. Biometry. W. H. Freeman and Company, San Francisco.
SOMERO, G. N., AND A. L. DEVRiES. 1967. Temperature tolerance of some Antarctic fishes. Science 156:
257-258.
VERNBERG, W. B., AND F. J. VERNBERG. 1966. Studies on the physiological variation between tropical
and temperate zone fiddler crabs of the genus Uca V. Effect of temperature on tissue respiration.
Com;?. Biochem. Physiol. 17: 118-126.
WEYMOUTH, F. W., AND H. C. MCMILLAN. 1931. The relative growth and mortality of the Pacific razor
clam (Siliqua panda Dixon) and their bearing on the commercial fishery. Bull. U. S. Bur. Fish.
46: 543-567.
WILLIAMS, G. C., R. K. KOEHN, AND J. B. MITTON. 1973. Genetic differentiation without isolation in the
American eel, Anguilla rostrata. Evolution 27: 192-204.
ZAMER, W. E., AND C. P. MANGUM. 1979. Irreversible non-genetic temperature adaptation of oxygen
uptake in clones of the sea anemone Haliplanella luciae (Verrill). Biol. Bull. 157: 536-547.
ZHIRMUNSK.Y, A. V. 1959. The thermostability of actinias and their ciliated epithelium in natural conditions
and in experimentally changed environmental temperatures. Tsitologiya 1: 270-276 (In Russian).
Reference: Biol. Bull. 165: 699-707. (December, 1983)
THE LATITUDINAL COMPENSATION HYPOTHESIS: GROWTH DATA
AND A MODEL OF LATITUDINAL GROWTH DIFFERENTIATION
BASED UPON ENERGY BUDGETS. II. INTRASPECIFIC COMPARISONS
BETWEEN SUBSPECIES OF OPHRYOTROCHA PUERILIS
(POLYCHAETA: DORVILLEIDAE)
JEFFREY S. LEVINTON AND ROSEMARY K. MONAHAN
Department of Ecology and Evolution, State University of New York, Stonv Brook, New York 11794
ABSTRACT
Individuals of two subspecies of Ophryotrocha puerilis (Polychaeta; Dorvilleidae)
were collected from differing thermal regimes, and cultures were maintained for over
a year. Despite common rearing, the two subspecies show substantial differences in
somatic growth rate. At 15°C, the warm-water subspecies grew more slowly, while
at 20°C growth for the two subspecies was not significantly different. At 24°C, the
warm-water subspecies grew more rapidly and suffered substantially less mortality
than the northern subspecies. These results conform to a model predicting genetic
differentiation of metabolic efficiency, leading to differences in growth efficiency among
populations adapting to thermally differentiated habitats. The problems faced by the
cold-water subspecies at 24°C conforms to expectations based upon natural habitat
temperatures.
INTRODUCTION
Many broadly distributed coastal marine species live in a strong thermal gradient
and would be expected to be subjected to natural selection to maximize growth
efficiency in the local thermal regime. If populations are sufficiently isolated, this
may result in a series of genetically distinct subpopulations whose growth characteristics
would differ even if reared under constant temperature conditions. In the first paper
of this series, Levinton (1983) suggested that a simple model of metabolic expenditure
and food (energy) intake would predict divergent temperature optima for subpopu-
lations living under different thermal regimes. Differences among sibling species of
the polychaete genus Ophryotrocha conform to such a model.
Here we show similar growth differences between two geographically separated
subspecies. We compare somatic growth rates of Atlantic and Mediterranean subspecies
of O. puerilis and show that, despite common rearing through several generations
under identical conditions, strong differences in growth rate persist between the two
populations. The differences, moreover, show an advantage that shifts from favoring
the high latitude population at low temperature, to favoring the low latitude population
at higher temperature. These results suggest that the thermal regime generates strong
genetic differentiation along the latitudinal gradient. Our results provide more direct
evidence of intraspecific latitudinal differences in temperature adaptation than do
recorded differentiation in, for example, allozyme polymorphisms (e.g., Levinton and
Suchanek, 1978).
Received 27 May 1983; accepted 29 August 1983.
699
700 J- S. LEVINTON AND R. K. MONAHAN
The organism
Ophryotrocha puerilis is a dorvilleid polychaete commonly collected in barely
subtidal and intertidal fouling communities in European waters. The species is a
protandrous hermaphrodite (sex reversal — from male to female), and occurs as two
subspecies, O. p. puerilis from Mediterranean waters and O. p. siberti from the Atlantic
coast of Europe (e.g., Bacci and LaGreca 1953). The two subspecies are reproductively
isolated to a variable degree, depending upon the nature of the cross, but incompatibility
is generally extensive (Akesson, 1975; 1977). Body size (estimated by number of
setigerous segments) at time of sex change differs between the subspecies. Although
size at sex change depends somewhat on temperature, O. p. puerilis switches at
approximately 18 setigerous segments (setigers) while O. p. siberti changes at 20
setigers (Bacci and LaGreca, 1953; Sella, 1978). The body length at which sex change
occurs is determined by a polygenic system, and selection experiments can change
the size at reversal in only a few generations (Bacci and Bortesi, 1961; Sella, 1980).
We used two populations collected by Dr. Gabriella Sella of the University of
Torino. The O. p. siberti stock was collected in 1978 at the Roscoff Marine station
(Brittany, north coast of France), while the O. p. puerilis culture was collected in the
harbor of Genoa, Italy in 1981. At Roscoff, the annual temperature range is 8.9-
15°C: the range is 12.5-24.2°C in Genoa (Sella, 1978). All stocks were kept at room
temperature (ca. 20°C) prior to being shipped to our laboratory in late 1981. We
kept the stocks at ca. 20°C until the summer of 1982 when the experiments were
performed; both stocks therefore existed for quite a long time under similar conditions.
A newborn individual worm becomes a female within about three weeks at 20°C;
therefore both stocks went through a number of generations in the laboratory. We
doubt that any field conditioning such as local acclimation to temperature could have
exerted effects on laboratory stocks over such a long period of time.
MATERIALS AND METHODS
A number of mating pairs were established for each subspecies by randomly
selecting individuals (consubspecifics) and placing pairs in individual glass bowls
provided with 30%o sterilized sea water and ground spinach as food (see Akesson,
1970 for instructions on the culture of Ophryotrocha). All mating pairs of each sub-
species were kept in an incubator at 20°C, on a 12:12 light/dark cycle. Pairs were
then monitored for egg case production. Juveniles of both subspecies hatched out of
egg cases after approximately 1 1 days at 20°C.
Progeny of five pairs of O. p. siberti and of four pairs of O. p. puerilis were chosen
for use in the experiment. On the day that most of the juveniles left each loose jelly
egg mass, 48 from each family were isolated in individual glass bowls (30%o sea water,
spinach for food). These bowls were then placed inside plastic boxes (with distilled
water on the bottom to slow evaporation). Of these 48 sibs per family, twelve progeny
each were moved into incubators at 1 5, 20, 24, and 28°C. This design placed members
of the same family under different temperature conditions, allowing an estimate of
among-family difference effects that might persist despite transfer to different tem-
peratures. The design also helps minimize the contribution of among-family differences
in confounding an estimate of between-subspecies differences. If completely different
families are placed in each and all dishes, then the among-family variance in growth
rate cannot be distinguished from a "bowl" effect that happens to make a given
family grow faster or slower, due to individual laboratory conditions. This is a problem
with the sibling species data presented in Levinton (1983).
LATITUDE AND INTRASPECIFIC GROWTH
701
The size (number of setigers) of the progeny was then measured every seven days
for four time periods. All hatched with zero setigers. Within the first week, all worms
of both subspecies kept at 28°C died so this part of the experiment was abandoned.
Water was changed when the worms were 2 1 days past hatching; food was replenished
every 7 days if needed (spinach was always provided in excess).
RESULTS
Figure 1 summarizes the results for all families of O. p. siberti (northern subspecies)
and O. p. puerilis. A decided shift in somatic growth rate occurs from low to high
temperature. At 15°C, the northern subspecies grows most rapidly, while at 20°C
growth rate is fairly similar for the two subspecies. At 24°C, however, the northern
20-
Lul
15-
LU
CO
CO
=3 10
O
CL
LU
C£
I-
LL)
CO
A P-20°
' S-20°
P= 0. p. puerilis
3-0. p. siberti
P-24°
21
AGE (days)
- 15
P-15
S-24
28
FIG. 1
FIGURE 1. Summary of growth data (means ± 95% confidence) for individuals of all families of the
two subspecies.
702 J. S. LEVINTON AND R. K. MONAHAN
subspecies grows very slowly while the southern subspecies grows much more rapidly.
The growth of the southern subspecies is less at 24°C than at 20°C. This indicates
that even the southern subspecies is nearing its upper thermal limit at 24 °C.
Growth plots for individual families show the pattern of growth differences between
the two subspecies (Fig. 2). At 1 5°C the growth of families from the southern subspecies
is less than that of the northern subspecies, with some overlap (Nested ANOVA,
F = 6.88, P < .05). At 20°C the families from both subspecies overlap substantially
(F = 0.42, Difference not significant). At 24°C, however, growth of the northern
subspecies is clearly depressed relative to the southern subspecies (F = 59.69,
P < .001).
The northern subspecies thus displays a growth disadvantage at higher temperature.
This disadvantage is also reflected in a noticeable incidence of setiger resorption and
generally poor nutritive condition. Not surprisingly, mortality in the experiment at
24°C was greater for O. p. siberti than for O. p. puerilis (Fig. 3). Most individuals
that survived grew poorly; a few, however, grew at rates rather similar to those of
the Mediterranean subspecies. This may represent genetic variation for the trait, but
we have not followed this up.
Although there are clear intersubspecific differences, we wondered if there were
significant among-family differences in growth rate, as reported in Levinton (1983).
To test for this we performed a three-way analysis of variance, using temperature,
family, and time as the variables. Since we had used the same families from a given
subspecies in all of the temperature treatments we could estimate whether the use of
different families caused an additional variance component. Table I shows the results
for both subspecies. In both cases significant among-family differences in growth rate
can be found when time and temperature are factored out.
In a sense, this analysis is problematical because the body size of a given individual
at a given time is not independent of the previous time. Thus the relative magnitude
of a family mean size may persist for more than one time period. To eliminate the
problem, we performed two-way analyses of variance at a given time, using temperature
and family as the variables. Table II shows the results for seven days and for 28 days.
At seven days, significant and persistent among-family differences in growth occur
despite rearing in several temperatures. At 28 days, however, no significant additional
variance component is generated by family difference. This change may represent
initial family differences that were eliminated subsequently by acclimation to new
common conditions over the course of the experiment, and, possibly, mortality of
more slowly-growing worms.
DISCUSSION
Our results demonstrate a shifting growth advantage consistent with the differing
thermal regimes of the two subspecies. They conform to an hypothesis which predicts
that local evolution should maximize metabolic efficiency and thus favor maximum
growth under local thermal conditions (Levinton, 1983). Thus, despite common
rearing for several generations under common conditions, the evolved differences
between the populations emerge as growth differences.
The latitudinal compensation hypothesis weighs heavily on the interpretation of
life-history differences among natural populations of a given species living along a
thermal (e.g., latitudinal) gradient. In any study of growth, body size, and age of first
reproduction, temperature may have to be considered as a primary influence on life-
history traits.
Although traditional studies of latitudinal differentiation in growth and metabolism
accounted for such a limitation (e.g., Weymouth and McMillan, 1931; Rao, 1953;
LATITUDE AND INTRASPECIFIC GROWTH
703
CO
t-
Z
LU
O
LU
CO
CO
o
g
i-
LU
CO
15'
i
18-
16.
i
i
, 'i
14,
i
i
i
12.
<
i
4
\
10
22 .
20 .
18
20'
\
20.
12 .
4 -
* * J
)
1
> 1
*
1
1
•
1
1
1
1
•
•
puerilis
siberti
FIGURE 2. Mean individual growth (number of setigerous segments after 28 days), ±95% confidence,
among the families of Ophryotrocha puerilis puerilis and O p. siberti at 15, 20, and 24°C.
704
J. S. LEVINTON AND R. K. MONAHAN
100n
£ 75-1
DC
o
I 50H
DC
25-
0
24°
0 7 14 21
TIME (DAYS)
FIGURE 3. Survival of individuals of the two subspecies at 24°C.
28
Vernberg and Vernberg, 1966, among others), recent workers have tended to assume
that latitudinal differences in life history patterns reflect differences in demography
which, in turn, select for different ages of first reproduction, investment in growth
TABLE I
Three-way analysis of variance for the growth experiment, testing for differences among families,
temperatures, and sampling times
Ophryotrocha puerilis puerilis:
Source of Variation DF
Temperature (T) 2
Families (F) 3
Sampling Week (W) 3
T X F Interaction 6
T X W Interaction 6
F X W Interaction 9
T X F X W Interaction 1 8
SS
318.74
42.29
1544.02
15.02
47.59
5.51
4.40
MS
159.37
14.10
514.67
2.50
7.93
0.61
0.25
651.79***
57.64***
2104.86***
10.24***
32.44***
2.51
Ophryotrocha puerilis siberti:
Source of Variation
DF
SS
MS
Temperature (T)
1
129.67
129.67
474.12***
Families (F)
4
53.01
13.25
48.45***
Sampling Week (W)
3
1510.14
503.38
1840.52***
T x F Interaction
4
1.00
0.25
0.91
T X W Interaction
3
16.22
5.41
19.77***
F X W Interaction
12
4.63
0.39
1.41
T X F X W Interaction
12
3.28
0.27
* P < .001.
Data for Ophryotrocha puerilis siberti exclude 24°C, due to low sample sizes.
LATITUDE AND INTRASPECIFIC GROWTH 705
TABLE II
Two-way analyses of variance considering variation in growth among temperatures
and families at 7 and 28 days
Family Temperatures
Subspecies Time (days) F P F P
O. p. puerilis
1
5.52
<.05
15.71 <.01
O. p. puerilis
28
2.84
NS
69.74 <.01
O. p. siberti
7
11.57
<.01
17.42 <.01
O. p. siberti
28
0.37
NS
31.28 <.01
NS: Not Significant (P > .05).
versus reproduction. In a study of the turban snail Tegula funebralis, Frank (1975)
concluded that the smaller reproductive size of individuals in low-latitude habitats
resulted from increased adult mortality relative to high latitudes. Such an interpretation
has also been used by Boehlert and Kappenman (1980) to explain latitudinal patterns
in size at reproductive maturity in a fish species. While adult mortality clearly can
influence life history tactics (Stearns, 1976; Charnov, 198 1 ), one cannot safely interpret
latitudinal patterns of life history change as being due to demography alone. Our
results and Levinton's (1983) model clearly show that temperature can strongly in-
fluence latitudinal variation in growth.
Some recent studies support the role of temperature in latitudinal patterns in life-
history tactics. For example, Searcy (1980) shows that latitudinal body size clines in
birds are best explained as an adaptation to conserve body heat. Birds living above
a certain temperature need not consume energy to cool the body. Below a certain
temperature for a given body volume, however, the rate of heat loss is not matched
by the rate of heat production of a bird that is "thermally neutral," i.e., producing
enough heat to maintain typical passerine body temperature. An increase in body
volume tends to reduce the rate of heat loss and therefore permits a bird to survive
lower temperatures with no additional metabolic cost.
Levinton and Lonsdale (1983) have examined latitudinal patterns of growth and
body size in the harpacticoid copepod Scottolana canadensis taken from localities
from Maine to Florida. They reared populations under common conditions through
several generations, and found strong differences in growth rate. These differences
reflected a growth advantage of high latitude derived populations at low temperatures
and the reverse at high temperature. Body size of northern-derived populations also
was larger than southern-derived individuals at all temperatures despite common
rearing. They suggest that both growth and body size are controlled by thermal
constraints.
Several studies of latitudinal differences in growth rates have explained the ac-
celeration of growth of high latitude (or altitude) populations, relative to low latitude
(altitude) populations at low temperatures as being a form of compensation (Ament,
1979; Bervan el al, 1979). The higher altitude forms grow more rapidly to compensate
for the effect of lowered temperature on poikilothermic activities. Clarke (1982) suggests
that slow growth in arctic forms reflects a strategy to deal with chronically low food
availability.
We might speculate that the difference in size of the male-female switch in sex
between subspecies may also reflect thermal limitations. The size of sex switch should
be determined by the relative fecundities of males and females as a function of
706 J S. LEVINTON AND R. K. MONAHAN
increasing size. Consider the case where temperature increases. Although temperature
increases metabolic demand, it also increases activity and, therefore, the rate of feeding.
As long as the animal is not near its upper thermal limit the energy gain in feeding
will increase disproportionately, relative to increased metabolic demand. Increasing
temperature will thus increase female fecundity for a given body size. If sperm are
energetically cheap to produce then there may be no difference in male fecundity
with differing temperature. Increasing temperature, therefore, will increase female
fecundity, relative to that of the male, at a given body size. This is sufficient to cause
evolution of a decreased size of male to female switch as a response to increased
temperature. Our prediction is complicated, however, by the ability of individuals of
this species to undergo repeated sex changes following the initial size-specific switch.
These secondary sex changes can be provoked by interactions among worms and
nutritional condition (Pfannenstiel, 1975, 1977; Berruti, 1980).
We conclude that the effects of temperature probably affect all life history features
of a poikilothermic organism. It therefore will be important in future studies to
account for temperature in studies of life histories, along with such factors as stochastic
processes of population extinction, mortality schedules, and other factors known to
govern the evolution of growth and reproductive strategies.
ACKNOWLEDGMENTS
This study was supported by National Science Foundation Grant OCE 80-18743
(Biological Oceanography). Contribution number 475 from the Graduate Studies in
Ecology and Evolution, State University of New York at Stony Brook.
LITERATURE CITED
AKESSON, B. 1970. Ophryotrocha labronica as a test animal for the study of marine pollution. Helgol. wiss.
Meeresunters. 20: 293-303.
AKESSON, B. 1975. Reproduction in the genus Ophryotrocha (Polychaeta, Dorvilleidae). Publ. Staz. Zool.
Napoli 39(suppl.): 377-398.
AKESSON, B. 1977. Crossbreeding and geographic races: experiments with the polychaete genus Ophryotrocha.
Mikro fauna Meeresboden 61: 11-18.
AMENT, A. 1979. Geographic variation in relation to life history in three species of the marine gastropod
genus Crepidula: growth rates of newly hatched larvae and juveniles. Pp. 61-76 in Reproductive
Ecology of Marine Invertebrates, S. E. Stancyk, ed. Univ. South Carolina Press, Columbia, SC.
BACCI, G., AND O. BORTESI. 1961. Existence of multiple sex genotypes in the hermaphrodite polychaete
worm Ophryotrocha puerilis siberti. Native 190: 838.
BACCI, G., AND M. LAGRECA. 1953. Genetic and morphological evidence for subspecific differences between
Naples and Plymouth populations of Ophryotrocha puerilis. Nature 111: 1115.
BERRUTI, G. 1980. The effect of starvation on the appearance of the definitive upper jaw in Ophryotrocha
puerilis (Annelida, Polychaeta). Boll. Zool. 47: 71-74.
BERVAN, K. A., D. E. GILL, AND S. J. SMITH-GILL. 1979. Countergradient selection in the green frog,
Rana clamitans. Evolution 33: 609-623.
BOEHLERT, G. W., AND R. F. KAPPENMAN. 1980. Variation of growth with latitude in two species of
rockfish (Sebastes pinniger and 5. diploproa) from the Northeast Pacific Ocean. Mar. Ecol. Prog.
Ser. 3: 1-10.
CHARNOV, E. L. 1981. Sex reversal in Pandalus borealis: effect of a shrimp fishery. Mar. Biol. Lett. 2: 53-
57.
CLARKE, A. 1982. Temperature and embryonic development in polar marine invertebrates. Int. J. Invert.
Reprod. 5: 71-82.
FRANK, P. W. 1975. Latitudinal variation in the life history features of the black turban snail Tegula
funebralis (Prosobranchia: Trochidae). Mar. Biol. 31: 181-192.
LEVINTON, J. S. 1983. The latitudinal compensation hypothesis: Growth data and a model of latitudinal
growth differentiation based upon energy budgets. I. Interspecific comparison of Ophrvotrocha
(Polychaeta: Dorvilleidae). Biol. Bull. 165: 686-698.
LATITUDE AND INTRASPECIFIC GROWTH 707
LEVINTON, J. S., AND D. J. LONSDALE. 1983. Latitudinal differentiation in copepod growth: physiological
and evolutionary significance. Proc. Nat. Acad. USA, submitted.
LEVINTON, J. S., AND T. H. SUCHANEK. 1978. Geographic differentiation, niche breadth, and genetic
differentiation at different geographic scales in the mussels, Mytilus californianus and M. edulis.
Mar. Biol. 49: 363-375.
PFANNENSTIEL, H.-D. 1975. Mutual influence on the sexual differentiation in the protandric polychaete
Ophryotrocha puerilix. Pp. 48-56 in Intersexuality in the Animal Kingdom, R. Reinboth, ed.
Springer- Verlag, Heidelberg.
PFANNENSTIEL, H.-D. 1977. Experimental analysis of the 'Paarkultureffekt' in the protandric polychaete,
Ophrvotrocha puerilis Clap. Mecz. J. E\p. Mar. Biol. Ecol. 28: 3 1 -40.
RAO, K. P. 1953. Rate of water propulsion in Mytilus californianus as a function of latitude. Biol. Bull.
104: 171-181.
SEARCY, W. T. 1980. Optimum body sizes at different ambient temperatures: an energetics explanation
of Bergmann's rule. J. Theor. Biol. 83: 579-583.
SELLA, G. 1 978. Responses adaptatives a differentes temperatures chez les deux sous-especes d 'Ophryotrocha
puerilis (Claparede et Mecznikow), au cours de la phase male. Can. Biol. Mar. 19: 1 17-125.
SELLA, G. 1980. The genetic system controlling the body length at which sex change occurs in Ophryotrocha
puerilis (Annelida, Polychaeta). Atti. Soc. Toscana Sci. Nat. P. V. Mem. Ser. B 87: 197-203.
STEARNS, S. C. 1976. Life history tactics: a review of the ideas. Q. Rev. Biol. 51: 3-47.
VERNBERG, W. B., AND F. J. VERNBERG. 1966. Studies on the physiological variation between tropical
and temperate zone fiddler crabs of the genus Uca. V. Effect of temperature on tissue respiration.
Cornp. Biochem. Physiol. 17: 118-126.
WEYMOUTH, F. W., AND H. C. MCMILLAN. 1931. The relative growth and mortality of the Pacific razor
clam (Siliqua patula Dixon) and their bearing on the commercial fishery. Bull. U. S. Bur. Fish.
46: 543-567.
Reference: Biol. Bull. 165: 708-722. (December, 1983)
ENERGY METABOLISM DURING AIR EXPOSURE AND RECOVERY
IN THE HIGH INTERTIDAL BIVALVE MOLLUSC GEUKENSIA
DEMISSA GRANOSISSIMA AND THE SUBTIDAL BIVALVE
MOLLUSC MODIOLUS SQUAMOSUS
C. V. NICCHITTA1 AND W. R. ELLINGTON2
Department of Biological Science, Florida State University, Tallahassee. Florida 32306
ABSTRACT
Metabolic responses to air exposure and recovery were investigated in the adductor
muscles of the high intertidal mussel Geukensia demissa granosissima and the subtidal
mussel Modiolus squamosus. Exposure to air for 12 h had no significant effect on
the levels of high energy phosphates (arginine phosphate, ATP) in the adductor muscles
of G. demissa granosissima, indicating minimal metabolic stress in this species. In
contrast, there was a considerable decline in arginine phosphate and ATP during air
exposure in the phasic and tonic adductor muscles of M. squamosus. In addition,
there was a substantial accumulation of alanine and succinate under these conditions.
Furthermore, D-lactate accumulated in the phasic muscle of M. squamosus during
air exposure. During recovery, there were transient accumulations of alanopine/
strombine in both G. demissa granosissima and M. squamosus. The differences in
metabolic responses between these two species reflect adaptations to specific micro-
habitats. It appears that metabolism in the posterior adductor muscle of G. demissa
granosissima is largely aerobic during air exposure. The subtidal species M. squamosus
displays a much greater reliance on anaerobic pathways of energy production under
these conditions.
INTRODUCTION
Bivalve molluscs are not structurally well adapted for aerial gas exchange (Lent,
1968). The gills show extensive modifications for filter feeding and, secondarily, for
gas exchange. The role of the gills in gas exchange may be quite reduced in some
species. Booth and Mangum (1978) showed that ligation of the aorta of the ribbed
mussel Modiolus demissus (Geukensia demissa) resulted in only a 15% decrease in
aquatic oxygen consumption. Thus, gas exchange in this species may take place
primarily over the generalized body surfaces. During exposure to air at low tides,
many marine bivalves appear to be capable of taking up atmospheric oxygen. Significant
rates of aerial oxygen consumption have been observed in Cerastoderma edule (Boyden,
1972), Mytilus edulis (Coleman, 1973), Modiolus modiolus (Coleman, 1976), and
Modiolus demissus (Booth and Mangum, 1978). Typically, rates of aerial gas ex-
change are lower than aquatic rates (Coleman, 1973; Bayne et al, 1976; Widdows
et ai, 1979).
The metabolic rates of bivalve molluscs exposed to air vary considerably between
species and, in a temporal sense, may vary considerably within an individual. For
Received 16 May 1983; accepted 15 August 1983.
1 Present address: Department of Biochemistry and Biophysics, University of Pennsylvania, Philadelphia,
Pennsylvania 19104.
2 To whom editorial correspondence and reprint requests should be sent.
708
ENERGY METABOLISM IN BIVALVE MOLLUSCS 709
instance, Pamatmat (1983) measured heat production rates during air exposure in
specimens ofGeitkensia demissa. Animals tended to show regular cycles of high rates
of heat production (valves presumably open, metabolism principally aerobic) followed
by low rates of heat production (valves presumably closed, metabolism principally
anaerobic). The period of the cycle varied from individual to individual (Pamatmat,
1983). The relative contributions of anaerobic energy yielding processes to the total
metabolic rate may depend on the previous acclimation history of the individual.
Shick and Widdows (1981) showed, using calorimetric techniques, that subtidally
acclimated specimens of M, edulis relied exclusively on anaerobic metabolism during
air exposure. Experiments with subtidally acclimated specimens of the cockle Cardium
edule indicated that metabolism was exclusively aerobic during air exposure. In con-
trast, anaerobic heat production accounted for 62% of the total heat production in
intertidally acclimated specimens of M. edulis exposed to air (Shick and Wid-
dows, 1981).
Anaerobic metabolism has been studied extensively in bivalve molluscs (de Zwaan,
1977). There are a variety of metabolic options available for energy production during
air exposure and anoxia. Lactate production is not common, although it is a major
end product in at least one bivalve mollusc (Gade, 1980). Typically, there is a si-
multaneous fermentation of glycogen and aspartate yielding succinate and alanine
as end products (Collicutt and Hochachka, 1977; Ebberink el ai, 1979). Aspartate
provides the carbon skeleton for succinate and the amino group used in alanine
formation. Assuming that once aspartate levels become depleted, further alanine
formation is minimal and succinate carbon is then derived exclusively from glycogen.
This metabolic transition may involve a shift at the phosphoenolpyruvate (PEP)
branchpoint involving increased activity of the enzyme phosphoenolpyruvate car-
boxykinase (PEPCK) (Ebberink et al, 1979). Recently, de Zwaan et al. (1982) ques-
tioned the role of PEPCK in the energy metabolism of the posterior adductor muscle
of M. edulis. During extended anoxia, the volatile fatty acid, propionate, has also
been shown to be a major end product in specimens of M. edulis (Kluytmans et al.,
1975, 1978).
In addition to lactate, alanine, succinate, and propionate, an entirely new class
of end products has recently been shown to accumulate during anoxia. Fields (1976)
discovered a cytoplasmic dehydrogenase in oyster tissues which utilized pyruvate and
an amino acid as substrates. The resulting products of the reaction were the imi-
nodicarboxylic acids, alanopine (alanine as substrate), and strombine (glycine as sub-
strate). Recently, it has been shown that strombine accumulates during air exposure
in the posterior adductor muscles of specimens of M. edulis (Zurburg et al., 1982;
de Zwaan et al., 1983).
Regardless of the qualitative nature of the end products produced during air
exposure, bivalve molluscs display great similarities with respect to the overall mag-
nitude of energy metabolism. A Pasteur effect is typically absent (de Zwaan, 1977).
Thus, there is no increase in glycolytic flux during anoxia and consequentially, rates
of ATP production fall. De Zwaan and Wijsman (1976) and Ebberink et al. (1979)
showed that the energy expenditure of the adductor muscle of M. edulis decreases
on the order of five fold during air exposure. The diminished energy demand tends
to maintain energy balance despite low rates of glycolytic flux.
Investigation into the metabolic events immediately following oxygen stress has
lagged far behind studies dealing with metabolism during air exposure. A variety of
metabolic events occur during recovery including recharging of high energy phosphates,
oxidation of end products, and resynthesis of anaerobic substrates. Typically, levels
of succinate, lactate, and alanine fall while aspartate levels rise (Gade and Meinardus,
710 C. V. NICCHITTA AND W. R. ELLINGTON
1981; Zurburg el ai, 1982). The resynthesis of ATP and the phosphagen, arginine
phosphate, also occurs during recovery. Most molluscs show a characteristic elevation
of oxygen consumption or oxygen debt following hypoxia reflecting, to some extent,
the enhanced energy demand of recovery (de Zwaan, 1977; de Vooys and de Zwaan,
1978). In addition, there may be enhanced glycolytic flux, as strombine accumulates
during recovery in at least one species (Zurburg et al., 1982; de Zwaan el al, 1983).
In the present study, we compare metabolic responses to air exposure and recovery
in two species of bivalve molluscs adapted to distinctly different micro-habitats. The
ribbed mussel Geukensia demissa granosissima is a high intertidal species which is
regularly exposed to air for hours or even days at a time. The mussel Modiolus
squamosus is a subtidal species. Populations of M. squamosus are exposed to air only
during exceptionally low tides. The present study shows dramatic differences in terms
of the metabolic responses of the two species to experimental air exposure. Specimens
of G. demissa granosissima appear to rely extensively on aerial gas exchange showing
only trivial accumulations of anaerobic end products. In contrast, specimens of M.
squamosus show substantial accumulations of anaerobic end products indicating a
reliance on anaerobic energy production during air exposure.
MATERIALS AND METHODS
Animals
Specimens of Geukensia demissa granosissima were collected in salt marshes at
Yent Bayou, Florida. Specimens of Modiolus squamosus were collected off Alligator
Point, Florida. Animals were maintained in running sea water (24-28°C, 30%o) at
the Florida State University Marine Laboratory, Turkey Point. Animals were used
in experiments four to seven days after collection.
Materials
Biochemicals were purchased from Sigma Chemical Company (St. Louis, Missouri)
and Boehringer Mannheim (Indianapolis, Indiana). All other chemicals were reagent
grade quality. Octopine dehydrogenase, alanopine dehydrogenase, and D-lactate de-
hydrogenase were purified by affinity chromatography from the adductor muscle of
the scallop Argopecten ir radians concentricus, the adductor muscle of the oyster Cras-
sostrea virginica, and muscle of the horseshoe crab Limulus polyphemus. These en-
zymes were used to assay for octopine, alanopine/strombine, and D-lactate, respec-
tively. Succinyl Co A synthase, used in succinate assays, was a gift of Dr. William
Bridger, Department of Biochemistry, University of Alberta, Edmonton, Alberta,
Canada.
Profile of adductor muscle enzyme activities
Activities of key glycolytic enzymes and citrate synthase were assayed in crude,
cell-free extracts of the posterior adductor muscle of G. demissa granosissima and
the phasic and tonic portions of the posterior adductor muscles of M. squamosus.
The following enzymes were assayed: phosphorylase (Plase), hexokinase (HK), phos-
phofructokinase (PFK), lactate dehydrogenase (LDH), alanopine dehydrogenase
(ADH), octopine dehydrogenase (ODH), glyceraldehyde-3-phosphate dehydrogenase
-3PDH), and citrate synthase (CS). Tissue was homogenized in nine volumes (w:v)
of extraction medium using a Brinkman Polytron tissue grinder and centrifuged at
ENERGY METABOLISM IN BIVALVE MOLLUSCS 711
10,000 X g for 20 min. The following extraction media were used: 50 mM trietha-
nolamine containing 1 mM EDTA, 1 mM MgCl2, and 30 mM 2-mercaptoethanol
at pH 7.4 for LDH, ODH, ADH, PK, PEPCK, and HK; 70 mMTris/HCl containing
1 mM EDTA and 5 mM MgSO4 at pH 8.2 for PFK; 100 mM triethanolamine
containing 7 mAf 2-mercaptoethanol at pH 7.0 for Plase; and 25 mM Tris/HCl
containing 1 mM EDTA at pH 7.5 for CS. Enzymes were assayed by standard pro-
cedures—Plase and G-3-PDH (de Zwaan et ai, 1980), PFK, HK, and PEPCK (Zammit
and Newsholme, 1976), PK, LDH, ODH, and ADH (Ellington, 1981), and CS (Sugden
and Newsholme, 1975). All assays were conducted in a Gilford 252-1 spectropho-
tometer at 25°C. Assays were initiated by the addition of substrate.
Metabolic responses to air exposure and recovery
Specimens of G. demissa granosissima and M. squamosus were collected and
maintained in running sea water for four days prior to experimentation. At zero time,
all animals were removed from the sea table and placed in a humidified (100%),
temperature controlled (27°C) chamber. A total of 130 specimens of G. demissa
granosissima and 144 specimens of M. squamosus were used in these experiments.
A zero time group of animals (n -- 10 for G. demissa granosissima, n = 12 for M.
squamosus) was randomly selected, and the posterior adductor muscles were excised
and frozen in liquid nitrogen. Phasic and tonic portions of the adductor muscle in
M. squamosus were frozen separately. At various time intervals during air exposure
(0.5 1, 2, 4, 7, and 12 h for G. demissa granosissima; 0.5, 1, 4, 7, and 12 h for M.
squamosus} subsets of either 10 animals (G. demissa granosissima) or 12 animals
(M. squamosus) were removed and posterior adductor muscles frozen. At the end of
12 h of air exposure, the remaining animals were returned to the sea table and subsets
of animals were removed at various time intervals (2, 4, 6, 8, 10, and 12 h) and
treated as above. All tissues were stored at -80°C. Tissues were processed and analyzed
within 36 h of tissue sampling.
Biochemical analyses of tissue samples
Tissue samples were fragmented using a mortar and pestle chilled in liquid nitrogen.
For each analysis, approximately 1 g of tissue representing the adductor muscles of
several animals was weighed and homogenized in 5 volumes (w:v) 6% perchloric acid
(4°C). The homogenates were centrifuged at 10,000 Xgfor 20 min and the supernatants
neutralized with 5 M KOH/0. 1 M KHCO3 . The neutralized extract was centrifuged
and the supernatant stored at — 80°C.
Arginine phosphate and ATP levels in the extracts were assayed within 3-5 h of
extract preparation. Arginine phosphate and ATP were assayed by the spectropho-
tometric assays of Lowry and Passonneau (1972) except that lobster arginine phos-
phokinase was substituted for creatine phosphokinase. ADP and AMP were assayed
according to Lowry and Passonneau (1972). Succinate was determined by the method
of Williamson (1974). Octopine, alanopine/strombine and D-lactate were assayed in
a reaction system consisting of 100 mM2-amino-2-methyl-l-propanol (pH 9.2) con-
taining 50 mM hydrazine, 4 mM NAD, and 10 mM EDTA. Assays were initiated
by the addition of 5 enzyme units of the appropriate enzyme. Alanine, glycine,
aspartate, and glutamate were determined using a Beckman model 120-1 automatic
amino acid analyzer. Propionate levels were determined by HPLC. One (1) ml of
the neutralized, perchloric acid extract was applied to a silica Sep-Pak (Waters, Inc.)
pretreated with 1 .0 ml ultra pure water followed by a 4.0 ml ultra pure hexane wash.
The sample was then washed with 2.0 ml of ultra pure hexane and the polar fraction
712
C. V. NICCHITTA AND W. R. ELLINGTON
eluted with 1 .0 ml of ultra pure water. Treated extracts were analyzed on a Waters
HPLC system using a BIO-RAD (Bio-Rad Laboratories, Richmond, California) ODS-
5 reversed phase column (250 mm X 4 mm, ID), isocratic elution (0.2 M KH2PO4,
pH 2.4), and UV detection (200 nm).
All metabolite data were analyzed for significance by one way ANOVA and a
least significant difference test (Freyer, 1966).
RESULTS
Profile of the activities of key glycolytic enzymes and citrate synthase
Activities of key glycolytic enzymes and citrate synthase in the adductor muscle
of G. demissa granosissima and the phasic and tonic portions of the adductor muscle
of M squamosus are listed in Table I. In general, enzyme activities were similar when
comparing the two species. However, ADH activity in both posterior adductor muscles
of M. squamosus was one order of magnitude greater than activity in G. demissa
granosissima. In addition, ODH was absent in the adductor muscles of M. squamosus.
The adductor muscles of both species had relatively low phosphorylase and hexokinase
TABLE I
Activities of key glycolytic enzymes and citrate synthase in the posterior adductor muscles ofG. demissa
granosissima and M. squamosus
Enzyme
Enzyme
activity1
M. squamosus
G. demissa
Lactate dehydrogenase
MP
MT
1.47 ± 0.37
0.84 ± 0.22
3.07
± 0.38
Octopine dehydrogenase
MP
MT
n/a
n/a
4.66
± 1.24
Alanopine dehydrogenase
MP
MT
13.56 ± 2.04
12.06 ± 2.76
1.02
± 0.04
Pyruvate kinase
MP
MT
2.49 ±0.16
2.26 ±0.11
1.07
±0.01
Phosphoenolpyruvate
carboxykinase
MP
MT
3.11 ±0.31
2.53 ± 0.15
4,41
± 0.03
Hexokinase
MP
MT
0.02 ± 0.00
0.02 ± 0.01
0.30
± 0.03
Citrate synthase
MP
MT
1.16 ± 0.08
0.96 ± 0.09
2.34
± 0.20
Glyceraldehyde-3-phosphate
dehydrogenase
MP
MT
42.75 ± 5.41
26.15 ± 3.12
41.05
± 3.42
Phosphorylase
MP
MT
0.87 ± 0.02
1.38 ± 0.15
1.24
± 0.18
Phosphofructokinase
MP
MT
4.27 ± 0.37
3.32 ±0.14
4.56
± 0.75
1 Enzyme activities are expressed as ^moles/(min • g wet wgt) at 25°C.
Each value represents a mean ± 1 S.D. (n = 4). MP = phasic adductor, Mt = tonic adductor, N/a
no activity.
ENERGY METABOLISM IN BIVALVE MOLLUSCS
713
activities implying reduced capacities for glycogen and glucose utilization. Enzyme
activities in the phasic and tonic portions of the posterior adductor muscle of M.
squamosus were virtually identical (Table I).
Metabolic responses to air exposure and recovery
Exposure to air for 1 2 h had no significant effect on the adenylate energy charge
(Fig. 1 ) and the levels of arginine phosphate and ATP (Fig. 2) in the posterior adductor
muscle of G. demissa granosissima. In contrast, adenylate energy charge and arginine
1.00i
0.75-
0.50-
Q25
o
o
>s
en
c
LU
c
<D
0
O
G. demissa
1.00 n
0.75-
Q50-
0.25-
0
M. squamosus
~i 1 1 r
0 l*
8
Time(h)
n r
12
r~
20
FIGURE 1. Alterations in the adenylate energy charge (ATP + '/2 ADP + ATP + ADP + AMP) in
the posterior adductor muscles of G. demissa granosissima and M. squamosus during air exposure and
recovery. Data for M. squamosus are given in terms of the phasic (solid circles) and tonic (open circles)
adductor muscles. The initial time point is depicted slightly to the left of zero. The arrow indicates the
onset of recovery. Each value is a mean ± 1 S.D. (n = 4).
714
C. V. NICCHITTA AND W. R. ELLINGTON
G demissa
3-
2-
1 -
0
0)
QJ
Q.
o
n
3-
2-
0
1 1 1 1 1 r
M squamosus (tonic)
n 1
i 1 1 r
3-
2-
1 -
0
0
M squamosus (phasic)
8 12
Time(h)
16 20
FIGURE 2. Effect of air exposure and recovery on the levels of arginine phosphate (open circles) and
solid circles) in the posterior adductor muscles of G. demissa granosissima and M. squamosus. Each
is a mean ± 1 S.D. (n = 4).
ENERGY METABOLISM IN BIVALVE MOLLUSCS
715
phosphate and ATP levels fell significantly in both portions of the adductor muscle
of M. squamosus (Figs. 1, 2). The greatest changes in these parameters occurred
during the first two hours of air exposure. Changes in the high energy phosphates
were most pronounced in the phasic adductor muscle of M. squamosus. During
recovery after air exposure, there continued to be no changes in high energy phosphates
in specimens of G. demissa granosissima (Figs. 1, 2). During recovery, the adenylates
returned to initial levels in the posterior adductor muscle of M. squamosus (Figs. 1,
601
4.5-
30-
1.5-
CD
S.
0
3-i
O
T 1
-40
-30
-20
-10
Lo
2-
1-
0
O
0
8 12
Time(h)
16
20
FIGURE 3. Effect of air exposure and recovery on the levels of alanine (open circles), aspartate (closed
circles), succinate (squares), and D-Iactate (triangles) in the posterior adductor muscle of G. demissa grano-
sissima. Each value is a mean ± 1 S.D. (n = 4).
716
C. V. NICCHITTA AND W. R. ELLINGTON
2). Arginine phosphate levels rose slowly during recovery but did not reach initial
levels after 12 h of recovery (Fig. 2).
There were no significant changes in the levels of alanine, aspartate, and D-lactate
en
o
9-
6-
3-
0J — i 1 1
81
6-
2-
0
0
8 12
Time(h)
16
20
T.U
FIGURE 4. Effect of air exposure and recovery on the levels of alanine, aspartate, succinate, and D-
lactate in the phasic adductor muscle of M. squamosus. Symbols are the same as in Figure 3. Each value
is a mean ± 1 S.D. (n = 4).
ENERGY METABOLISM IN BIVALVE MOLLUSCS
717
during air exposure and recovery in the posterior adductor muscle of G. demissa
granosissima (Fig. 3). There was a transient accumulation of succinate during the
early period of air exposure, but succinate levels returned to the initial levels by the
end of air exposure.
en
2.
I
1
9-
6-
3-
0
o
i i i
8 12
Time (h)
16
20
FIGURE 5. Effect of air exposure and recovery on the levels of alanine, aspartate, succinate, and D-
lactate in the tonic adductor muscle of M. squamosus. Symbols are the same as in Figure 3. Each value
is a mean ± 1 S.D. (n = 4).
718
C. V. NICCHITTA AND W. R. ELLINGTON
G.demissg
5-
M.squamcsus (tonic)
en
£
-5
en
o
CL
O
U-
3-
2-
1-
0
-\ r
5-j
4-
3-
2-
1-
0-
0
1 r
M.SQuamosus ( phasic)
Q
8 12
Time(h)
1 1
i i l l
16 20 21*
ENERGY METABOLISM IN BIVALVE MOLLUSCS 719
There were pronounced changes in metabolite levels in the posterior adductor
muscles of M. squamosus. In the phasic adductor muscle, aspartate levels declined
throughout air exposure and there was nearly a stoichiometric increase in alanine
levels (Fig. 4). There was a linear accumulation of succinate and D-lactate in the
phasic adductor muscle (Fig. 4). A similar pattern of aspartate depletion and succinate
and alanine accumulation was observed in the tonic adductor muscle of M squamosus
(Fig. 5). In contrast to the phasic adductor, the accumulation of D-lactate was low
in the tonic adductor muscle during air exposure. The general patterns of recovery
were similar in the phasic and tonic adductor muscle of M. squamosus. Succinate
was rapidly cleared with initial levels being attained after 2-4 h of recovery (Figs. 4,
5). Aspartate levels increased during recovery and there was a gradual decline in
alanine. After 12 h of recovery, alanine and aspartate levels still differed considerably
from pre-air exposure levels. In the case of the phasic adductor muscle of M. squamosus,
D-lactate levels slowly declined to initial levels during recovery (Fig. 4).
Alanopine/strombine accumulated during both air exposure and recovery in the
adductor muscles of G. demissa granosissima and M. squamosus (Fig. 6). In the
posterior adductor muscle of G. demissa granosissima there was an initial increase
in alanopine/strombine during air exposure followed by a gradual decline. Alanopine/
strombine levels then increased two-fold during recovery. In both the phasic and
tonic adductor muscles of M. squamosus, alanopine/strombine accumulated through-
out air exposure (Fig. 6). At the onset of recovery, there was a transient decline in
alanopine/strombine followed by a period of further increase during the midpoint of
the recovery period.
No significant changes in the levels of glycine and glutamate were observed in
the adductor muscles of G. demissa granosissima and M. squamosus. In addition,
there was no accumulation of octopine in either species. Propionate levels remained
low during both air exposure and recovery.
DISCUSSION
The results of this study show that there can be considerable intergeneric differences
in terms of metabolic responses to air exposure in bivalve molluscs. The high intertidal
mussel Geukensia demissa granosissima characteristically undergoes air gaping under
these conditions. In contrast, the subtidal mussel Modiolus squamosus typically main-
tains tightly sealed valves during air exposure and displays air gaping only after
extended periods of exposure. The consequences of these different responses to air
exposure are strongly reflected in the patterns of energy metabolism in the tissues of
these two species.
Air exposure for 12 h produced minimal metabolic stress on specimens of G.
demissa granosissima as is evidenced by the lack of changes in high energy phosphates
in the posterior adductor muscle. Although succinate and alanopine/strombine did
accumulate during air exposure, the magnitude of the accumulation is small compared
to that seen in other bivalve molluscs such as Mytilus edulis (de Zwaan el al., 1983)
and M. squamosus (this study). Thus, it appears that the anaerobic contribution to
energy metabolism during the first 12 h of air exposure is minimal. Alanine and
succinate accumulate to high levels in G. demissa after extended periods (>36 h) of
incubation in oxygen free sea water (Ho and ZubkofF, 1982). Thus, this species has
the capability of producing these end products under sufficiently stressful conditions.
FIGURE 6. Effect of air exposure and recovery on the levels of alanopine/strombine in the posterior
adductor muscles of G. demissa granosissima and M. squamosus. Each value is a mean ± 1 S.D. (n = 4).
720 C. V. NICCHITTA AND W. R. ELLINGTON
Specimens of G. demissa appear to be able to maintain significant rates of oxygen
uptake during air exposure (Booth and Mangum, 1978). However, aerial oxygen
consumption in this species is substantially less than aquatic oxygen consumption.
Since rates of aerobic energy production are reduced during air exposure and there
appears to be no large-scale utilization of anaerobic energy-producing pathways, the
overall rates of ATP production in G. demissa granosissima posterior adductor muscle
must fall during air exposure. Since the high energy phosphate levels are constant
during air exposure, it is evident that overall rates of energy demand in the adductor
muscle fall under these conditions. Thus, the apparent metabolic responses of the
mussel G. demissa granosissima involve aerial gas exchange coupled with an overall
reduction in the rates of ATP utilization in the posterior adductor muscle.
Air exposure produced dramatic alterations in the high energy phosphate levels
in the phasic and tonic adductor muscles of M. squamosus. These alterations in high
energy phosphates were similar in magnitude to what has been observed during anoxia
in the tissues of a number of molluscs including the posterior adductor muscle of
M. edulis (Ebberink et al, 1979), the foot muscle of the cockle Cardium tiiberciilatum
(Ga'de, 1980), and the ventricle of the whelk Busycon contrarium (Ellington, 1981).
The simultaneous depletion of aspartate, and accumulation of succinate and
alanine in both adductor muscles, indicates that glycogen and aspartate were fermented
in M. squamosus during air exposure. This phenomenon has been consistently observed
in a variety of molluscs (Collicutt and Hochachka, 1977; Ebberink et al., 1979;
Ellington, 1981). Collicutt and Hochachka (1977) predicted that both succinate and
alanine accumulation should occur in a 1:1 ratio with aspartate depletion. However,
all previous studies have shown that the amount of alanine accumulated was sub-
stantially greater than succinate. In the present study, the alanine:succinate accu-
mulation ratio during the first 4 h of air exposure was 2. 1 in the phasic adductor
muscle of M. squamosus. However, in the tonic adductor muscle, the accumulation
ratio was less than one during the first 4 h of air exposure and approached unity only
after 12 h of exposure. Recently, de Zwaan et al. (1983) have explained accumulation
ratios greater than one by suggesting that the mitochondrial malic enzyme is involved
in shunting some aspartate-derived carbon in the direction of alanine synthesis. The
rather different alanine:succinate accumulation ratios between the phasic and tonic
adductor muscles of M squamosus reflect variations in the metabolic disposition of
malate derived from aspartate. There is also the possibility that some of the succinate
production is derived from glycogen by the PEPCK route.
The accumulation of D-lactate in the phasic muscle and lack of accumulation in
the tonic muscle is rather surprising in that the activities of LDH are virtually identical
in the tissues. However, it must be noted that the decreases in high energy phosphates
were much more pronounced in the phasic adductor muscle of M. squamosus. In
addition, absolute levels of accumulation of alanine and succinate were much higher.
Thus, rates of energy demand in the phasic muscle may be greater than the tonic
muscle under these conditions. The lactate accumulation reflects increased glycolytic
flux in this tissue. The simultaneous accumulation of lactate and succinate has also
been observed in the foot muscle of the cockle Cardium edule (Ga'de and Meinar-
dus, 1981).
Recovery in the posterior adductor muscles of M. squamosus was characterized
by the rapid clearance of succinate. Lactate was also cleared rapidly in the phasic
muscle. Levels of ATP were rapidly restored. Arginine phosphate and aspartate slowly
increased during the 12 h of recovery. Similar phenomena have been observed during
recovery in the tissues of M. edulis (de Zwaan et al., 1983) and C. edule (Ga'de and
sinardus, 1981). In the present study, the time courses of succinate removal and
ENERGY METABOLISM IN BIVALVE MOLLUSCS 721
aspartate resynthesis were distinctly different indicating that there was probably no
direct metabolic link during recovery between the two processes.
In specimens of both M. squamosiis and G. demissa granosissima there was a
transient production of alanopine/strombine during recovery from air exposure. Sim-
ilarly, the bulk of strombine production in M. edulis occurred during recovery (Zurburg
et al, 1982; de Zwaan et al, 1983). De Zwaan et al. (1983) found that the PO2 levels
in the hemolymph of the adductor muscle rapidly approached normoxic values during
recovery. Thus strombine, a putative end product of anaerobic metabolism, was
produced under essentially aerobic conditions. De Zwaan et al. (1983) rationalized
this paradox by suggesting that energy demands exceed the limited capacity of aerobic
ATP yielding processes in the tissue. Thus, there is an increase in glycolytic flux to
meet the energy demands leading to strombine formation. The production of alan-
opine/strombine in the posterior adductor of M. squamosiis during recovery can be
easily interpreted by this argument. The production of alanopine/strombine in this
species is coincident with the period of recharging of the adenylate pool. The post
air exposure production of ananopine/strombine in G. demissa granosissima is more
difficult to explain since there were no changes in high energy phosphates. However,
increased energy demands might also result from other ATP requiring processes such
as the possibility of increased contractile activity of the adductor muscle during
recovery. It would be of great interest to measure valve movements during recovery
in G. demissa granosissima.
The overall results of this study show that energy metabolism during 12 h of air
exposure in the posterior adductor muscle of G. demissa granosissima is largely
aerobic. Booth and Mangum (1978) suggested that the metabolism of the adductor
muscle of G. demissa is largely anaerobic even in normoxic sea water. However, our
results show that the role of anaerobic energy metabolism is minimal even under
conditions of air exposure. This suggests that aerial gas exchange is sufficient to
maintain adequate rates of ATP production. Furthermore, apparent reductions in
energy demand tend to maintain energy balance in this tissue. In contrast, there are
substantial decreases in high energy phosphates and an extensive reliance on anaerobic
energy yielding processes during air exposure in the phasic and tonic adductor muscles
of M. squamosiis. These metabolic responses are probably due to a reduced capacity
for aerial gas exchange, and, perhaps, smaller reductions in energy demands during
air exposure. The patterns of aspartate and glycogen fermentation are similar to what
has been observed in other molluscs. The differences in metabolic responses of G.
demissa granosissima and M. squamosiis to air exposure reflect differences in ad-
aptation in micro-habitats of chronic versus infrequent air exposure.
ACKNOWLEDGMENTS
This work was partially supported by NSF Grant #PCM-8202370 to W.R.E.
LITERATURE CITED
BAYNE, B. L., C. J. BAYNE, T. C. CAREFOOT, AND R. J. THOMPSON. 1976. The physiological ecology of
Mvtilus californianus Conrad. 2. Adaptations to low oxygen tension and air exposure. Oecologia
22: 229-250.
BOOTH, C. E., AND C. P. MANGUM. 1978. Oxygen uptake and transport in the lamellibranch mollusk
Modiolus demissus. Physiol. Zoo/. 51: 17-32.
BOYDEN, C. R. 1972. Aerial respiration in the cockle Cerastoderma edule in relation to temperature. Comp.
Biochem. Physiol. 43A: 697-712.
COLEMAN, N. 1973. The oxygen consumption of Mvtilus edulis in air. Comp. Biochem. Physiol. 45A:
393-400.
722 C. V. NICCHITTA AND W. R. ELLINGTON
COLEMAN, N. 1976. The aerial respiration of Modiolus modiolus. Comp. Biochem. Physiol. 54 A: 401-406.
COLLICUTT, J. M., AND P. W. HOCHACHKA. 1977. The anaerobic oyster heart: coupling of glucose and
aspartate fermentation. J. Comp. Physiol. 115: 147-157.
EBBERINK, R. H. M., W. ZURBURG, AND D. I. ZANDEE. 1979. The energy demand of the posterior adductor
muscle of Mytilus edulis in catch during air exposure. Mar. Biol. Lett. 1: 23-31.
ELLINGTON, W. R. 1981. Energy metabolism during hypoxia in the isolated, perfused ventricle of the
whelk Busycon contrarium Conrad. J. Comp. Physiol. 142: 457-464.
FIELDS, J. H. A. 1976. A dehydrogenase requiring alanine and pyruvate as substrates from oyster adductor
muscle. Fed. Proc. 35: 1687.
FREYER, H. C. 1966. Concepts and Methods of Experimental Statistics. Allyn and Bacon, Boston, 602 pp.
GADE, G. 1980. The energy metabolism of the foot muscle of the jumping cockle, Cardium tuberculatum:
sustained anoxia versus muscular activity. J. Comp. Physiol. 137: 177-182.
GADE, G., AND G. MEINARDUS. 1981. Anaerobic metabolism in the common cockle Cardium edule V.
Changes in the levels of metabolites in the foot during aerobic recovery after anoxia. Mar. Biol.
65: 113-116.
Ho, M., AND P. L. ZUBK.OFF. 1982. Anaerobic metabolism of the ribbed mussel Geukensia demissa. Comp.
Biochem. Physiol. 73B: 931-936.
KLUYTMANS, J. H., P. R. VEENHOF, AND A. DE ZWAAN. 1975. Anaerobic production of fatty acids in the
sea mussel Mytilus edulis L. / Comp. Physiol. 104: 71-78.
KLUYTMANS, J. H., M. VANGRAFT, J. JANUS, AND H. PIETERS. 1978. Production and excretion of volatile
fatty acids in the sea mussel Mytilus edulis. J. Comp. Physiol. 123: 163-167.
LENT, C. N. 1968. Air gaping by the ribbed mussel, Modiolus demissus (Dillwyn): Effects and adaptive
significance. Biol. Bull. 134: 60-73.
LOWRY, O. H., ANDJ. V. PASSONNEAU. 1972. A Flexible System of Enzymatic Analysis. Academic Press,
New York. 291 pp.
PAMATMAT, M. M. 1983. Measuring aerobic and anaerobic metabolism of benthic infauna under natural
conditions. J. Exp. Zool. (in press).
SHICK, J. M., AND J. WIDDOWS. 1981. Direct and indirect calorimetric measurement of metabolic rate in
bivalve molluscs during air exposure. Am. Zool. 21: 985.
SUGDEN, P. H., AND E. A. NEWSHOLME. 1975. Activities of citrate synthase, NAD linked and NADP
linked isocitrate dehydrogenase, glutamate dehydrogenase, aspartate aminotransferase and alanine
aminotransferase in nervous tissue from vertebrates and invertebrates. Biochem. J. 150: 105-1 1 1.
VOOYS, G. G. N. DE, AND A. DE ZWAAN. 1978. The rate of oxygen consumption and ammonia excretion
by Mvtilus edulis after various periods of exposure to air. Comp. Biochem. Phvsiol. 60A: 343-
347. '
WIDDOWS, J., B. L. BAYNE, D. R. LIVINGSTONE, R. E. NEWELL, AND?. DONK.IN. 1979. Physiological and
biochemical responses of bivalve molluscs to air. Comp. Biochem. Physiol. 62A: 301-308.
WILLIAMSON, J. R. 1974. Succinate. Pp. 1616-1621 in Methods of Enzymatic Analysis, Vol. 3, H. U.
Bergmeyer, eds., Academic Press, New York.
ZAMMIT, V. A., AND E. A. NEWSHOLME. 1976. The maximum activities of hexokinase, phosphorylase,
phosphofructokinase, glycerol phosphate dehydrogenase, lactate dehydrogenase, octopine dehy-
drogenase, phosphoenolpyruvate carboxykinase, nucleoside diphosphate kinase, glutamate oxal-
oacetate transaminase and arginine kinase in relation to carbohydrate utilization in muscles from
invertebrates. Biochem. J. 160: 447-462.
ZURBURG, W., A. M. T. DE BONT, AND A. DE ZWAAN. 1982. Recovery from exposure to air and the
occurrence of strombine in different organs of the sea mussel Mvtilus edulis. Mol. Phvsiol. 2:
135-147.
ZWAAN, A. DE. 1977. Anaerobic energy metabolism in bivalve molluscs. Oceanogr. Mar. Biol. Ann. Rev.
15: 103-187.
ZWAAN, A. DE, ANDT. C. N. WIJSMAN. 1976. Anaerobic metabolism in bivalvia (mollusca). Characteristics
of anaerobic metabolism. Comp. Biochem. Physiol. 54B: 313-324.
ZWAAN, A. DE., D. R. LIVINGSTONE, AND R. J. THOMPSON. 1980. Physiological and biochemical aspects
of the valve snap and valve closure responses in the giant scallop Placopecten magellanicus. II.
Biochemistry. / Comp. Physiol. 137: 105-114.
ZWAAN, A. DE., A. M. T. DE BONT, AND A. VERHOEVEN. 1982. Anaerobic energy metabolism in isolated
adductor muscle of the sea mussel Mytilus edulis L. J. Comp. Physiol. 149: 137-143.
ZWAAN, A. DE., A. M. T. DE BONT, W. ZURBURG, B. L. BAYNE, AND D. R. LIVINGSTONE. 1983. On the
role of stombine formation in the energy metabolism of adductor muscle of a sessile bivalve. /
Comp. Physiol. 149: 557-563.
Reference: Biol. Bull. 165: 723-732. (December, 1983)
SCANNING ELECTRON MICROSCOPY OF THE REGENERATED
SHELL OF THE MARINE ARCHAEOGASTROPOD, TEGULA
CHARLENE REED-MILLER
Department of Geology, Florida State University, Tallahassee, FL 32306
ABSTRACT
A window was cut in the first body whorl of the marine snail, Tegula, to induce
shell regeneration. At various intervals after the shell window was cut, the window
with the regenerated material and the shell surrounding it were prepared for scanning
electron microscopy. Initial crystal deposition occurred in association with an organic
matrix and appeared as small, spindle-shaped crystals formed by the aggregation of
needle-like subunits. The spindles were frequently aggregated into stellate clusters
that coalesced to form a sheet of mineralized tissue. After about two months of
regeneration, dumbbell-shaped crystal aggregates and spherulites were apparent on
the surface of the regenerated shell. The regenerated shell assumed a normal structure
after at least four months of regeneration.
Crystal deposition also occurred on the normal shell bordering the shell window.
The crystals assumed several forms, and their orientation appeared to be determined
by the microtopography of the underlying shell.
INTRODUCTION
Molluscan shell mineralization is the result of a complex and delicate association
of biological, chemical, and physical processes. The result of the interaction of these
factors is not always the same, even in a single animal. The degree of organic versus
inorganic control of mineralization in the molluscan shell is an example of variability
in structure determined by the interplay of these three processes. Molluscan growth
surfaces show variation in organic and inorganic mechanisms of crystallization. Organic
suppression of natural crystal form of the outer (distal) shell layer was much less than
in the inner three shell layers of an archaeogastropod, Cittarium pica (Wise and Hay,
1968a, b). The same was found to be true for five species of the archaeogastropod
genus, Tegula (Reed-Miller, 198 la). The aragonitic crystals of the nacreous shell layer
are often present in tabular or diminished "c" axis form. This differs from the usual
conformation of inorganically precipitated aragonite, elongate twinned prisms, and
represents another example of organic control of crystal morphology.
The mineralized product formed during shell regeneration can be similar to, or
quite different from the ultrastructure of the normal shell. This emphasizes again
structural range of mineralized tissue (Saleuddin and Wilbur, 1969; Wilbur, 1972;
Wong and Saleuddin, 1972). Earlier reports showed differences in the structure of
regenerated shell compared to the normal shell of Tegula (Reed-Miller et al., 1980;
Reed-Miller, 198 la). The region of the shell involved in regeneration is considered
to be another area of active calcification and mineralization. Since the area of least
suppression of natural crystal form occurred at the lip, or growing edge of the shell
in some archaeogastropods, including Tegula (see above), it was of interest to look
at the crystal structure in regenerated Tegula shell. The initial ultrastructural changes
Received 18 July 1983; accepted 26 September 1983.
723
FIGURE 1. Regenerated material in the shell window, showing doubly-pointed crystallites grouped
into bundles or rosettes on an organic membrane. One week of regeneration. Bar = 50 ^m.
FIGURE 2. Higher magnification of spindle-shaped crystals similar to those shown in Figure 1. Note
the elongated, needle-like subunits that make up the spindles. One week of regeneration. Bar = 10 ^m.
FIGURE 3. Rosette-shaped assemblages of crystalline spindles. Note the underlying layer of coalesced
crystals. Two weeks of regeneration. Bar = 10 pm.
FIGURE 4. A sheet of mineralized tissue formed by the coalescence of rosette-shaped crystal aggregates.
^"hree weeks of regeneration. Bar = 20 nm.
724
REGENERATED TEGULA SHELL
725
FIGURE 5. Spherulites formed of radiating clusters of needles. Three weeks of regeneration. Bar
= 10 Mm.
FIGURE 6. Regenerated shell with dumbbell-shaped crystal aggregates (D). Note the organic matrix
(M). Two months of regeneration. Bar = 100
in the mantle, foot, and hepatopancreas during shell regeneration in this marine snail
have been reported (Reed-Miller, 1983). The present study was undertaken to describe
the ultrastructure of regenerated shell in Tegula, and to outline a possible mechanism
for the crystal formation.
Preliminary accounts of this work were presented to the American Society of
Zoologists (Reed-Miller, 1981b; 1982) and to the American Malacological Union.
MATERIALS AND METHODS
Tegula funebralis and Tegula eiseni were obtained from the Pacific Biomarine
Laboratories, Inc., Venice, California. They were maintained in aquaria in filtered,
aerated sea water from the Gulf of Mexico (32 ppt) at 15°C. The animals were fed
marine algae from a laboratory culture.
A 4 mm2 section of shell was carefully removed from the first body whorl of the
shell using a Dremel "Moto-tool," jeweler's saw and a triangular file. Care was taken
not to injure the underlying tissue. The opening in the shell, or window, was covered
with a small piece of plastic coverslip, and covered with warm dental wax, sealing
the window from the external environment.
The regenerated material was removed from the animals (the procedure follows
below) at intervals from 6 hours to 6 months after the shell window was cut. These
were six, 12, and 18 hours; one, two, three, seven, and ten days; two weeks; and then
on a weekly basis up to six months. The experiments were repeated three times with
at least four experimental animals examined each time.
FIGURE 7. The edge of the shell window showing spindle-shaped crystals dotting the normal shell
(S) and forming the regenerated shell (R). One month of regeneration. Bar = 10 ^m.
FIGURE 8. An area of the shell near the window (W) showing the clustering of spindle-shaped crystals
into spherulites. Arrow points to one large spherule. The regenerated shell has been removed. Two months
of regeneration. Bar = 100 ^m.
FIGURE 9. Inside of the shell at the juxtaposition of regenerated shell (R) and normal shell (S). Note
the pavement of small crystals obscuring the normal shell. 2'/2 months of regeneration. Bar = 100 nm.
FIGURE 10. Edge of the shell window viewed from the inside of the shell with the regenerated material
726
REGENERATED TEGULA SHELL
727
FIGURE 1 1. Small polygonal crystallites that were occasionally seen on the normal shell surrounding
the shell window. Three months of regeneration. Bar = 4 ^m.
FIGURE 12. Elongated trends of needle clusters that were seen on the normal shell bordering the
window. Three months of regeneration. Bar = 5
Scanning electron microscopy
The soft parts were removed from the shell, and the shell was preserved in 70%
ethanol, until it was prepared for scanning electron microscopy. The shell was then
carefully cut around the window with a rotary rock saw until a small frame of shell
(about 3 mm wide) surrounded the window on all sides. This frame and the shell
window with the regenerated material were rinsed with distilled water and air dried.
The samples were mounted on aluminum scanning electron microscopy stubs with
nail polish, and coated with 100-200 A of gold-palladium (60:40), using an E5100
Polaron Sputter Coaler. The material was observed with a Cambridge S4-10 scanning
electron microscope operated at 20 kV.
RESULTS
Most of the regenerated shell of Tegula was built up from spindle-shaped crystals
associated with an organic matrix (Fig. 1 ). The spindles were made up of smaller,
elongated crystallites (Fig. 2). The doubly-pointed crystallites grew and formed radiating
(R) on the left. Note the large radial clusters of crystals emanating from the normal shell (S) on the right.
Arrows show some of the contacts between the regenerated shell and growth from the normal shell. Organic
matrix is visible overlying some of the regenerated shell in the window. Two months of regeneration. Bar
= 100 ^m.
FIGURE 13. The edge of the shell window showing radiating clusters of rod-shaped crystals on the
shell (S) surrounding the window. The regenerated shell (R) has separated from the normal shell in this
micrograph. Four months of regeneration. Bar = 20 ^m.
FIGURE 14. Higher magnification of rod-shaped crystals similar to those shown in Figure 13. The
organic matrix has collapsed over the tops of some of the crystals. Four months of regeneration. Bar = 10 ^m.
FIGURE 15. An area of shell where the nacreous layer was fractured during the removal of the shell
idow. Small crystallites pave the surface of the fractured shell, and the hexagonal outlines of the nacre
:s are visible underneath them. The shell window is just out of view at the top of the figure. Three
>nths of regeneration. Bar = 20 nm.
728
REGENERATED TEGULA SHELL 729
clusters or rosettes (Fig. 3) that eventually coalesced into a mineralized sheet (Fig.
4). Spherules formed of needle-like crystals were also observed (Fig. 5). After about
two months of regeneration, dumbbell-shaped crystal aggregates and spherulites as-
sociated with an organic matrix were predominant on the surface of the regenerated
material (Fig. 6).
Frequently, mineral was deposited on the normal shell surrounding the shell
window. This occurred on the edges of the window both inside (next to the mantle)
and outside of the shell. Typically the crystals were small and spindle-shaped (Fig.
7), and covered about one to two mm of the normal shell bordering the window (Fig.
8). The area of shell next to the mantle usually showed thicker deposition than the
region on the outside of the shell (Fig. 9). After about two months of shell regeneration,
crystals of regenerated shell inside the window and crystals growing from the frame
around the window made contact in some places (Fig. 10).
The microstructure of the crystallites deposited on the normal shell varied somewhat
from the doubly-pointed crystals described for regenerated shell. Polygonal (Fig. 1 1 )
and elongated aggregate needles (Fig. 12) were common. After four months of re-
generation, the crystals on the edge of the shell window were large and rod-shaped,
and were usually assembled in radiating clusters (Fig. 13). The rods were composed
of smaller subunits (Fig. 14).
Figure 1 5 shows crystal deposition over nacreous shell. Small crystallites dot the
shell, and outlines of the nacre tablets are discernable. The "c" axes of the crystallites
deposited along the edges of the normal nacre tablets are slightly more elongated than
those axes of the crystallites deposited on the more central regions of the tablets (Fig.
16). The regenerated shell attained a normal ultrastructural appearance after at least
four months of regeneration.
DISCUSSION
There is a striking resemblance between the crystalline structures reported in this
paper and structures described in other molluscan shells and for inorganically pre-
cipitated aragonite. This similarity has led to a four part hypothesis for the phases
of shell regeneration in Tegula.
1. Aragonitic needles are precipitated from a carbonate-rich solution onto an
organic matrix where they grow and form regenerated shell. The regenerated shell of
Tegula was built up from aragonitic needle clusters that formed dual tapered crystal
spindles. According to this part of the hypothesis, however, regeneration involves
precipitation from a solution, and the exact area of deposition may not be limited
to the shell window. This was found to be the case for Tegula. Crystallites were found
on a small region of the nacreous shell bordering the window. The crystallites were
typically smaller than the underlying nacre tablets, and in some cases, appeared to
conform to or be guided by the pattern imposed by the shape of the individual nacre
tablets (See Figs. 15, 16). Schroeder (1973) examined Pleistocene gastropod shells
and found that apparently inorganically precipitated aragonite needles lined the in-
teriors of the shells. The needles were oriented in two directions, determined by the
underlying crossed lamellar shell structure. Meenakshi et al. (1974a) showed that the
substrate microtopography influenced calcification patterns during shell regeneration
in Ota/a lactea, a land snail. Alexandersson ( 1 974) stated that even during inorganic
FIGURE 16. Higher magnification of Figure 15 showing elongate "c" axes of the crystallites on the
edges of the nacre tablets. Three months of regeneration. Bar = 2
730 C. REED-MILLER
precipitation, the organic matrices and matrix derivatives have some control over
the form of skeletal carbonates.
The crystals described in and around the regenerated shell closely resemble the
morphology of inorganically precipitated aragonite crystals (See Ginsburg and Schroe-
der, 1973 for a description of inorganically precipitated aragonite). In fact, Wind and
Wise (1976) noted in their study of spine mineralization in the archaeogastropod
Guildifordia triumphans, that it was virtually impossible to determine where organically
precipitated aragonite ended and inorganically precipitated aragonite began. Note
Figures 7, 9, and 10 in this paper which are micrographs of mineralization close to
and around the edge of the shell window. It is impossible to discern whether these
crystals are of an organic or an inorganic origin. Similar aragonitic crystals have been
described filling in and lining gastropod shells in cup shaped algal reefs (Ginsburg et
al, 1971; Schroeder, 1972a, b; Ginsburg and Schroeder, 1973), forming the skeletons
of one order of green algae (Marszalek, 1971), and as algal cement (Alexanders-
son, 1974).
2. The needles aggregate to form doubly-pointed bundles, or spindle-shaped crystals
associated with an organic matrix. Spindle-shaped crystals have been described in
the regenerated shell of other molluscs. For example, Blackwelder and Watabe (1977)
and Meenakshi et al. (1974b) reported the occurrence of spindle-shaped crystals in
the regenerated shells of the freshwater gastropod, Pomacea paludosa, and the cepha-
lopod, Nautilus macromphalus. In addition, crystals morphologically similar to those
described in the regenerated shells have been described in calcified byssi of the bivalve,
Anomia simplex, and on the surface of the lithodesma of another bivalve, Lvonsia
floridana (Prezant, 1982).
The random orientation of the crystal spindles in the regenerated shell of Tegula
parallels the description of the formation of the growth stops and spine diaphragms
in Guildifordia triumphans (Wind and Wise, 1976). These authors pointed out that
the unpatterned disposition of the spindles indicated that they probably began forming
in the extrapallial fluid, and settled at random.
3. The spindle-shaped crystals form spherules in one of two ways as outlined by
Watabe ( 198 1 ). First, by additional growth, the spindles become grouped into stellate-
or rosette-shaped aggregates that eventually become spherules. Rosettes of spindle-
shaped crystals were a prominent component of the regenerated shell in Tegula. They
coalesced to form a mineralized sheet in the shell window. Spherulitic aggregates of
crystals have been observed in other molluscs where shell is being filled in or repaired.
Wind and Wise (1976) describe "elongate trends of radiating aragonite needle clusters"
filling in the spine cavities of Guildifordia triumphans, and Watabe (1981, Fig. 5)
showed spherulites of aragonite formed during early shell regeneration in the terrestrial
snail, Cepaea nemoralis. Moreover, these aggregate crystals have been found in the
normal shells of the archaeogastropod, Cittarium pica (Wise and Hay, 1968a, b;
Erben, 1971).
The second possibility for the mechanism of spherule formation is by the addition
of needles to the ends of the spindle-shaped crystals, forming a dumbbell shape.
Filling in the midregion of the dumbbell with more needles would result in radial
development and spherule formation. After about two months of shell regeneration,
large dumbbell-shaped crystals as well as spherules were evident in the regenerated
shell of Tegula. These crystal structures were also evident in the regenerated shell of
Mytilus edulis, a marine bivalve, and Pomacea paludosa, a freshwater snail (Uozumi
and Suzuki, 1979; Blackwelder and Watabe, 1977).
The results of the present study indicate that the stellate- or rosette-shaped clusters
f crystal spindles occur during early shell regeneration, and the dumbbell-shaped
REGENERATED TEGULA SHELL 731
aggregates are present during a later stage of regeneration. This is not a definitive
statement for all shell regeneration, but examples such as those ofMytilus and Pomacea
show that these crystal types can occur under a wide range of conditions in regener-
ated shell.
4. Finally, the crystals derived from the rosette-like or the dumbbell-shaped crystal
aggregates are closely apposed, and competitional growth results in their coalescence
and the formation of a spherulitic prismatic type of shell layer. Micrographs of the
regenerated shell after at least three months of regeneration show this type of layer
in Tegula (Reed-Miller, unpub.). This shell structure has also been seen in the re-
generated shell of Pomacea paludosa (Blackwelder and Watabe, 1977), the shells of
Cittarium pica (Wise and Hay, 1968a, b; Erben, 1971), Nautilus (Erben et al,
1969; Mutvei, 1972; Meenakshi et al., 1974b), and in bivalve ligaments (Mano and
Watabe, 1979).
In summary, the sequence of changes throughout shell repair in Tegula is as
follows. The initial crystal deposition occurs in association with an organic matrix
and appears as small, spindle-shaped crystals formed by the aggregation of needle-
like subunits. The spindles are frequently aggregated into stellate clusters that coalesce
to form a sheet of mineralized tissue. After about two months, dumbbell-shaped
crystal aggregates and spherulites are apparent on the surface of the regenerated shell.
A normal shell structure is present after at least four months of regeneration. Crystal
deposition also occurs on the normal shell surrounding the window.
A salient question arises from this study. What degree of control does an animal
have over the microarchitecture of regenerated shell if at any stage the ultrastructural
appearance is similar to that described for inorganically precipitated mineral?
ACKNOWLEDGMENTS
This work was supported by N.I.H. Grant #DE05491. William I. Miller, III pro-
vided excellent assistance with the scanning electron microscopy. Dennis Cassidy
graciously allowed me to use the darkroom in the Antarctic Research Facility in the
Department of Geology at FSU. I am grateful to Dr. S. W. Wise, Jr. for helpful
discussions of the interpretation of some of the micrographs. This is contribution
number 2 1 1 from the Tallahassee, Sopchoppy and Gulf Coast Marine Biological
Association.
LITERATURE CITED
ALEXANDERSSON, T. 1974. Carbonate cementation in coralline algal nodules in the Skagerak, North Sea:
Biochemical precipitation in undersaturated waters. J. Sediment. Petrol. 44: 7-26.
BLACKWELDER, P. L., AND N. WATABE. 1977. Studies on shell regeneration. II. The fine structure of normal
and regenerated shell of the freshwater snail, Pomacea paludosa. Biomineralization 9: 1-10.
ERBEN, H. K. 1971. Anorganische und organische schalenkompenten bei Cittarium pica (L.) (Archaeo-
gastropoda). Biomineralization 3: 51-64.
ERBEN, H. K., G. FLAJS, AND A. SIEHL. 1969. Die fruhontogenetische Entwicklung der schalenstrucktur
ectocohleaten Cephalopoden. Paleontogr. Am. 132: 1-54.
GINSBURG, R. N., D. S. MARSZALEK, AND N. SCHNEIDERMANN. 1971. Ultrastructure of carbonate cements
in a Holocene algal reef of Bermuda. J. Sediment. Petrol. 41: 472-482.
GINSBURG, R. N., AND J. H. SCHROEDER. 1973. Growth and submarine fossilization of algal cup reefs,
Bermuda. Sedimentology 20: 575-614.
MANO, K., AND N. WATABE. 1979. Scanning electron microscope observations on the calcified layers of
the ligament ofMercenaria mercenaria (L.) and Brachydontes exustus (L.). Pleycypoda. Biominer-
alization 10: 25-32.
MARSZALEK, D. S. 1971. Skeletal ultrastructure of sediment producing green algae. Pp. 273-280 in Scanning
Electron Microscopy 1971 (part 1), O. Johari, ed. IIT Research Institute, Chicago.
732 C. REED-MILLER
MEENAKSHI, V. R., G. DONNAY, P. L. BLACKWELDER, AND K. M. WILBUR. 1974a. The influence of
substrata on calcification patterns in molluscan shell. Calcif. Tissue Res. 15: 31-44.
MEENAKSHI, V. R., A. W. MARTIN, AND K. M. WILBUR. 1974b. Shell repair in Nautilus macromphalus.
Mar. Biol. 27: 27-35.
MUTVEI, H. 1972. Ultrastructural studies on cephalopod shells. Part I. The septa and siphonal tube in
Nautilus. Bull. Geol. Inst. Univ. Ups. 8: 237-261.
PREZANT, R. S. 1982. Inorganic and biogenic deposition of calcium carbonate in molluscan accessory
structures. Am. Zool. 22: 982.
REED-MILLER, C., S. W. WISE, JR., AND H. J. SIEGEL. 1980. Correlated techniques for identifying mineral
phases and morphologies in regenerated archaeogastropod shell: A progress report. Pp. 73-76 in
The Mechanisms of Biomineralization in Animals and Plants, M. Omori and N. Watabe, eds.
Tokay University Press, Tokyo.
REED-MILLER, C. 198 la. Shell structures in five species of the marine snail, Tegula. Pp. 243-251 in Scanning
Electron Microscopy 1981, IV, O. Johari, ed. IIT Research Institute, Chicago.
REED-MILLER, C. 1981b. Transmission and scanning electron microscopy of gastropod shell regeneration.
Am. Zool. 21: 1037.
REED-MILLER, C. 1982. Crystalline and amorphous calcium in the tissues of shell regenerating snails. Am.
Zool. 22: 982.
REED-MILLER, C. 1983. The initial calcification process in shell-regenerating Tegula (Archaeogastropoda).
Biol. Bull. 165: 265-275.
SALEUDDIN, A. S. M., AND K. M. WILBUR. 1969. Shell regeneration in Helix pomatia. Can. J. Zool. 47:
51-53.
SCHROEDER, J. H. 1972a. Calcified filaments of an endolithic alga in recent Bermuda reefs. Nens. Jb. Geol.
Palaont. Mh. 1: 16-33.
SCHROEDER, J. H. 1972b. Fabrics and sequences of submarine carbonate cements in Holocene Bermuda
cup reefs. Geol. Rdsch. 61: 708-730.
SCHROEDER, J. H. 1973. Submarine and vadose cements in Pleistocene Bermuda reef rock. Sediment.
Geol. 10: 179-204.
UOZUMI, S., AND S. SUZUKI. 1979. "Organic membrane-shell" and initial calcification in shell regeneration.
J. Fac. Sci. Hokkaido Univ. Ser. IV. 19: 37-74.
WATABE, N. 1981. Crystal growth of calcium carbonate in the invertebrates. Pp. 99-147 in Inorganic
Biological Crystal Growth, B. R. Pamplain, ed. Pergamon Press, Oxford.
WILBUR, K. M. 1972. Shell formation in molluscs. Pp. 103-145 in Chemical Zoology. Vol. VII, Mollusca,
M. Florkin and B. T. Scheer, eds. Academic Press, New York.
WIND, F. H., AND S. W. WISE, JR. 1976. Organic vs. inorganic processes in archaeogastropod shell min-
eralization. Pp. 369-387 in The Mechanisms of Mineralization in the Invertebrates and Plants,
N. Watabe and K. M. Wilbur, eds. Univ. South Carolina Press, Columbia.
WISE, S. W., JR., AND W. W. HAY. 1968a. Scanning electron microscopy of molluscan shell ultrastructure.
I. Techniques for polished and etched sections. Trans. Am. Microsc. Soc. 87: 411-418.
WISE, S. W., JR., AND W. W. HAY. 1968b. Scanning electron microscopy of molluscan shell ultrastructure.
II. Observation of growth surfaces. Trans. Am. Microsc. Soc. 87: 419-430.
WONG, V., AND A. S. M. SALEUDDIN. 1972. Fine structure of normal and regenerated shell of Helisoma
duryi duryi. Can. J. Zool. 50: 1563-1568.
Reference: Biol. Bull. 165: 733-744. (December, 1983)
MORPHOLOGY AND GENETICS OF REJECTION REACTIONS
BETWEEN OOZOOIDS FROM THE TUNICATE
BOTRYLLUS SCHLOSSERI
VIRGINIA L. SCOFIELD12 AND LAUREN S. NAGASHIMA
^Marine Biological Laboratory, Woods Hole, Massachusetts 02543, Hopkins Marine Station of Stanford
University, Pacific Grove, California 93950. and Laboratory of Experimental Oncology, Department of
Pathology, Stanford University School of Medicine, Stanford, California 94305
ABSTRACT
Botryllus rejection reactions were followed in pairs of oozooids placed together
immediately after initiation of metamorphosis. Within twelve hours, both compatible
and incompatible oozooid pairs underwent tunic fusion and initiation of ampullar
tip-to-side contact. Vascular fusion followed within two days between compatible
pairs, while the fusion sequence was interrupted in the incompatible pairs by a rapid
cytotoxic rejection response. Events occurring within and outside the ampullae in
rejections were effector responses whose consequences were separation of the ampullae
and isolation of the involved tissues from the bodies of the oozooids. Genetics ex-
periments suggested that the four distinct types of rejection reflect a hierarchy of
histoincompatibility in this system.
INTRODUCTION
Recent interest in colonial tunicates has centered around the phenomenon of
colony specificity, which is the capacity for self-nonself distinction leading to fusion
or rejection between colonies. In Botryllus, this histocompatibility discrimination is
controlled by a single multiallelic Mendelian locus (Oka and Watanabe, 1960; Sab-
badin, 1962) that resembles loci within the vertebrate major histocompatibility com-
plex, or MHC (Scofield et ai, 1982a). We have undertaken studies to determine
whether genes controlling allogeneic recognition in Botryllus are homologous to those
within the MHC. To complement our molecular studies, we have examined Botryllus
rejection responses in live preparations of rejecting oozooids, using differential in-
terference contrast (Zeiss-Nomarski) microscopy.
Botryllus colonies are clones of individuals, or zooids, enclosed in a common
tunic. Each zooid is parabiosed to all the others through a colonial vascular network
that is terminated at the colony periphery by bulbous ampullae. Tanaka and Watanabe
(1973) have shown that fusions and rejections between colonies are contact responses
between their ampullae. All the individuals in a colony arise by budding from the
"founder" individual, or oozooid, that is established by metamorphosis of a swimming
tadpole-like larva. Oozooids possess eight microampullae, and paired oozooids undergo
vascular fusions and rejections similar to those occurring between grown colonies
(Scofield et ai, 1982a, b).
When the separated growing edges of the same Botryllus colony meet, the tunic
(test) dissolves, and the opposite ampullae interdigitate to form tip-to-side contacts
(Tanaka and Watanabe, 1973). This sequence of events is part of the morphogenetic
"program" that establishes blood flow between the contacted blood vessels (Katow
Received 15 March 1983; accepted 23 September 1983.
"" To whom correspondence should be addressed.
2
733
734 V. L. SCOFIELD AND L. S. NAGASHIMA
and Watanabe, 1980). Fusion also proceeds, without interruption, between colonies
sharing at least one allele at the fusion locus (Oka and Watanabe, 1960). For colonies
sharing no fusibility alleles, however, the fusion sequence is aborted after ampullar
contact, and is followed by a cytotoxic rejection (Tanaka and Watanabe, 1973).
Recently, Taneda and Watanabe ( 1982a, b, c) firmly established that the allorecognition
elements that allow fusion, or cause rejection, are humoral and cellular elements in
the blood. To outline the sequence of cellular events that follow allorecognition and
lead to a completed rejection response, we followed rejections in vitro in paired
oozooids.
In other invertebrates (Ivker, 1972), as in vertebrates (Gotze, 1977), polymorphic
histocompatibility gene systems show a hierarchy, manifested by differences in the
timing and severity of rejection responses that depend upon particular alleles possessed
by the contacted cells. To determine whether such a hierarchy exists for Botryllus
fusibility alleles, we subjected colonies to different kinds of genetic crosses, and scored
rejection "types" for the offspring.
MATERIALS AND METHODS
Colonies of B. schlosseri were gathered from the Eel Pond in Woods Hole, Mas-
sachusetts, and maintained with constant aeration in beakers of filtered sea water.
Tadpole larvae were gathered by placing coverslips along the waterline inside the
beakers, where they attached and underwent metamorphosis to form natural pairs.
For time-lapse observations, coverslips carrying oozooid pairs were inverted over
a drop of sea water on a glass microscope slide. Observations were made using Zeiss-
Nomarski optics. Between observations, the coverslips were cultured in their original
beakers.
Genetic crosses were carried out in the same beakers that were used to culture
single colonies. Three sets of experiments were done. First, colonies already carrying
developing embryos ("wild-fertilized" colonies) were collected and cultured until the
developing tadpoles hatched and metamorphosed to form oozooid pairs. For "defined"
crosses, pairs of colonies were placed in beakers, where eggs of one colony were
fertilized only by sperm from the crossing partner. For "self" crosses, colonies were
isolated and self-fertilizations were allowed to proceed (Scofield el ai, 1982a). After
each cross, colonies containing fertilized eggs were cultured in isolation until the F,
larvae hatched.
RESULTS
The thin oozooid preparations allowed easy visualization of ampullar junctions
under the microscope. Rejections and fusions occurred readily between paired oozooids
within two days of hatching and metamorphosis. Within 12-24 hours of contact,
blood flow was established and connecting vessels formed between fusible oozooids
(Fig. 1A). Likewise, rejection reactions usually were completed by one day after
initiation of ampullar contact. The characteristic feature of oozooid rejections was a
bright golden-brown necrotic zone (Fig. IB).
Figure 2A shows a normal ampulla photographed at its point of attachment to
an inverted glass coverslip. The surfaces of the "tip" cells are smooth, and the sur-
rounding tunic contains only the interconnected "test cells" (Fig. 2A). In rejecting
oozooid pairs, by contrast, the ampullae and the surrounding tunic showed striking
alterations. After a period of tip-to-side contact, ampullar reseparation was followed
rapidly by movement of blood cells through the "tipping" ampullar tip into the tunic
(Fig. 2B). Closer examination of the cytotoxic mass revealed concave "holes" in the
BOTRYLLUS REJECTION REACTIONS
735
B
1mm
FIGURE 1 . Fused and rejected Botryllus oozooid pairs. A. Fused oozooids, showing the connecting
blood vessels (arrows) at the site of a prior tip-to-side contact. B. Rejected oozooids, showing the necrotic
zone and an autoamputated ampulla (arrows).
tip cells (Fig. 2C). In some pairs, blood flow inside the involved ampullae slowed to
a stop. Emboli broken from these clotted masses frequently plugged the proximal
end of the ampulla (Fig. 3A) at sites where amputation eventually occurred (see
below). Examination of the blood cells released into the tunic revealed that the first
to appear there had the distinct berry-like appearance of morula cells (Fig. 3B). Their
vacuoles had turned a dark brown. After deposition into the tunic, morula cell dis-
integration was accompanied by condensation of fibers at the site (Fig. 2C).
Other morula cells, morula cell precursors (signet-ring and compartment cells),
and granular amoebocytes were shed into the tunic as the ampullae retreated from
the contact point. These, however, remained transparent by transmitted light (Fig.
2C). Some developed processes and moved away from the rejection site, while others
contributed to the cytotoxic mass (Fig. 2C). It is clear from Figures 2 and 3 that
736
V. L. SCOFIELD AND L. S. NAGASHIMA
amp
tc
B
BOTRYLLUS REJECTION REACTIONS 737
rejection reactions following allogeneic contacts in Botryllus are extremely destructive
to surrounding tissues.
A surprising finding was that different oozooid pairs from the same hatching gave
very different rejection responses. Although necrotic regions always appeared between
rejected oozooids, a striking difference in timing of rejection events and gross ap-
pearance of the cytotoxic lesion became evident after examination of many pairs.
The time between establishment of ampullar junctions and rejection was highly vari-
able, ranging between about 30 minutes and approximately 12 hours. Completed
rejection responses could be placed into one of four categories (1-4; Fig. 4, Table I)
that were distinguished easily by reflected light. The several forms taken by oozooid
rejections in this study appear in Figure 4.
"Type 1" rejections showed very slight bleeding from the "tipping" ampulla
following ampullar reseparation. In most instances, careful inspection of the retreating
ampullae was necessary to visualize the few golden-brown cells bled from their tips.
This sometimes was accompanied by visible "sticking" of the rounded tip cells onto
the "side" ampulla at the prior contact site (Fig. 4A). The "type 2" response was a
more extensive bleeding of the "tipping" ampulla, again with the ampulla itself re-
maining sealed and generally intact. In both these types of bleeding responses, the
characteristic brown color reaction was seen in the rejection lesion, but not within
the ampullae.
The third type of rejection (type 3) was autoamputation (Fig. 4B), occurring with
or without ampullar bleeding from the tip. For these rejections, the entire amputated
ampulla, and its contents, turned brown. The rejection type "4" was ampullar dis-
integration, where the ampullar contents and epithelium became part of the colored
rejection mass (Fig. 4C).
It appeared that these rejection types represented a continuum of responses, with
the differences being a function of the extent to which the ampullae moved through
the fusion sequence before it was aborted (for example, a brisk response to a rapid
allorecognition might account for both the minor and more extensive bleeding re-
sponses, while more extreme ampullar reactions — amputation or disintegration —
would result from more extensive mixing of allogeneic blood elements).
Because the distinct responses seemed to reflect different thresholds for "effective"
allorecognition, we proposed that these differences actually reflect a nested hierarchy
of histoincompatibility for the many fusibility alleles. If so, different oozooid pairs
with the same combinations of fusibility haplotypes would be expected to give the
same kind of rejection response. An oozooid microassay (Scofield el ai, 1982a) was
used to test this hypothesis with different genetic crosses (Fig. 5).
Since wild colonies usually are heterozygotic at one Mendelian gene for fusion
at which there are many alleles (Oka and Watanabe, 1960), any given colony can be
named AB at this locus (Fig. 5, top) and the diploid progeny of that colony will be
A or B with respect to the maternal fusibility allele (Scofield et ai, 1982a). Certain
predictions can be made regarding genotypes of F, oozooid pairings that give rejections.
FIGURE 2. Anatomy of normal ampullae and of ampullae participating in rejection reactions. A.
Normal ampullar tip (amp), surrounding tunic, and test cells (tc) in a Botryllus oozooid. The ampullar tip
cells are columnar, vacuolated, and tightly interconnected. B. Ampullar withdrawal following a rejecting
tip-to-side contact. Blood cells (be) can be seen moving from the "tipping" ampulla (top) into the tunic
where they undergo cytotoxic interactions and cause fiber deposition. The "side" ampulla (bottom) also
is filled with clumped blood cells. C. Changes in ampullar tip cells after a rejecting tip-to-side contact.
Rounded "holes" or depressions (dep) appear in the tip cells of the interacting ampullae. The brown fibrous
barrier deposited by blood cells in the tunic (Jb) appears at the top. A and B, X500. C, X1000.
738
V. L. SCOFIELD AND L. S. NAGASHIMA
A
FIGURE 3. Effector responses in oozooid blood vessels and tunic after a rejection reaction. A. Embolus
I :'mb) of clotted blood cells and fibers preventing backflow of blood through the proximal end of an ampulla
•licipating in a rejection response. B. Ferrocytes (fc), deposited into the tunic from the ampullar tip at
:ht, have vacuoles which have turned dark red-brown. XlOOO.
BOTRYLLUS REJECTION REACTIONS 739
For example, if a colony is fertilized in the natural environment by sperm from many
different colonies, the randomly combined A and B oozooids yield 50% fusing and
50% rejecting pairs (the chance that any two share a paternal allele is small. Fig. 5,
left). Because many different sperm fertilize in such "wild" crosses, rejections between
the progeny oozooids involve many different allelic combinations (there are 50-100
fusibility alleles in natural populations — Schlumpberger and Scofield, unpub.). If
rejection type depends upon fusibility alleles, then the pairs of rejected progeny from
these wild crosses should show some frequency distribution of all four rejection types
(Fig. 5, left).
If the same maternal AB colony is crossed by only one other colony, CD (Fig.
5, center) then only C and D sperm fertilize; the F, progeny are of four types, and
that 25% of the progeny pairs which reject are of only two haplotypic combinations.
Thus only one or two different rejection types should be represented in the paired
progeny of a "denned" cross. If the AB colony is self-crossed, on the other hand, the
rejected 12.5% of the experimental pairs are of only one allelic combination (Fig. 5,
right); thus only one rejection type should be found.
Results from these experiments appear in Table I. As expected, wild-fertilized
colonies yielded oozooids which, when paired, showed 50% fusions and 50% rejections
(numerical data not shown; Fig. 5, left). Rejection types were distributed fairly evenly
over all four categories. The denned crosses, on the other hand, hatched progeny
whose pairs gave 75% fusions and 25% rejections (Fig. 5, center). The rejected pairs
from these crosses generally showed only one or two rejection types; exceptions were
seen only in the progeny of two crosses (denned crosses 6 and 7) where ampullar
amputation in some pairs was accompanied by bleeding from the tips. Self-crossed
colonies produced progeny whose pairs gave very few rejections, both because they
represent only 12.5% of the total pairs (Fig. 5, right) and because inbreeding depression
reduces the total number of hatched larvae (Sabbadin, 1971; Scofield et al, 1982a).
However, those rejections were all of one type in three experiments (Table I).
DISCUSSION
Fusions and rejections between Botryllus oozooids appear to be similar to those
occurring between adult colonies (Tanaka and Watanabe, 1973; Katow and Watanabe,
1980). For incompatible pairs, the ampullae move into position for fusion, as they
do for compatible pairs, but the process is interrupted abruptly by a cytotoxic effector
cascade.
In the present study, large holes were observed in the tip cells of rejecting oozooid
ampullae (Fig. 2C). Whether these were formed as part of the aborted fusion sequence
(and perhaps were the means by which blood exchange leading to rejection was made)
or were released endocytotic vacuoles transporting blood cells into the tunic (DeSanto,
1968) remains to be determined. In these cases, however, blood cell stasis and clumping
became apparent soon after ampullar contact was established (Fig. 4A-C). This suggests
that blood exchange of some kind must occur before rejection can begin, and, indeed,
the first result of contact between ampullae (compatible or incompatible) appears to
be tip cell alteration. Electron-microscopic examination of fusing ampullar junctions
has revealed "fenestrations" in the tip cells (Katow and Watanabe, 1980). After
rejections, likewise, India Ink injected into a retreating ampulla was shown to leak
through the tip cells into the tunic (Taneda and Watanabe, 1982c). The results of
the rejection reaction activated by mixing of allogeneic blood elements are: ( 1 ) rapid
isolation of the involved structures, and (2) eventual reseparation of the allogeneic
colonies.
740
V. L. SCOFIELD AND L. S. NAGASHIMA
BOTRYLLUS REJECTION REACTIONS
74;
TABLE I
Percentages of each of four types of rejection (Types 1-4) in the paired F, progeny of colonies fertilized
(1) by many different paternal colonies in the natural environment ("wild-fertilizations". Fig. 5, left), (2)
by one paternal colony ("defined" crosses, Fig. 5, center) and (3) by self sperm ("self". Fig. 5, right)
Rejection type
Crosses'
n (pairs)
1
2
3
4
Wild
50
13
13
11
13
46
13
11
9
13
23
5
6
4
8
50
15
14
9
12
Defined
5
1
4
3
1
2
8
1
7
11
11
6
6
16
8
5
3
5
1
3
1
5
4
1
9
6
3
6
4
2
Self
3
3
2
2
5
5
1 Rejections are typed as ( 1 ) slight bleeding; (2) severe bleeding; (3) ampullar autoamputation; and (4)
ampullar disintegration. For details, see text and Figure 4.
Botryllus provides one of only two known examples of genetically controlled
allorecognition and response in tunicates. As shown for the solitary tunicate Halo-
cynthia (Fuke, 1980; Fuke and Numakunai, 1981), allogeneic mixtures of Botryllus
blood cells undergo rapid contact-mediated cytolysis (Scofield, in prep.)- The most
striking features of the in vitro and in vivo reactions between allogeneic Botryllus
blood cells are their speed, the lack of requirement for an induction period, and the
characteristic golden-brown color of the rejection lesion itself. The best clues as to
cellular mechanisms for Botryllus alloreactivity may come from recent studies with
vertebrates.
In mammals, a class of natural killer (NK) cells has been described (Herberman,
1982). Such cells have native capacities for rapid, nonimmune recognition and killing
of cells of certain tumor lines, and may play a role in rejection of transplanted
allogeneic blood cells (Rolstad el al, 1983). Like neutrophils and monocytes, but
unlike cytolytic T lymphocytes, vertebrate NK cells appear to employ reduced oxygen
FIGURE 4. Different types of cytotoxic response following a rejection reaction. A. Bleeding after
partial completion of the fusion sequence. Blood cells are moving out of the "tipping" ampulla as it retreates
(above right, amp). The site of prior tunic fusion (//) and tip-to-side contact (ts) are marked clearly. B.
Ampullar autoamputation following a rejecting tip-to-side contact. The pinching site is surrounded by
morula cells (me). C. Ampullar disintegration at the site of a prior tip-to-side contact (ts). Morula cells
(me) are adhered to the blood vessels of the "side" ampulla (top). X1000.
742
V. L. SCOFIELD AND L. S. NAGASHIMA
AB Colony
eggs A or B
Ax y
diptoid oozooids
Bx y
Paired oozoids:
l
Wild fertilization
x y=C,D,E n
Ax y
B. y
Defined cross (ABxCD)
x y^CorD
AC AO BC BD
Self cross (AB*AB)
B. y
AC
AD
BC
BD
Re,
AB
BB "•"
50% fusions
Reject ions = A? to B?
(Many allelic combinations)
75% fusions
Reject ions AC to SP. AD to BC.
(Two allelic combinations)
875% fusions
Rejections^ AA to BB
(One allelic combination)
FIGURE 5. Schematic diagram showing genetic crosses performed in this study, progeny genotype
ratios, and haplotypes represented in rejecting oozooid progeny pairs. Wild colonies are heterozygotic at
one locus for fusibility, at which there are many alleles segregating in natural populations (50-100 in North
American Botryllus species; Schlumpberger and Scofield, unpub.). If the mother colony is designated AB
at the fusion locus (top), and the fertilizing sperm alleles designated x. .y, the oozooid progeny of any
genetic crossing will be Ax. .y or Bx. .y in 1 : 1 proportions. Left: Wild fertilizations: many different fertilizing
sperm (from an unknown number of paternal colonies) fertilize the A and B eggs. If rejection type is
determined by fusibility haplotypes, all four rejection types should be seen in the paired offspring. Center:
Cross-fertilizations: the maternal AB colony is fertilized by sperm carrying one of two fusibility alleles (C
or D); rejections among the paired offspring will be AC to BD or AD to BC: therefore, one or two rejection
types should be seen among the paired F, progeny. Right: Self-fertilizations: the only rejecting haplotype
combination is AA to BB; thus, only one rejection type is expected.
intermediates in their cytolytic pathways (Roder et al, 1982). We have found that
mixed allogeneic Botryllus blood cells release both hydrogen peroxide and ferrous
iron (Poenie and Scofield, in prep.), and that peroxidase appears in the tunic area
around rejecting ampullae (Nynas-McCoy, unpub.). Ascidian morula cells carry the
transition metals vanadium, niobium, or iron (Goodbody, 1974; Rowley, 1982). In
Botryllus, the morula cells contain reduced iron and sulfuric acid (Milanesi and
Burighel, 1979). The morula cells are conspicuous participants in rejection lesions,
where their transparent vacuoles turn a dark red-brown (Fig. 3B). Since this color
reaction may reflect a change in the oxidation state of the contained iron, it is tempting
5 speculate that tunicate transition metals participate in allogeneic effector reactions
y performing a catalytic function. All tunicates have large amounts of bound iodine
BOTRYLLUS REJECTION REACTIONS 743
in the blood and tunic matrix (Barrington, 1975). It is interesting to note, therefore,
that hydrogen peroxide, ferrous sulfate, and potassium iodide together can generate
cytotoxic iodide (I • ) and hydroxyl (OH • ) radicals (Klebanoff, 1982). If tunicate metal
ions initiate free radical-generating reactions, such intermediates could participate,
as they may in vertebrates, in killing of bacteria or allogeneic cells. In Botryllus, for
example, they might polymerize fibers from tunic or blood-borne precursors for
clotting or encapsulation functions. Discovery of such a role(s) for tunicate transition
metals might help to solve the long-standing mystery of their adaptive function (Good-
body, 1974).
Our observation of broad heterogeneity in rejection types in Botn'llus is reminiscent
of findings by Koyama and Watanabe (1982) with Perophora sagamiensis, where two
distinct types of rejection were observed. Our studies suggest that colony specificity
in B. schlosseri occurs on a continuum, where the time required for response varies
for different pairs of interacting alleles. If the hierarchy of alleles in Botryllus reflects
diverse thresholds for initiation of the rejection reaction, such differences might likewise
explain interspecies variations noted by Koyama and Watanabe (1982) for different
botryllid ascidians (see also Scofield, 1983).
The clear differences between the "acute" rejections described in this study and
the "chronic" reactions occurring subsequent to fusion in some colony pairs (Mukai,
1967; Sabbadin and Zaniolo, 1979; Saito and Watanabe, 1982; Taneda and Watanabe,
1982c; Scofield, 1983) offer the intriguing possibility that in Botryllus, as in mammals,
there may be two systems for cellular defense: the rapid, NK-like, native reaction
described above, and a slower, induced response. Many different cell types appear to
participate in in vivo or in vitro reactions between fully allogeneic blood cells (Scofield,
in prep.). By contrast, the "delayed" response to semiallogeneic cells is significantly
attenuated by X-irradiation of the recipient colony, a treatment that affects the numbers
of lymphocyte-like cells (Taneda and Watanabe, 1982c).
It appears that both types of Botryllus allorecognition are controlled in some way
by genes within the fusibility complex. The rapid, "acute" response serves a primary
protective function against allogeneic invasion. Reactions between cells mixed after
fusion, on the other hand, may prevent continued resource sharing between distantly
related colonies that happen to share one fusibility allele. If the induction period for
this response (about two weeks; Taneda and Watanabe, 1982c) reflects the time
required for activation or expansion of clones of specific alloreactive cells, it may be
significant that the genetics of these semiallogeneic reactions resemble those for ver-
tebrate allograft rejection by cytotoxic T lymphocytes.
LITERATURE CITED
BARRINGTON, E. J. W. 1975. Problems of iodine binding in ascidians. Pp. 129-158 in Protochordates,
E. J. W. Barrington and R. P. S. Jefferies, eds. Academic Press, London.
DESANTO, R. S. 1968. The Histochemistry. Fine Structure, and Ecology of the Synthesis of the Test in
Botryllus schlosseri (Pallas) Savigny. Ph.D. Thesis, Columbia University.
FUKE, M. T. 1980. "Contact reactions" between xenogeneic or allogeneic coelomic cells of solitary ascidians.
Biol. Bull. 158: 304-318.
FUKE, M. T., AND T. NUMAKUNAI. 1981. Allogeneic cellular reactions between intraspecific types of a
solitary ascidian, Halocynthia roretzi. Dev. Comp. Immunol. 6: 253-261.
GOODBODY, I. 1974. The physiology of ascidians. Adv. Mar. Biol. 12: 1-149.
GOTZE, D. 1977. The major histocompatibility complex in man and animals. Springer- Verlag, New York.
HERBERMAN, R. B. 1982. Overview and perspectives: natural resistance mechanisms. Pp. 799-788 in
Marcophages and Natural Killer Cells: Regulation and Function, S. J. Normann and E. Sorkin,
eds. Plenum, New York.
IVKER, F. S. 1972. A hierarchy of histoincompatibility in Hydraclinia echinata. Biol. Bull. 143: 162-166.
744 V. L. SCOFIELD AND L. S. NAGASHIMA
KATOW, H., AND H. WATANABE. 1 980. Fine structure effusion reaction in the compound ascidian Botryllus
primigenus Oka. Dev. Biol. 76: 1-14.
KLEBANOFF, S. J. 1982. The iron-H2O2-iodide cytotoxic system. J. Exp. Med. 256: 1262-1267.
KOYAMA, H., AND H. WATANABE. 1982. Colony specificity in the ascidian Perophora sagamiensis. Biol.
Bull. 162: 171-186.
MILANESI, C., AND P. BuRiGHEL. 1979. Blood cell ultrastructure of the ascidian Botryllus schlosseri. I.
Hemoblast, granulocytes, macrophage, morula cell and nephorocyte. Ada Zool. (Stockholm) 9:
135-147.
MUKAI, H., AND H. WATANABE. 1 974. On the occurrence of colony specificity in some compound ascidians.
Biol. Bull. 147:411-424.
OKA, H., AND H. WATANABE. 1960. Problems of colony-specificity in compound ascidians. Bull. Mar.
Biol. Sta. Asamushi. Tohukti Univ. 10: 153-155.
RODER, J. C., S. J. HELFAND, J. WERKMEISTER, R. MCGARRY, T. J. BEAUMONT, AND A. DUWE. 1982.
Oxygen intermediates are triggered early in the cytolytic pathway of human NK cells. Nature
298: 569-572.
ROLSTAD, B., S. POSSUM, B. TONNESEN, AND W. L. FORD. 1983. Athymic nude rats resist graft- versus-
host disease because they reject allogeneic lymphocytes. Transplant. Proc. XV(2): 1656-1658.
ROWLEY, A. F. 1982. The blood cells of dona intestinalis: an electron probe X-ray microanalytical study.
J. Mar. Biol. Assoc. U. K. 62: 607-620.
SABBADIN, A. 1962. La basi genetiche della capacita di fusione fra colonie in Botrvllus schlosseri (Ascidacea).
Rend. Accad. Lincei 32: 1031-1035.
SABBADIN, A. 1971. Self and cross-fertilization in the compound ascidian Botrvllus schlosseri. Dev. Biol.
24: 379-391.
SABBADIN, A., ANDG. ZANIOLO. 1979. Sexual differentiation and germ cell transfer in the colonial ascidian
Botryllus schlosseri. J. Exp. Zool. 207: 289-304.
SAITO, Y., AND H. WATANABE. 1 982. Colony specificity in the compound ascidian, Botrvllus scalaris. Proc.
Jpn. Acad. 58(Ser. B): 105-108.
SCOFIELD, V. L., J. M. SCHLUMPBERGER, L. A. WEST, AND I. L. WEISSMAN. 1982a. Protochordate allo-
recognition controlled by a MHC-like gene complex. Nature 295: 499-505.
SCOFIELD, V. L., J. M. SCHLUMPBERGER, AND I. L. WEISSMAN. 1982b. Colony specificity in the colonial
tunicate Botryllus and the origins of vertebrate immunity. Am. Zool. 22: 783-794.
SCOFIELD, V. L. 1983. Control of fusibility in colonial tunicates. In Cell Receptors and Cell Communication
in Invertebrates, A. H. Greenberg, ed. Marcel Dekker, New York. (In press.)
TANAKA, K., ANDH. WATANABE. 1973. Allogeneic inhibition in a compound ascidian, Botryllus primigenus
Oka. I. Processes and features of "noninfusion" reaction. Cell. Immunol. 1: 410-426.
TANEDA, Y., ANDH. WATANABE. 1982a. Studies on colony specificity in the compound ascidian Botryllus
primigenus Oka. I. Initiation of "nonfusion" reaction with special reference to blood cells infiltration.
Dev. Comp. Immunol. 6: 253-261.
TANEDA, Y., ANDH. WATANABE. 1982b. Studies on colony specificity in the compound ascidian, Botryllus
primigenus Oka. II. In vivo bioassay for analyzing the mechanism of "noninfusion" reaction. Dev.
Comp. Immunol. 6: 243-252.
TANEDA, Y., AND H. WATANABE. 1982c. Effects of X-irradiation on colony specificity in the compound
ascidian, Botryllus primigenus Oka. Dev. Comp. Immunol. 6: 665-673.
Reference: Bio/. Bull. 165: 745-757. (December, 1983)
FEEDING STRUCTURES, BEHAVIOR, AND MICROHABITAT OF
ECHINOCYAMUS PUSILLUS (ECHINOIDEA: CLYPEASTEROIDA)
MALCOLM TELFORD, ANTONY S. HAROLD, AND RICH MOOI
Department of Zoology, University of Toronto, Ontario, Canada, M5S 1A1
ABSTRACT
In the Firth of Lome, Scotland, Echinocyamus pusillus was found most abundantly
in highly variable, poorly sorted substrates at depths of 10-20 m. It was common in
areas exposed to extensive wave and tidal current activity, but absent in fine sediments
in sheltered areas. In size, feeding mechanism, and behavior, the species is highly
adapted for nestling in the interstices between relatively large pebbles. The feeding
mechanism is atypical for clypeasteroids: substrate particles with attached organisms
are selected and transported by the suckered podia. At the mouth, particles are held
in place and slowly rotated by the free margin of the peristomial membrane, while
the teeth strip away diatoms and organic debris. The peristomial membrane and
ciliation of spines and podia are shown in scanning electron micrpgraphs of critical
point dried material. The histology of these structures is described with special reference
to mucus secretion. High resolution SEM micrographs show mucus secreting pores
among the epithelial microvilli of suckered and buccal podia but not in the epithelium
of miliary spines. The suggestion that E. pusillus might represent a sand dollar ancestor
is discussed. The evidence presented supports the view that it is specialized rather
than primitive.
INTRODUCTION
The fibulariids are a family of very small clypeasteroids including two principal
living genera, Fibularia and Echinocyamus. These genera have most often been re-
garded as specialized rather than primitive forms (Clark, 1914; Mortensen, 1948).
The family is thought to be most closely related to the Laganidae and Rotulidae
(Mortensen, 1948;Hyman, 1955; Durham and Melville, 1957, inter alia). That species
of Echinocyamus are specialized has not been accepted universally. Cuenot (1941)
considered E. pusillus to be clearly primitive and proposed that the family Fibulariidae
was the evolutionary point of departure leading to the more advanced clypeasteroids.
Phelan (1977) seems to support this interpretation as does the phylogenetic dendrogram
shown by Durham and Melville (1957). Most recently Ghiold (1982), following a
study of such external structures as spines, podia, and pedicellariae, concluded that
E. pusillus was not a true sand dollar and suggested that it and similar forms "...
may represent an ancestral stage of the sand dollars."
The fundamental body forms of the Clypeasteroida appear to be shaped by hy-
drodynamic forces (Telford, 1981; Telford and Harold, 1982; and Telford, in press)
and by the requirements of their peculiar rocking-sieve feeding mechanism (Goodbody,
1960; Seilacher, 1979; Mooi and Telford, 1982). In assessing the status of Echino-
cyamus, correct interpretation of morphology depends in part on understanding the
chosen habitat and the feeding mechanism. Neither Nichols (1959), in the most
extensive study of the morphology of Echinocyamus, nor Ghiold (1982) were able
Received 17 February 1983; accepted 8 August 1983.
745
746 M. TELFORD ET AL.
to make direct observations of feeding. Therefore, previous knowledge of this process
has been based only on inference. The species has been reported in "shelly gravel"
(Nichols, 1959). Ghiold (1982) made a laboratory study of burrowing and locomotory
activity in different sized sediment particles, but no complete sieve analyses of natural
substrates have been given. Wolff (1968) reported that E. pusillus was abundant in
the North Sea in "relatively coarse sands," with a median grain size of 210-460 nm.
No further details of substrate composition were provided.
We present an account of the feeding mechanism of E. pusillus from direct
laboratory observations. Additional morphological details of the structures involved,
based on scanning electron microscopy (SEM) and histological examination, are
provided. Substrate analysis, SEM examination of natural substrate material and
analysis of gut contents are also presented in an attempt to explain local distribution,
within the Firth of Lome, Scotland.
MATERIALS AND METHODS
Collection
Specimens of Echinocyamus pusillus were collected by dredge with 9 1 .4 cm X 30.5
cm rectangular mouth, fitted with a double layer of 6 mm string mesh, inside a
protective heavy rope mesh and by Petersen grab, 36.8 X 33.0 cm. Animals were
washed from the substrate by gently swirling with water in a plastic bowl, the method
was analogous to gold panning but in this case we retained the lightweight urchins.
Specimens were either fixed immediately or returned live to the laboratory where
they were maintained in natural substrate material washed by constant running sea
water at approximately 4°C.
Live observations
To observe feeding, the methods of Mooi and Telford (1982) were used. Animals
were placed in darkened glass chambers constructed from microscope slides. They
were given a thin layer of natural substrate with the larger particles (>4.0 mm)
removed. Observations were made using a stereomicroscope focussed on an inclined
front-silvered mirror beneath the chamber. Illumination was provided by a fiber optic
light. Substrate particles handled by feeding animals were measured in situ by ocular
micrometer. Gut contents of fresh animals were examined under the light microscope
and those of preserved animals by SEM.
Specimen preparation
All material used in this study was fixed for 12 hours in 2% gluteraldehyde in
filtered sea water. Specimens were then briefly rinsed and stored in 2% formalin in
filtered sea water. Relaxation was difficult but best results were obtained by gradually
transfering specimens to 3.5% Epsom salts in distilled water. Suckered podia were
also well relaxed by propylene phenoxytol-saturated sea water. For histology, specimens
were decalcified in Bouin's solution for 24 to 48 h (Mooi and Telford, 1982). Paraffin
sections were cut at 4 ^m and stained with Milligan's trichrome and Mallory-Hei-
denhain rapid one-step azan for general histology. Toluidine blue and PAS were used
to investigate secretory structures. All histological procedures followed the methods
of Humason (1967). For SEM, whole or dissected specimens were transferred through
a graded series to pure acetone, critical point dried with carbon dioxide in a SORVALL
bomb, and sputter coated with gold in a SEM-PREP II (Nannotech Thin Films,
FEEDING IN ECHINOCYAMUS PUSILLUS
747
England). Substrate particles and gut contents were gently washed in distilled water
to remove salts, strewn on stubs, and freeze-dried before sputter coating.
Substrate
Samples were taken from the grab and dredged material. They were dried at 80°C,
weighed, and then ashed at 400°C for 30 min. After cooling and reweighing, the
samples were passed through screens of mesh size 12.50, 4.00, 2.00, 1.00, 0.50, and
0.25 mm, into a collecting pan. Each fraction was weighed separately and expressed
as a percentage of the total. Organic content, calculated from weight lost during
ashing, was similarly expressed as a percentage of the total dry weight. Repeated
ashing confirmed that organic material was fully oxidized in the initial 30 minute
period. Estimates of shell (biogenic) to mineral (abiogenic) particle ratios were made
by frequency from light microscope observation. Whenever possible, the generic origin
of the biogenic material was noted.
RESULTS
Habitat
Echinocyamus pusillus (4-14 mm length) was collected at several sites in the Firth
of Lome (Fig. 1), at depths of 10 to 200 m. They were most abundant in shallow
water, 10-20 m, and very sparse below about 50 m. We found no E. pusillus living
in sheltered areas such as Loch Creran nor in the lee of islands but they tended to
be common in locations which, according to the West Coast of Scotland Pilot (1949),
FIGURE 1. Collection sites in the Firth of Lome, West Scotland. Solid symbols mark locations where
Echinocyamus pusillus was obtained by dredging; open symbols indicate absence.
748 M. TELFORD ET AL.
are exposed to strong currents. Greatest numbers (over 100 in a 10 minute dredge
haul) were obtained along the exposed side of a promontory on the north side of a
small bay, Camas Nathais (56°29rN, 05°28'W). Six of seven grab samples taken from
this site contained live E. pusillus. Analysis of particle sizes showed the substrate to
be poorly sorted and highly variable (Table I). The Petersen grab tends to bias samples
towards the smaller particle fractions, failing to collect the larger pebbles or rocks,
sometimes returning to the surface empty. Several samples from the same location,
were taken separately from the dredged material, from which unknown amounts of
the finer particles had been lost during collection. A comparison of these with the
grab material is shown in Table I. The fraction >12.5 mm included pebbles up to
100 mm, rarely to 200 mm. Bigger pebbles, often with attached macrophytic algae,
were not included. Estimates of the ratio of shell to mineral particles, in different
fractions of the samples, showed considerable variability (Table II). However, the
shell component tended to be greatest around the 0.50 mm fraction. Maximum
numbers of animals were found when the shell:mineral ratio in this size range was
between 2:1 and 10:1, that is, about 67% to 91% shell material. Identification of
organisms contributing to the shell component was possible in the larger particle
fractions (Table II). Above 0.5 mm the shell comes primarily from the locally dominant
pelecypods and gastropods. Below 0.5 mm, echinoderm spines and forams make
significant contributions. Organic contents of the substrate were likewise rather variable,
ranging from 0.8% to 2.7% by dry weight. We did not detect any correlation between
organic content and the proportion of biogenic substrate particles. Substrate samples
collected from several other sites in the Firth of Lome and their characteristics were
also very variable (Table III). Echinocyamus pusillus was not found in muddy sub-
strates, with high proportions of fine particles (<0.25 mm).
General behavior
E. pusillus occupies the spaces between relatively large pebbles (> 12.5 mm) where
it either nestles (Nichols, 1959) or burrows (Ghiold, 1982) in pockets of sandy gravel.
Substrate particles were moved over both the oral and aboral surfaces of exposed
animals, by the action of suckered podia. Once covered, individuals retained a complete
canopy of particles held firmly against the spine tips by the suckered podia, even
when they were fully buried. The animals retracted their podia and gradually released
particles when disturbed. During hours of daylight E. pusillus moved about very little
but during darkness they often relocated themselves. Between the spines, water currents
flow towards the peristome and from there to the periproct, as described previously
(Nichols, 1959). These currents are generated by bands of cilia along the shafts of
miliary spines, placed at right angles to the current flow (Fig. 2F); additional cilia are
TABLE I
Comparison of substrate samples taken by Petersen grab (n = 7) and dredge (n = 9) from the shallow
bay, Camas Nathais (marked by two triangular symbols in Fig. 1)
Source
>12.5 mm
>4.0 mm
>2.0 mm
>1.0 mm
>0.5 mm
>0.25 mm
<0.25 mm
Grab
13.1
±18.2
10.5
±10.3
11.3
±6.9
19.7
±11.9
17.8
±10.4
13.9
±10.3
13.7
±9.8
Dredge
12.0
±9.1
13.7
±5.9
18.5
±6.6
25.7
±5.0
16.2
±4.6
9.8
±4.9
4.6
±2.8
Mean particle fractions (±S.D.) expressed as percent dry weight.
FEEDING IN ECHINOCYAMUS PUSILLUS
749
TABLE II
Shell to mineral particle ratios by number, expressed as percentages, and origin of biogenic material for
substrate samples in which Echinocyamus pusillus was abundant
Dredge substrate samples:
Fraction
Range
Median
4.0 mm
15-90
55
2.0 mm
30-85
70
1.0 mm
50-90
70
0.5 mm
30-95
75
0.25 mm
25-95
60
Petersen grab samples:
Range
Median
5-80
20
2-75
55
1-80
60
1-90
75
10-85
70
Origin
Pelecypoda
Gastropoda
Polychaeta
Echinodermata
Pelecypoda
Gastropoda
Echinodermata
Pelecypoda
Gastropoda
Echinodermata
Echinodermata
Foraminifera
Foraminifera
Echinodermata
Pelecypoda: A starte, \'enerupis, Chlamys, Ensis. Cardiitm.
Gastropoda: Turritella, Patella, Calliostorna.
Echinodermata: Echinus, Psammechinus, Echinocyamus, Echinocardium.
Median values for shell material give a good indication of the "typical" substrate.
located in relatively shorter bands on the primary spines, where they are restricted
to the spine bases (Fig. 2B).
Feeding
During feeding, substrate particles were picked up and initially transported by
the suckered podia which actively explored the substrate. Under the experimental
conditions it was not possible to see whether podia on the aboral surface contributed
equally to this process. Particles were held by the combined action of the sucker and
secreted mucus: occasionally particles adhered to podia even when their suckers were
fully expanded and visible. The handled particles ranged from 0.25 to 1 mm but
were mostly about 0.5 mm in maximum dimension. The animals manipulated biogenic
TABLE III
Particle size fractions as percent dry weight for dredged substrate samples in which Echinocyamus
pusillus was abundant, present, or absent
> 12. 55 mm >4.0 mm >2.0 mm > 1.0 mm >0.5 mm >0.25 mm <0.25 mm
Abundant
52.5
44.6
2.5
0.3
0.1
0.1
0.1
13.0
21.6
29.9
26.3
8.6
0.4
0.3
47.7
41.7
8.6
1.3
0.4
0.2
0.2
27.2
61.5
2.0
1.9
5.8
1.4
0.2
Present
73.5
8.9
3.2
3.4
6.1
2.9
1.8
49.2
25.3
17.6
6.0
1.1
0.5
0.4
17.8
23.2
25.8
21.7
9.2
1.9
0.5
Absent
44.7
22.9
1.5
2.7
3.9
4.2
20.2
53.2
8.3
4.0
5.5
7.5
7.1
14.2
0.0
6.6
0.5
1.9
1.6
2.7
85.9
Results from single dredge hauls; samples too variable to justify calculation of means.
M. TELFORD ET AL
FIGURE 2. Scanning electron micrographs of Echinocyamus pusillus. (A) Tiered arrangement of
circum-oral spines covering mouth; arrow indicates large, paired buccal podia. (B) Base of primary spine
showing short band of cilia. (C) Tips of suckered podia showing cilia surrounding central nipple. (D)
Sensory pad at tip of buccal podium, with scattered sensory cilia. (E) Mouth with circum-oral spines removed
to show lips and protrusion of lantern teeth. (F) Distal end of miliary spine with band of cilia.
and native mineral particles in proportion to their occurrence in the substrate. Particles
travelled towards the mouth area from podium to podium, until they reached the
fringing circum-oral spines (Fig. 2A). These spines are arranged in two or three tiers,
shorter spines near the mouth, longer ones further away, so that all of their tips can
just reach the mouth itself. Five pairs of large buccal podia (Figs. 2A, D), much less
FEEDING IN ECHINOCYAMUS PUSILLUS 751
active than suckered podia but nonetheless highly extensible, surround the mouth.
Fully elongated, they extend more than halfway across the peristome and can readily
reach into the mouth. Substrate particles arriving at the mouth region were received
by the circum-oral spines and slowly manipulated into the mouth itself. During this
process the particles were delicately explored by the buccal podia which collected
loose organic material from the surface, or material dislodged by the action of the
spines. This material was sometimes passed directly into the mouth by the podia and
sometimes by the spines themselves. The greatest bulk of food, however, was obtained
by the gnawing and scraping activity of the lantern teeth (Fig. 2E). Substrate particles
were held against the teeth and, assisted by the circum-oral spines, they were carefully
revolved by the free edge of the peristomial membrane, which functioned as a set of
five mobile lips. When the particle had been stripped clean it was finally released and
fell away from the mouth.
Anatomy of feeding structures
The tips of the suckered podia bear a ring of sensory cilia surrounding a central
nipple (Fig. 2C) with more scattered cilia distributed outside the ring. Inside the ring
of cilia there are numerous small secretory cells (10 jum in length) which stain brightly
in azan (as noted by Nichols, 1959) and in Milligan's trichrome. Longer, very narrow
secretory cells (15-17 ^m) on the margin of the disk, outside the ring of cilia, are
toluidine blue and PAS positive. These larger cells, which were not described by
Nichols (1959), are difficult to detect in E. pusillus but are more conspicuous in other
clypeasteroids (Mooi, 1983). Both types have external pores from which substances
are exuded. These and other aspects of the detailed anatomy of the suckered podia
have been treated by Nichols (1959) and Mooi (1983). The buccal podia (Fig. 3) also
show features not observed by Nichols (1959). Many short sensory cilia are scattered
over the large sensory pad (Figs. 2D, 4A). They are not confined to an outer ring,
nor are they especially more numerous around the margin of the pad. The epithelium
covering all surfaces of the spines and podia, including the sensory pad of the buccal
podia, is densely supplied with microvilli. No cuticle is visible by SEM and the
structure reported by Nichols (1959) is most probably the surface layer of microvilli.
Around the sensory pad and extending towards its center there are numerous small
pores (0.2-0.3 /um) among the microvilli (Figs. 4A, B) as in the suckered podia. Both
PAS and triple stained sections show secretory cells wedged into the fibrous material
of the sensory pad. Although the miliary spines are alleged to secrete mucus (Ghiold,
1982) we were unable to find any evidence of it. The tips of the miliary spines are
covered by smooth, uninterupted epithelium without any pores (Fig. 4C). No secretory
pores could be found along the spine shafts. Histological sections show that the lumen
of the miliary spines is packed with darkly staining nuclei and granular material,
which is quite unlike the secretory cells found in the podia. The peristomial membrane
is flexible and allows the lantern teeth to protrude slightly (Fig. 2E). It is thickened
into lips (Figs. 5, 6) which grip substrate material during feeding. Histological ex-
amination shows a substantial layer of collagenous connective tissue which stains
blue with azan and green with Milligan's trichrome. This layer is much more developed
than in other clypeasteroids and is covered by epidermis which contains thickened
areas of ciliated, secretory tissue, especially near the mouth opening. Secretions from
these cells in the lips most likely assist in holding particles during feeding. As in the
tips of podia, this thickened epithelium is reinforced by supporting fibers. The lips
are operated by two layers of muscle, located on the inner surface of the peristomial
membrane (Fig. 6). An outer layer of circumferential fibers act as sphincter muscles
752
M. TELFORD ET AL.
100pm
JECRETORY CELL
;ILIUM
UPPORTING FIBRE
NUCLEUS IN EPITHELIUM
NERVE RING
HEAVILY
NUCLEATED
COELOMIC EPITHELIUM EPITHELIUM
STEM RETRACTOR MUSCLE *•«
CIRCULAR CONNECTIVE TISSUE
LONGITUDINAL CONNECTIVE TISSUE
FIGURE 3. Section of buccal podium. The sensory pad consists of thickened, heavily nucleated
epithelium with scattered cilia. Numerous secretory cells, squeezed between the epithelial cells (see inset),
have short ducts opening among the epithelial microvilli (also see Figs. 4A, B).
to close the lips. An inner layer of radial muscles attached to the stereom of the
peristome opens the lips.
Gut contents and substrate particles
Substrate particles selected by the suckered podia during feeding were often covered
with organic material. Under light microscopy much of this appeared to be amorphous,
flocculent stuff, but some diatoms and other algae were visible. Washed material
prepared for SEM lacked most of the amorphous component but extremely numerous
diatoms were found on many particles (Fig. 4D). Both light microscope and SEM
examination of the gut contents of E. pusillus revealed fragmented and whole diatoms,
small pieces of echinoderm spines, sponge spicules, forams, pieces of crustacean
cuticle and setae, fragments of multicellular algae, assorted pieces of organic debris,
and a few mineral fragments smaller than 0.25 mm. Diatoms made up much the
greatest part of the recognizable material in the gut. Those identified included species
of Navic ula, Nitzschia, Pinnularia, Pleurosigma, Fragilaria, and Cocconeis. Several
unidentified diatoms were also present.
DISCUSSION
The occurrence of Echinocyamus pusillus in shelly gravel or sand has been well
icumented (Mortensen, 1948; Nichols, 1959; Ghiold, 1982). Our observations in
FEEDING IN ECHINOCYAMUS PUSILLUS
753
FIGURE 4. Scanning electron micrographs of Echinocyamus pusillus. (A) Cilia, microvilli, and secretory
pores in sensory pad of buccal podium. (B) Microvilli and secretory pores (as in Fig. 4A). (C) Epithelial
microvilli of miliary spine: no secretory pores were found anywhere on the spines. (D) Diatoms attached
in shallow hollows of sand grain. Numerous mucilaginous threads mark earlier sites of attachment.
the Firth of Lome suggest that the species occurs most commonly on substrates
exposed to extensive wave and tidal current activity. These substrates may be disturbed
and turned over frequently by current action and are, presumably, relatively well
aerated. Echinocyamus pusillus was scarce or absent in fine, muddy sediments in
sheltered areas (Fig. 1 and Table III) although Wolff (1968) was of the opinion that
it might occur on such substrates. Other investigators (cited above) have emphasized
the shell component of the substrate. The significance of this, if any, is difficult to
determine. We have found E. pusillus to be abundant in gravelly substrates virtually
free of shell debris and in substrates where shell rubble constitutes 90% or more of
the particles (Table III). It seems likely that a wide range of particle sizes, including
large pebbles with finer material between, and strong current exposure are the critical
requirements. In the laboratory, E. pusillus ceased feeding when water flow, and
hence oxygenation of the substrate, was low. In such active environments, shell debris
may accumulate or even originate more readily from neighboring mollusc populations.
We saw no evidence that the shell component was used preferentially by E. pusillus
nor that the resident flora was greater than that on abiogenic particles. This observation
is further supported by the fact that substrate organic contents were not related to
the shell: mineral particle ratios. In fact, the most shelly substrates included both the
lowest and highest percentages of organic material. SEM examination of substrate
particles shows numerous diatoms, including many of those found in the gut of E.
pusillus. Those shown in the SEM micrograph (Fig. 4D), are mostly attached in
hollows of the grain surface, as noted by Meadows and Anderson (1968). The mi-
754
M. TELFORD ET AL.
1.0mm
-PHARYNX
COMMINATOR
MUSCLE
TOOTH
AURICLE
ANTERN
RETRACTOR
MUSCLE
CILIATED
SECRETORY
EPITHELIUM
UCCAL
MILIARY
SPINE
PODIUM
CIRCUMORAL
SPINE
PERISTOMIAL
MEMBRANE
(LIP)
SUCKERE
PODIUM
ORAL"
SPINE
FIGURE 5. Cross section through mouth of Echinocyamus pusillus showing thickened peristomial
membrane. The free margin of the membrane serves as a set of mobile lips which hold substrate particles
in place while the lantern teeth strip away diatoms.
crograph also shows remnants of many more mucilaginous threads where diatoms
were formerly attached.
The feeding mechanism of E. pusillus is markedly atypical of clypeasteroids, as
Nichols (1959) correctly surmised. The use of the suckered podia to collect and
transport food-bearing particles and, most especially, the use of the lantern teeth, is
more characteristic of regular echinoids than any other group. The action of the lips
at the margin of the peristomial membrane was quite unexpected and is unlike any
mechanism previously described in feeding of clypeasteroids, such as sand dollars. It
should, however, be noted here that sand dollars make extensive use of their accessory
podia in drawing particles onto the sieving mechanism of the aboral surface (Goodbody,
1960; Bell and Frey, 1969; Mooi and Telford, 1982). Furthermore, Clypeaster rosaceus,
another aberrant clypeasteroid, uses both the suckered podia and lantern teeth in a
similar fashion.
Contrary to the opinion of Nichols (1959), the buccal podia do not seem to be
solely sensory in function. Nichols did not observe secretory cells in these podia but
in our sections they were present, in and around the sensory pad (Fig. 3). The pores
seen among the microvilli (Fig. 4A, B) correspond in position with these cells and
could be secretory outlets. These pores were never visible in areas lacking secretory
cells. In addition to a major sensory function, the buccal podia are used also in
collecting and transferring some of the food into the mouth. This use invites comparison
with the feeding of spatangoids but the functional similarity is superficial, resting
mostly on secretion of sticky substances. The simple paired buccal podia of fibulariids
in no way approach the sophistication of the highly modified spatangoid feeding
organs.
Ghiold (1982) reported the presence of large mucus secreting pores at the tips of
miliary spines. In this study, histology did not show any evidence of secretory material
FEEDING IN ECHINOCYAMUS PUSILLUS
755
ESOPHAGEAL EPIDERMIS
CIRCUMFERENTIAL]
RADIALJ-MUSCLE
LAYERS
CONNECTIVE TISSUEV^
XV -^ — ^ -^ \\
SECRETORY
CELL
SUPPORTING
FIBER
100pm
OUTER
EPIDERMIS
—I
91 tj
FIGURE 6. Section of edge of peristomial membrane (lip). The surface of the lip region is covered
by thick secretory epithelium with scattered cilia. The lips are retracted by an inner layer of radial muscle
fibers and closed by circumferential fibers.
in these spines. The lumena are filled with darkly staining nuclei and granules which
do not react like secretory material with PAS or toluidine blue. Furthermore, SEM
showed the total absence of pores in miliary spine epithelium (Fig. 4C). The large,
terminal pores shown by Ghiold (1982) are undoubtedly artefacts due to poor specimen
preparation: air-dried material is unsuitable for cellular details, such as microvilli,
secretory pores, or cilia. The relatively large holes and depressions along the spine
shafts (Figs. 2B, F) correspond with openings in the underlying stereom. The absence
of secretory cells or granules within the spines, indicates that these openings, which
might be artefacts, are not secretory pores.
Sand dollars such as Leodia and Mellita are thought to use the primary and
miliary spines as a two-tiered sieve mechanism (Goodbody, 1960; Bell and Frey,
1969; Seilacher, 1979; Lane and Lawrence, 1982) which dislodges diatoms and organic
debris from substrate particles. This material is then collected by ciliary currents and
perhaps mucus secretion, moved to the mouth along well-defined food grooves, and
there ingested. No such mechanism exists in E. pusillus. It is equipped with some of
756 M. TELFORD ET AL.
the requisite structures but lacks others. There is a very clear differentiation between
primary and miliary spines and the distribution of cilia on them is almost identical
to that of Echinarachnius parma (Mooi and Telford, 1982). Ghiold (1982) has hy-
pothesized that early clypeasteroids exploited surface cleansing currents as a new
feeding system and that spine differentiation in Echinocyamus represents pre-adap-
tation in an early stage of the evolutionary development of this new mechanism. He
offered no explanation of the possible adaptive significance of spine differentiation
during this "pre-adaptational" stage. It is curious that the miliary spines of E. pusillus
are more sharply differentiated and have more elaborate crowns than those of almost
any other clypeasteroid. Others with highly differentiated miliary spines, although of
somewhat different form, include the rotulids and mellitids, which are generally con-
ceded to be advanced forms. Thus, according to this feature, E. pusillus could be
regarded as advanced, not primitive. Departure from the characteristic mode of food
transport in the clypeasteroids may also be considered as a secondary, specialized
feature. The absence of any vestige of the food grooves or of a podial arrangement
reminiscent of them, raises some interesting questions about the possible point of
evolutionary divergence of the Fibulariidae. As remarked earlier, the family is generally
placed close to the Laganidae, which have distinct but short food grooves, and the
Rotulidae in which the grooves are much branched. Other clypeasteroids which have
developed secondary feeding mechanisms, such as Dendraster excentricus (Timko,
1976; O'Neill, 1978), have retained clear food grooves. This species, of course, readily
feeds in the conventional mode as well as in the upright posture.
In summary, we tend to agree with the early opinion of Clark (1914) that Echino-
cyamus pusillus is a specialized, not a primitive species. Spine differentiation and
ciliation are characteristics shared with all clypeasteroids, which makes it unlikely
that Echinocyamus could in any sense represent an ancestral form of the true sand
dollars. Their small size is most probably an adaptation to existence in pockets of
sediment between frequently moving pebbles or stones, on substrates worked by
currents. The rocking sieve mechanism described for some species, appears to work
best with the relatively fine particles found in well-sorted substrates in which sand
dollars most commonly occur. The small surface area of the specialized fibulariids
provides insufficient spines to make an effective sieve. They rely, instead, on the
collection of individual particles from which food material can be stripped by the
lantern teeth.
ACKNOWLEDGMENTS
This work has been supported by the Natural Sciences and Engineering Research
Council of Canada through Operating Grant #A 4696. We gratefully acknowledge
support from the British Council who provided a travel grant under the ALIS program
to M.T. We also wish to thank the staff of the Dunstaffnage Marine Research Lab-
oratory, Scotland, and in particular Alan Ansell and Ian Drummond, for assistance
and hospitality. We are indebted to Eric Lin, Department of Zoology, University of
Toronto, for technical assistance with SEM.
LITERATURE CITED
BELL, B. M., AND R. W. FREY. 1969. Observations on ecology and the feeding and burrowing mechanisms
of Afellita quinquiesperforata (Leske). J. Paleontol. 43: 553-560.
CLARK, H. L. 1914. Hawaiian and other Pacific Echini. The Clypeasteridae, Arachnoididae, Laganidae,
Fibulariidae and Scutellidae. Mem. Mus. Comp. Zool., Harvard College, XLVI No.l.
CUENOT, M. L. 1941. Un paradox evolutif: la neotenie chez les Oursins. Compt. Rend, de I'Acad. des Sci.
Feb. 1941.
FEEDING IN ECHINOCYAMUS PUSILLUS 757
DURHAM, J. W., AND R. V. MELVILLE. 1957. A classification of echinoids. J. Paleontol. 31: 242-272.
GHIOLD, J. 1982. Observations on the clypeasteroid Echinocyamus pusillus (O. F. Muller). /. Exp. Mar.
Biol. Ecol. 61: 57-74.
GOODBODY, I. 1960. The feeding mechanism in the sand dollar Mellita sexiesperforata (Leske). Biol. Bull.
119: 80-86.
HUMASON, G. L. 1967. Animal Tissue Techniques. W. H. Freeman and Co., San Francisco. 569 pp.
HYMAN, L. 1955. The Invertebrates 4: Echinodermata. McGraw-Hill Co., NY. 763 pp.
LANE, J. M., AND J. M. LAWRENCE. 1982. Food, feeding and absorption efficiencies of the sand dollar,
Mellita quinguiesperforata (Leske). Estuarine Coastal Shelf Sci. 14: 421-431.
MEADOWS, P. S., AND J. G. ANDERSON. 1968. Micro-organisms attached to marine sand grains. J. Mar.
Biol.Assoc. U.K. 48: 161-175.
Mooi, R. 1983. Morphology, Diversity and Function of Non-respiratory Podia oj Clypeasteroids (Echino-
dermata: Echinoidea). M. Sc. Dissertation, University of Toronto.
Mooi, R., AND M. TELFORD. 1982. The feeding mechanism of the sand dollar Echinarachnius parma
(Lamarck). Proc. Int. Echinoderms Conf., Tampa Bay (1981). Pp 51-57.
MORTENSEN, Th. 1948. A Monograph of the Echinoidea IV: 2. Clypeasteroida. C. A. Reitzel, Copenhagen.
471 pp.
NICHOLS, D. 1959. The histology and activities of the tube-feet of Echinocyamus pusillus. Q. J. Microsc.
Sci. 100: 539-555.
O'NEILL, P. L. 1978. Hydrodynamic analysis of feeding in sand dollars. Oecologia 34: 157-174.
PHELAN, T. F. 1977. Comments on the water vascular system, food grooves, and ancestry of the clypeasteroid
echinoids. Bull. Mar. Sci. 27: 400-422.
SEILACHER, A. 1979. Constructional morphology of sand dollars. Paleobiology 5: 191-221.
TELFORD, M. 1981. A hydrodynamic interpretation of sand dollar morphology. Bull. Mar. Sci. 31: 605-
622.
TELFORD, M. An experimental analysis of lunule function in the sand dollar Mellita quinquiesperforata
(Leske). Mar. Biol. (in press).
TELFORD, M., AND A. S. HAROLD. 1982. Lift, drag and camber in the northern sand dollar, Echinarachnius
parma. Proc. Int. Echinoderms Conf., Tampa Bay (1981). Pp 235-241.
TlMKO, P. L. 1976. Sand dollars as suspension feeders: a new description of feeding in Dendraster excentricus.
Biol. Bull. 151: 247-259.
West Coast of Scotland Pilot, 9th Ed. 1949. Hydrographic Dept., Admiralty, London.
WOLFF, W. J. 1968. The Echinodermata of the estuarine region of the rivers Rhine, Meuse, and Scheldt,
with a list of species occurring in the coastal waters of the Netherlands. Neth. J. Sea Res. 4: 59-
85.
Reference: Biol. Bull. 165: 758-777. (December, 1983)
THE ROLES OF HEMOCYTES IN TANNING DURING THE MOLTING
CYCLE: A HISTOCHEMICAL STUDY OF THE FIDDLER CRAB,
UCA PUGILATOR
LINDA L. VACCA1 AND MILTON FINGERMAN2
^Department of Anatomy, The University of Kansas Medical Center, Kansas City, Kansas 66103 and
2 Department of Biology, Tulane University, New Orleans, Louisiana 70118
ABSTRACT
Histochemical data support the previous biochemical finding that the blood is a
major site for the production of proteinaceous and diphenolic substances for tanning
of the cuticle in the fiddler crab, Uca pugilator. Five types of hemocytes are described.
Specifically in tanning, the hyaline cells (cystocytes) appear responsible for the pro-
duction of diphenolic tanning agents whereas the granulocytes synthesize the proteins
involved. Other types of hemocyte may be transitional forms involved in clotting
(intermediate cells). Various histochemical reactions for each type of hemocyte and
the cuticle are recorded throughout the molting cycle, and appear cyclic. The data
suggest there is hormonal control of the cyclic events during the tanning process.
INTRODUCTION
At least in some arthropods, sclerotinization consists of two major processes: (1)
the biosynthesis of tanning agents (N-acetyldopamine and N-acetylnoradrenalin) from
their amino acid precursors (tyrosine and phenylalanine), and (2) the subsequent
incorporation of the newly formed tanning agents into the cuticle (Brunet, 1965;
Koeppe, 1971; Vacca and Fingerman, 1975a, b). In the cockroach, the synthesis of
the tanning agent, N-acetyldopamine, begins within the hemocytes (Whitehead, 1969).
However, in crustaceans, the synthesis site of the tanning agents remains unknown.
Crustaceans, like insects, maintain high metabolic pools of free amino acids within
the hemolymph (Awapara, 1962; Florkin and Schoffeniels, 1965). In the hemolymph
of the crab, Carcinus maenas, most of the free amino acid pool is concentrated within
the hemocytes (Evans, 1972). The blood cells, although they provide only 1% of the
total blood volume, contain 58% of the total free amino acid concentration. In this
way, the blood cells maintain a steep gradient against the serum; but the purpose of
this gradient remains obscure. Presumably some of the free amino acids could serve
as precursors for tanning agents and their protein carriers.
Early workers regarded one type of crustacean hemocyte, the granulocyte, as a
carrier of metabolites (Tail and Gunn, 1918). However, more recent evidence supports
other functions also, including phagocytosis, wound agglutination, blood coagulation,
parasitic encapsulation, basement membrane formation, and storage of glycoproteins
(George and Nichols, 1948; Dumont et al, 1966; Bang, 1967; Wood and Visentin,
1967; Strutman and Dolliver, 1968; Busselen, 1970; Wood et al, 1971; Ravindranath,
1980). On the other hand, these data fail to explain why the clotting ability of the
blood is minimal at ecdysis, precisely when the soft-shelled animal is most susceptible
Received 14 February 1983; accepted 16 September 1983.
Abbreviations: Az-Eo, azure-eosin; DAS, diazosulfanilic acid; DAS-AzA, diazosulfanilic acid pH 1
re A; DOPA, dihydroxyphenylalanine; Fell, ferrous iron; Felll, ferric iron; NQS, beta-naphthoquinone-
odium sulfonate; PAS, periodic acid-Schiff; PCB, post-coupled benzylidine; RNA, ribonucleic acid.
758
HEMOCYTES AND TANNING 759
to injury and infection (Bang, 1967; Levin, 1967; Strutman and Dolliver, 1968).
Since the hemocytes do not clot well during ecdysis, they may be involved with yet
another and more important function during this period, namely tanning.
Reportedly, cyclic fluctuations occur during the molting cycle in: (a) the enzymatic
activity of blood phenoloxidase (Pinhey, 1930; Decleir and Vercauteren, 1965; Sum-
mers, 1967); (b) the numbers of circulating hemocytes (Bruntz, 1907; Kollman, 1908;
Marrec, 1944); and (c) the appearance of carrier proteins which transport tanning
agents from the hemolymph into the cuticle at ecdysis (Vacca and Fingerman, 1975a,
b). These cycles suggest that the hemocytes of crustaceans may have a special function
which is intimately associated with the tannning process. That the hemocytes can
penetrate the epithelium and synthesize protein during the secretion of the proecdysial
cuticle in the crayfish Orconectes limosus (Keller and Adelung, 1970) further implies
that they are involved in the tanning process. The present investigation explores this
possibility in the fiddler crab, Uca pugilator, by a histochemical study of the hemocytes
during the molting cycle. The histochemical reactions of the developing exoskeleton
are correlated.
MATERIALS AND METHODS
All observations were made on fiddler crabs (Uca pugilator} during various stages
of the molting cycle. Stock male and female fiddler crabs were maintained individually
in finger-bowls containing enough artificial sea water (Instant Ocean, Aquarium Sys-
tems, Inc.) to cover the bottom 1 cm deep. The water was changed every 2-3 days
after the animals were fed a few flakes of oatmeal. Crabs were induced into a precocious
proecdysial period and eventual ecdysis as previously described (Vacca and Fingerman,
1975a, b) by removing both eyestalks (Brown and Cunningham, 1939) or by auto-
tomizing several legs (Skinner and Graham, 1972; Fingerman and Fingerman, 1974).
The intermolt crabs were intact specimens that had undergone ecdysis (induced by
limb removal) and limb regeneration at least 1 month prior to use.
Stages of molt were determined according to Guyselman (1953). Proecdysial an-
imals were selected from eyestalkless or autotomized crabs. They showed external
evidence of apolysis, a bluish gray opalescence on the carapace. Forty-four specimens
were selected at different stages in the molting cycle including ecdysis, various times
of postecdysis (5, 10, 24 and 48 h), proecdysis, and intermolt. The crabs were fixed
in toto either by injection of, or immersion in various fixatives including 10% neutral
phosphate-buffered formalin and 6% neutral phosphate-buffered glutaraldehyde to
which 6% NaCl was added; formalin-acetic acid-salt (10%:5%:5%); chloro-
form:methanol (2:1); and Barnett and Bourne silver fixative (Lillie, 1965).
The crabs were bisected to allow rapid entry of the fixatives, and were fixed for
24 hours. After a thorough washing, they were dehydrated in graded alcohols, and
cleared in xylene. Tissues were then embedded in paraffin in vacua.
Tissue sections (6-8 nm) were stained with azure-eosin (Az-Eo), pH 4.5, and
examined for numbers and types of blood cells. The extinction coefficient of basic
dye uptake by the hemocytes was determined with toluidine blue 0 (0.1%) at pH 1
through 3.
Other histochemical tests included: the periodic acid-Schiff (PAS) reaction for the
identification of 1,2-glycols (Mowry, 1963); black Bauer and black periodic techniques
for aldehyde detection (Lillie, 1965); and Sudan black B for the localization of lipids
(Lillie, 1965).
In conjunction with these procedures, various blockades were used. Acetylation
was accomplished after 3 hours at 60°C in a 2:3 mixture of acetic anhydride: pyridine
760 L. L. VACCA AND M. FINGERMAN
(Barka and Anderson, 1963), to distinguish lipid from other PAS-reactive substances.
Deacetylation was performed by immersing tissue sections in ammonium hydrox-
ide:ethanol (1:4) for 24 hours (Lillie, 1965). Incubation in saliva (1-3 h) was used to
identify glycogen. To distinguish bacteria from other intra- and extracellular inclusions,
ribonucleic acid (RNA) was extracted by incubating tissue sections in KOH (1% in
70% ethanol, 15-20 min.).
Several diazotized dyes were prepared for the demonstration of proteins and
phenols (Lillie, 1965). These included: diazosafranin, pH 3.2 for serotonin (Lillie et
al, 1973a), or pH 7.8 for proteins; and diazosulfanilic acid, followed by pH 1 azure
A (DAS-AzA), for norepinephrine or another primary catecholamine (Lillie et al,
1973b). Lack of extraction of the colored tissue sites by acid (0.1 TV HC1 for 24 h at
room temperature) verified azo-coupling.
Blocking procedures were used in conjunction with the localization of phenols.
Oxidation was carried out with periodic acid (1%, 30 minutes); reduction with 5%
sodium dithionite (2 or 4 two-hour incubations). Ferrous chloride (FeCl2, O.I M, 2
h), freshly prepared by the method of Lillie et al. (1971), was used to block histidine
staining by the DAS-AzA technique, was previously demonstrated in mammalian
erythrocytes (see Lillie et al., 1973b, c).
Indole derivatives were visualized by the post-coupled benzylidine (PCB) reaction
(Glenner and Lillie, 1957). The beta-napthoquinone-4-sodium sulfonate (NQS) method
of Lillie et al. (1971) was used to demonstrate sites rich in arginine. The Morel-Sisley
procedure for the demonstration of tyrosine was also applied (Lillie, 1965). The
reaction for tyrosine was blocked by pretreatment (6 h at room temperature) with
tetranitromethane (0.1 ml in 10 ml pyridine to which 20 ml 0.1 N HC1 was added).
To demonstrate amino groups, slides were mordanted for two hours in FeCl2,
then stained with neutral hematoxylin, with and without prior deamination. Deam-
ination was accomplished over a 24 hour period at 4°C in a mixture of 14% sodium
nitrite in 2 TV acetic acid.
Ferric ferricyanide was used to identify reducing sites. To distinguish phenolic
sites from iron reaction, sections were reacted with acid ferri- and ferrocyanide. To
differentiate between sites of reduction and oxidation respectively sections were first
mordanted in FeCU (0. 1 M; 2 h), then reacted with acid ferri- and ferrocyanide.
Sections were incubated in acid silver (0.1 M AgNO3 in 0.01 M acetate buffer,
pH 5.0) in the dark (24 h at room temperature) to demonstrate further the presence
of reducing substances (Lillie, 1957). Additionally, ammoniacal silver procedures were
applied to the tissue sections for 10 minutes and 24 hours in the dark at room
temperature (Lillie, 1965).
Several procedures were used to localize copper. These included Clara's (Mallory's
neutral) hematoxylin (Lillie, 1965) and ammoniacal rubeanic acid, with and without
mordanting in a copper sulfate solution (2.5% in 50% alcohol for 2 h).
RESULTS
During all stages of the molting cycle, two main types of blood cells could be
distinguished histochemically by the presence or absence of acidophilic cytoplasmic
granules (Az-Eo; Figs. 1 A, 2 A): (a) large hemocytes containing numerous acidophilic
(eosinophilic) granules in an abundant acidophilic cytoplasm were recognized as
granulocytes and (b) smaller agranular cells exhibiting a scanty pale basophilic cy-
toplasm around an intensely basophilic nucleus were identified as hyaline cells (also
vnown as cystocytes). On rare occasions, a third type of blood cell could be seen (Fig.
2} which seemed to be an "intermediate" or transitional type. It resembled the hyaline
m u
1C
\
i
2B
2A
2C
FIGURE 1. Two main types of hemocyte can be identified in the blood of Uca: granulocytes (g) and
hyaline cells (h). In Figure 1A these hemocytes aggregate in great numbers near the epidermis and soft
cuticle (cut) of a crab fixed in a buffered glutaraldehyde-salt fixative. Granulocytes and hyaline cells appear
scattered within the eosinophilic serum which contains numerous granules (arrow). Certain granules exhibit
basophilia; others exhibit acidophilia. Az-Eo, pH 4.5. MEL, melanophore. X430. Figure IB shows the
positive reaction for arginine in the granulocytes (g) and the negative reaction in the hyaline cells (h). NQS.
X600. Figure 1C demonstrates reducing substances in small (immature?) granulocytes (arrows) found deep
within the hemocoel. Ferric ferricyanide. mu, muscle. X600.
FIGURE 2. Besides hyaline cells and granulocytes, a third type of hemocyte (intermediate or transitional
type) can be seen within the hemocoel. In Figure 2A a specimen fixed in formalin:acid:salt during late
intermolt, the intermediate type (transitional cystocyte) hemocyte (i, arrow) approximates the size of the
hyaline cell (h) and exhibits tiny unstained granules within a less extensive cytoplasm than the granulocytes
(g). Az-Eo, pH 4.5. X900. Figure 2B shows intermediate cells (arrows) releasing tiny proteinaceous granules
which have azo-coupled with DAS. The hyaline cell (h) contains diphenols. The hemocytes appeared in a
specimen fixed 10-15 hours postecdysis in buffered formalin-salt. X600. Figure 2C represents a diagrammatic
interpretation of the intermediate cells rupturing and releasing their granules, thereby forming a cytoplasmic
network which may function in clotting.
761
762
L. L. VACCA AND M. FINGERMAN
cell in size and nucleancytoplasmic ratio, but it contained a number of cytoplasmic
granules like the granulocyte. However, the granules were smaller than those observed
in the granulocyte and were refractory to staining with acid and base dyes; also they
occurred within an unstained cytoplasm. Two additional types of granule-containing
(transitional?) cells, large and small granular cells contained swollen granules which
were discerned by other histochemical procedures (Figs. 3 and 4).
v
•
FIGURE 3. Small aggregates of hyaline cells (h) occur in the hemocoel of a crab fixed five hours
postecdysis in buffered glutaraldehyde-salt. Granulocytes (g) are rare. Two large, flattened hemocytes exhibit
an extensive and faintly basophilic cytoplasm which contains numerous swollen granules. Intensely eosin-
ophilic, smaller granules surround a reticulate nucleus; peripheral granules are larger and slightly basophilic
(arrows). These hemocytes have been identified as large granular cells (Igc's). Az-Eo, pH 4.5. X900.
FIGURE 4. Hyaline cells (h) and granulocytes (g) accumulate within the hemocoel beneath the epidermis
underlying the newly formed cuticle of a crab fixed between 10 and 15 hours postecdysis in buffered
formalin-salt. The granulocyte contains histidine in the cytoplasm and granules. The small hyaline cells
possess a diphenol; a few appeared unstained as if they had released their phenolic contents. The arrow
points to a small granular cell which contains swollen proteinaceous granules of undetermined function.
A leucophore (L) passes across the field at right beneath the epidermis. DAS-AzA. X900.
HEMOCYTES AND TANNING 763
Fluctuations in the number of hemocytes during the molting cycle
To determine whether fluctuations occurred in the number of hemocytes during
the molting cycle, counts were made in the tissue sections taken from each tissue
block. The sections were examined microscopically using low magnification (100X)
for an area heavily populated with hemocytes. Using high magnification (450X) two
counts were made of the hemocytes in that area.
The granulocytes and hyaline cells were counted; their relative numbers varied
with the stages of the molting cycle. During intermolt and proecdysis, there were
twice as many granulocytes as hyaline cells. At ecdysis, the numbers of both gran-
ulocytes and hyaline cells increased: two-fold and ten-fold, respectively. Thus, the
proportion of granulocytes and hyaline cells (2/5) was the inverse of that in the earlier
two stages. By 5-10 hours postecdysis, the hyaline cells outnumbered the granulocytes
by 10:1. However, 24-48 hours postecdysis, the numbers of both types of hemocyte
gradually declined. The decline was more severe among the hyaline cells, which still
prevailed over granulocytes by 2: 1 by 24 hours postecdysis.
During intermolt and proecdysis, the hemocytes were usually floating freely in
the hemocoel. At ecdysis and throughout postecdysis, numerous hemocytes aggregated
beneath the epidermal cells and penetrated the epidermal layer, approaching the newly
formed cuticle (Fig. 1). In sections of crabs fixed at ecdysis and during early (5-10
h) postecdysis, numerous hyaline cells were packed together into large nodules floating
near the epidermis, or occasionally freely within the hemocoel. In some specimens,
small aggregates were formed near the epidermis by hyaline cells surrounding an
occasional granulocyte (Fig. 3). Among the small aggregates, a fourth type of hemocyte
could be identified as a large granular cell (Fig. 3). The large granular cells contained
two types of swollen granules within a flattened faintly-basophilic cytoplasm; pale
acidophilic granules encircled the nucleus, whereas basophilic granules populated the
extensive peripheral cytoplasm. The pale basophilic nucleus had a reticulate chromatin
network and contained an intensely basophilic nucleolus.
Histochemical observations — the hemocytes and the serum
Basophilia, acidophilia, glycogen, 1,2-glycols, lipids, and aldehydes. The basophilic
staining of the hyaline cell cytoplasm became extinguished at pH 3 (Table I). At this
pH the granules within granulocytes stained metachromatically; granule staining be-
came abolished at pH 2 and basophilic nuclei and melanophore granules (still apparent
at pH 1) could be visualized.
By PAS staining, the granulocytes contained 1,2-glycols which concentrated within
the granules; the cytoplasm reacted moderately. By contrast, the hyaline cells appeared
negative. The serum showed transient reactions which varied with the stages of the
molting cycle: during early postecdysis the serum became filled with 1 ,2-glycols and
numerous intensely PAS-positive granules like those in the granulocytes. Intensely
PAS-positive granules also appeared within the epidermal cells and tegumental glands
during this period. At the other stages of the molting cycle, the serum, epidermal
cells, and glands became devoid of the presumed glycoprotein(s).
Fixation of crabs in chloroform:methanol freed the tissues of lipids, but no change
occurred in the PAS reactions of the granulocytes (cytoplasm and granules) or the
"serum granules." The serum exhibited reduced staining by PAS which could be
ascribed to extracted lipids, but no sudanophilia could be demonstrated. Further
proof that PAS stained non-lipid substances was obtained when acetylation abolished
764
L. L. VACCA AND M. F1NGERMAN
TABLE I
Summary of histochemical reactions in the hemocytes, serum, cuticle, and melanophores
of the fiddler crab*
Histochemical reactions
Azo-
Stain Induced Native
Cell or tissue Baso- Acido- extinc- Glyco- 1 ,2- aide- aide-
component philia philia tion gen Glycols Lipids hydes hydes
Proteins
Sero-
tonin
Hyaline cell
pH 3
to -
Granulocyte
pH 2
± to -
to -
Serum
± to - (hard)
>PH3 ± * ± ± ~ +<soft> -
Exocuticle
hard
soft
+ + (endo only) >pH 3 ± to +
>pH 3 - ± - - - +
Melanophore
granules
+ pH 1 ++ +
* Note: The words "hard" and "soft" refer roughly to the state of the cuticle during the molting cycle. "Exocuticle"
was taken as representative for histochemical changes also occurring in the endocuticle (endo) during the molting cycle
which often appear in parallel but at different times. Results were recorded separately when a difference in staining capacity
was noted. The symbols indicate strength of the histochemical reaction: + + , intensely positive; +, positive; ±, moderately
positive; +, mildly positive; — , negative.
the reactions in the granulocytes, some of the serum granules, and reduced the PAS
reaction in the serum itself; deacetylation partially restored the reactions. Digestion
of glycogen from the tissue sections did not change the PAS reactions within the
granulocytes. However, some of the "serum granules" showed reduced staining and
therefore contained glycogen.
The induced aldehyde groups detected by black Bauer and black periodic techniques
were intensely visualized within the granulocytes. In contrast, the hyaline cells reacted
mildly or sometimes not at all.
Native (free) aldehydes were detectable (by direct application of SchifT reagent, 1
hour, to the tissue sections) in granulocyte cytoplasm, especially the perinuclear region,
during intermolt, proecdysis and late postecdysis, but not during early postecdysis.
The intracellular granules did not stain. Likewise, the hyaline cells and serum did
not react.
Diazotization reactions for aromatic end-groups. The aromatic end-groups of pro-
teins azo-coupled intensely (diazosafranin pH 7.8) within the granulocytes during
most of the molting cycle (except late postecdysis) but only mildly in the hyaline
cells. Interestingly at 48 hours postecdysis the granulocytes lost the ability to azo-
couple as if they had released the responsible proteins. During this period, the serum
showed increased reactions as if it had received the proteins released from the gran-
locytes. However, during proecdysis the serum must not have contained these proteins
se it did not react. Subsequent extraction of the azo-coupled tissue sections in
HEMOCYTES AND TANNING
765
TABLE I (Continued)
coupling
Protein End-Groups
Catechol-
amines
Histi- Tyro- Reducing
dine Tryptophan Arginine sine Amino substances
Copper
+ + to -
_ - (hard)
+ to - + +
+ (soft)
+
-
+ (soft)
+ + (hard)
+ (soft)
± to - (hard)
+ (soft)
+ (hard) ± to + (hard)
+ to + (soft) ± to - (soft)
+ to + (hard)
+ (soft)
+ + to ± +
+ to - ±
± to - + to - T ++ to ± ++ to + + to +
+
(cytoplasm + + )
+ + (hard)
+ + (soft)
(cytoplasm +)
dilute HC1 failed to remove the tightly bound dye. Serotonin could not be detected
using diazosafranin, pH 3.2 (Lillie et ai, 1973).
By azo-coupling with DAS-AzA, a primary catecholamine was demonstrated within
the hyaline cells (Lillie et al, 1973b, c) during most stages of the molting cycle (Fig.
4). During postecdysis, the phenolic substance gradually disappeared. Early in pos-
tecdysis, intact hyaline cells near or within the epidermal net azo-coupled mildly, as
if they were losing their former contents. A phenol visualized in the serum during
intermolt and proecdysis was still detectable early in postecdysis. However, by 48
hours postecdysis, the phenol in the serum became substantially reduced, and also
disappeared from the hyaline cells.
Using DAS-AzA, two additional granule-containing hemocytes could be identified:
small granular cells the size of hyaline cells (Fig. 4), and the large granular cells
previously identified by Az-Eo (Fig. 3). The small and large granular cells contained
swollen granules which exhibited intense azo-coupling (Figs. 4, 5). The large granular
cells increased their numbers during early postecdysis (Fig. 5A) when two forms
became apparent: cellular forms possessed a distinct cell shape and a nucleus (Fig.
5B); amorphous forms had a more extended cytoplasm and no nucleus (Fig. 5C). By
15-24 hours postecdysis, the large granular cells aligned along the epidermis (Figs.
6A, B). Morphologically they resembled melanophores, except they contained larger
(swollen) granules.
Intermediate ("transitional") cells, fixed in the process of rupturing, spewed forth
from their cytoplasm numerous tiny granules which azo-coupled with DAS-AzA (Fig.
2B, C). These granules approximated the size of bacteria. However, prior extraction
with KOH did not remove azo-coupling capacity. Therefore, RNA was not responsible.
Furthermore, no gram-positive material was demonstrable. Epidermal melanophore
granules also azo-coupled intensely (Fig. 6C). Surrounding them, the cytoplasm of
the melanophores azo-coupled as if it contained a phenol. The sites of azo-coupling
766
L. L. VACCA AND M. FINGERMAN
•
h
5A
Jb^
i
V
••y.^'fc"---"^'/''
•/jr*J:»V''
' •'• .'A <•/•
' ?:'
6C
FIGURE 5. The number of large granular cells (Igc) increases in the hemocoel, especially near the
epidermis, after ecdysis. The specimen was fixed 10-15 hours postecdysis in buffered formalin-salt. Figure
5A shows two types of large granular cells: type 1 (Igc,) possesses discrete cytoplasmic boundaries and a
nucleus; type 2 (Igc2) has an amorphous cytoplasm and no nucleus can be seen. Both types contain
characteristic swollen granules whose protein matrix has azo-coupled with DAS-AzA. The hyaline cell (h)
is much smaller and contains diphenols. Figures 5B and 5C show diagrammatic interpretations of the two
types of large granular cells shown in Figure 5A. Figure 5B shows the distinct cellular shape of an Igc, .
Large swollen granules (dark circles) surround the nucleus (clear space). Figure 5C shows the amorphous
cytoplasm of an Igc2 which also contains swollen granules (dark circles). No nucleus can be seen perhaps
indicating that these large granular cells are degenerating and releasing their contents into the serum. DAS-
AzA. X600.
FIGURE 6. The relationship between the large granular hemocytes and melanophores is uncertain.
are 6A shows three large granular cells (Igc) near the epidermis of a crab fixed 5-10 hours postecdysis
HEMOCYTES AND TANNING 767
described above could not be decolorized by prolonged extraction with dilute HC1.
Verification of a diphenol: oxidation-reduction experiment. After prior oxidation,
the phenols detected within the hyaline cells and serum converted into quinones and
could not azo-couple with DAS-AzA. Pretreatment with dithionite reduced quinones
into phenols which could then azo-couple. After brief, 4 hours, dithionite treatment,
the suspected phenols in the hyaline cells and the serum curiously azo-coupled less
intensely; however, prolonged dithionite (8 h) treatment rendered the staining more
intense at both sites. When dithionite-reduced tissue sections were oxidized, diphenols
became visualized again in the serum and hyaline cells. The intracellular granules
described by DAS-AzA within the small and large granular cells, and intermediate
cells remained unaffected by oxidation or reduction. Often the reducing solution
extracted the granules from epidermal melanophores; the azo-coupled cytoplasm,
unaffected by reduction, was rendered negative by oxidation, verifying its phenolic
content.
Demonstration of histidine and amino groups. The color of the granulocytes (cy-
toplasm and granules) and some of the serum granules after DAS-AzA (Fig. 4) re-
sembled that of erythrocytes containing histidine (Lillie el al, 1973b, c). Pretreatment
with FeCl2 blocked the reaction, confirming the presence of histidine (Lillie et al.,
1971). Oxidation and reduction rendered the sites more intense. Increased histidine
reactions occurred in the granulocyte and serum granules during intermolt, proecdysis,
and early postecdysis. By this and other reactions (PAS, diazosafranin), some of the
serum granules may be identical to (and released from) the granulocytes during the
molting cycle.
Tryptophan (PCS reaction) appeared in the granulocytes during early postecdysis.
However, by late postecdysis, the reaction decreased as if the cells released their
proteinaceous contents. Like the granulocytes, the serum became positive during early
postecdysis, but reacted less intensely 24 hours postecdysis, as well as during intermolt
and proecdysis. The hyaline cells reacted mildly, or not at all, throughout the molt-
ing cycle.
Small amounts of arginine (NQS reaction, Fig. IB) and large amounts of tyrosine
could be visualized within the granulocytes during all stages of the molting cycle. In
contrast the hyaline cells and serum exhibited reactions which were negative for
arginine and mild for tyrosine at all times. Pretreatment of tissue sections with tet-
ranitromethane selectively abolished the staining for tyrosine.
The presence of amino groups (with and without prior deamination) was verified
by the uptake of iron (Fell) subsequently visualized by hematoxylin or by acidophilia
(Az-Eo). The granulocytes stained intensely; their staining could be abolished by prior
deamination. Amino groups visualized in the serum during intermolt and proecdysis
were deaminated inconsistently; their presence varied during postecdysis, possibly
indicating the disappearance of a protein during this time. Hyaline cells showed mild
reactions at all times.
Reactions for reducing sites. The hyaline cells, granulocytes, and serum showed
cyclic reactions for reducing substances during the molting cycle. Minimal at ecdysis,
the reducing substances could not be visualized within hyaline cells at any other time.
in buffered formalin-salt. A type 1 large granular cell (an intact cell) with nucleus is shown at the right
with the label Igc; type 2 large granular cells (amorphous without nucleus) appear at arrows to left. X600.
Figure 6B diagrammatically depicts at higher magnification the type 1 large granular cell shown in Figure
6A. X900. Figure 6C shows mature melanophores (M) near epidermis and cuticle of a crab fixed 10 hours
postecdysis in buffered formalin-salt. Diphenols azo-couple in the exo- (e) and endocuticle (cut), and are
present in the cytoplasm of the melanophore. The melanophore granules are also proteinaceous but tiny
compared with the swollen granules in the large granular cells in Figure 6A. DAS-AzA. X600.
768 L. L. VACCA AND M. FINGERMAN
Granulocytes remained positive between ecdysis and 5 hours postecdysis; but also
became negative during the later stages of postecdysis. During late postecdysis, the
serum contained reducing substances (released from the granulocytes?), but lost them
cyclically during early postecdysis, intermolt, and proecdysis. The tegumental glands
also reacted strongly during postecdysis. The reducing substances described above
were visualized by ferricyanide, Clara's hematoxylin, and Fell acid ferrocyanide re-
actions.
With ferric ferricyanide, positive sites also appeared within the granules of small
and large granular cells, intermediate cells, and the serum (including large swollen
granules). The intracellular granules and the swollen serum granules were not affected
by oxidation or reduction, and therefore they do not contain reducing groups. The
serum granules having the size of those within granulocytes were affected by oxidation
and reduction.
Melanophores contained numerous small granules which blackened character-
istically in ammoniacal silver after 10 minutes and stained intensely with ferricyanide.
The granules were contained by a positive cytoplasm. Oxidation intensified the fer-
ricyanide (quinhydrone) reaction in the granules and masked the visualization of
reducing phenols in the cytoplasm. Reduction restored the original reactions, and
enabled the visualization of reducing substances (some probably phenols) within the
formerly-negative hyaline cells, granulocytes, and serum. Oxidation rendered the sites
negative once again.
Surprisingly, epidermal melanophore granules intensified their natural brown color
by incubation in Clara's solution during postecdysis, but not during proecdysis. Re-
ducing substances may be responsible for the transient reaction.
Copper-rich sites. Identification of copper-containing sites failed using ammoniacal
rubeanic acid. Using Clara's hematoxylin, copper could be visualized midly in the
hyaline cells: more intensely in the granulocytes. In copper uptake studies, the hem-
ocytes took up copper to a moderate degree. The reactions of the granulocytes and
serum varied with the molting cycle: during postecdysis, the granulocytes stained
intensely. The serum took up copper during intermolt, proecdysis, and late postecdysis.
However, during early postecdysis, the serum itself showed weaker copper uptake;
but intensely positive serum granules (released from the granulocytes?) could be seen.
Histochemical observations — the cuticle
Basophilia, acidophilia, glycogen, 1,2-glycols, lipids, aldehydes, and azo-coupling.
Using the various histochemical procedures described above, we recorded the changes
in the staining of the cuticle during the molting cycle. For the structure of the exo-
skeleton. Skinner's (1962) terminology was applied. An upper thin lipid epicuticle
was distinguished from the procuticle below. The procuticle was divided into an upper
thin exocuticle (pigmented layer) and, beneath it a thicker endocuticle (calcified layer).
Epi- and exocuticle form during proecdysis; and endocuticle during postecdysis (Skin-
ner, 1962). During late postecdysis, a deeper layer forms over the epidermis which
is thin and uncalcified called the membranous (uncalcified) layer, and which does
not undergo further modification by calcification or quinone tanning. In this study,
the membranous layer was rarely seen.
Curiously, the exocuticle of freshly molted crabs did not stain with Az-Eo. However,
by late postecdysis, the exocuticle and underlying endocuticle attained a weak ba-
sophilia. Characterizing the influx of acidic substances, this basophilia increased as
the entire procuticle became wider and hardened. The new endocuticle, formed by
24 hours postecdysis, remained unstained by Az-Eo up to 48 hours postecdysis. Later,
HEMOCYTES AND TANNING 769
when more fully formed and hardened, it became acidophilic, as if basic substances
had penetrated.
The epicuticle which covered the fully tanned exocuticle exhibited basophilia.
However, at ecdysis, the epicuticle became acidophilic. It contained aldehydes by the
black Bauer and black periodic methods; sudanophilia was absent.
By PAS staining, the new exocuticle contained 1,2-glycols variably during proec-
dysis, ecdysis, and late postecdysis. This reaction became abolished by acetylation,
but it was not restored by deacetylation. Exo- and endocuticle did not react early in
postecdysis. The reaction was negative in intermolt crabs. No glycogen, aldehydes,
nor lipids were detected.
Using various diazonium salts, the cyclic appearance of aromatic protein end-
groups was detected within the cuticle's protein matrix. The exocuticle layer azo-
coupled mildly during proecdysis and ecdysis indicating proteins exist in low con-
centration. The azo-coupling of the cuticle proteins intensified during early postecdysis.
During late postecdysis, more aromatic end groups appeared in the outermost exo-
cuticle than in the newer endocuticle layer. The epicuticle did not azo-couple. Sub-
sequent extraction in weak HC1 failed to change the results.
Diphenols in cuticle. The reaction of the cuticle to the DAS-AzA showed variations
in the staining for phenols throughout the molting cycle. At intermolt and proecdysis,
the fully-formed and hardened (quinonized) procuticle did not azo-couple. In contrast,
phenols penetrated the soft exocuticle at ecdysis and azo-coupled intensely (Fig. 7A).
The azo-coupling capacity of phenols in the exocuticle decreased as tanning progressed
during early postecdysis; by 1 0 hours postecdysis only small amounts could be detected.
By late postecdysis, the new endocuticle still reacted moderately for phenols. However,
its azo-coupling capacity continued to decrease during late postecdysis as the width
and hardening of the procuticle increased (Fig. 7B). The epicuticle showed intense
reactions for phenols during the entire molting cycle (Fig. 7B).
cut
7C
FIGURE 7. The cuticle changes its reaction for diphenols before and after tanning takes place. Figure
7A shows the intense azo-coupling of phenols in the soft exo- (e) and endocuticle (cut) of a crab fixed at
ecdysis in buffered formalin-salt. Both the exo- and endocuticle react. Phenol-laden hyaline cells and
histidine-rich granulocytes (not shown) occur in the hemocoel and gather close to the epidermal cells (ep)
which also contain phenols during this time. Figure 7B shows the nonreactive tanned endocuticle (cut) of
a crab fixed in formalin-acetic acid-salt 24-48 hours postecdysis. Exocuticle (e) reacts only mildly. Epidermal
cells (ep) still react at this time. Waxy epicuticle (arrowhead) azo-couples throughout the molting cycle.
DAS-AzA. Figure 7C shows the reducing capacity of the soft cuticle (cut) after fixation in Barnett-Bourne
silver solution. The cuticle loses its reducing capacity as it tans. Reducing substances (non-phenolic) also
appear in granulocytes; hyaline cells contain phenols which react mildly or not at all. Epidermal cells (ep)
are negative. X600.
770 L. L. VACCA AND M. FINGERMAN
Temporally, the azo-coupling of the phenols in the endocuticle paralleled that in
the older exocuticle, but occurred at later times. If the endocuticle began to form
soon after ecdysis, its azo-coupling capacity coincided closely with that of the exocuticle
layer (Fig. 7A). During late postecdysis, the endocuticle became non-reactive prior
to the older exocuticle above it (Fig. 7B). The reactions probably depend on the extent
to which phenols penetrate and become quinonized during tanning.
Oxidation and reduction procedures verified the presence of a diphenol within
the cuticle. After oxidation, the diphenol in the cuticle became quinonized and no
longer reacted. After reduction (4 h), tanned cuticles which did not azo-couple with
DAS-AzA exhibited the presence of a diphenol.
Demonstration ofhistidine and amino group. Tryptophan and arginine could not
be detected during proecdysis and ecdysis, although sometimes the exocuticle of newly
molted crabs showed a mild reaction for arginine. Small amounts of tyrosine (selectively
abolished by tetranitromethane) were detected in the exocuticle and epicuticle, but
not in the endocuticle.
Prior incubation in FeCl2 reduced the DAS-AzA reaction in the cuticle, indicating
the presence of histidine. The presence of amino groups in cuticle was verified by
Fell uptake stained by hematoxylin or Az-Eo, with and without prior deamination.
Amino groups stained intensely but sporadically in the exocuticle between ecdysis
and 10 hours postecdysis; the endocuticle reacted less intensely. The data show that
histochemical changes in the cuticle occur early in postecdysis, and imply that protein(s)
penetrate at this time.
Reducing substances in cuticle. Short (10 min) incubations in ammoniacal silver
gave no reaction in the cuticle; after 24 hours, reducing sites (possibly phenols) became
moderately visible. No reaction occurred in acid silver. Interestingly, the cuticle of
crabs fixed during ecdysis and early postecdysis in Barnett-Bourne solution strongly
reduced silver (Fig. 7C). Fully formed and hardened cuticle of intermolt exhibited
the mild reduction of silver.
The epicuticle contained substances which reduced silver during postecdysis, but
not at other times in the molting cycle. Curiously, no reducing substances were
detected in the epicuticle with ferric ferricyanide.
With ferricyanide, reducing substances in the cuticle varied cyclically with the
molting cycle. During intermolt and proecdysis, the hardened cuticle did not react.
In contrast, reducing substances penetrated the new exocuticle at ecdysis and reacted
intensely; the endocuticle reacted less intensely. The reducing substances were not
detected during postecdysis. Reduction reversed the results obtained in the negative
(quinonized) cuticles of specimens fixed during postecdysis and visualized reducing
substances. However, no change was induced within the fully-quinonized cuticles of
intermolt, proecdysis, and late postecdysis. Oxidation of reducing substances present
in the cuticle at ecdysis rendered them negative.
Oxidizing substances (visualized by Fell-acid ferricyanide) were mildly or not
detectable in the endocuticle and the epicuticle. The exocuticle reacted intensely.
Using Clara's hematoxylin, copper was moderately visualized in the fully-formed,
hardened intermolt cuticle. During postecdysis, the visualization of copper in the
exocuticle decreased continuously. Curiously, intense amounts of copper were seen
in the endocuticle during late postecdysis. The epicuticle did not react.
DISCUSSION
By histochemistry and morphology, the present report identifies five types of
-aocytes in the hemocoel of the fiddler crab. The two most commonly encountered
s are a small agranular hyaline cell, or cystocyte, characterized by a scanty basophilic
HEMOCYTES AND TANNING 771
cytoplasm encircling a densely basophilic nucleus; and a larger granulocyte containing
numerous eosinophilic granules within an eosinophilic cytoplasm. The other three
hemocytes were: an intermediate cell, partially resembling the hyaline cell and the
granulocyte, and thus appearing to be a transitional stage in the granulocyte maturation
process (Toney, 1958; Ravindranath. 1980); a small granular cell, and a large granular
cell. Perhaps the latter two also represent transitional stages in the granulocyte mat-
uration process (see Ravindranath, 1980, for review). However, their functions are
unknown.
The granulocytes of several arthropod species transform their shape and degranulate
on exposure to air into intermediate cells and hyaline cells (Wharton Jones, 1846;
Hardy, 1892; Vranckx and Durliat, 1977). Degranulation after swelling has been
associated with clotting in Limulus and Homarus (Dumont et al. 1966; Hearing,
1969). In vertebrates, degranulation may result from cell injury, autolysis, aging, death
(Deruby, 1918; Myers and Dewolf-Glade, 1964). The present data show evidence for
degranulation of intermediate cells (Figs. 2B, C), and large granular cells (Figs. 5A,
C) and include indirect data for degranulation of small granular cells (personal ob-
servations) and granulocytes (histochemically by their resemblance to serum granules).
The hyaline cells and the granulocytes of Uca pugilator may be involved in tanning
at certain points in the molting cycle. Counts of the numbers of hemocytes show
cyclic events occur and verify earlier work that both granulocytes and hyaline cells
increase their numbers to a peak at ecdysis (Kollman, 1908). The present report
indicates the granulocytes predominate before ecdysis; the hyaline cells after ecdysis.
Histochemically, these hemocytes cyclically contain protein end groups and diphenols
respectively which seem to be shuttled into the serum and new exocuticle at ecdysis.
Biochemically, the blood appears to be the main site of tanning agent synthesis.
Using paper chromatography, Vacca and Fingerman (1975a) identified N-acetyldo-
pamine and N-acetylnoradrenalin as metabolites of labeled dopamine (as well as their
beta-glucosides) which appear in the blood of the fiddler crab, Uca pugilator, during
ecdysis. Subsequent incorporation into the cuticle suggests the N-acetylated dopamine
metabolites attach to the glucosides and then act as tanning agents. Prior to cuticle
incorporation, they become attached to two large blood proteins (>400,000 d and
~ 1 50,000 d) which transport the tanning agents into the soft cuticle. The appearance
of free glucosides and attached carrier proteins in the blood is cyclic and corresponds
to the incorporation of label into the cuticle during postecdysis (Vacca and Finger-
man, 1975b).
Histochemically, the diphenolic substance(s) visualized in the hyaline cells at
ecdysis and early postecdysis, when the hyaline cells occur in large numbers, may
represent the tanning agent(s) or precursor(s). Probably a primary catecholamine,
candidate tanning diphenols include norepinephrine (as demonstrated histochemically
in the adrenal medulla by Lillie et al., 1973a), DOPA, dopamine, N-acetyldopamine,
and N-acetylnorepinephrine. Interestingly, the hyaline cells lose the diphenol during
late postecdysis, as the new cuticle tans. They appear in large numbers near the
epidermis, looking empty as if their contents had been released. Like the hyaline cell,
the serum contains a phenol during intermolt, proecdysis, and early postecdysis.
However, by 48 hours postecdysis, its presence becomes diminished. Speculatively,
the diphenol in the serum originates from the hyaline cells, and enters the soft cuticle
during ecdysis and early postecdysis.
During late postecdysis, the diphenol cannot be visualized in the hardened ex-
ocuticle without reduction by dithionite. Presumably, the diphenol acts as a tanning
agent and cross-links with the cuticle protein matrix, transforming into the non-
reactive quinone form during late postecdysis, intermolt, and proecdysis.
Vacca and Fingerman (1975b) speculated that a permeability factor enables the
772 L. L. VACCA AND M. FINGERMAN
rapid transfer of tanning agents from the blood (hemocytes and serum) into the cuticle
during early ecdysis. Precedence for the hormonal control of tanning comes from
insect studies: ecdysone and ecdysterone accelerate the formation of dopamine from
precursor tyrosine within the hemocytes of tsetse fly puparia in vitro (Whitehead,
197 1 ); bursicon stimulates hemocyte permeability in the initial stages of tanning agent
synthesis, thereby enabling them to overcome a concentration barrier to tyrosine
(Whitehead, 1970). Bursicon also stimulates lysine uptake by the cuticle (Fogal and
Fraenkel, 1969). The diuretic hormone of the American cockroach enables the removal
of excess liquid from the blood (via Malpighian tubules) during postecdysis, and also
enhances the uptake of compounds such as tyrosine by the hemocytes and epidermal
cells (Mills and Whitehead, 1970). Among the crustaceans, ecdysone triggers protein
synthesis within the hemocytes of the crayfish Orconectes limosus during proecdysis
(Keller and Adelung, 1970).
Various histochemical procedures visualize proteins, amino end groups, and amino
acids within the granulocytes. Cyclic histochemical reactions imply that these hem-
ocytes serve in the production of proteins during the molting cycle. Some of these
groups (arginine and tyrosine, lysine and histidine) occur in the granulocytes (cytoplasm
and granules) throughout the molting cycle, and can be visualized as structural elements
of the protein matrix of the cuticle and the granules of the granulocytes. Other end
groups appear cyclically: tryptophan (also appearing in serum) became visible in the
granulocytes and serum during early postecdysis. Although detectable in the cuticle
matrix throughout the molting cycle, lysine, histidine, and aromatic protein end-
groups become histochemically intense and probably enter the cuticle during early
postecdysis when it is still soft. Indeed, water-soluble proteins extracted from insect
cuticle exhibit the free end groups of lysine: but the same groups cannot be dem-
onstrated in sclerotinized cuticle (Hackman, 1953). The visualization of additional
protein end-groups in the cuticle may represent the incorporated protein carriers
detected biochemically (Vacca and Fingerman, 1975a, b). The tanning protein in the
hemolymph of the insect Manduca sexta is immunologically identical to cuticle
protein (Koeppe and Gilbert, 1973). Unfortunately, the precise relationship between
the cuticle protein matrix of Uca and the proteins carried by the granulocytes cannot
be precisely determined from the present data.
The granules of the granulocytes contain basic (amino) end groups (lysine, arginine,
and histidine). The reactions in serum suggest that these granule constituents are
released after ecdysis as well. As the new exoskeleton forms, numerous free granules
appear in the serum thereby encouraging the speculation that the granulocytes release
their proteinaceous granules, as well as a cytoplasmic protein, into the serum during
early postecdysis. By 48 hours postecdysis, the release process seems to be complete.
Diverse serum granules were detected histochemically in Uca and have been reported
in other arthropod hemocytes by histochemical and ultrastructural studies (see Rav-
indranath, 1980, for review). The different granules may represent stages in the co-
agulation process (Ravindranath, 1980), or may possess different functions including
basement membrane formation, wound healing (Ravindranath, 1980), or tanning.
The diverse functions may account for some of the staining variations of granules
visualized in this report within the granulocytes, large and small granular cells, and
"serum granules."
Granulocytes which contain reducing substances (probably non-phenolic) during
most of the molting cycle, become unreactive at ecdysis and 5 hours later, as if their
reducing substances become released. Deep within the hemocoel, small, perhaps im-
mature, granulocytes still react intensely (Fig. 1C). The data coincide with the synthesis
and release of a weakly acidic glycoprotein (perhaps a carrier which contains sulfhydryl
or other reducing groups) during early postecdysis; alternatively, protein synthesis
HEMOCYTES AND TANNING 773
becomes blocked or breakdown increases. Minute amounts of native aldehyde de-
tectable within the granulocytes during intermolt, proecdysis, and late postecdysis
also disappear during early postecdysis.
Surprisingly, serum (apart from its contained granules) contains few soluble re-
ducing substances during postecdysis, when biochemically it sequesters both tanning
phenols and protein carriers (Vacca and Fingerman, 1975a). Perhaps the weak his-
tochemical reaction reflects their transient presence; alternatively these substances
are not detectable because they are bound to glucosides (Vacca and Fingerman, 1 975a)
or to the granules released from the granulocytes.
In other arthropod species, hemolymph proteins appear cyclically during the
molting cycle. Carcinus blood contains a glycoprotein throughout the molting cycle
which disappears at ecdysis and then reappears 10 days later (Busselen, 1970), the
time during which sclerotinization is complete in Uca. Its appearance and maintenance
depends upon the nutritional status of the organism. Gecarinus also possesses a blood
protein involved in clotting which becomes barely detectable during postecdysis
(Strutman and Dolliver, 1968). The present study shows that intermediate cells rupture
easily and spill their proteinaceous granules into the serum during early postecdysis.
These cells may be involved in clotting. Curiously, the ability of the blood to clot is
minimal during postecdysis (Strutman and Dolliver, 1968). A noteworthy speculation
as to why the animal is at such a disadvantage when it is most susceptible to injury
and infection might be that most of the hemocytes instead become involved in the
synthesis of other substances (proteins and diphenols) to be used for sclerotinization.
Indeed, this function would take priority in order to reinstate the animal into its
protective shell after growth.
The mechanism of hemocyte degeneration may play a significant role in supplying
tanning agents to the cuticle. By late postecdysis, the serum and hyaline cells of Uca
become exhausted of diphenols. With the loss of their diphenols, few hyaline cells
remain intact and their numbers diminish severely. In addition, granulocytes release
their granules and reduce in number during early postecdysis. Hypothetical tanning
hormone(s) might increase hemocyte permeability to substances, thereby causing
swelling and eventual lysis. This mechanism could account for the numerous granules
visualized in the serum during postecdysis, and has been proposed for the numerical
decrease in the hemocytes during postecdysis (Marrec, 1944). Histochemical and
ultrastructural evidence exists for the disintegration and vesiculation of lipoprotein
cells and nuclei with subsequent streaming-in of neighboring hemocytes during proec-
dysis and postecdysis in the crab Paratelphusa (Adiyodi and Adiyodi, 1972). Cell
explosion of hyaline cells and granule release by intermediate cells and granulocytes
have been postulated as mechanisms of clotting (Hardy, 1892; Tail and Gunn, 1918;
Wood et al., 1971; see review by Ravindranath, 1980), and tyrosinase liberation
(Pinhey, 1930).
Based on his in vitro studies. Summers (1968) proposed that the epidermis, not
the blood, is the site of tanning agent synthesis. We now present evidence that diphenols
(presumably tanning agents) appear in the epidermis transiently between ecdysis and
early postecdysis. These data, as well as previous biochemical evidence (Vacca and
Fingerman, 1975b), suggest that the epidermis is a site of translocation rather than
synthesis. Degeneration as a mechanism of tanning would cause the hemocytes to
release their tyrosine-metabolizing enzymes, and would account for Summer's findings
that most of the tyrosinase enzyme activity occurs in the plasma, and not the hemocytes,
of the fiddler crab (Summers, 1967).
Histochemically, our study shows the hyaline cells in Uca take up copper, a
component of hemocyte tyrosinase (Pinhey, 1930). Functionally, tyrosinase oxidizes
phenols (tanning agents) to quinones which then act as strong oxidizing agents. Under
774 L. L. VACCA AND M. FINGERMAN
pathological conditions, quinones respond to injury and infection by forming melanin
upon coagulation (Taylor, 1969). Therefore, despite a deficient clotting mechanism,
perhaps the soft-shelled crab possesses the enzyme complex within the hemocytes
and eventually the serum, during tanning as a ready system for defense. Indeed, the
granulocytes also take up copper (especially during early postecdysis when the serum
is least reactive), and contain a substance which can oxidize Clara's hematoxylin.
Koeppe (1971) proposed that tyrosinase is the actual protein carrier of tanning
agents in insects. Unfortunately, no oxidizing capacity could be detected histochem-
ically in the serum, and though the exocuticle oxidizes Fell (by acid ferrocyanide
reaction) at ecdysis, it resists copper uptake. Therefore, we propose another blood
protein complex may be involved in the tanning process. The tegumental glands,
known to possess tyrosinase, also oxidize Fell and may function in an aspect of
tanning (Stevenson, 1963a, b) which histochemically relates to the epicuticle. Indeed,
multiple mechanisms may exist as implied by variable basophilia or acidophilia of
exo- and endocuticle.
Interestingly, reducing sites were demonstrated in the melanophore granules at
ecdysis (intensified ferricyanide reaction after oxidation). Azo-coupling reactions
demonstrate the presence of a protein matrix which contains amines, diphenols, and
indoles. After ecdysis, these histochemical reactions changed. The data imply that
amine and phenolic (tanning?) substances enter a protein-granule matrix within the
melanophores during ecdysis, when tanning agents and permeability factors are avail-
able. Indeed melanin may form at the end of the tanning process as the result of a
biochemical equilibrium displacement of tanning agent biosynthesis.
Histochemically, the epidermal melanophore granules resemble melanin by their
argentophilia, strong basophilia, ability to take up iron, insolubility in organic solvents,
and their negative PAS, acid fast, and lipid reactions. However, Noel (1982) reports
that Uca melanophore pigment is ommochrome. Indeed the present study does not
define the granules as melanin. Dithionite used for reduction extracted many me-
lanophore granules; those which remained could not be reduced even after 8 hours.
The epidermal melanophores exhibited the presence of a cytoplasmic diphenol
throughout the molting cycle. Unlike the melanophore granules, perhaps the cytoplasm
maintains a separate pool of diphenols in continuous supply. It is tempting to speculate
that the production of diphenolic tanning agents within the hemocytes might relate
biochemically to the process of pigment formation.
Indeed certain histochemical reactions of the granules visualized in the small and
large granular cells resembled those of the melanophore granules. The appearance of
numerous large granular cells near the epidermis during early postecdysis suggests a
relationship with melanophore formation, which has not yet been defined.
Interestingly, the large granular cells azo-couple (DAS-AzA) only during ecdysis
and postecdysis. Apparently, they sequester tanning agents from the blood and deposit
them on the protein matrix of the contained swollen granules. In this way, the large
granular cells, like the melanophores, may function in the disposal of excess tanning
agents remaining in the blood during postecdysis. This mechanism of disposal provides
an alternative to glucoside formation which masks the tanning agents prior to use
(Vacca and Fingerman, 1975a, b).
In conclusion, these histochemical data parallel previous biochemical findings
which document the cyclic appearance of a protein-bound phenolic tanning agent
in the blood of Uca pugilator (Vacca and Fingerman, 1975a, b). We now show that
granulocytes contain protein end groups in their cytoplasm and granules which might
represent portions of the protein carrier complex: the hyaline cells contain diphenols
which could act as tanning agents. The reducing groups of proteins and diphenols
HEMOCYTES AND TANNING 775
become demonstrable in the cuticle during the early postecdysis and cannot be vi-
sualized by 10 hours postecdysis when tanning (quinonization) takes place. Addi-
tionally, the cyclic presence of proteins and diphenols in the blood cells, serum, and
cuticle indicates the existence of a tanning hormone (perhaps more than one) which
enables the cyclic synthesis of the diphenolic tanning agents and protein carriers, and
appropriately shuttles them from the hemocytes into the serum and cuticle during
the molting cycle. Fingerman and Yamamoto (1964) provided evidence that tanning
of the cuticle of the dwarf crayfish, Cambarellus shufeldti, is hormonally controlled.
However, the tanning hormone(s) remain(s) to be identified in crustaceans.
ACKNOWLEDGMENTS
Completion of this work was partially supported by a grant from the Committee
to Combat Huntington's Disease (KUMC Grant No. 9712-01) and the University of
Kansas Biomedical Research Support Grant (KU SO7-RR05373) (L.L.V.). The work
was initiated at Tulane University, Department of Biology, in partial fulfillment for
the Ph.D. degree (L.L.V.).
LITERATURE CITED
ADIYODI, R. G., AND K. G. ADIYODI. 1972. Histochemistry of the integumentary connective tissue of a
crab in relation to the molt cycle. J. Histochem. Cytochern. 20: 545.
AWAPARA, J. 1962. Free amino acids in invertebrates: A comparative study of their distribution. In Amino
Acids Pools, J. T. Holden, ed. Elsevier, Amsterdam.
BANG, F. 1967. Serological responses among invertebrates other than insects. Fed. Proc. 26: 1680-1684.
BARKA, T., AND P. J. ANDERSON. 1963. Histochemistry: theory, practice, and bibliography. Harper and
Row, New York. 660 pp.
BROWN, F. A., JR., AND O. CUNNINGHAM. 1939. Influence of the sinus gland of crustaceans on normal
viability and ecdysis. Biol. Bull. 77: 104-1 14.
BRUNET, P. C. J. 1965. The metabolism of aromatic compounds. Pp. 49-78 in Aspects of Insect Biochemistry,
T. W. Goodwin, ed. Academic Press, New York.
BRUNTZ, L. 1907. Etudes sur les organes lymphoides phagocytairien et excreteurs des Crustaces superieurs.
Arch. Zoo/. Exp. Gen. 1: 1-67. '
BUSSELEN, P. 1970. Effects of moulting cycle and nutritional conditions on haemolymph proteins in
Carcinus maenas. Comp. Biochem. Physiol. 37: 73-83.
DECLEIR, W., AND R. VERCAUTEREN. 1965. Activite phenoloxidasique dans les leucocytes due crabe au
cours du cycle d' intermue. Cah. Biol. Mar. 6: 163-172.
DERUBY, K. G. 1918. A study on autolysis of animal tissue. J. Biol. Chem. 35: 179-219.
DUMONT, J. N., E. ANDERSON, ANDG. WINNER. 1966. Some cytological characteristics of the hemocytes
of Limulus during clotting. J. Morphol. 119: 181-208.
EVANS, P. D. 1972. The free amino acid pool of the hemocytes of Carcinus maenas (L). /. Exp. Biol. 56:
501-507.
FINGERMAN, M., AND S. W. FINGERMAN. 1974. The effects of limb removal on the rates of ecdysis of
eyed and eyestalkless fiddler crabs, Uca pugilator. Zoo/. Jb. Physiol. 78: 301-309.
FINGERMAN, M., AND Y. YAMAMOTO. 1964. Endocrine control of tanning in the crayfish exoskeleton.
Science 144: 1462.
FLORKIN, M., AND E. SCHOFFENIELS. 1965. Euryhalinity and the concept of physiological radiation. In
Studies in Comparative Biochemistrv, K. A. Munday, ed. Macmillan (Pergamon Press) Co., New
York.
FOGAL, W. H., ANDG. FRAENKEL. 1969. Melanin in the puparium and adult integument of the fleshfly,
Sarcophaga bullata. J. Insect Physiol. 15: 1235-1247.
GEORGE, W. C., AND J. NICHOLS. 1948. A study of the blood of some Crustacea. J. Morphol. 83: 425-
443.
GLENNER, G. G., AND R. D. LILLIE. 1957. The histochemical demonstration of indole derivatives by the
post-coupled p-dimethylaminobenzylidene reaction. /. Histochem. Cytochern. 5: 279-296.
GUYSELMAN, J. B. 1953. An analysis of the molting process in the fiddler crab, Uca pugilator. Biol. Bull.
104: 115-137.
HACK.MAN, R. H. 1953. Chemistry of the insect cuticle. II. Hardening and darkening of the cuticle. Biochem.
J. 54: 367-377.
776 L. L. VACCA AND M. FINGERMAN
HARDY, W. B. 1892. The blood corpuscles of the Crustacea together with a suggestion as to the origin of
the Crustacean fibrin ferment. /. Physiol. 13: 165-190.
HEARING, V. G. 1969. Demonstration of acid phosphatase activity in the granules of the blood cells of
the lobster, Hornarus americanus. Chesapeake Sci. 10: 24-28.
KELLER, R., ANDD. ADELUNG. 1970. Vergleichende morphologische und physiologische Untersuchungen
des Integumentgewebes und des Hautungshormongehaltes beim Flusskrebs Orconectes limosus
wahrend eines Hautungszyklus. Wilhelm Roux' Arch. 164: 209-221.
KOEPPE, J. K. 1971. Cuticle Sclerotinization: Metabolites of Dopamine and a High Molecular Weight
Carrier System of Diphenol Glucosides in the American Cockroach. Doctoral Dissertation, Tulane
University.
KOEPPE, J. K., AND L. I. GILBERT. 1973. Immunochemical evidence for the transport of haemolymph
protein into the cuticle of Manduca sexta. J. Insect Physiol. 19: 615-624.
KOLLMAN, M. 1908. Recherches sur les leucocytes et le tissue lymphoide des invertebres. Ann. Sci. Nat.
Zool. 7: 1-240.
LEVIN, J. 1967. Blood coagulation and endotoxin in invertebrates. Fed. Proc. 26: 1707-1712.
LILLIE, R. D. 1957. Metal reduction reactions of the melanins: histochemical studies. J. Histochem. Cvtochem.
5: 325-333.
LILLIE, R. D. 1965. Histopathologic Technic and Practical Histochemistry, 3rd Ed. McGraw-Hill Book
Co., New York. 715 pp.
LILLIE, R. D., P. PIZZOLATO, H. C. DESSAUER, AND P. DONALDSON. 1971. Histochemical reaction at
tissue arginine sites with alkaline solutions of beta-napthoquinone-4-sodium sulfonate and other
oquinones and oxidized o-diphenols. A possible mechanism of the Sakaguchi reaction. J. Histochem.
Cvtochem. 19: 487-497.
LILLIE, R. D., P. PIZZOLATO, L. L. VACCA, R. A. CATALANO, AND P. T. DONALDSON. 1973a. Use of
diazosulfanilic acid, pH 1 azure A sequence of adrenal medulla and the effects of prior oxidation
and reduction of the reaction in several species: man, monkey, hog, dog, cat, guinea pig, rat,
mouse, gerbil, opossum. J. Histochem. Cvtochem. 21: 448-454.
LILLIE, R. D., P. PIZZOLATO, L. L. VACCA, R. A. CATALANO, AND P. T. DONALDSON. 1973b. The pH
range of the diazosafranin reaction of rat and other mast cells. J. Histochem. Cvtochem. 21: 441-
447.
LILLIE, R. D., P. PIZZOLATO, L. L. VACCA, R. A. CATALANO, AND?. T. DONALDSON. 1973c. Histochemical
azo-coupling reactions: a catecholamine in enterochromaffin cells in place of or in addition to 5-
hydroxytryptamine. J. Histochem. Cvtochem. 21: 455-463.
MARREC, M. 1944. L'organe lymphocytogene des Crustaces Decapodes et son activite cyclique. Bull. Inst.
Oceanogr. Monaco 867: 1-4.
MILLS, R. R., AND D. L. WHITEHEAD. 1970. Hormonal control of tanning in the American cockroach:
Changes in blood cell permeability during ecdysis. J. Insect Physiol. 16: 331-340.
MOWRY, R. W. 1963. The special value of methods that color both acidic and vicinal hydroxyl groups in
the histochemical study of mucins. With revised directions for colloidal iron stain, the use of
Alcian Blue GBX and their combinations with the periodic acid-Schiff reactions. Ann. New York
Acad. Sci. 106: 402-423.
MYERS, D. K., AND DEWOLFE-GLADE. 1964. Some effects of cationic condensing agents on rat thymocyte.
Exp. Cell Res. 33: 344-349.
NOEL, P. 1982. Does melanin occur in crab's melanophores? Fourth European Workshop on Melanin
Pigmentation (Abst.), Sept. 21-23, 1982.
PINHEY, K. G. 1930. Tyrosinase in crustacean blood. J. Exp. Biol. 7: 19-36.
RAVINDRANATH, M. H. 1980. Haemocytes in haemolymph coagulation of arthropods. Biol. Rev. 55: 139-
170.
SKINNER, D. M. 1962. The structure and metabolism of a crustacean integumentary tissue during a molting
cycle. Biol. Bull. 123: 635-647.
SKINNER, D. M., AND D. E. GRAHAM. 1972. Loss of limbs as a stimulus to ecdysis in Brachyura (true
crabs). Biol. Bull. 143: 222-233.
STEVENSON, J. R. 1963a. Phenols in the cuticle of the fiddler crab, Uca pugnax and their possible function
in sclerotinization. Biol. Bull. 125: 392-393.
STEVENSON, J. R. 1963b. A preliminary investigation of sclerotin precursors in crustacean cuticles. Am.
Zool. 3: 253.
STRUTMAN, L. V., AND M. DOLLIVER. 1968. The mechanism of coagulation in Gecarcinus lateralis. Am.
Zool. 8:481-489.
SUMMERS, N. M. 1967. Cuticle sclerotinization of blood phenoloxidase in the fiddler crab, Uca pugnax.
Comp. Biochem. Physiol. 23: 129-138.
SUMMERS, N. M. 1968. The conversion of tyrosine to catecholamines and the biogenesis of N-acetyldopamine
in isolated epidermis of the fiddler crab, Uca pugilator. Comp. Biochem. Physiol. 26: 259-269.
TAIT, J., AND J. D. GUNN. 1918. The blood ofAstacusfluviatilis: a study in crustacean blood, with special
reference to coagulation and phagocytosis. Q. J. Exp. Physiol. 12: 35-80.
HEMOCYTES AND TANNING 777
TAYLOR, R. L. 1969. A suggested role for the polyphenol-phenoloxidase system in invertebrate immunity.
J. Invert. Pathol. 14: 427-428.
TONEY, M. E. 1958. Morphology of the blood cells of some Crustacea. Growth 22: 35-50.
VACCA, L. L., AND M. FINGERMAN. 1975a. The mechanism of tanning in the fiddler crab, Uca pugilator.
I. Tanning agents and protein carrier in the blood during ecdysis. Comp. Biochem. Phvsiol. 51B:
475-481.
VACCA, L. L., AND M. FINGERMAN. 1975b. The mechanism of tanning in the fiddler crab, Uca pugilator.
II. The cyclic appearance of tanning agents and attached carrier proteins in the blood during the
molting cycle. Comp. Biochem. Physiol. 51 B: 483-487.
VRANCKX, R., AND M. DURLIAT. 1977. Etude des hemocytes circulants de Crustaces Decapodes. Correlations
entre polymorphisime et mode de prelevement. C. R. Acad. Set. Paris 285B: 1045-1047.
WHARTON-JONES, T. 1846. The blood corpuscle considered in its different phases of development in the
animal series. Memoir II — Invertebrate. Phil. Trans. R. Soc. Loud. 136: 89-101.
WHITEHEAD, D. L. 1969. New evidence for the control mechanism of sclerotinization in insects. Nature
224: 721-723.
WHITEHEAD, D. L. 1970. L-Dopa decarboxylase in the haemocytes of Diptera. FEBS Lett. 1: 263-266.
WHITEHEAD, D. L. 1971. Studies on the control of pupariation in tsetse flies. Trans. R. Soc. Trap. Med.
Hyg. 65: 224-225.
WOOD, P. J., AND L. P. VISENTEN. 1967. Histology and histochemical observations of the hemolymph
cells in the crayfish, Orconectes virilis. J. Morphol. 123: 559-568.
WOOD, P. J., J. PODLEWSK.I, AND T. E. SHENK. 1971. Cytochemical observations of hemolymph cells
during coagulation in the crayfish, Orconectes virilis. J. Morphol. 134: 479-489.
Reference: Biol. Bull. 165: 778-790. (December, 1983)
REGENERATION OF INJURIES AMONG JAMAICAN GORGONIANS:
THE ROLES OF COLONY PHYSIOLOGY AND ENVIRONMENT
CHARLES M. WAHLE*
Department of Earth and Planetary Sciences, The Johns
Hopkins University, Baltimore, Maryland 21218
ABSTRACT
The consequences of injury to reef dwelling colonial animals are determined partly
by rates of regeneration of lost tissues. These experiments examined two potential
influences on regeneration rates of Jamaican gorgonians: 1 ) intrinsic physiological
and energetic differences among co-occurring, conspecific colonies differing in size,
reproductive phase, or injury location; and 2) differential responses among three
plexaurid species to changing environmental variables across their depth range. In
Plexaura homomalla, regeneration rate varied with the location of injury within
colonies, but was unexpectedly independent of either colony size or reproductive
phase. In addition, colonies of P. homomalla, Eunicea mammosa, and Plexaurella
dichotoma differed in relative ability to regenerate equivalent injuries in different reef
zones across their depth range.
"There is one fact in the life-history of corals which the study of processes of repair
clearly brings out, and it is this, that all the methods of regeneration are more for
the life-saving of the colony than of the individual.
Wood-Jones, 1912
INTRODUCTION
Injury is common among arborescent Caribbean gorgonians during both cata-
strophic (e.g., hurricane) and routine conditions (Cary, 1914, 1918; Bayer, 1961;
Kinzie, 1970, 1973, 1974; Birkeland 1974; Birkeland and Gregory, 1975; Kitting,
1975; Wahle, 1980; Woodley et al, 1981). Moreover, injuries to Jamaican gorgonians
can exhibit complex variation in both frequency and pattern among colonies living
in different reef zones (Woodley et al., 198 1 ; Wahle, in prep.). Although many injuries
are limited initially to a few cm of tissue (Cary, 1914; Kinzie, 1970, 1974), their
effects on colonies can be subtle, delayed, and extensive. They can range from pro-
portional reduction in the number of feeding, reproductive, and defensive polyps
(e.g., Jackson, 1977, 1979), to disruption of colony-wide physiological integration
(Bayer, 1961, 1973; Wainwright and Dillon, 1969; Preston and Preston, 1975; Mur-
dock, 1978a, b), and eventually to complete overgrowth by encrusting organisms
(Kinzie, 1970, 1974; Wahle, 1980; and references therein). The ultimate extent and
duration of the various effects of injury are determined largely by the time required
to regenerate the lost tissue and cover the internal, proteinaceous axis (Kinzie, 1970,
1974; Kitting, 1975; Lang da Silveira and van't Hof, 1977; for other taxa: Glynn,
1976; Bak et al., 1977; Jackson, 1977; Jackson and Palumbi, 1979; Palumbi and
Jackson, 1982). Consequently, any intrinsic or extrinsic variable(s) affecting rates of
Received 18 November 1982; accepted 8 August 1983.
* Present address: Department of Zoology, The University of Texas, Austin, Texas 78712.
778
GORGONIAN REGENERATION 779
regeneration may indirectly but profoundly affect the survivorship, fecundity, and
general ability of the colony to perform basic biological and ecological functions.
In this paper, I examine two sources of variation in the in situ ability of Jamaican
arborescent gorgonians to regenerate injuries simulating those occurring routinely in
nature (Woodley el al., 198 1 ). The first experiment considered the influence of colony
physiology and energetics on the regeneration rates in the common plexaurid. Plexaura
homomalla. Specifically, it tested the separate effects of colony size, colony reproductive
phase, and location of injury on rates of regeneration of equal sized wounds placed
on co-occurring colonies. The second experiment examined the relative regeneration
rates of three common plexaurid species across their depth ranges. It contrasted
regeneration rates of equivalent injuries among replicate colonies of P. homomalla,
Plexaurella dichotoma, and Eunicea mammosa in three reef zones in and near Dis-
covery Bay, Jamaica.
The two experiments differed in rationale, chronology, methods, and implication.
Therefore, I first describe methods common to both, followed by separate treatments
of specific methods, results, and discussion for each experiment.
MATERIALS AND METHODS
Sites
Regeneration experiments were conducted in situ in four reef zones in and near
Discovery Bay, Jamaica during winter 1977 and summer 1978 (Fig. 1). The zones
differed primarily in depth, exposure to waves, and gorgonian abundance (Kinzie,
1970, 1973; Woodley et al., 1981). Many of the habitat characteristics described
below, particularly the topography and structure of the benthic communities, were
significantly altered by the passage of Hurricane Allen in August 1980 (Woodley et
al., 1981). Consequently, these descriptions apply to pre-storm conditions only.
The Mixed Zone (Fig. 1, site 1) is a shallow (7 m) hardground seaward of the
reef crest containing an abundant and diverse gorgonian fauna (mean colony density
of 14.6/m2; see Woodley et al., 1981 for survey methods). The East Fore Reef Terrace
(Fig. 1, site 2; henceforth called the Terrace) is a gently sloping plain at 15 m,
characterized by thickets ofAcropora cervicornis, scattered massive corals and sponges,
and a diverse gorgonian assemblage (mean densities of 2.9 colonies/m2). The Rear
Zone (Fig. 1, site 3) lies slightly west of the mouth of Discovery Bay and immediately
leeward of the reef crest in depths ranging from 0.5 m to 1.5 m. Gorgonians were
relatively rare, with mean colony densities of 0.3/m2. The Shallow East Fore Reef
(Fig. 1, site 4) lies southeast of the Terrace (site 2) and immediately east of the mouth
of the bay. Although no quantitative surveys were conducted here, the site was similar
to the Mixed Zone (site 1 ) in most respects relevant to this study.
Species
The three species chosen for these experiments span a range of polyp and colony
morphologies characteristic of common, reef dwelling Caribbean plexaurids (Bayer,
1961; Kinzie, 1970). All occur as adult colonies in each zone and are frequently
among the dominant members of gorgonian assemblages throughout the Caribbean
and southwestern Atlantic (Bayer, 1961; Kinzie, 1970, 1973; Opresko, 1973).
Plexaura homomalla is perhaps the most studied of the Caribbean gorgonians
(e.g. Cary, 1914, 1918; Kinzie, 1970, 1973; Bayer and Weinheimer, 1974, and papers
therein; Wahle 1980). Its colonies are relatively large (roughly 1 m) with either planar
or bushy branching patterns (Kinzie 1970, 1974). Colonies possess relatively thick
780
C. M. WAHLE
"t
DISCOVERY
BAY
FIGURE 1. Location of experimental sites in four reef zones in and near Discovery Bay, Jamaica
(inset): (1) Mixed Zone, (2) East Fore Reef Terrace. (3) Rear Zone, and (4) Shallow East Fore Reef. Dotted
Line: reef crest; arrows: direction of prevailing winds and swells; asterix: D.B.M.L.
coenenchyme, and have small polyps with light spicular ornamentation on the verrucae
(analogous to calyces among scleractinians; Bayer, 1961). Plexaurella dichotoma forms
large (roughly 1.5 m), dichotomously branched colonies with thick coenenchyme,
and long, unarmored polyps. Eunicea mammosa grows as relatively small (0.5 m),
planar, candelabra-shaped colonies with thin coenenchyme and moderately long exert
polyps which have heavily armored verrucae.
Techniques
Experimental injuries were placed on colonies in situ by carefully removing, with
a scalpel, all tissue and sclerites (including the axial sheath) from around the internal,
proteinaceous axis. The number, size, location, and timing of experimental injuries
varied between the two experiments and are described separately below. Within each
comparison of regeneration times, all injuries were equal in size (1.0 or 2.0 cm,
measured by vernier calipers to within 0.1 mm) and were initiated simultaneously
(within 48 hours of each other, unless otherwise specified).
The extent of tissue regrowth was recorded daily at 0700 hours. Regeneration
was deemed complete when the internal axis was completely covered by gorgonian
tissue and was no longer susceptible to fouling. The data, which do not satisfy the
assumptions of analysis of variance (Sokal and Rohlf, 1969; Zar, 1974), were analyzed
using non-parametric tests (i.e., Kruskal-Wallis and Mann- Whitney).
GORGONIAN REGENERATION
781
RESULTS
Effects of colony physiology on regeneration in Plexaura homomalla
The following experiment examined the separate effects of colony size, colony
reproductive phase, and injury location on regeneration rate among colonies of Plex-
aura homomalla on the Shallow East Fore Reef (Fig. 1, site 4). The experimental
design (depicted schematically in Fig. 2) consisted of four paired comparisons of
regeneration times (Fig. 2, bottom). A standard, simulated natural injury on 5 replicate
control colonies (Fig. 2; labeled control) was compared to each of four other treatment
groups differing from the controls in only one of the following variables: size, repro-
ductive phase, or injury location (2 treatments). Controls consisted of 5 large (40-
60 cm in height and width) replicate colonies, each with a single, one cm injury
TISSUE
1 CM.
AXIS
1
control
COLONY SIZE : LARGE
REPRODUCTIVE: YES
LOCATION : PRIMARY
MEAN NO. DAYS : 4.60
STD. DEV. : 0.55
NO. INJURIES 5
SMALL
YES
PRIMARY
4.80
0.45
5
LARGE
NO
PRIMARY
4.00
0.00
4
LARGE
YES
MID-COLONY
5.00
0.00
5
COMPARISONS :
SIZE (p>0.l)
REPRODUCTIVE PHASE (p>0.05)
LOCATION: PRIMARY vs. MID-COLONY (p> 0.2)
LOCATION: PRIMARY vs. TIP ( p< O.OOS)
FIGURE 2. Colony physiology experiment: experimental design and results. Top: type of experimental
injuries. Center: schematic of control and experimental colonies, each described below by three variables
(size, reproductive phase, and injury location) and by results of regeneration experiments (mean number
of days to regenerate, standard deviation, and total number of replicate injuries). Bottom: results of paired
comparisons between controls and four treatment groups differing by the indicated variable (using Mann-
Whitney tests).
782 C. M. WAHLE
placed mid-way down a terminal (primary) branch (Fig. 2, top and center, left).
Methods and results for each of the four paired comparison are described below and
in Figure 2.
With the exception of colonies in the non-reproductive treatment (described below),
all experimental injuries were made in mid-July, 1978. Experimental colonies were
equivalent in all obvious respects and were specifically chosen to have no external
evidence of previous injury or other abnormalities. Consequently, this experiment
controlled for many physiological and methodological variables potentially affecting
regeneration rates within a species, including colony condition. I necessarily assumed
that any other potential sources of variation affected all treatments equally or negligibly.
Colony size and regeneration. Connell (1973) and others (Fishelson 1973; Loya,
1976; Bak et ai, 1977) suggested that colony size might affect regeneration rates
among scleractinians by limiting the availability of energy for regrowth within small
colonies. Hence, assuming that energy is limiting to plexuarid gorgonians, one would
predict slower regeneration rates (longer regeneration times) among small injured
colonies differing from the larger controls only in colony size (10-20 cm versus 40-
60 cm in height and width).
The results (paired comparison labeled Size, bottom of Fig. 2) showed that, while
the trend in regeneration time was slightly in the predicted direction, small colonies
did not regenerate significantly slower than large controls (4.80 versus 4.60 days;
Mann- Whitney one-tailed test; P > 0. 1 ). Thus, under these conditions, the presumed
energetic differences between gorgonian colonies differing in size by up to 36-fold
(calculated as height X width) had no significant effect on their ability to regenerate
lost tissue.
Colony reproductive phase and regeneration. It has been suggested for a variety
of solitary and colonial taxa that, to the extent that energy is limiting, regeneration
and sexual reproduction may compete for energy and thus may be mutually inhibitory.
For example, repeated injury and regeneration may reduce subsequent sexual repro-
duction (bivalves: Trevaillion et ai, 1970; ectoprocts and sponges: Jackson, 1979;
Jackson and Palumbi, 1979; zoanthids: Karlson, 1981, 1983). This experiment tested
the converse hypotheses: that P. homomalla colonies at the peak of reproductive
activity (controls) should regenerate slower than comparable but non-reproductive
colonies not undergoing the simultaneous cost of gametogenesis. P. homomalla un-
dergoes an annual reproductive cycle with gametogenesis peaking in late-June to mid-
July (Goldberg and Hamilton, 1974; confirmed in Jamaica by in situ dissection and
observation of gametes). Non-reproductive colonies used in this experiment were
equivalent in all respects to the reproductive controls except that they were injured
in December, 1977, when gametes were lacking or poorly developed (Goldberg and
Hamilton 1974).
The results (paired comparison labeled Reproductive Phase in Fig. 2) show that,
although the non-reproductive colonies followed the predicted trend of slightly faster
regeneration rates (4.0 versus 4.60 days), the difference was not significant (Mann-
Whitney one-tailed test; P > 0.05). Thus, under these conditions, the ability of P.
homomalla to regenerate simulated, natural injuries was not significantly reduced by
the presumed energetic costs of simultaneous gametogenesis.
Location of injury and regeneration. Natural injuries do not occur randomly
within gorgonian colonies on northern Jamaican reefs. Rather, they tend to be con-
centrated on the colony periphery, and particularly on terminal or primary branches
(Wahle, in prep.). This experiment compared the regeneration rates of injuries placed
in three common locations on colonies of P. homomalla: primary branches (control),
branch tips, and mid-colony (Fig. 2, top and center).
GORGONIAN REGENERATION 783
The results (Fig. 2, bottom) revealed no significant difference in regeneration rate
between injuries in mid-colony and those on primary branches (5.00 versus 4.60 days;
Mann-Whitney two-tailed test; P > 0.2). In contrast, injuries on branch tips regenerated
significantly slower than those on primary branches of the controls (8.80 versus 4.60
days; Mann- Whitney one-tailed test, P < 0.005). Presumably, this two-fold difference
in regeneration rate exists because injuries on branch tips have only one tissue front
contributing to regrowth compared to two for injuries elsewhere in the colony (sensu
Lang da Silveira and van't Hof, 1977, for P. flexuosd).
Among certain well-studied colonial taxa such as the ectoprocts, regenerative
ability varies within colonies due to astogenetic gradients in zooid morphology and
condition (Jackson and Palumbi, 1979; Palumbi and Jackson, 1982). That equivalent
injuries (i.e., mid-colony versus primary) did not vary in regeneration rate within
these gorgonian colonies may reflect the apparent lack of comparable differentiation
of function among polyps in the shallow water, Caribbean gorgonians (Bayer, 1961,
1973). Nevertheless, natural injuries to holdfasts and basal tissues often fail to regenerate
(Gary, 1914, 1918; pers. obs. after Hurricane Allen, see Woodley et a/., 1981). This
pattern, combined with the two-fold difference in regeneration rates between injuries
on branch tips and those elsewhere on the colony, suggests that any systematic or
predictable variation in the location of injury within colonies may seriously influence
the ecological consequences of those injuries (particularly if on branch tips; Wahle
in prep.).
Effects of species and reef zones on regeneration
Morphological and physiological differences among colonial taxa may be reflected
in their ability to replace lost tissues (e.g., for ectoprocts and sponges: Jackson and
Palumbi, 1979). In addition, many plexaurid gorgonian species have relatively broad
depth ranges across Caribbean reefs (Bayer, 1961; Kinzie 1970, 1973, 1974; Opresko,
1973). Consequently, conspecific colonies which are potentially within the same
breeding population may experience very different environmental conditions, such
as the availability of food or light (Kinzie, 1970) or the frequency of natural injury
(Woodley et al., 1981). Variation in these environmental conditions may in turn
affect colony physiology, and specifically, rates of regeneration. This experiment ex-
amined regeneration rates among colonies of three common and morphologically
distinct plexaurids living in three reef zones in and near Discovery Bay, Jamaica.
Three related questions were addressed: 1) does a species' regenerative ability vary
across its depth range; 2) do co-occurring species differ in regeneration rates within
the same habitat; and 3) do the relative regeneration rates of the three species remain
constant across their depth range, or are they differentially affected by changes in
environmental conditions?
The species used in this experiment were Plexaura homomalla, Plexaurella di-
chotoma, and Eunicea mamrnosa (described in Methods). The experiments were
conducted in the Mixed Zone, the East Fore Reef Terrace (Terrace), and the Rear
Zone (Fig. 1, sites 1, 2, and 3 respectively), during late December, 1977. The exper-
imental design consisted of placing two equivalent injuries on each of three replicate
colonies for each of the three species in each of the three zones (Fig. 3; initial number
of colonies = 27, number of injuries = 54). Final sample sizes, after loss of 2 replicate
branches to storms and human disturbance, and elimination of 3 taxonomically
ambiguous Eunicea spp. colonies, are given in Table I. Experimental injuries were
made by completely removing 2 cm of tissue and sclerites mid- way down two terminal
(primary) branches on each colony (Fig. 3). Paired injuries were placed on opposite
784
C. M. WAHLE
TISSUE
2 CM
AXIS
TISSUE
PLEXAURELLA
DICHOTOMA
PLEXAURA
HOMOMALLA
EUNICEA
MAMMOSA
2 INJURIES x 3COLONIESX 3 SPECIES x 3 ZONES
FIGURE 3. Species-Reef Zone experiment: experimental design showing type of injury (top), location
of injuries on schematic representations of three species (center), and initial sample sizes (bottom).
sides of the colony to maximize their physiological independence and to minimize
the potential for any influence of integration among regions of the colony (see Bayer,
1973; Murdock, 1978a, b). All experimental colonies were chosen for maximum size
and minimum evidence of previous injury, and were thus presumed to be in optimal
physiological condition within each zone.
TABLE I
Regeneration time (in days) of injuries on colonies of three species in three reef zones in and near
Discover}' Bav*
Species
Reef zones
Differences
within
species
Mixed
Terr.
Rear
Plexaura homomalla
x:
(6.83)
(8.60)
(7.67)
P < 0.05
s:
1.17
1.34
0.52
n:
6
5
6
Plexaurella dichotoma
x:
(6.67)
(9.00)
(10.60)
P < 0.005
s:
0.52
1.55
0.89
n:
6
6
5
Eunicea mammosa
x:
(6.00)
(6.00)
(7.75)
P > 0.05
s:
0.00
0.82
1.26
n:
4
4
4
Differences within zones
P> 0.1
P < 0.01
P < 0.01
* Values are: x: mean regeneration time in days (parentheses); s: standard deviation; n: total number
of injuries; P: significance levels for six Kruskal-Wallis tests of regeneration time within species or zones.
GORGONIAN REGENERATION 785
The effect of environment on specific regeneration rates. Within two of the three
species examined (Plexuara homomalla, Plexaurella dichotoma), rates of regeneration
differed significantly across the three reef zones (Table I; rows). The exception was
seen among colonies of Eunicea mammosa, which showed no significant variation
in regeneration rate across the reef (Kruskal-Wallis two-tailed test; P > 0.05). For
each species (rows. Tables I and II) regeneration rates were generally fastest in the
Mixed Zone, and slowest in the Rear Zone.
For each of the three reef zones, an overall, grand mean regeneration time was
calculated for all co-occurring colonies, regardless of species. These three grand means
were then ranked, with the zone having the fastest overall regeneration rate (least
time) given primary rank (Table II, bottom row). These zone-specific regeneration
ranks showed the same pattern as did the data for the individal species: increasing
from Mixed to Rear Zone.
Differences in regeneration rate among co-occurring species. Within two of the
three reef zones examined (Terrace and Rear Zone), regeneration rates differed sig-
nificantly among the three co-occurring gorgonian species (Table I; columns). The
exception was in the Mixed Zone, where regeneration rates of the three co-occurring
species did not differ significantly (Kruskal-Wallis two-tailed test; P > 0.1). The relative
regeneration rates (i.e., fastest, intermediate, slowest) of the three species differed
from zone to zone across the reef (Tables I and II, columns). In general however,
the overall species ranking (Table II, right column) showed fastest rates of regeneration
among colonies of Eunicea mammosa, followed by Plexaura homomalla and Plex-
aurella dichotoma.
DISCUSSION
Colony physiology and regeneration rates in P. homomalla
All organisms must allocate presumably limited energy to various biological func-
tions such as growth, reproduction, regeneration and maintenance (e.g., Charnov and
Schaffer, 1973; Schaffer and Gadgill, 1975; Williams, 1975; Jackson, 1977; Stearns,
TABLE II
Mean regeneration rates of three species, ranked (*) for: conspecijic colonies in different reef zones
(rows); different species in the same zone (columns, parentheses); overall species rank (**); and
overall zone rank ("**)
Reef zones
Rank
Overall
Species
across
Mixed
Terr.
Rear species rank
Plexaura homomalla
zone:
1
3
2 2
species:
(3)
(2)
(1)
Plexuarella dichotoma
zone:
1
2
3 3
species:
(2)
(3)
(3)
Eunicea mammosa
zone:
1.5
1.5
3 1
species:
(1)
(1)
(2)
Overall zone rank
1
2
3
* Ranks increase with decreasing mean regeneration rate (i.e., 1, Fast; 3, slow).
** Overall ranks calculated as the ranked grand mean regeneration rate for each species across zones
(right), and for each zone across species (bottom).
786 C. M. WAHLE
1977; Jackson and Palumbi, 1979; Karlson, 1981). The results of these experiments
on regeneration among gorgonian colonies differing in some of the above variables
(Fig. 2) suggest that energetic tradeoffs among competing biological functions may
be more complex than previously thought for reef corals (Connell, 1973; Fishelson,
1973; Loya, 1976; Bak et ai, 1977).
For example, under the levels of injury tested in these experiments, rates of
regeneration were independent of both colony size and reproductive phase (Fig. 2).
These results, which contradict predictions based on simple energetic models, may
have at least three possible and not necessarily mutually exclusive explanations. First,
energy may seldom be limiting among reef-dwelling plexuarid gorgonians. This pos-
sibility will remain untestable until more is known about sources of nutrition and
the energetic costs of growth, reproduction, and regeneration among these colonial
animals. Second, the frequency and potential impact of injury on colony fitness may
be sufficiently great to have selected for maintaining a permanent capacity to replace
lost tissue, independent of other simultaneous energetic demands. Thus, for example,
while the allocation of energy or other limited materials (sensit Lang da Silveira and
van't Hof, 1977) may oscillate over time between growth and reproduction, gorgonians
may possess a permanent and independent reserve available for future regeneration.
Third, this experiment measured rates of regeneration under normal, but relatively
low levels of injury as compared to those occurring during catastrophic storms such
as hurricanes (Woodley et al, 198 1 ). Moreover, experimental colonies were specifically
chosen to have no evidence of previous injury or abnormalities which could potentially
affect regenerative ability (sensu Lang da Silveira and van't Hof, 1977; Jackson and
Palumbi, 1979; Palumbi and Jackson, 1982). Finally, the paired comparisons of
regeneration rates consistently showed the predicted trends, but failed to differ sig-
nificantly. Combined, these factors suggest that the effects on regeneration of colony
size and reproductive phase (and perhaps of other aspects of colony energetics) may
not become apparent until the intensity of injury (either natural or experimental) is
considerably higher than that tested here. For example, gorgonians seem to be able
to regenerate efficiently under low levels of natural injury (Kinzie, 1970; Birkeland
and Gregory, 1975; Kitting, 1975, and references therein). However, repeated injury
and regeneration among colonies ofPlexauraflexousa can inhibit future regeneration
by depleting a critical population of rate-limiting, interstitial and transitional cells
(Lang da Silveira and van't Hof, 1977). Thus, the predicted energetic constraints on
regeneration may become important mainly among colonies with large, numerous,
or repeated injuries. Such conditions could occur either routinely, in certain frequently
disturbed reef zones, or during hurricanes (Gary, 1914, 1918; Woodley et al., 1981).
Species, environment, and regeneration rates
Many common, Caribbean plexaurids extend in depth range across a variety of
reef zones and environmental conditions (Bayer, 1961; Kinzie, 1970, 1973; Opresko,
1973). Among the three species examined, the influence of the environment on rates
of regeneration was varied and complex (Tables I and II). Within two of the three
species examined (P. homomalla, P. dichotoma), conspecific colonies differed signif-
icantly in regeneration rate across their depth range. In addition, within two of three
reef zones examined (Terrace and Rear), co-occurring colonies of the three species
differed significantly in regeneration rate. Moreover, the relative rankings of overall
regeneration rates changed from zone to zone (Table II, across columns), and from
species to species (Table II, among rows). This changing pattern suggests a potential
interaction between species and environment on regeneration rate (hypothesis 3,
above; Sokal and Rohlf, 1969).
GORGONIAN REGENERATION 787
Despite this degree of variation among species and reef zones, rates of regeneration
were fastest and did not vary among co-occurring species in the Mixed Zone. This
pattern suggests that the three species might have inherently similar regenerative
capacities, but are differentially affected by changes in environment across their depth
range. While the relation between environment and regeneration is undoubtedly
complex, involving many variables, it may be influenced by frequencies of routine
injury across the reef, and in the associated, cumulative energetic costs of repeated
regeneration (sensit Lang da Silveira and van't Hof, 1977; Potts, 1977). For example,
rates of regeneration (Tables I and II, bottom row) were fastest in the Mixed Zone,
where previously surveyed natural injuries were relatively uncommon, and were slowest
in the Rear Zone, where most gorgonian colonies were injured relatively heavily
(Woodley el al., 1981).
Injury and regeneration as ecological processes
The ability to regenerate lost tissue and skeleton is common to most of the marine
invertebrate taxa which inhabit coral reefs (Mattson, 1976). Regeneration functions
both as an integral part of the life history (Moment, 1951; Tardent, 1965), and as a
response to injury (Wood-Jones, 1912; Gary, 1914, 1918; Kawaguti, 1937; Bayer,
1961; Mangum, 1964; Ebert, 1968; Kinzie, 1970, 1974; Trevaillion el al, 1970;
Connell, 1973; Fishelson, 1973; Birkeland and Gregory, 1975; Glynn, 1976; Loya,
1976; Bak el al., 1977; Jackson, 1977, 1979; Lang da Silveira and van't Hof, 1977;
Potts, 1977; Jackson and Palumbi, 1979; Hughes and Jackson, 1980; Karlson, 1981,
1983; Palumbi and Jackson, 1982; Hughes, 1983).
Despite its ubiquity however, the ecological role of regeneration remains relatively
obscure, in part because neither injury nor regeneration immediately affect colony
survivorship. Rather, their effects on colony fitness, and on the structure of sessile
assemblages, may be subtle (and intimately related to colony physiology), delayed,
and highly variable among different colonial taxa.
For example, regeneration of lost tissues often precludes the settlement of fouling
organisms onto areas of exposed internal skeleton within the injured colony. Such
fouling can have two important ecological consequences. First, settlement of com-
petitive superiors can lead to the eventual overgrowth of the entire colony (Kinzie,
1970; Jackson and Palumbi, 1979; Palumbi and Jackson, 1982; and references therein).
This potential relationship between injury, regeneration, survivorship, and abundance
may have influenced patterns of gorgonian abundance in the three Jamaican reef
zones examined here. Gorgonians were most common in the Mixed Zone (14.6
colonies/m2), where frequencies of natural injury were low (Woodley el al., 1981)
and rates of regeneration were fast (Tables I and II). Conversely, gorgonian abundances
were low (0.3 colonies/m2) in the Rear Zone, where frequencies of injury were high
(Woodley el al., 1981) and rates of regeneration were slow (Tables I and II). Clearly
however, these patterns are probably affected by many other variables as well.
The second consequence of fouling is the immediate addition of new organisms
to benthic assemblages. By preventing recruitment of other organisms onto surviving
colonies, regeneration may profoundly influence the structure and composition of
benthic communities (Bak el al, 1977; Jackson and Palumbi, 1979; Palumbi and
Jackson, 1982). The impact of such fouling, however, will vary with the size, growth
form, and competitive ability of the fouling taxa in relation to that of the injured
colony. For example, recruitment of encrusting organisms (e.g., bryozoans, forami-
niferans, and crustose algae) onto similar taxa living in the relatively two-dimensional,
cryptic community (Jackson, 1979) may have much greater effects on community
788 C. M. WAHLE
structure (Jackson and Palumbi, 1979; Palumbi and Jackson, 1982) than would
fouling by comparable organisms onto the larger colonial animals of the open reef.
Although all injuries in these experiments were fouled by various encrusting taxa
(filamentous algae, athecate hydroids; with varying effects on regeneration rate) all
gorgonian colonies were able to fully regenerate over these fouling organisms (see
also Bak et al, 1977).
Thus, among reef communities differing in scale (e.g., cryptic versus open reef;
Jackson, 1979), similar processes of injury, regeneration, and fouling may have very
different ecological consequences. In the cryptic community, the major ecological
effect of injury may be its influence on recruitment of comparable organisms into
the community (Jackson and Palumbi, 1979; Palumbi and Jackson, 1982). On the
open reef, where many fouling taxa are small relative to injured gorgonians, corals,
or sponges, the primary influence of injury and regeneration may be more on colony
physiology than on colony numbers. For example, in many gorgonian species, both
behavior and reproduction are integrated and synchronized among most polyps within
the colony (behavior: Wainwright and Dillon, 1969; Bayer, 1973; Preston and Preston,
1975; reproduction: Bayer, 1973, 1974; Goldberg and Hamiton, 1974). Injuries have
the potential to temporarily or permanently disrupt these and other aspects of a
colony-wide physiological integration by isolating distal regions of the colony from
the main body of polyps (Wahle, 1983). Thus, a major role of regeneration among
reef dwelling gorgonians, and among other open reef colonial taxa, may be to restore
colony-wide integration of critical biological and ecological functions disrupted by
injury.
ACKNOWLEDGMENTS
I am indebted to many people and institutions for help in this work: to D. Gerhart
for tireless field assistance; to C. Cook for field work and support; to the National
Science Foundation, Johns Hopkins University, and Sigma Xi for funding; to the
Department of Zoology, University of Texas for facilities; to the staff of D.B.M.L.;
to L. Buss, E. Chornesky, P. Gilman, T. Hughes, J. Jackson, J. Lang, S. Palumbi,
P. Ringold, C. Slocum, S. Stanley, G. Wellington, S. Woodin, and an anonymous
reviewer for helpful reviews; to E. Chornesky for invaluable criticism and support;
and particularly to J. Jackson for stimulating and critical discussions of injury and
regeneration. This is contribution number 195 of the Discovery Bay Marine Lab,
University of the West Indies. This paper is in partial fullfilment of the requirements
for the Ph.D. at J.H.U.
LITERATURE CITED
BAK, R., J. BROUNS, AND F. HAYS. 1977. Regeneration and aspects of spatial competition in the scleractinian
corals Agaricia agaricites and Montastraea annular is. Proc. 3rd Int. Coral Reef Svmp. Pp. 143-
148.
BAYER, F. 1961. The shallow water Octocorallia of the West Indian region. Martinus Nijhoff, Netherlands.
373 pp.
BAYER, F. 1973. Colonial organization in Octocorals. Pp. 69-94 in Animal Colonies, R. Boardman, A.
Cheetham, and W. Oliver, eds. Hutchinson and Ross, Inc.
BAYER, F., AND A. WEINHEIMER. 1974. Prostoglandins from Plexuara homomalla: ecology, utilization and
conservation of a major marine resource. A symposium. Studies in Tropical Oceanography, vol.
12. Univ. of Miami Press. 165 pp.
BIRK.ELAND, C. 1974. The effect of wave action on the population dynamics ofGorgonia ventalina Linneaus.
Pp. 115-126 in Studies in Tropical Oceanography, vol. 12. Univ. of Miami Press.
BIRKELAND, C., AND B. GREGORY. 1975. Foraging behavior and rates of feeding of the gastropod Cyphoma
gibbosum. Sci. Bull. Mus. Nat. Hist. Los Angeles County.
GORGONIAN REGENERATION 789
GARY, L. 1914. Observations on the growth rate and oecology of Gorgonians. Publ. of Carnegie Inst. of
Washington. 182: 79-90.
GARY, L. 1918. The gorgonaceae as a factor in the formation of coral reefs. Publ. Carnegie Inst. Washington
213: 343-362.
CHARNOV, E., AND W. SCHAFFER. 1973. Life history consequences of natural selection: Cole's result
revisited. Am. Nat. 107: 791-793.
CONNELL, J. 1973. Population ecology of reef building corals. Pp. 205-246 in The Biology and Geology
of Coral Reefs. Biology J, vol. 2, O. Jones and R. Endean, eds. Academic Press.
EBERT, T. 1968. Growth rates of the sea urchin Strongylocenlrotus purpuratus related to food availability
and spine abrasion. Ecology 49(6): 1075-1091.
FISHELSON, L. 1973. Ecological and biological phenomena influencing coral species composition on the
reef tables at Eilat (Gulf of Aquaba, Red Sea). Mar. Biol. 19: 183-196.
GLYNN, P. 1976. Some physical and biological determinants of coral community structure in the Eastern
Pacific. Ecol. Monogr. 46: 431-456.
GOLDBERG, W., AND R. HAMILTON. 1974. The sexual cycle of Plexaura homomalla. Pp. 58-61 in Studies
in Tropical Oceanography, vol. 12. Univ. of Miami Press.
HUGHES, R. 1983. Evolutionary ecology of colonial reef-organisms, with particular reference to corals.
Biol. J. Linnean Soc. 20: 39-58.
HUGHES, T., AND J. JACKSON. 1980. Do corals lie about their age? Science 209: 713-715.
JACKSON, J. 1977. Competition on marine hard substrata: the adaptive significance of solitary and colonial
strategies. Am. Nat. 111(980): 743-767.
JACKSON, J. 1979. Morphological strategies of sessile animals. Pp. 499-555 in Biology and Systernatics of
Colonial Organisms, G. Larwood and B. Rosen, eds. Academic Press, London.
JACKSON, J., AND S. PALUMBI. 1979. Regeneration and partial predation in cryptic coral reef environments:
preliminary experiments on sponges and ectoprocts. Pp. 303-308 in Biologie des Spongiaires. C.
Levi and N. Boury-Esnault, eds. Colloq. Int. C.N.R.S., Vol. 291.
KARLSON, R. 1981. Reproductive patterns in Zoanthus spp. from Discovery Bay, Jamaica. Proc. 4lh Int.
Coral ReefSymp. 2: 699-704.
KARLSON, R. 1983. Disturbance and monopolization of a spatial resource by Zoanthus sociatus (Coelenterata,
Anthozoa). Bull. Mar. Sci. 33(1): 118-131.
KAWAGUTI, S. 1937. On the physiology of reef corals III. Regeneration and phototropism in reef corals.
Palau Trap. Biol. Sta. Studies 2: 209-218.
KJNZIE, R. 1970. The Ecology of the Gorgonians (Cnidaria: Octocorallia) of Discovery Bay, Jamaica. Ph.D.
Thesis, Yale University, New Haven.
KJNZIE, R. 1973. The zonation of West Indian gorgonians. Bull. Mar. Sci. 23(1): 93-155.
KINZIE, R. 1974. Plexaura homomalla: the biology and ecology of a harvestable marine resource. Pp. 22-
38 in Studies in Tropical Oceanography, vol. 12 University of Miami Press.
KITTING, C. 1975. The impact of molluscs feeding on some West Indian gorgonians. Bull. Am. Malacol.
Union (abstract).
LANG DA SILVEIRA, AND F. VAN'T HOP. 1977. Regeneration in the gorgonian Plexaura flexuosa. (Cnidaria:
Octocorallia). Bijdragen tot de Dicrkunde. 47(1): 99-108.
LOYA, Y. 1976. Skeletal regeneration in a Red Sea scleractinian coral population. Nature 261: 5560.
MANGUM, C. 1964. Studies on speciation in maldanid polychaetes of the North Atlantic coast. II. Distribution
and competitive interactions of five sympatric species. Limnol. Oceanogr. 9: 12-26.
MATTSON, P. 1976. Regeneration. Bobbs Merrill Co., Inc. Indianappolis. 176 pp.
MOMENT, G. 1951. Simultaneous anteror and posterior regeneration and other growth phenomena in
Maldanid Polychaetes. J. Exp. Zool. 117: 1.
MURDOCK, G. 1978a. Digestion, assimilation, and transport of food in the gastrovascular cavity of a
gorgonian octocoral (Cnidaria: Anthoza). Bull. Afar. Sci. 28: 354-362.
MURDOCK, G. 1978b. Circulation and digestion of food in the gastrovascular system of gorgonian octocorals
(Cnidaria: Anthozoa). Bull. Mar. Sci. 28: 363-370.
OPRESKO, D. 1973. Abundance and distribution of shallow-water gorgonians in the area of Miami, Florida.
Bull. Mar. Sci. 23(3): 535-537.
PALUMBI, S., AND J. JACKSON. 1982. Ecology of cryptic coral reef communities. II. Recovery from small
disturbance events by encrusting bryozoa: the influence of "host" species and lesion size. J. Exp.
Alar. Biol. Ecol. 64: 103-1 15.
POTTS, D. 1977. Suppression of coral populations by filamentous algae within damselfish territories. J.
Exp. Mar. Biol. Ecol. 28: 207-216.
PRESTON, E., AND J. PRESTON. 1975. Ecological structure in a West Indian gorgonian fauna. Bull. Mar.
Sci. 25(2): 248-258.
SCHAFFER, W., AND M. GADGIL. 1975. Selection for optimal life histories in plants. Pp. 142-15 1 in Ecology
and Evolution of Communities. M. Cody and J. Diamond, eds. Belknap Press, MA.
790 C. M. WAHLE
SOKAL, R., AND F. ROHLF. 1969. Biometry. W. H. Freeman and Co., San Francisco. 776 pp.
STEARNS, S. 1977. The evolution of life history traits: a critique of the theory and a review of the data.
Annu. Rev. Ecol. Syst. 8: 145-171.
TARDENT, P. 1965. Regeneration in Animals and Related Problems, V. Kiortsis and H. Trampusch, eds.
North Holland Publ. Co., Amsterdam.
TREVAILLION, A., R. EDWARDS, AND J. STEELE. 1970. Dynamics of a benthic bivalve. Pp. 285-295 in
Marine Food Chains, J. Steele, ed. Oliver and Boyd, Edinburgh.
WAHLE, C. 1980. Detection, pursuit, and overgrowth of tropical gorgonians by milleporid hydrocorals:
Perseus and Medusa revisited. Science 209: 689-69 1 .
WAHLE, C. 1983. The roles of age, size and injury in sexual reproduction among Jamaican gorgonians.
Am. Zool. 23(4).
WAINWRIGHT, S., AND J. DILLON. 1969. On the orientation of sea fans (genus Gorgonia). Biol. Bull. 136:
130-139.
WILLIAMS, G. 1975. Sex and Evolution. Princeton Univ. Press, Princeton. 200 pp.
WOOD-JONES, F. 1912. Corals and Atolls. Reeve and Co., Ltd. London.
WOODLEY, J., E. CHORNESKY, P. CLIFFORD, J. JACKSON, L. KAUFMAN, N. KNOWLTON, J. LANG, M.
PEARSON, J. PORTER, M. ROONEY, K. RYLAARSDAM, V. TUNNICLIFFE, C. WAHLE, J. WULFF,
A. CURTIS, M. DALLMEYER, B. JUPP, M. KOEHL, J. NEIGEL, AND E. SIDES. 1981. Hurricane
Allen's impact on Jamaican coral reefs. Science 214: 749-755.
ZAR, J. 1974. Biostalistical Analysis. Prentice-Hall. New Jersey. 620 pp.
Reference: Biol. Bull. 165: 791-810. (December, 1983)
MARINE BIOLUMINESCENCE SPECTRA MEASURED WITH AN
OPTICAL MULTICHANNEL DETECTION SYSTEM
EDITH A. WIDDER, MICHAEL I. LATZ, AND JAMES F. CASE
Department of Biological Sciences and the Marine Sciences Institute, University of California,
Santa Barbara, California 93106
ABSTRACT
The emission spectra of 70 bioluminescent marine species were measured with
a computer controlled optical multichannel analyzer (OMA). A 350 nm spectral
window is simultaneously measured using a linear array of 700 silicon photodiodes,
coupled by fiber optics to a microchannel plate image intensifier on which a poly-
chromator generated spectrum is focused. Collection optics include a quartz fiber
optic bundle which allows spectra to be measured from single photophores. Since
corrections are not required for temporal variations in emissions, it was possible to
acquire spectra of transient luminescent events that would be difficult or impossible
to record with conventional techniques. Use of this system at sea on freshly trawled
material and in the laboratory has permitted acquisition of a large collection of
bioluminescence spectra of precision rarely obtained previously with such material.
Among unusual spectral features revealed were organisms capable of emitting more
than one color, including: Umbellula magniflora and Stachyptilum superbum (Pen-
natulacea), Para:oanthus lucificum (Zoantharia), and Cleidopus gloria-maris (Pisces).
Evidence is presented that the narrow bandwidth of the emission spectrum for Ar-
gyropelecus affinis (Pisces) is due to filters in the photophores.
INTRODUCTION
Measurements of bioluminescence spectra, especially from fragile marine organ-
isms, are complicated by the frequently dim and transient nature of their luminescence.
Recently developed intensified optical multichannel detectors, capable of simultaneous
measurement of an entire spectral region, are well suited to overcoming these diffi-
culties. Here we describe the use of one such optical multichannel analyzer (OMA)
system sufficiently robust for use at sea as well as in the laboratory, and present the
first collection of bioluminescence spectra acquired with it. This system is adaptable
to the extreme range of variables encountered in taking spectra of living systems,
namely potentially broad spectral range, wide range of possible luminous intensities,
erratic time-intensity characteristics of light emission, and the variable size and structure
of the light emitting tissues.
MATERIALS AND METHODS
Theory of operation
The detector (Table I and Fig. 1) is an intensified silicon photodiode (ISPD) linear
array placed in the image plane of a polychromator. Light focused on the entrance
Received 12 August 1983; accepted 21 September 1983.
Abbreviations: FWHM, full width at half maximum; ISPD, intensified silicon photodiode; NBS,
National Bureau of Standards; OMA, optical multichannel analyzer; S.D., standard deviation; S/N, signal
to noise ratio; UV, ultraviolet; Xmax, wavelength at peak emission.
791
792
E. A. WIDDER ET AL.
slit of the polychromator is dispersed by the diffraction grating across the 700 intensified
detector elements (25 ^m X 2.5 mm each) of the ISPD linear array. Photons striking
the reverse-biased diodes create electron-hole pairs in the semiconductor material
which discharge the equivalent capacitance of the diode. When the array is scanned,
the amount of recharging required by each diode (pixel) is a measure of the number
of electron-hole pairs formed. Since pixel position can be directly related to the
wavelength of incident photons, the charge per pixel represents a spectrum of the
light focused on the entrance slit.
Light may be directly integrated on the detector for periods ranging from 16 ms
to more than 20 s. Since the entire spectral window under examination is measured
simultaneously rather than sequentially, as in the conventional spectrophotometer,
there is no need to correct for time dependent changes in the emission, which are
common in bioluminescence. The limit of integration is determined by the dark
current which is kept low by thermoelectric cooling of the detector. Computer control
(Digital Equipment Corporation LSI-1 1) of integration time provides the sensitivity
and dynamic range necessary for dealing with the wide range of luminous intensities
encountered in bioluminescent organisms.
Collection optics
Because the size and structure of light emitting sources varied widely in the
organisms studied, considerable flexibility was required of the polychromator collection
optics. Initially, we used the convenient method of positioning the organism in front
TABLE I
Instrumentation
Detector
Detector controller
Computer console
Plotter
Polychromator
Collection optics
EG&G-PARC Model 1420 intensified silicon photodiode (ISPD) linear array,
composed of 1024 diodes (each 25 ^m x 2.5 mm). The microchannel plate
intensifier is fiber optically coupled to 700 of the detector elements.
EG&G-PARC Model 1218.
EG&G-PARC Model 1215 and Model 1217 outboard disc drive. Allows storage
and manipulation of up to 60 spectra per diskette.
Hewlett-Packard 7045B X-Y recorder.
ISA Model HR-320, 0.32 m Czerny-Turner f/4.8 with a 58 x 58 mm 152
grooves/mm grating blazed at 250 nm. Reciprocal linear dispersion, 0.49 nm/
diode.
Two all-quartz systems, aperture matched to the polychromator and used to focus
the image of the source onto the entrance slit. Large aperture system: a 50 mm
objective lens fixed at 100 mm from the 75 mm field lens. Fiber optic system:
a Welch Allyn circle to line converter, 50 cm ultraviolet light pipe, with a 2mm
bundle diameter terminating in a linear output 7 mm x 0.9 mm, focused with
a 10 mm cylindrical lens.
Calibration system Oriel Model 6047 low pressure mercury spectral lamp.
Optronics Model 245H 45 watt quartz tungsten-halogen standard of total and
spectral irradiance.
Optronics Model 65 precision current source.
Optronics Model UV-40 40 watt deuterium arc standard of spectral irradiance.
Optronics Model 45 deuterium lamp precision current source.
BIOLUMINESCENCE SPECTRA
793
of the entrance slit without input optics (Seliger el ai, 1964; Swift el al, 1977). Since
this procedure may degrade the resultant spectrum by scattering unfocused, stray
light in the polychromator, two interchangeable quartz lens systems were developed
to focus the luminescent source at the polychromator entrance slit. For large lumi-
nescent sources we used a 50 mm objective lens and a 75 mm field lens mounted
in a fixed tube, baffled and aperture-matched to the polychromator (Fig. 1A). The
input port for the tube was 1.76 cm in diameter and designed so that a luminescent
source positioned in the input plane produced a focused image at the entrance slit
of the polychromator. For smaller luminescent sources such as photophores (ca. 1
mm diameter) or other discrete luminescent regions of organisms, we used a 2 mm
diameter quartz fiber optic terminated in a 7 mm X 0.9 mm rectangular output
focused on the slit by a 10 mm focal length fused silica cylindrical condenser lens.
This system was also baffled and aperture matched to the polychromator (Fig. IB).
FOCUSING MIRROR
PHOTOCATHODE
MCP
I PHOSPHOR
SPD ARRAY
<"--- — — \
COLLIMATING MIRROR
DIAPHRAGM
cb
ENTRANCE
SLIT
APERTURE
CYLINDRICAL
LENS
FIBER
OPTIC
DETECTOR
CONTROLLER
DIFFRACTION
'GRATING
COMPUTER
CONSOLE
ENTRANCE SLIT
•FIELD LENS
•OBJECTIVE
LENS
DIFFUSER
OR ORGANISM
LIGHT PROOF BOX
X-Y
RECORDER
BAFFLE
•STANDARD
LAMP
•OPTICAL
BENCH
STANDARD
LAMP
POWER
SUPPLY
FIGURE 1. Schematic of apparatus for measuring emission spectra in calibration configuration. (A)
Bioluminescence or standard lamp output is focused by the input optics on the entrance slit of the poly-
chromator. The spectrum from the diffraction grating falls on the cathode of the detector, where it is
intensified and then detected by a linear array of silicon photodiodes. The detector is operated through the
detector controller under computer control. Spectral distributions are stored on floppy disk, analyzed, and
then plotted on an X-Y recorder. The system is calibrated by tungsten-halogen and deuterium standard
lamps and a mercury spectral lamp. The output from these calibration sources is collimated by a series of
baffles and diffused by a quartz ground disc before entering the input optics to the polychromator. (B)
Alternate collection optics consist of a 2 mm diameter quartz fiber optic light pipe terminating in a linear
output, focused on the entrance slit by a cylindrical lens. Not drawn to scale.
794 E. A. WIDDER ET AL.
Sensitivity
The combination of the microchannel plate intensifier with the long integration
times attainable with the silicon photodiode array allow attainment of high sensitivity.
The minimum detectable signal (signal to noise ratio = 2) for an ISPD at the normal
operating temperature of 0°C is 4 photons/s/diode with an integration time of 10 s
(Talmi, 1982). Insertion of the f/4.8 polychromator and collection optics between
the ISPD and the source resulted in a minimum detectable signal at the input of the
fiber optic of 7,150 photons/s at 475 nm with a 1 mm slit and 20 s integration time.
With the double lens system, which had an input area 8 1 times that of the fiber optic,
45,000 photons/s at 475 nm were required.
This level of sensitivity permitted measurement of the spectrum of one flash of
a single dinoflagellate, Pyrocystis fusiformis, and, using integration times of 10 to
20 s, it was possible to measure the spectrum of any source visible to the dark-adapted
human eye.
Calibration
Since the input optics and spectral window were changed frequently while working
at sea, a method of field calibration was necessary. A low-pressure mercury spectral
lamp (Table I) was used for wavelength calibration. A spectral irradiance standard
and precision current source (Table I) were used to correct for nonuniformities in
channel-to-channel sensitivity and for detector and polychromator efficiencies. The
calibration function was generated as the ratio of the measured spectrum to the true
spectrum (Fig. 2A), which was determined from the NBS referenced calibration data
supplied with the lamp. To be accurate, the calibration function must be generated
under the same conditions as the spectrum to which it will be applied. To do this
the unfocused standard lamp beam was collimated, and then diffused by a quartz
ground glass of known transmission. This diffuser served as the radiant source to the
polychromator and was positioned at the input of the optical system at what would
be the plane of focus of the bioluminescent organism or tissue.
Stray light, often a problem in single stage polychromators, is a fraction of a
percent of the total irradiance (Talmi, 1982) and is insignificant at the low intensities
characteristic of bioluminescence. However, the intense red emission of the tungsten-
halogen standard lamp produced stray light that was a significant percentage of the
lamp's much weaker emissions below 400 nm. To insure accuracy of the correction
curve in the near UV, we used an NBS referenced deuterium arc lamp to generate
the correction function below 400 nm.
Data analysis
Spectra were stored on floppy disks, with automatic subtraction of the dark charge,
background spectrum. Postexperiment manipulation of data primarily involved di-
vision of the emission spectrum by the correction curve stored in memory and digital
smoothing of the data using a Savitzky-Golay least-square polynomial algorithm
(Savitzky and Golay, 1964; Edwards and Willson, 1974) (Fig. 2B). The running least
squares fit was to a second degree polynomial over a 25 channel smoothing range.
This smoothing range was well below the recommended value of 70% of the narrowest
spectral feature observed (Edwards and Willson, 1974) and, therefore, facilitated iden-
tification of such spectral features as Xmax and FWHM without decreasing resolution.
The smoothing function was applied from 1 to 10 times depending on the signal to
noise ratio (S/N) of the spectrum. The signal to noise ratio of each spectrum was
BIOLUMINESCENCE SPECTRA
795
400
450
500
550
600
650
700
400
450 500 550 600
WAVELENGTH (nm)
650
700
FIGURE 2. (A) Real versus ideal curves for spectral output of the tungsten-halogen standard lamp.
Derived from 100 scans, each 16.6 ms in duration, using the double lens collection optics with 0.025 mm
entrance slit. Relative intensity is shown as a function of wavelength, (a) Ideal curve calculated from a
third degree polynomial curve fit to the data supplied with the lamp, (b) Standard lamp spectral output as
measured with OMA. Fluctuations in the measured spectrum are the result of non-uniformities in the
sensitivity of the system. A correction curve is generated by dividing curve (b) by curve (a). (B) Effects of
data correction and analysis on emission from a colony of the tunicate, Pyrosoma atlanlicum, using double
lens optics and 1 mm slit, (a) Uncorrected spectrum. Apparent bimodality is explained by fluctuation in
the real standard lamp curve ( A-b), with the result that the corrected spectrum (b) is unimodal. The corrected
spectrum which has been smoothed five times (c) has a Xmax = 491 nm, FWHM = 96, and S/N = 99.
Each curve is shown on a similar relative scale but is displaced vertically for clarity.
computed as the ratio of the signal at Xmax to the root mean square noise over the
whole spectral range (calculated by first subtracting the smoothed from the unsmoothed
corrected spectrum). Other computer operations included calibration of channel
number to wavelength value and generation of the ideal standard lamp curve using
a third degree polynomial curve fit to the data supplied with the lamps.
796
E. A. WIDDER ET AL.
Accuracy and resolution
Since bioluminescent emissions occur over the entire visible spectrum, it is desirable
to examine as broad a spectral range as possible when examining new organisms.
Consequently we have used a 1 52 grooves/mm plane grating which provides a spectral
coverage of about 350 nm and a spectral bandwidth of 0.5 nm/diode. Geometric
registration on the ISPD is excellent, ±1 diode (Talmi and Simpson, 1980); therefore,
wavelength accuracy was ±0.5 nm with the 152 grooves/mm grating. Resolution was
a function of the grating and the slit width and was empirically determined as the
product of the reciprocal linear dispersion and the FWHM of one of the mercury
lines (Felkel and Pardue, 1979). Experimentally determined resolution was: 2 nm
using the 0.025 mm slit, 3 nm with the 0.1 mm slit, and with the 1 mm slit it was
9 nm using the fiber optic input and 20 nm using the lens system or no collection
optics. Resolution and accuracy were also a function of the relative brightness of the
source. In order to examine this effect, a C14 activated phosphor disc (Xmax 524 nm)
was attached to the input of the double lens system and five readings were taken with
the 1 mm slit at each of several different integration times. At integration times
producing signal to noise ratios of 150 and above, the average standard deviation
was less than 1.5 nm for Xmax and less than 0.8 nm for FWHM measurements. With
signal to noise ratios between 30 and 150, the average S.D. was 6.5 nm for Xmax and
0.9 nm for FWHM. At S/N ratios below 30, the S.D. of Xmax and FWHM measurements
were 19 and 10 nm respectively. Examples of spectra with S/N ratios within these
three different ranges are shown in Figure 3A. For signal to noise ratios above 150
the smoothing function was applied only once, for ratios between 30 and 150 it was
applied a maximum of 5 times, and below 30 it was applied a maximum of 10 times
(Fig. 3B).
B
350 400 450 500 550 600 650
WAVELENGTH (nm)
700 350 400 450 500 550 600
WAVELENGTH (nm)
J
650 700
FIGURE 3. Determination of accuracy as a function of source intensity using a constant output C1'
activated phosphor. Using the double lens optics with 1 mm slit, different integration times yielded spectra
with different signal to noise ratios. (A) Corrected spectra with (a) S/N = 229, (b) S/N = 71, and (c) S/N
= 11. With the lowest signal to noise ratio (c), noise spikes are pronounced. (B) Smoothed spectra from
(A). The curve with S/N = 229 (a) was smoothed once, Xmax = 524 nm, FWHM = 74 nm. (b) Curve with
S/N = 71 underwent five smoothings, Xmax = 524 nm, FWHM = 73 nm. (c) Curve with lowest signal to
noise ratio (S/N = 1 1) was smoothed 10 times, Xmax = 530 nm, FWHM = 75 nm. Note noise bumps. All
six curves are plotted on a relative linear intensity scale.
BIOLUMINESCENCE SPECTRA 797
Collection and handling of organisms
Deep-living organisms were collected from the Catalina, East Cortez, San Clemente,
Santa Barbara, and Velero Basins off the coast of Southern California during 1982
and 1983. Benthic samples were taken with a 5 ft beam trawl and a near-bottom
beam trawl with a 2 X 10 ft mouth opening; midwater collections were made with
an opening-closing midwater Tucker trawl ( 10 X 10 ft opening) fitted with an opaque,
thermally-insulated cod-end. The pelagic holothurian Scotoanassa was collected by
the deep submersible "Alvin." Sorted animals were placed in chilled sea water (4-
8°C) and maintained in light-proof coolers until use. All measurements were made
within 4 h after collection. Coastal and subtidal animals were collected locally in the
Santa Barbara Basin, near Santa Barbara, or near Scripps Institution of Oceanography,
La Jolla, California, by trawling or SCUBA diving and maintained in the laboratory
in aquaria with flow-through, sand-filtered sea water (18°C). A few specimens were
obtained from laboratory cultures or a commercial aquarium. Specimens were placed
for testing in quartz glassware containing sea water or held in air and positioned in
front of the detector collection optics. In some cases, bioluminescence was stimulated
or enhanced by 1 X 10~3Mnorepinephrine, 1 X 10~4 g/ml serotonin, or 2% hydrogen
peroxide. Otherwise, unstimulated or mechanically stimulated emissions were mea-
sured. Specimens were preserved in 5 or 10% buffered formalin for subsequent iden-
tification. Reference specimens have been deposited in the Invertebrate Zoology Col-
lection, Santa Barbara Museum of Natural History.
RESULTS
Spectral data
Table II lists 70 bioluminescent species from which we have obtained spectra
using the OMA. They are arranged taxonomically and the spectral features listed are
the wavelength at peak emission (Xmax) and the width at half the maximum value
(FWHM). The signal to noise ratio (S/N) is included as a measure of relative accuracy.
Whenever a given organism was available on more than one occasion, spectra were
taken. Multiple spectra were averaged and the mean with standard deviation was
determined. Spectra were grouped for averaging in three S/N ranges: above 150,
between 1 50 and 30, and below 30. Readings below 30 were not listed if better
measurements were available. Measurements made at different slit widths were kept
separate, and in cases where different collection optics had a significant effect on the
values, they were listed separately. This was most apparent when the fiber bundle
collection optics were used with the 1 mm slit. The greater resolution of the fiber
bundle at this slit width was due to its 0.9 mm linear output width and was most
apparent for spectra with narrow bandwidths.
Over the year during which these spectra were collected an effort was made to
insure accurate calibration and correction factors. One obvious independent check
of the entire system is the reproducibility of the spectral data and comparison of our
measurements with accurately known bioluminescence spectra from other laboratories.
Renilla has an extremely stable emission spectrum (Wampler et al, 1973) and as a
result it has been suggested that it might serve as an "emission standard ... for
routine calibration checks" (Wampler, 1978). Following this advice, we measured
specimens of the local species, Renilla kollikeri, many times throughout the year with
different collection optics, slit widths, and polychromator settings (Table II). The
values measured remained in good agreement and also compared well with mea-
surements made in other laboratories (Reynolds, 1978; Wampler et al., 1973).
798
E. A. WIDDER ET AL.
u
06
U U
0 0
^^ ^N
(N f>4
£
£ 0
JD -0.
c/3 E
o o ooo o —
O O O O O O O
0 — 0 -
— o —do — d
o — o — —
— d — d
00
r~- fN
Z
+1
+1 +1
c/f
ro TJ- »/"»—• r*"i ^f fN
t^ fN OO ^ ^J" O^ f~t
r~~~ r**i r^~ ^^ ^^ ^^ *"^
Tf fN •/"> ^O
m m r-i <^i —
fl fN —
UJ
m
*j
O
rr, O
H
X ?
+1
+1 +1
•^ c
iTs. ^__^ ^-^ ^- ^, ^. ^^
Tj" — * — ™ VO *^^ '^ ^"
m r-i oo TJ-
^^\ oo f^^ ^^ ^^ r^ r^~
CO r*% co ro
?5
Sq
fN
?7 ?T
*«°°
^^
0 0
"^'X
+l
+1 +1
— ! — r*^ — i • — ' <N r*^
\c o •/-> oo rn MO — '
— fN fN fN
oo oo oo oo oo ON o*s
O\ TJ- ov OO OO OO OO
TJ- Tt ^f 'S-
s
to
3
3
--« ^J
-s:
u
'o
£
C8 g -«
1 1 1
ell 'g
3 C J- .$-
•?. aj sj ^
•^ ^ « «
1. 1
1 - Hi Us
l ! Ill 111
Phylum Dinophyta
ocystis fusiform is
ocystis lunula
111
^ ^ "o^^"o'^
^ ^ % £
>~- b
C C
BIOLUMINESCENCE SPECTRA
799
<n a
•- P.
o
2 a
oo^2
e e u£
O O T3 O
o o • o
.o x> e 2 a a
o u x>
.D « ^
— O—
o — d
o «' n* « w ig <u ii j5 «' f~ I ^ n ,-j n i a_ T -i-
O O O Oppppp — O O O O O pppO
' ' ' ' ' '
OO
+1
r-
+1
OO
oo
— l^w^ OOt^ — 1^— SO OO
_,r4r<^ <-«-!r^rn<Nrs|«<"> —
+1 +1 +1 +1 +1 +1 +1 +1 +1 +1
vOOr^<^ — 1" r~ <N vO — — OO so — u-i
(~~O — <N TJ- oo r-- so ONP^I^I^ >^> oo —
+1
mo
SO
Tl-T}- O<N— — — — O
+1 +1 +1 +1 +! +1 +1 +1 +1
SO ON (NV-irnw^OO<N— — in SO w^
l~* r— fN(NfNC^ror»^(N(N m fN <N rs
+1 +1
+1
O SO —
+1
O —
-^-r-4
+1 +1 +1 +1 +1
r«-> m fN OvONONONCTvONOOOO
r- so so oooooooo
+1 +1 +1 +1
OO — O O
o — — —
ft w~i «^ vi
(x| O
+1 +1
+1 +1
— ' O OS O ON OO
OOONI~-OOON
'C
-§
-S
|
I i
-
l
-2
u
i
I
i
'
=?-
-3 K
11!
~-C)^
t
I
I
I
I
I
I
1
800
E. A. WIDDER ET AL.
C3
•o
•c
o
E
a
g"
&
•43
i
3
-^ °°
— y/
>-• 3
fc"«
C ^?
O ^
c c c c c
-:
0 C/}
oooo o
O C
E 1
,i— i
C/1 ti-i
.— .— .— ._ .— OOOO O
U.UHU-UM — OOOJQJ OJ
"O*O"O"O "O WDC/5l/lt/5 t/5
0 U
X; u o u -C
S ?
« xinno £^i^S^!i^^
o oooo oooooo
op— . p p op— _p p
O OO fN
CM
OO OO Tf — 1^1
z
+1
+1 +1 +1 +1 +1
i/"") « ^^ \^ OO ^^ ^^ ^^ ^J1 ~~"
>•— V.
TJ- */^ T^- tn VO ^J" ^ ^^ OO OO f^i
•^^ ^^ r~^~ r~*~ "^ oo v~i r**i r*~i ^o
s
••d
c
o
U
1^
u-i — — <^l JN
~ S ^
+1
+ 1 +1 +1 +1 +1
" I £
^ £5
r^ ^^ OO ^^ f^i f^t f^> ^*"i vO ^_^ I
m fNOOvOv oooooooooor-
oooooooo oo r-r-~r^i^ oo
H
? ~
sci
s
<•••> rs <N oo
* c/5
ra
oo ^ O <^> f^
i+i
+1
+1 +1 +1 +1 +1
•^'X
^^ ^^ OJ ^^ ^^ ^^ ^^ f^l ^-1 ^^ ^^
O oor-o oooooooooooo
i/~l </^ </^ O^ VO O^ ^^^ '"'^ OO ^J"
VO *^^ ^O ^O VO OO OO OO OO OO
*" S
a i § -S
y
'o
£
1 ^ -i %
•K '$ - 1 -| ^
^ 0 g- | a
tS ^ 3 E -a ^ |
« '5 s •*= - S. ^ ^ •
2'-! 1 1 •= -~ ^ "6 &
O '•— ^J
t, vc a<3
_c C1 sr
•So ^ -5
5 ^3 Q.
•^ S! ^
^* -^ ~ ^5
E -S? '3> -S a
^5 5 fe. ^
^ ^ 3 -3 §•
<3 ^;' u G- $~ '-Su :^ :^u
§22 -LJ ^S 25
Q S^_ s^ ^ ^ ^
S, ft, ^ < QQ 05 cq
OH S 3 % -^
s? s? s s
Q ^C! Q
BIOLUM1NESCENCE SPECTRA
801
1
I
03
t«
0
0
J3
photophoi
photophor
<S1
<U
J3
0
07;
u
£H
"Vw ^ ^J
•5 o o
^ J3 J2
~ Q. a
•S o o
<U O O
t- _c _c
lotophore '
c
Q.
0
a
o
Sg
i-
O
Q.
O
0
x:
o a a
a.
C3
03
op
O
J=
0
j=
"5
o
"5
o
o
J=
0
J3
a
o
o 4^ C^
Q (/) S/3
a
UH
O
°
a
a
o
£
0
£.
Q.
a
0
j=
^ "c "c
c
C
o
—
_o
_ap
•o
o
"5
3
3
V)
^2
3
"3
o.
a<J= «c
'^
It
L_
^
.e
n.
tS
•o
o
Q
{^i
•5
1
1
o
5
03
|.§|
.«
u
o
O
a
E
3
E
]?
1/J
3
<C
<C OJ <U
x
<u
o o
o ra ™ tT] « « «' M a rc ra
O OOO O O OOOOO
+1
ON
+1
no
+1
r--
— oo
+1 +1
+1
+1
— \£i ^^
O-) '"^i r""""
+1
ON
+1
O
r-
— o
oo r^
SO
+1
oo
r-
ON
+ 1
oo
t—
ON
o — —
+1 +1 +1
&\ r^~ r^ r~~-
m
+1
m
+1
ON
rn
(N
+1
O
r-
+1
n
r-
ri
r—
+1
O
+1
— — O
V} »/-> »/^
O
+1
O — —
u~> 04 r~
r^*i r*"i r^")
+i +i +i
— O <^ —
r-i oo oo oo
VD
VO —
— o
+1 +1
^O ^>O
~
•2
s:
>j
~cL
Ji
o
o
^
^
.5
, rattrayi
1
.2
f
s:
:S'
0.
2
•§
|
5*3
~
o
^
E
_3
O
i
2
1
a
1
•^
•Sz
.a
|
^g
o
o
5
-«
|
£
1
S
R,
Cu
.§•
I
3
^
1
2
~~
?
^
£;
g
&«
Si
"2
s:
_**a
^
S
T-*
2
£•,
^
I-P
t
fc
-o
T<
c"-
th
>>,
5J
03
l_
I
.Ell
S "I I
E ^ %
^ ^ ^
?? "&• Q
sF ° ^
802
E. A. WIDDER ET AL.
(/i
E
'£
<u
o
06
"o
o
1
(U
E
00
'5.
^
u n .O
u o
175 E
p p
p p p
O —
— o
sO
(N ro
Z
+1
+1 +1
+1
C/2
r-- so
1/1 I" I/")
^\ r*^
**
•^J- ro r~
oo m
s
.c
s
sj
UJ
_)
03
K ?
> £
+1
O OO
r- ON
+1 +1
iy~j rO OO
+1
ON ON
H
+1
?S
sci
— ??
fN r7
g*
» !/5
>»^ 's-^'
•• — •
j+i
— O
+1 +1
+1 +1
+1
?s
i/-> m >/•>
rn OO (N
— l~r)
SO OO ^"
^" ^" "~1
OO OO ON
ON OO
«
IT*
a
to
u
'o
u
ex
Pannychia moselyii
Ophiophol is cf. longispint
Phylum Chordata
Subphylum Tunicati
Oikopleura dioica
Pyrosoma atlanticum
Pyrosomella cf. verticillai
TD
1
O
X
rn
<u
<L>
<L>
u-
I-
CO
^a
O
x:
0
x:
4-*
a
o.
to . ,
(U to
o
o
I- u
0 fe
x: o
af:
0
•§.
o
Q.
0 g-
•5; °
o *••
3
"ol
•*-*
X! 0
°"§.
!£
o
i
IS
o
to —
to
to
0 C
D. ol
0
a
0
o.
C
!
"c
•a'
^
0
"ol
*o
TD
c
•o
01
0
'E
to
<U
%
2
C
O
J=
"ol
B
o
CO
^•4
u-
Q.
C
J=
^
o
•a
£
O
<U
D.
0
0
O
jn
cn
LO
c*
to
x:
a
|
<u
D.
D.
'|
D.
0
o
11
It
a
C
IU
c
T3
-a
3
TD
a
bbb b
b
o o o o o o o
+1
OO —
— ON
u-> ON
+1
TT «/i
O r-
rvl
oo O </">
00
+1
^- ON
+1
OO ON
r-- — ON
r~- m ON
r-~ oo so
+1
(N SO TT
OO OO OO
OO
r-
+1
ON so — m
r- r- oo oo
oo o r--
r~ r- ON
ol
2
o
s: ^
^ E!
^ tj*-
1 ^
I
oo
§
I
S
a
s:
|
a
i
.^ ?J
§
-5 a
a
-s:
a.
CO
5
o
BIOLUMINESCENCE SPECTRA
803
<u
QJ U
~ "5 'c .-r
s
11
3 3
"O "^
llMI
Z3
C
i/3
m ^. -^ o
. r • r
c ^ c c . ,
OJ
r photophore
(tophores
s a
go «
llll
1
o
"ol
•a
c c
a 03
OO
0 0 c
— — 03
03 03 OO
tn
C
03
00
"in u u O -S
^.§.2. o o
0 D. D. -15 o
-§.22^-1
0 0 0 03 "•
1 •§. •& s 3
~^" t , ,^i
ex rr rr o X)
photophores3
c c
0 0
£ cT.. £ £ ^-
O-Q QJ 0.0 &Q ^*~
•c — -c 'c £
c § c c c
.2 -a 2.2 .2
OO
Q
00
c
^o
i^Q
al photophor
o a
'C •*•
1, W —
« S — g.
3
r\
3 3 O
03 03 —
O O 03
"o3
CU CU 2 T
15 i: i: 2 o
03
OJ <L>
U.
15
S ^
as
1 a 3 o
.E 5 S a
1
03 03 TD
<i i 3
.E .£ S
3
03
UN C C Oi «U
t; <u «j 3 3
C > > in to
> — — — fN
l_
C
u
03
1
03
il
E'~ O C 'c 'c
D. 03 C C
D.
Q
Oi
o
JD
b b
b b b b
b
bob
b
O O O O O
o
O
0
0 0
b b b b b
b
b
m m
oo
—
in fN
m
m f>
ON
<n
+1
+1 +1
+1
+1 +1
+1
+1
+ 1
oo oo
O oo >n NO
ON >/"> O
oo
rf Tfr —
O ON f-
NO
oo ON m fN m
>/•)
OO
o
o
ON
NO O
— O ON — <**")
_ ^
ir!
fN
r^j r*"i — — ' fN fvj
fN
— fN
o
0
+1
+1 +1
+1
+1 +1
+1
+1
+1
fN r~
fN — O —
fN
fN fN TJ-
fN
f"l */^ f*^ r~^ ON
NO
Tj-
O
fN «n
OO ^O f^ f^ f^
_
-J.
oo r-
sO ^O NO ON
NO
^O ^O *^D
NO
f^ I — (^ NO NO
r—
"""
OO
ON OO
oo oo *N rj r\
O
ON
<u
2 ^
2 c
fc 'C
^_ ^
fiT r^
Zf
Ci
_ a
t« gj
rr-i
CN
00 .E
+1
+1
o a
^ w
TD u.
r-
fM
>, O
fN
vO rn
s
r*-> fN~
5
o
m
^
x; c
fN
fN r*}
o
ro —
„_
CN — r~- ^T —
0
xT x?
+1
+1 +1
+1
+1 +1
+i
,.000 —
+1 >n m in in
+1
a> c (u
fN —
ON ON ON —
ON
Tf \.o ro
oo
ON "^ !/1 NO — '
— 1
0
r-
NO NO
^) r*"i i/~) ro c^l
NO
r*^
O '3 O
NO NO NO NO
NO
r^- r-~ r-
OO OO OO OO ON
Ov
oo
r-
o o
— ^^ V") C^ CN|
—
ON
5 o 5
^T ^f
T^~ Tj- ^ ^f
rj- rj- ^-
Tfr
Tj T^- ^ Tj- Tj"
^-
^.
Tj-
>n >n
t/^ IO */~! */^ tO
l/~l
Tj"
03 ^ 03
•C O J-
C i- C
1J 4J 1>
I— I—
C/3 • O ^ O
0 « ^ ^ ^ V5
"•^ '£ P -a "° "C
o §• ^ g J3 "o
lidlll!
S
fll O •— t/5 4J S/5 ' "™
Idiacanthus antrostomu
Triphoturus mexicanus
Lampanyctus ritteri
Stenobrachius leucopsat
Lampadena urophaos
§
1
Porichthys myriaster
Oneirodes acanthi as
Oneirodes sp.
Cleidopus gloria-maris
Anomalops katoplron
3 Fiber bundle colli
h Double lens colle
c No collection opt
' Bioluminescence
2 Bioluminescence
3 Bioluminescence
* Secondary peaks
804 E. A. WIDDER ET AL.
Spectral range
In general, spectra were confined to the range of 400 nm to 700 nm. Occasionally
spectra extended into the near UV. The greatest emission noted below 400 nm was
by the squid Abraliopsis falco (Fig. 4E). Minor emission below 400 nm was also seen
in the crustaceans Scina and Sergestes (Figs. 4A and 4C). The range of spectral
maxima measured extended from 439 nm for Scina to 574 nm for Parazoanthus
(Table II).
Spectral shapes
Examples of the observed variety in spectral shapes are shown in Figure 4. Most
spectra were structureless and unimodal (Fig. 4A), with bandwidths ranging between
26 nm (Argyropelecus affinis, Fig. 4A) and 100 nm (Pyrosoma atlanticum. Fig. 2B)
with an average of 75 to 80 nm. The emission spectrum for Argyropelecus was notable
for its unusually narrow bandwidth. Among the other organisms measured, the nar-
rowest bandwidths among the structureless unimodal emissions were those from the
caudal organs of the myctophids (Triphoturus mexicanus, Lampanyctus ritteri, Sten-
obrachius leucopsarus, and Lampadena urophaos) (Table II). Emission spectra for
these fish exhibited bandwidths of about 62 nm.
It is known that the photophores of Argyropelecus consist of deeply placed tissue
that transmits light to the ventral surface through light-pipe-like structures which
contain pigmented filters (Denton et al, 1970). To determine if the filters in the
ventral photophores are responsible for the unusual narrowness of the emission spec-
trum, a 1 mm strip was cut from the anterior photophores of one specimen. The
emission spectrum from this region was measured with the fiber bundle collection
optics and compared with a spectrum taken from the posterior uncut region of the
same specimen. Removal of the ventral anterior strip increased the bandwidth more
than 20 nm over the spectrum measured from the posterior intact photophores
(Fig. 5).
Essentially unimodal spectra were observed with some structural complexity,
commonly seen as a long wavelength shoulder as in all the pennatulids and dinofla-
gellates measured (Fig. 4B) and less commonly with a short wavelength shoulder as
seen in Sergestes, Parazoanthus and the Y-l strain of Vibrio fischeri (Fig. 4C).
Porichthys exhibited the only emission spectrum with bimodal peaks of approx-
imately equal intensity (Fig. 4D). All squids measured had bimodal emission spectra
with short wavelength secondary peaks and long wavelength shoulders (Fig. 4E). The
only trimodal spectrum measured was that of the brittle star Ophiopholis which had
both short and long wavelength secondary peaks (Fig. 4F).
Spectral variation within species
In several instances different emission spectra were observed from different colonies
of the same species. The most notable example of this was the pennatulid, Stachyptilum
superbum (Table II). One 60 min beam trawl at 600 m in the Santa Barbara Basin
yielded hundreds of these sea pens. Visual inspection revealed that approximately 1
out of every 100 colonies had a much yellower emission than the majority. Mea-
surements of the emission spectra demonstrated a 30 nm difference in emission
maxima and a slightly broader bandwidth for the yellow emitters. No morphological
differences were found to distinguish the two variants.
Another example of this phenomenon was seen in a colony of Parazoanthus
luciftcum. The emission spectrum for the colony as initially measured with the double
BIOLUMINESCENCE SPECTRA
805
1 -
z
LLJ
z -
>
H
<
C/5
Z
z
UJ
>
H
<
700
700
400
450 500 550 600
WAVELENGTH (nm)
650
700 350
400
450 500 550 600
WAVELENGTH (nm)
650
700
FIGURE 4. Some representative emission spectra of marine organisms. Relative intensity is shown
with respect to wavelength. (A) Structureless unimodal distributions, (a) Scina cf. rattravi, Xmax = 439 nm,
FWHM = 70 nm, S/N = 48; (b) Vargula hilgendorfii, Xmax = 465 nm, FWHM = 83 nm, S/N = 101; (c)
Argyropelecus affinis, Xmax = 487 nm, FWHM = 26 nm, S/N = 125; (d) Cleiodopus gloria-maris adult,
Xmax = 506 nm, FWHM = 92 nm, S/N = 264. The polychromator spectral window was set for 400-750
nm for (d) and 350-700 nm for (a-c). (B) Unimodal distributions with one or more long wavelength
shoulders, (a) Pvrocvstis noctiluca, Xmax = 472 nm, FWHM = 35 nm, S/N = 180; (b) Pennatula phosphorea,
Xmax = 500 nm, FWHM = 53 nm, S/N = 45; (c) Renilla kollikeri, Xmax = 509 nm, FWHM = 22 nm,
S/N == 199; (d) Stachyptilum superbum. Xmax = 533 nm, FWHM = 58 nm, S/N == 78. (C) Unimodal
distributions with short wavelength shoulder, (a) Sergestes similis, Xmax = 469 nm, FWHM = 62 nm,
S/N = 18; (b) Vibrio fischeri Y-l strain at 20°C, Xmax = 540 nm, FWHM = 81 nm, S/N = 333; (c)
Parazoanthus lucificum, Xmax = 574 nm, FWHM = 94 nm, S/N = 56. The bumps in (a) are due to random
fluctuations (noise) which are present in all spectra but are only apparent at low signal to noise ratios. (D)
Bimodal emission spectra of Porichthys notatus. (a) Spectrum from ventral photophores viewed by double
lens optics, Xmax = 488, 504 nm, FWHM == 76 nm, S/N = 130; (b) spectrum from single photophore
measured with fiber optic, Xmax = 484, 501 nm, FWHM = 75 nm, S/N = 64. The curves are standardized
to the same scale but are vertically displaced. (E) Emission spectra of squids displaying a short wavelength
secondary peak and long wavelength shoulders, (a) Abraliopsis falco, Xmax = 421, 466 nm, FWHM = 105
nm, S/N = 12; (b) Cranchia scabra. Xmax = 483, 511 nm, FWHM = 77 nm, S/N = 63. (F) Trimodal
emission spectrum of Ophiopholis cf. longispina, Xmax = 483, 512, 545 nm, FWHM = 102 nm, S/N = 42.
806
E. A. WIDDER ET AL.
1 r
z
LU
LJJ
>
LU
DC
400
450
500
550
600
WAVELENGTH (nm)
FIGURE 5. Effect of filters in the photophores of a single specimen ofArgyropelecus affinis on emission
spectra. Measured with fiber collection optics; relative intensity displayed as function of wavelength, (a)
Emission of posterior photophores, Xmax = 486 nm, FWHM = 29 nm, S/N = 15; (b) spectral emission of
anterior photophores with ventral tip excised, Xmax = 484 nm, FWHM = 53 nm, S/N = 9.
lens collection optics appeared to be bimodal (Fig. 6A). Examination with the fiber
bundle collection optics showed that some polyps produced different unimodal emis-
sion spectra (Fig. 6B) with a difference in emission maxima of 70 nm. Due to its
dimness, this spectral shift was much more difficult to distinguish visually than the
30 nm difference displayed by different Stachyptilum colonies. A different colony of
the same Parazoanthus species had only one emission spectrum which matched the
shorter wavelength spectrum of the two color colony.
In the pinecone fish Cleidopus gloria-maris there is a visible color difference
between the bacterial light organ of juveniles as compared to adults. Comparison of
the emission spectra confirmed a 15 nm short wavelength shift in the adults. The
fiber bundle collection optics demonstrated the presence of a pronounced emission
gradient across the juvenile light organ that was essentially absent in the adult (Table
II). Light from the anterior region of the juvenile light organ had a Xmax of 555 nm
compared to a Xmax of 523 nm measured from the mid region and 516 nm in the
posterior region. In the adult, measurements from equivalent regions across a single
organ were 506 nm, 506 nm, and 503 nm from anterior to posterior. These emission
spectra of both juvenile and adult are very different from the spectrum measured
from Vibrio fischeri (Xmax 492 nm, Table II), the bacterium isolated from light organs
of this species (Fitzgerald, 1977).
A similar color gradient was seen in the sea pen Umbellula (Table II). Luminescence
at the base of the stalk was bright green (Xmax = 500 nm) with the narrow bandwidth
and long wavelength shoulder typical of in vivo pennatulid emissions (Morin and
Hastings, 197 Ib; Wampler el al, 1973). Proceeding up the stalk the emission became
broader, bluer, and dimmer until at the top of the stalk it had a broad, structureless
spectrum with a Xmax of 470 nm.
The Y-l strain of Vibrio fischeri is also capable of more than one emission color
luby and Nealson, 1977). At 20°C or below, emission is unimodal with a short
wavelength shoulder and a Xmax of 540 nm. However, upon heating the emission
BIOLUMINESCENCE SPECTRA
807
CO
z
LU
H
LU
>
H
LU
CC
J
450 500 550 600 650 700 750
CO
Z
LU
h-
LU
>
LU
DC
450 500 550 600 650 700 750
WAVELENGTH (nm)
FIGURE 6. Emission spectra of a colony of Parazoanthus luciftciim. Relative intensity is shown with
respect to wavelength. (A) Apparent bimodal spectral distribution from entire colony, measured with double
lens collection optics, Xmax = 509, 537 nm, FWHM = 140 nm, S/N = 96. (B) Emissions from individuals
of colony measured with fiber collection optics. In this colony, two different unimodal distributions were
produced, (a) Xmax = 502 nm, FWHM = 108 nm, S/N == 145; (b) Xmax = 574 nm, FWHM = 94 nm,
S/N = 56.
gradually shifts until at 28 °C and above the emission is structureless with a Xr
495 nm.
of
DISCUSSION
The number of published bioluminescence emission spectra is a small percentage
of the total number of luminescent marine species known to exist and the number
of accurately determined spectra is probably even smaller. This is largely due to the
difficulty of measuring bioluminescence by conventional spectrophotometric tech-
niques. The fragility of marine luminescent organisms demands that the instrumen-
tation be brought to them and the nature of the instrumentation thus far in common
use makes this difficult.
808 E. A. WIDDER ET AL.
The earliest practical spectrophotometric systems used in bioluminescence work
were designed for high sensitivity at the expense of resolution. Nicol (e.g., 1958,
1960), using paired photomultipliers, one to correct for total energy variation and
the other measuring spectral regions through a series of colored niters, was able to
measure dim, relatively long time-course sources such as myctophid photophores,
but resolution was limited to approximately 25 nm. Morin and Hastings (197 la, b)
utilized a grating monochromator with calibrated photomultiplier to give a calculated
bandwidth of 6.6 nm with 4 mm slits. Using a similar system with 2 mm slits, Swift
et al. (1977) were able to resolve emission peaks 3.5 nm apart. Even though scanning
time across the spectrum with the latter system was as short as 3.1 s (Biggley et al.,
1981), there was variability in consecutive measurements due to modulations in
emission intensity. Other high resolution systems currently in use include Reynolds1
(1978) photographic spectroscope-intensifier system and Wampler's spectrofluorometer
on-line computer system (Wampler and DeSa, 1971).* The Reynolds system allows
simultaneous registration of a wide spectral range (400-600 nm) but in its present
configuration it is not amenable to ship-board use and data reduction is time-con-
suming. The spectrofluorometer system of Wampler utilizes computer software to
facilitate data collection, storage, and analysis, but requires scan times of 8 s or longer
(Wampler et al., 1971, 1973). For this reason, techniques such as quick freezing and
subsequent thawing have been employed to generate steady bioluminescence over
the scanning period.
The OMA system, as we have employed it, has three properties essential for
determining the emission spectra of living specimens: high sensitivity, high resolution,
and simultaneous light collection. It, therefore, represents a practical solution to many
problems which have plagued bioluminescence emission spectroscopy, especially with
regards to extreme temporal variations of the emissions. For example, luminescent
flashes with very fast kinetics such as those produced by myctophid caudal organs
(as contrasted with photophores) were easily measured with the OMA in the present
study. No published spectra exist for these bright luminescent organs, presumably
because irregular flashes of such short duration (60-80 ms, Barnes and Case, 1974)
have been impossible to measure with a scanning spectrophotometer.
This system also has the advantage of being able to measure spectra from very
localized sources owing to the collection optics employed. The fiber collection optics
provide a spatial resolution that has not been previously available. The combination
of high sensitivity and spatial resolution allowed convenient measurement of the
spectrum of a single photophore (Fig. 4D) and makes the system ideally suited for
studying organisms capable of multichromatic emissions. The bichromatic Para-
zoanthus lucificum colony is a dramatic example of the need for this kind of resolving
power. Bioluminescence from the colony produced a bimodal spectrum that localized
measurements from single polyps resolved into two unimodal peaks. The highest
degree of spacial resolution was attained with the pinecone fish, Cleidopus gloria-
marls, where a gradient of emission was clearly resolved in a light organ measuring
only 4X2 mm (Haneda, 1966).
In some organisms the chemistry of the luminescent system is responsible for the
different colors of emissions. For example, in Umbellula the difference in emission
between the base and the top of the stalk may be accounted for by different ratios
of the two emitters responsible for pennatulid bioluminescence (Wampler et al., 197 1 ,
1973). The emission patterns present in Umbellula could involve an increasing con-
Note: Herring (in press) describes results obtained with a paired scanning photomultiplier system
(Collier et al., 1979). This device has a scan time of 30s and a stated accuracy of ±4 nm.
BIOLUMINESCENCE SPECTRA 809
centration of the green fluorescent protein emitter from stalk tip to base that gradually
masks the dim blue luciferin emission. The 533 nm emission spectrum from some
colonies of the pennatulid Stachyptilum is not readily explainable on the basis of the
known luciferin and green fluorescent protein emitters responsible for pennatulid
luminescence. This emission could be due to a different, undescribed emitter.
Optical filtering may also alter the color of some emissions. In Argyropelecus
filters narrow the bandwidth of the bioluminescence emission. This may facilitate
counterillumination since the maximum and bandwidth of the filtered emission are
very similar to that measured for oceanic downwelling irradiance (Young et al., 1980).
Filtering may also be responsible for the color of the bacterial light organ of the
pinecone fish, Cleidopus gloria-rnaris. The use of filters in the light organ accounts
for the difference between the blue-green emission from the intact light organ and
the blue luminescence of the bacterial isolates from such organs (Haneda, 1966). The
presence of a gradient of emission across the juvenile light organ that is absent in
the adults also seems to be due to optical filtering.
The primary value of accumulating a large library of corrected spectra is to classify
emitter types. Evidence exists that similar spectra may be due to a common emitter
(Wampler et al., 1973). It remains to be seen what chemical relationships exist between
organisms that share a common spectral fine structure but emit at different wavelengths,
such as the pennatulids already discussed (Fig. 4B) or the squids of Figure 4E. It is
also possible that similar spectra in unrelated species may reveal dietary dependencies,
although so far where such dependencies have been demonstrated the spectra of
predator and prey have been markedly dissimilar (Tsuji et al., 1975; Frank, Widder,
Latz and Case, in press).
ACKNOWLEDGMENTS
The authors are indebted to Mark Lowenstine who assisted with installation of
the OMA, provided advice on data processing, and has since maintained the system
in working order. We also thank Dr. Jeremy Lerner of Instruments SA for suggesting
the design of the collection optics. For assistance with animal collection we thank
Dr. Peter Anderson, Shane Anderson, Dr. Alissa Arp, Dr. James Childress, John
Favuzzi, Tamara Frank, Bill Lowell, and the captains and crews of the "R/V VELERO
IV" and "R/V NEW HORIZON." Dr. B. M. Sweeney provided unialgal cultures of
all the dinoflagellates and the bacterial isolates of Abyssicola macrochir and Coelor-
hynchus japonicus; Dr. Kenneth Nealson generously allowed us to work in his lab-
oratory and provided the remainder of the bacterial cultures. We are grateful to him
and to Sea World, San Diego, for providing specimens ofAnomalops and Cleidopus
from their displays and to Scripps Aquarium for providing the Parazoanthus colonies.
Specimens of Oikopleura were kindly provided by Dr. Charles Gait. Specimens were
identified with the assistance of Dr. F. G. Hochberg, Jr. and Paul Scott of the Santa
Barbara Museum of Natural History, Dr. Lawrence Madin of Woods Hole Ocean-
ographic Institution, Dr. Robert Carney of Moss Landing Marine Laboratory, and
Dr. Andrew Lisner. This work was supported by a grant from the Office of Naval
Research (ONR contract number N00014-75-C-0242), the FBN fund, and faculty
research funds from the University of California, Santa Barbara. Work at sea was
supported by a grant from the National Science Foundation (OCE 81-10154) to
J. J. Childress.
LITERATURE CITED
BARNES, A. T., AND J. F. CASE. 1974. The luminescence of lanternfish (Myctophidae): Spontaneous activity
and responses to mechanical, electrical, and chemical stimulation. J. Exp. Mar. Biol. Ecol. 15:
203-221.
810 E. A. WIDDER ET AL.
BIGGLEY, W. H.,T. A. NAPORA, AND E. SWIFT. 1981. The color of bioluminescent secretions from decapod
prawns in the genera Oplophorus and Systellaspis (Caridea). Pp. 66-7 1 in Bioluminescence Current
Perspectives, K. H. Nealson, ed. Burgess Publishing Co., Minneapolis.
COLLIER, A. F., R. J. P. BURNHAM, AND P. J. HERRING. 1979. A system for the collection of comparative
emission spectra suitable for shipboard use. J. Mar. Biol. Assoc. U.K. 59: 489-495.
DENTON, E. J., J. B. GILPIN-BROWN, AND P. G. WRIGHT. 1970. On the "niters" in the photophores of
mesopelagic fish and on a fish emitting red light and especially sensitive to red light. J. Phvsiol.
208: 72P-73P.
EDWARDS, T. H., AND P. D. WILLSON. 1974. Digital least squares smoothing of spectra. Appl. Spectrosc.
28: 541-545.
FELKEL, H. L., JR., AND H. L. PARDUE. 1979. Simultaneous multielement determinations by atomic
absorption and atomic emission with a computerized echelle spectrometer/imaging detector system.
Pp. 59-96 in Multichannel Image Detectors, Y. Talmi, ed. Am. Chem. Soc. Symp. Ser. Washington,
DC.
FITZGERALD, J. M. 1977. Classification of luminescent bacteria from the light organ of the Australian
pinecone fish, Cleidopus gloria-maris. Arch. Microbiol. 112: 153-156.
FRANK, T. M., E. A. WIDDER, M. I. LATZ, AND J. F. CASE. Dietary maintenance of bioluminescence in
a deep-sea mysid. / Exp. Biol. (in press).
HANEDA, Y. 1966. On a luminous organ of the Australian pine-cone fish, Cleidopus gloria-maris De Vis.
Pp. 547-555 in Bioluminescence in Progress, F. H. Johnson and Y. Haneda, eds. Princeton
University Press, Princeton.
HERRING, P. J. The spectral characteristics of luminous marine organisms. Proc. R. Soc. Lond. B. (in
press).
MORIN, J. G., AND J. W. HASTINGS. 197 la. Biochemistry of the bioluminescence of colonial hydroids and
other coelenterates. J. Cell Phvsiol. 77: 305-3 1 1 .
MORIN, J. G., AND J. W. HASTINGS. 1971b. Energy transfer in a bioluminescent system. J. Cell. Phvsiol.
77: 313-318.
NICOL, J. A. C. 1958. Observations on luminescence in pelagic animals. J. Mar. Biol. Assoc. U. K. 37:
705-752.
NICOL, J. A. C. 1960. Spectral composition of the light of the lantern fish Mvctophum punctaturn. J. Mar.
Biol. Assoc. U. K. 39: 27-32.
REYNOLDS, G. T. 1978. Application of photosensitive devices to bioluminescence studies. Pholochem.
Photobiol. 27: 405-421.
RUBY, E. G., AND K. H. NEALSON. 1977. A luminous bacterium that emits yellow light. Science 196: 432-
434.
SAVITZKY, A., AND M. J. E. GOLAY. 1964. Smoothing and differentiation of data by simplified least squares
procedures. Anal. Chem. 36: 1627-1639.
SELIGER, H. H., J. B. BUCK, W. G. FASTIE, AND W. D. MCELROY. 1964. The spectral distribution of
firefly light. J. Gen. Phvsiol. 48: 95-104.
SWIFT, E., W. H. BIGGLEY, AND T. A. NAPORA. 1977. The bioluminescence emission spectra of Pyrosoma
atlanlicum, P. spinosum (Tunicata), Euphausia tenera (Crustacea) and Gonostoma sp. (Pisces).
J Mar. Biol. Assoc. U. K. 57: 817-823.
TALMI, Y. 1982. Spectrophotometry and spectrofluorometry with the self-scanned photodiode array. Appl.
Spectrosc. 36: 1-18.
TALMI, Y., AND R. W. SIMPSON. 1980. Self-scanned photodiode array: a multichannel spectrometric detector.
Appl. Opt. 19: 1401-1414.
TSUJI, F. I., B. G. NAFPAKTITIS, T. GOTO, M. J. CORMIER, J. E. WAMPLER, AND J. M. ANDERSON. 1975.
Spectral characteristics of the bioluminescence induced in the marine fish, Porichthys nolatus, by
Cypridina (Ostracod) luciferin. Mol. Cell. Biochem. 9: 3-8.
WAMPLER, J. E. 1978. Measurements and physical characteristics of luminescence. Pp. 1-48 in Biolumi-
nescence in Action, P. J. Herring, ed. Academic Press, London.
WAMPLER, J. E., AND R. DESA. 1971. An on-line spectrofluorimeter system for rapid collection of absolute
luminescence spectra. Appl. Spectros. 25(6): 623-627.
WAMPLER, J. E., K. HORI, J. W. LEE, AND M. J. CORMIER. 197 1 . Structured bioluminescence. Two emitters
during both the in vitro and the in vivo bioluminescence of the sea pansy, Renilla. Biochemistry
10: 2903-2909.
WAMPLER, J. E., Y. D. KARKHANIS, J. G. MORIN, AND M. J. CORMIER. 1973. Similarities in the biolu-
minescence from the Pennatulacea. Biochim. Biophys. Ada 314: 104-109.
YOUNG, R. E., E. M. KAMPA, S. D. MAYNARD, F. M. MENCHER, AND C. F. E. ROPER. 1980. Counter-
illumination and the upper depth limits of midwater animals. Deep-Sea Res 27A: 671-691.
Reference: Biol. Bull. 165: 811-815. (December, 1983)
SPATIAL AND TEMPORAL PATTERNS OF MITOSIS IN THE CELLS
OF THE AXIAL POLYP OF THE REEF CORAL
ACROPORA CERVICORNIS
ELIZABETH H. GLADFELTER
West Indies Laboratory, Teague Bay, Chrisliansted, St. Croix, USVI, and Department of Biology,
University of California, Los Angeles, California 90024
ABSTRACT
The fluorescent stain DAPI was used to observe mitoses in the endoderm and
the calicoblastic ectoderm of the axial polyp of the reef coral Acropora cervicornis.
A diel periodicity in the mitotic index (denned as the percentage of cells in some
stage of mitosis) of each tissue occurred with a maximum of about 2% at midnight
and a minimum of 0.5% at midday. Dividing cells were located from the tip of the
column (when the polyp was contracted into the calyx) to 10 mm proximal to this
point suggesting that there is no narrow zone of proliferating cells. The magnitude
of the mitotic indices of these tissues suggests that it may account for the observed
daily growth rate of ca. 300 ^m in the axial polyp.
INTRODUCTION
Coral growth has been the focus of numerous studies (e.g., Buddemeier and
Kinzie, 1976; Highsmith, 1979; Gladfelter 1982, 1983; Wellington and Glynn, 1983);
most investigators have measured some parameter of skeletal growth. Rates of tissue
production have been inferred from linear increases in the skeleton (Lewis, 1982),
but direct measurements have not been made. In acroporid corals, for example, an
increase in the length of an axial polyp as the skeleton extends could initially involve
only a change in the shape of the cells, i.e., elongation. Eventually, however, the
production of new tissue must involve cell division and subsequent growth.
Site and frequency of cell division have been investigated in other cnidarians,
e.g.. Hydra spp. (David and Campbell, 1972; Neckelmann, 1982); colonial hydroids
(Hale, 1964; Campbell, 1968); anemones (Singer, 1971; Minasian 1980); and scler-
actinian corals (Cheney 1973). Some authors reported that cell proliferation occurs
only or at least primarily in the ectoderm (Hale, 1964; Singer, 1971; Cheney, 1973)
while others stated that cell division occurs in both the endoderm and the ectoderm
(Campbell, 1968; David and Campbell, 1972; Minasian, 1980). Whether this dis-
crepancy in the site of cell proliferation is due to species specific differences or to
incorrect interpretation of data has not yet been resolved (Davis, 1971).
In this study, the site and diel periodicity of mitoses in the endoderm and the
calicoblastic ectoderm of the axial polyp of the reef coral Acropora cervicornis were
determined.
MATERIALS AND METHODS
Coral tips were collected at 0600, 0900, 1200, 1500, 1800, 2100, 2400, and 0200
from a depth of 1 1 m in Buck Island Channel, St. Croix, U. S. Virgin Islands. The
tips were transferred immediately to the West Indies Laboratory. Within 30 min of
Received 16 May 1983; accepted 2 September 1983.
811
812 E. H. GLADFELTER
collection, the specimens were fixed in 10% formalin and stored. The corals tips were
decalcified in 10% EDTA in 0.03 M NaOH for a day. Each tip was trimmed to a 1
cm length. Tips were dehydrated in a graded series of ethyl alcohol, cleared in toluene,
and embedded in Paraplast (m.p. 57-59°C); each step required 15 min.
Longitudinal sections, 10 /im thick were cut from prepared tissue blocks with a
microtome. Sections through the midsection of the polyp were saved and placed on
glass slides coated with 1% gelatin. The tissue on the slides was rehydrated (2 min
per step); after 4 min in distilled water, a drop of DAPI (4'-6-Diamidino-2-Phenylindole;
1 Mg * nil"1 distilled water; Russell et al, 1975) was placed on the tissue and a coverslip
placed on the slide.
The slide was examined within minutes by epifluorescence microscopy as described
by Neckelmann (1982). The nuclei of all cells appear fluorescent with mitotic figures
staining brightly. Zooxanthellae fluoresce red. The slide was first surveyed using the
25X objective. Fields viewed with the 63X objective were sampled from the distal
tip of the polyp to 2 mm below the tip. The percentage of cells in some phase of
mitosis (i.e., late prophase, metaphase, anaphase, and telophase) was determined for
the endoderm and the calicoblastic ectoderm for each field sampled. Enough fields
were counted on each tip until ca. 1000 endodermal cells and ca. 750 calicoblastic
ectodermal cells were examined. Two tips were thus examined for each time of
collection and an average value determined.
RESULTS
The tissues at the tip of the axial polyp ofAcropora cervicornis are clearly outlined
by the fluorescence of their nuclei when stained with DAPI. In these longitudinal
sections, the outer ectoderm had the highest density of cells; the positions of the
nematocysts and spirocysts are also clearly visible. The nuclei of the cells of the outer
ectoderm overlapped so frequently that an accurate determination of cells in some
phase of mitosis in this tissue layer was not feasible; some mitotic figures were observed,
however.
The calicoblastic ectoderm and the endoderm appear to have a high density of
cells covering the distal tip of the skeleton; as the conformation of these cells changes
with distance from tip, from columnar to squamous (Gladfelter, 1983) the nuclei
become spaced further apart.
Most of the nuclei in all the tissues at all times of day were in interphase (i.e.,
some stage of the cell cycle other than mitosis). The nuclei in this condition stained
brightly, but diffusely when compared to nuclei where mitotic figures were present.
Occasionally, dividing zooxanthellae were also seen.
Mitosis occurs in both the endoderm and the calicoblastic ectoderm. To determine
the frequency of mitosis in each tissue layer, I calculated a mitotic index (M.I.) for
each tissue at each time of day. The mitotic index is the percentage of total cells in
some stage of mitosis. The results, expressed as an average and a range (Fig. 1 ) show
that the diel pattern of mitotic division in the cell populations is moderately syn-
chronous. The M.I. of the endoderm is high (>1%) from 1800 through 0600, peaking
at 2% at 2400, and low (<0.7%) from 0900 through 1 500. The M.I. of the calicoblastic
ectoderm also shows a peak of ca. 2% at 2400, but the range of values is much greater,
and the peak much sharper than seen in the endoderm. All values (with the exception
of almost overlapping values at 0900) are higher in the endoderm than the calicoblastic
ectoderm.
Frequency of division as a function of distance from tip was not quantified.
Observations indicate that dividing cells occur not only right at the tip (among the
CELL DIVISION IN AXIAL POLYP OF ACROPOR.4
813
0.4 -
1200 1500 1800
2100 2400 0200
Time of day
0600 0900 1200
FIGURE 1. Diel pattern of percentage division (M.I.) of cells from the endoderm (solid circles) and
cells from the calicoblastic ectoderm (open circles). Each point is the average of two determinations from
each time period; the vertical bars indicate the range of the two values. Note that the two values for the
calicoblastic ectoderm at 1 500 were the same, hence no range bar.
columnar cells) but also 2 mm from the tip (the extent of the region surveyed for
the determination of mitotic indices) and up to 10 mm from the tip as well. No
narrow zone of cell proliferation is apparent; cells divide at random points throughout
the column.
DISCUSSION
The axial polyp of Acropora cervicornis contains dividing cells from the tip to at
least 10 mm proximal to the tip in both the endoderm and the calicoblastic ectoderm.
These results are similar to those found by David and Campbell (1972) for the
hydrozoan polyp Hydra attenuata and by Minasian (1980) for the anthozoan (actini-
arian) polyp Haliplanella luciae. Both studies described proliferating cells among all
epithelial layers. In addition, David and Campbell (1972) showed that the number
of divisions observed in the endodermal and ectodermal tissue were enough to account
for the observed growth of those cell populations; migration of cells from one epithelial
layer to another probably did not occur. Until more is known about the cell cycle
kinetics of A. cervicornis, it cannot be stated with certainty that the mitotic indices
observed in this study would result in a sufficient increase in cell population to account
for the observed growth rate of the polyp. However, the magnitude of the mitotic
indices in the endoderm and the calicoblastic ectoderm of A. cervicornis is the same
as that seen in Hydra attenuata (David and Campbell, 1972) and H. viridis (Neck-
elmann, 1982) suggesting that if the cell cycle kinetics are similar to those described
for H. attenuata (David and Campbell, 1972), which result in a cell population
doubling time of 3 days, then these observed mitotic events are probably enough to
account for the rapid axial growth of A. cervicornis.
Campbell (1968) found that cell division occurred in both the ectoderm and the
endoderm of the extending stolons of Proboscidactyla, a colonial hydroid. The rate
of elongation along a growth axis in this situation is comparable to that seen in A.
814 E. H. GLADFELTER
cervicornis. However Hale (1964), described cell division primarily among the ec-
todermal cells in the stolons of another hydroid, Clytia. Cheney (1973) used tritiated
thymidine to label proliferating cells and found labeled cells primarily among cells
of the column epidermis (ectoderm) and of the polyps and coenosarc of the reef coral
Pocillopora damicornis. The internal tissues incorporated little, if any, label. This
might reflect a label uptake problem rather than a true picture of the sites of cell
proliferation in coral tissues. The internal tissues probably take up label from the
fluid in the coelenteron; under experimental conditions this fluid might not exchange
rapidly with the external medium.
The frequency of mitosis in the endoderm and the calicoblastic ectoderm of
Acropora cervicornis has a diel periodicity, with a peak at midnight. David and Camp-
bell (1972) found a diel periodicity in the mitotic index of ectodermal and endodermal
cells of Hydra attenuata; they correlated the midnight peak with a daily feeding regime
at 1000 each morning. Neckelmann (1982) also found a diel periodicity in the mitotic
index of endodermal cells in H. viridis. The peak occurred ca. 10-12 h after feeding;
no peak occurred in starved controls. The A. cervicornis in the present study were
collected from field populations. Normal feeding in these coral colonies probably
occurs on a diel cycle set by food availability. Demersal plankton, an important food
source for corals (Porter, 1974; Aldredge and King, 1977) is thought to be most
abundant, i.e., available for consumption, at dusk and especially at dawn (Glynn,
1973). Johannes and Tepley (1974) found that the peak feeding activity of Porites
lobata (another reef coral with small polyps as in A. cervicornis} occurs at dawn. Thus
the diel cycle in mitotic index of cells in the axial polyp of A. cervicornis might be
set by a naturally occurring cycle in feeding behavior. It could also be related to the
diel periodicity of a reef coral's other carbon source, i.e., photosynthate transferred
from the zooxanthellae (Muscatine et ai, 1981).
In another system in which the growth of the organism is dependent primarily
on increase in cell number, i.e., freshwater planarians, Baguna (1974) demonstrated
a rapid increase in mitotic activity following feeding. He hypothesized that the cell
population of a planarian contains a number of cells in the G2 state of the cell cycle.
These cells are awaiting the proper stimulus (e.g., food) to divide; the precise mechanism
of how food stimulates cell division is unclear. Presumably, reef corals receiving a
daily pulse of organic carbon (from zooplankton or zooxanthellae) might have a daily
peak of cell division; those deprived of this normal nutritional regime should show
decreased mitotic activity.
In this study, a diel cycle in mitosis was revealed. It suggests a periodicity in polyp
elongation; daily extension is 300 /urn. Skeletal growth in A. cervicornis has a diel
pattern (Gladfelter, 1983). Extension in another branching acroporid is at least as
rapid during the night as in the day (Barnes and Crossland, 1980). The factors which
set these diel cycles in both tissue and skeletal growth are unknown.
To summarize, endodermal and calicoblastic ectodermal cells are in stages of
mitosis in the column of the axial polyp of Acropora cervicornis. The magnitude of
the mitotic indices of these cell populations are on the order of 0.5%-2% and vary
in a diel pattern. Cell division in each tissue layer is probably enough for the observed
rate of growth of these cell populations, resulting in a daily elongation rate of
300 urn.
ACKNOWLEDGMENTS
I would like to thank W. B. Gladfelter and many West Indies Laboratory un-
dergraduate students for assistance in collection of samples and L. Muscatine for
CELL DIVISION IN AXIAL POLYP OF ACROPORA 815
reviewing the manuscript. T. James, N. Neckelmann, G. M. Parker, and F. Wilkerson
contributed valuable suggestions about using DAPI to observe dividing cells. Support
during the final part of this study was provided through the Meta McBride Haupt
Dissertation Fellowship of the American Association of University Women.
This is West Indies Laboratory Contribution No. 98.
LITERATURE CITED
ALLDREDGE, A. L., AND J. M. KING. 1977. Distribution, abundance and substrate preferences of demersal
reef zooplankton at Lizard Island Lagoon, Great Barrier Reef. Mar. Biol. 41: 317-333.
BAGUNA, J. 1974. Dramatic mitotic response in planarians after feeding, and a hypothesis for the control
mechanism. J. Exp. Zool. 190: 117-122.
BARNES, D. J., AND C. J. CROSSLAND. 1980. Diurnal and seasonal variations in the growth of a staghorn
coral measured by time-lapse photography. Limnol. Oceanogr. 25: 113-117.
BUDDEMEIER, R. W., AND R. A. KJNZIE. 1976. Coral growth. Oceanogr. Mar. Biol. Ann. Rev. 14: 183-
225.
CAMPBELL, R. D. 1968. Colony growth and pattern in the two-tentacled hydroid, Proboscidactvla flavicirrata.
Biol. Bull. 135: 96-104.
CHENEY, D. P. 1973. Cell proliferation as an index of growth in corals: incorporation of 3H-thymidine.
Publ. Seto Mar. Biol. Lab. 20: 285-297.
DAVID, C. N., AND R. D. CAMPBELL. 1972. Cell cycle kinetics and development of Hvdra attenuata. I.
Epithelial cells. J. Cell Sci. 11: 557-568.
DAVIS, L. V. 1971. Growth and development of colonial hydroids. Pp. 16-36 in Experimental Coelenterate
Biology, H. M. Lenhoff, L. Muscatine, and L. V. Davis, eds. University of Hawaii Press, Honolulu.
GLADFELTER, E. H. 1982. Skeletal development in Acropora cervicornis. I. Patterns of calcium carbonate
deposition in the axial corallite. Coral Reefs 1: 45-51.
GLADFELTER, E. H. 1983. Skeletal Development and Related Aspects of the Reef Coral Acropora cervicornis.
Ph.D. Dissertation, University of California, Los Angeles.
GLYNN, P. W. 1973. Ecology of a Caribbean coral reef. The Porites reef-flat biotype. Pt. II. Plankton
community with evidence for depletion. Mar. Biol. 22: 1-22.
HALE, L. J. 1964. Cell movements, cell division and growth in the hydroid Clvtia johnstoni. J. Embrvol
Exp. Morph. 12: 517-538.
HIGHSMITH, R. C. 1979. Coral rates and environmental control of density banding. J. Exp. Mar. Biol.
Ecol. 37: 105-125.
JOHANNES, R. E., AND L. TEPLEY. 1974. Examination of feeding of the reef coral Porites lobata in situ
using time-lapse photography. Proc. 2nd Int'l Coral ReefSymp. I: 127-131.
LEWIS, J. B. 1982. Estimates of secondary production of reef corals. Proc. 4th Int'l Coral Reef Svmp. 2:
369-374.
MIN ASIAN, JR, L. L. 1980. The distribution of proliferating cells in an anthozoan polyp, Haliplanella luciae
(Actiniaria: Acontiara), as indicated by 3H-thymidine incorporation. Pp. 415-420 in Developmental
and Cellular Biology of Coelenterates, P. Tardent and R. Tardent, eds. Elsevier/North-Holland
Biomedical Press, Holland.
MUSCATINE, L., L. R. McCLOSKEY, AND R. E. MARIAN. 1981. Estimating the daily contribution of carbon
from zooxanthellae to animal respiration. Limnol. Oceanogr. 26(4): 601-611.
NECKELMANN, N. S. 1982. Regulation of Numbers of Symbiotic Algae in the Digestive Cell of Hydra viri-
dis. Ph.D. Dissertation. University of California, Los Angeles.
PORTER, J. W. 1974. Zooplankton feeding by the Caribbean reef-building coral Montastrea annularis. Proc.
2nd Int'l Coral ReefSymp. I: 1 1 1-125.
RUSSELL, W. C., C. NEWMAN, AND D. H. WILLIAMSON. 1975. A simple cytochemical technique for
demonstration of DNA in cells infected with mycoplasmas and viruses. Nature 253: 461-462.
SINGER, I. I. 1971. Tentacular and oral-disc regeneration in the sea anemone Aiptasia diaphana III. Au-
toradiographic analysis of patterns of tritiated thymidine uptake. J. Embrvol. Exp. Morphol. 26:
253-270.
WELLINGTON, G. M., AND P. W. GLYNN. 1983. Environmental influence on skeletal banding in Eastern
Pacific (Panama) corals. Coral Reefs 1(4): 215-222.
INDEX
Accumulation of late H2b histone mRNA in sea
urchin embryogenesis, 501
Acid-base balance, 582
Acid phosphatase, 241
Acid precipitation, 507
AOCERMAN, E. B., AND H. L. KORNBERG, Mutants
of Escherichia coli affected in "inducer exclu-
sion," 520
ACK.ERMAN, JOSEF, Current flow around Zostera
marina plants and flowers: implications for
submarine pollination, 504
Acquisition of a collection of western north Atlantic
fishes (Pisces) by the Gray Reference Collection,
Marine Biological Laboratory, Woods Hole,
MA, The, 505
Acritarchs, 241
Acropora cen'icornis. 619
Actin microfilaments are a major cytoskeletal com-
ponent in squid axoplasm, 489
ADELMAN, W. J., JR., AND ALAN J. HODGE, Pseu-
dostereoscopy allows direct visualization of the
velocity distribution of particles undergoing fast
axonal transport, 523
ADELMAN, W. J., JR., see Alan J. Hodge, 527
Adhesion, 502
Aeolidia papillosa, 394
Aequorin, 529
Afferent neurons, 527
Age of first reproduction in Melampus bidentatus:
the effects of overwintering degrowth and repair,
511
Aging, skate ocular lens, 499
Aggression, intrasexual in Metridium, 416
AGUDELO, MARIA I., KENNETH KUSTIN, GUY C.
MCLEOD, WILLIAM E. ROBINSON, AND ROB-
ERT T. WANG, Iron accumulation in tunicate
blood cells. I. Distribution and oxidation state
of iron in the blood of Botenia ovifera, Slycla
clava, and Molgula manhattensis, 100
Air exposure, 582. 708
ALATALO, PHILIP, AND CARL J. BERG, JR., Popu-
lation ecology of the Caribbean bivalve Asaphis
deflorata (Linne, 1758), 504
ALBERTE, R. S., see W. C. Dennison, 507, and L.
Mazzella, 508
Alcyonium siderium Verrill, 286
Algal mats, 509
ALKON, DANIEL L., see Alan M. Kuzirian, 528, Izja
Lenderhendler, 528, Louis Leibovitz, 535, and
Serge Gart, 525
ALLEN, NINA STROMGREN, AND WENDY F. Boss,
Studies of the isolation and calcium-induced
fusion of fusogenic wild carrot protoplasts, 487
u.LEN, ROBERT, D., DOUGLAS T. BROWN, SUSAN
P. GILBERT, AND HIDESHI FUJIWAKE, Trans-
port of vesicles along filaments dissociated from
squid axoplasm, 523
ALLIEGRO, M. C., AND H. SCHUEL, Is there specificity
in the induction of polyspermy in sea urchins
by protease inhibitors? 512
Allogromia. 489, 497
Almyracuma proximoaili, 370
Alpha and Beta tubulin gene linkage, 488
Alpha-2-macroglobulin, 495
ALVAREZ, Luis, see Marta Bretos, 559
Alzheimer's disease, 530
Amino acid, 532
Amoebomastigote, 241
cAMP: a possible intracellular transmitter of cir-
cadian rhythms in Lirnulus photoreceptors, 540
Anaerobic chitin degradation as a carbon and hy-
drogen source for sulfate reduction and meth-
anogenesis in salt marsh bacteria, 505
Anaerobic metabolism, 708
Analysis of hemolymph oxygen levels and acid-base
status during emersion "in situ' in the red rock
crab, Cancer productus, 582
ANAYA-VALAZQUEZ, L. F., AND K.-P. CHANG,
Phagocytosis and intralysosomal killing of
Leishmania mexicana by Entamoeba histoly-
tica. 534
ANDERSON, CATHLEEN, see Gerald Weissmann, 503
ANDERSON, WINSTON A., AND WILLIAM R. ECK-
BERG, A Cytological analysis of fertilization in
Chaetopterus pergamentaceus, 1 1 0
Animal husbandry, 528
Animalization, 502
Anthozoan, 569
Antigen masking, 537
Antigens on both mechanical and lung stage schis-
tosomula of Schistosoma mansoni are masked
by host molecules, 537
Anti-tubulin, 515
ANTONELLIS, BLENDA, see Seymour Zigman, 499
Arbacia punctulata, 499, 500, 513, 516, 517
ARMSTRONG, PETER B., JAMES P. QUIGLEY, AND
JACK LEVIN, A proteinase inhibitor released
from the Limulus amebocyte during exocytosis,
488
ARMSTRONG, PETER B., see James P. Quigley, 495
Arsenazo III, 529, 542
Anemia tunisiana, 506
Ascidian embryos, 501
Ascidian-/Vo<:/2/0rott symbiosis: the role of larval
photoadaptations in midday larval release and
settlement, 221
Ascidians
coordination by epithelial conduction, 209
symbiosis with Prochloron, 221
Associative learning in Hermissenda crassicornis
(Gastropoda): evidence that light (the CS) takes
on characteristics of rotation (the UCS), 528
816
INDEX TO VOLUME 165
817
ATEMA, JELLA, see Marilyn Spalding, 532
Atlantic, 505
ATPase. 494
ATP-reactivated models of ctenophore comb plates,
497
AUGUSTINE, GEORGE J., see Stephen J. Smith, 532
Axon, 529
Axonal protein synthesis, 526
Axoplasm, 489
AYLING, AVRIL L., Growth and regeneration rates
in thinly encrusting demospongiae from tem-
perate waters, 343
B
Baclofen, 523
Bacteria
enteric, 522
manganese oxidizing, 520
photosynthetic. 509
spore forming, 520
symbiotic, 521
Bacterial clearance in urchins, 473
Bacterial taxis, 509
Baja California, 24 1
BAKER, ROBERT, see Stephen M. Highstein, 527
Balamis, 330
BARLOW, ROBERT B., JR., LEONARD KASS, VIVIAN
MANCINI, AND JANICE L. PELLETIER, Vision
in Limulus mating behavior: tests for detection
and form discrimination, 539
BARLOW, ROBERT B., JR., see Leonard Kass, 540,
and Jennifer Marler, 541
BARNES, WAYNE M., see Eric R. Ward, 498
Barometric pressure, 509
BARRY, SUSAN, R., Presynaptic action of baclofen,
a GABA analog, at the crayfish neuromuscular
junction, 523
Bathypelagic, 182
BAYNE, CHRISTOPHER, J., see Mary A. Yui, 473
Behavior, 539, 745
squid, 637
BENAYAHU, Y., AND Y. LOYA, Surface brooding in
the Red Sea soft coral Parerythropodium fiilvum
fiih'um (Forskal, 1775), 353
BERG, CARL J. JR., KATHERINE S. ORR, AND JEFFRY
B. MITTON, Genetic variation in the queen
conch, Strombus gigas. across its geographic
range. Preliminary results, 504
BERG, CARL J. JR., see Philip Alatalo, 504
BlCKELL, LOUISE R., AND STEPHEN C. KEMPF, Lar-
val and metamorphic morphogenesis in the nu-
dibranch Melibe Leonina (Mollusca: Opistho-
branchia), 1 19
Binding of MC-gossypol by Arbacia sperm, 516
Biology of Fissure/la maxima Sowerby (Mollusca:
Archaeogastropoda) in northern Chile. 2. Notes
on its reproduction. The, 559
Bioluminescence, 444, 522
Bioluminescence spectra, 791
Biomineralization, 723
Biotinylated DNA probes, 537
Bivalves, 504, 521, 708
Black line disease, 429
Blebbing, 502
Blood cells of tunicates, 100
BLUM, FREDERIC, MARGARET NACHTIGALL, AND
WALTER TROLL, Superoxide dismutase bio-
mimetic compounds prevent fertilization in
Arbacia punctulata eggs, 513
Boltenia ovifera, see tunicates
BOOKMAN, RICHARD J., Slow rearrangements of
membrane bound, halogenated fluoresceins
produce altered K+ currents in squid axon, 524
BOOTH, CHARLES E., see Peter L. deFur, 582
BORNBUSCH, ALAN H., The acquisition of a collec-
tion of western north Atlantic fishes (Pisces) by
the Gray Reference Collection, Marine Biolog-
ical Laboratory, Woods Hole, MA, 505
BORNSLAEGER, ELAYNE, see Richard Cornall, 513
Boss, WENDY F., see Nina Stromgren Allen, 487
Botryllus rejection reactions, 733
BOWER, J. M., see R. Llinas, 529
BOYER, BARBARA C., see Paul P. Palaszewski, 502
BOYER, JOSEPH N., AND RALPH S. WOLFE, Anaer-
obic chitin degradation as a carbon and hydro-
gen source for sulfate reduction and methan-
ogenesis in salt marsh bacteria, 505
BRAY, SARAH, AND TIM HUNT, Developmental
studies of a major mRNA in Arbacia punctulata,
499
BRAY, SARAH, see Elizabeth L. George, 515
BRENCHLEY, G. A., AND J. T. CARLTON, Compet-
itive displacement of native mud snails by in-
troduced periwinkles in the New England in-
tertidal zone, 543
BRETOS, MARTA, ITALO TESORIERI, AND Luis AL-
VAREZ, The biology ofFissurella maxima Sow-
erby (Mollusca: Archaeogastropoda) in northern
Chile. 2. Notes on its reproduction, 559
BROWN, DOUGLAS T., see Robert D. Allen, 523
BROWNE, ROBERT, Speciation in the brine shrimp
Artemia: cross-breeding between sexual Medi-
terranean populations, 506
BURGOS, M. H., see Eimei Sato, 516, and S. J. Segal,
517
Burrowing response, 509
CaCo3, 723
Calcification in Tegula, 265
Calcium, 503, 542
action potential, 529
-activated protease, 531
buffer, 541
changes, 498
regulation in mitosis, 495
transport enzyme, 495
Calcium activated channels in the mechanically sen-
sitive abfrontal ciliated cells ofMytiliis gill, 496
Calcium transients during early development in single
starfish (Asterias forbesi) oocytes and eggs, 514
818
INDEX TO VOLUME 165
Calcium transients during fertilization in single sea
urchin eggs, 514
Calcium transients in voltage clamped presynaptic
terminals, 532
Callinectes sapidus, 32 1
Capitella sibling species, 538
Capitella spp. eggs and follicle cells, 379
CARACO, NINA, AND IVAN VALIELA, Iron and phos-
phorus cycling in a permanently stratified
coastal pond, 506
Carbohydrate transport, 520
Carbon dioxide, 5 1 1
Carcinoscorpius rotundicauda, see horseshoe crabs
Caribbean, 504
CARLTON, J. T., see G. A. Brenchley, 543
CARSON, MONICA, AND REX L. CHISHOLM, Isolation
and characterization of tubulin clones from
Dictyostelium discoidium. 488
Caryology, see karyology
CASE, JAMES F., see Edith A. Widder, 791
CASTENHOLZ, RICHARD W., see Alan W. Decho,
507, Lisa Muehlstein, 521, and Kenneth M.
Noll, 509
CATALANO, E., see R. Vitturi, 450
CAVANAUGH, COLLEEN M., see Tricia A. Mitchell,
521
Cell
cycle, 513
shape maintenance, 489
Cell division in axial polyp of Acropora. 8 1 1
Cell-cell recognition and adhesion during embryo-
genesis in the sea urchin, 502
CENTONZE, V. E., AND J. L. TRAVIS, Immunoflu-
orescence ofAllogromia reticulopodia, 489
CENTONZE, V. E., see J. L. Travis, 497
Cerebratulus. 516
Chaetopierus. 1 1 0, 5 1 4, 5 1 5, 5 1 8
CHANG, DONALD C, ICHIJI TASAK.I, AND TOM S.
REESE, Structure of the squid axon membrane
as seen after freeze-fracture, 524
CHANG, K.-P., see L. F. Anaya-Valazquez, 534, and
L. Rivas, 536
Changes in histone synthesis during Arhacia devel-
opment, 500
Characterization and isolation of a homologue of
alpha-2-macroglobulin from the plasma of the
horseshoe crab Limulus, 495
Characterization of the major surface antigen of
Plasmodiwn falciparum merozoites, 535
Characterization of Trypanosoma brucei tubulin
genes, 490
CHARLTON, MILTON P., see Stephen J. Smith, 532
Chemical communication in Rhithropanopeus, 154
Chemical scavenging, 506
Chemoreception, 532
Chemotaxis, 419
CHILDRESS, JAMES J., see Page Hiller-Adams, 182
Chionqicetes bairdi, see tanner crab
CHISHOLM, REX L., see Monica Carson, 488
Chitin, 505
! oHnergic agonists at the giant synapse, 533
RNESKY, ELIZABETH A., Induced development
of sweeper tentacles on the reef coral Agaricia
agaricites: a response to direct competition, 569
Chthamalus, 330
Ciliary
movement, 496
reversal, 497
Circadian pacemaker neurons, 540
Circadian rhythms, 540
Circulation of fluids in the gastrovascular system of
the reef coral Acropora cervicornis, 619
CLAPIN, D. F., see J. Metuzals, 530
Classical conditioning, 528
Closed systems, 637
Clumping organisms, 512
Cnidarian, 569, 778
Coelomic fluid, 473
Coelomocytes, 473
COHEN, ROCHELLE S., NASRIN HAGHIGHAT, AND
GEORGE D. PAPPAS, Fine structure of synapses
and synaptosomes of the squid (Loligo pealei)
optic lobe, 525
COHEN, ROCHELLE S., see Harish C. Pant, 531
COHEN, WILLIAM D., AND JACQUELYN JOSEPH-SIL-
VERSTEIN, Marginal band function in the dog-
fish erythrocyte, 489
COHEN, WILLIAM D., see Jacqueline Joseph-Silver-
stein, 492
Colony size, 778
Comparative immunology, 473
Competition, 569
Competitive displacement of native mud snails by
introduced periwinkles in the New England in-
tertidal zone, 543
Composition and function of the cytoskeleton in
"blood clam" erythrocytes, 492
CONLON, RONALD A., see Albrecht Von Brunn, 519
Constraint surface, 305
Control of egg hatching in the crab Rhithropanopeus
harrisii (Gould), 154
Coordination of compound ascidians by epithelial
conduction in the colonial blood vessels, 209
Coral
disease, 429
larval settlement, 286
reef, 569, 619
sweeper tentacle development, 569
CORNALL, RICHARD, ELAYNE BORNSLAEGER, AND
TIM HUNT, What makes cyclin cycle? 513
COSTLOW, JOHN D., see John A. Freeman, 409
Crabs, Rhithropanopeus. 139, 154
CRONIN, THOMAS W., AND RICHARD B. FORWARD,
JR., Vertical migration rhythms of newly
hatched larvae of the estuarine crab, Rhithro-
panopeus harrisii. \ 39
Crustacea, 409, 582
Crustacean reproductive biology, behavior, 370
CSEKO, YARA, see Giullermo Romero, 537
Ctenophore, 491, 497
Cumacean dimorphism and behavior, 370
Current clamp of photoreceptors and pacemaker
neurons in eye of Bulla. 540
Current flow around Zostera marina plants and
INDEX TO VOLUME 165
819
flowers: implications for submarine pollination,
504
Cysts, 241
Cytological analysis of fertilization in Chaetopterus
pergamentaceus, A, 110
Cytoskeleton, 489, 492, 514
D
DAPI, 811
Dark-field microscopy, 491
DECHO, ALAN W., AND RICHARD W. CASTENHOLZ,
Interactions of harpacticoid copepods and pho-
tosynthetic microbes in the salt marsh, 507
Decomposition, 509
Decapoda, 582
Deep-sea, 167, 182
Deforestation in the Amazon Basin measured by
satellite: a release of CO-> to the atmosphere,
511
DEFUR, PETER L., BRIAN R. MCMAHON, AND
CHARLES E. BOOTH, Analysis of hemolymph
oxygen levels and acid-base status during emer-
sion "in situ' in the red rock crab. Cancer prod-
uctus, 582
DENNISON, W. C, AND R. S. ALBERTE, Growth
responses ofZostera marina (eelgrass) to in situ
manipulations of sediment nitrogen availability,
507
DENNISON, W. C., see L. Mazzella, 508
Depolarizing and desensitizing actions of glutami-
nergic and cholinergic agonists at the squid giant
synapse, 533
Desiccation of tolerance, 241
Design and construction of a benchtop reactor to
model an anaerobic/oxic wastewater treatment
system. The, 510
Detection of Leishmania kinetoplast DNA using
biotinylated DNA probes, 537
Detection of membrane signals correlated with sen-
sory excitation of phototactic Halobacterium
halobium, 540
Detritus, quality and age, 510
Deuterostome immunology, 473
Development, 491, 518
Developmental regulation, 500
Developmental studies of a major maternal mRNA
in Arbacia punctulata, 499
Diapause, 5 1 1
Dictyostelium discoidium. 488
Diel patterns of coral cell division, 8 1 1
Diel vertical movements of bacteria in intertidal
streams of Sippewissett Marsh, 509
Direct development, 591
Disease, in Hermissenda, 535
DMSO facilitation, 527
cDNA clone, 499
DOBBELAERE, D. A. E., see A. S. Fairneld, 535
Dogfish erythrocyte, 489
Dopamine, 758
Drosophila melanogaster. 490
DUNCAN, THOMAS K., Sexual dimorphism and re-
productive behavior in Almyracuma proxi-
moculi (Crustacea: Cumacea): the effect of hab-
itat, 370
DUNCAN, JENNIFER, BRUCE PETERSON, AND SALLY
MARQUIS, The sensitivity of freshwaters of Cape
Cod, Massachusetts to acid precipitation, 507
DUNHAM, PHILIP, see Gerald Weissmann, 503
DUNN, DAPHNE FAUTIN, see Steven E. Wedi, 458
DUNN, KENNETH, see L. B. Slobodkin, 305
E
EAGLES, P. A. M., see J. Metuzals, 530
Ecdysis, 758
Ecdysteroid liters during the molt cycle of the blue
crab resemble those of other Crustacea, 32 1
Echinocyamus pusillus, 745
Echinoderm immunology: bacterial clearance by the
sea urchin Strongylocentrotus purpuratus, 473
ECKBERG, WILLIAM R., AND GEORGE M. LANG-
FORD, Isolation of cytoskeletons from Chae-
topterus eggs, 5 1 4
ECKBERG, WILLIAM R., see Winston A. Anderson,
110
ECKELBARGER, KEVIN J., AND JUDITH P. GRASSLE,
Ultrastructural differences in the eggs and ovar-
ian follicle cells ofCapitella (Polychaeta) Sibling
species, 379
Ecology of coral reefs, 569
Effect of age and quality of detritus on growth of
the salt marsh snail, Melanipus bidentatus, 510
Effects of feeding, feeding history, and food depri-
vation on respiration and excretion rates of the
bathypelagic mysid Gnathophausia ingens, 182
Effects of intracellular calcium/EGTA on the pho-
toactivation of Linntlus ventral photoreceptors.
The, 541
Efferent neurons, 527
Egg, 514
Egg hatching in Rhithropanopeus, 154
Ehrlich ascites tumor cells, 496
EHRLICH, BARBARA E., CATHY R. SCHEN, AND
JOHN L. SPUDICH, Detection of membrane sig-
nals correlated with sensory excitation of pho-
totactic Halobacterium halobium, 540
EISELE LESLIE, see Jennifer J. Marler, 541
EISEN, A., G. T. REYNOLDS, S. WIELAND, AND
D. P. KIEHART, Calcium transients during fer-
tilization in single sea urchin eggs, 514
EISEN, A., G. T. REYNOLDS, S. WIELAND, AND D.
P. KIEHART, Calcium transients during fertil-
ization in single sea urchin eggs, 514
EKAPANYAKUL, G., A. FLISSER, A. Ko, AND D.
HARN, IgE monoclonal antibodies produced
from mice immunized with irradiated cercariae
of Schistosoma mansoni, 534
Electrochemical potential, 496
Electron microscopy
of calcification process, 265
of protein crystals, 530
ELLINGTON, W. R., see C. V. Nicchitta, 708
ELLIS, GORDON W., see Shinya Inoue, 492
820
INDEX TO VOLUME 165
Embryology, hydrozoan, 591
Embryonic shell formation, 394
Emersion, 582
Emission spectra, 791
Energetics, 197, 532, 778
Enhancement of the appearance of lateral projections
on negatively stained microtubules after glu-
taraldehyde — tannic acid fixation, 493
Energy budgets, 686, 699
Energy metabolism during air exposure and recovery
in the high intertidal bivalve mollusc Geiikensia
demissa granossisima and the subtidal bivalve
mollusc Modiolus squamosus, 708
Energy metabolism pathways of hydrothermal vent
animals: Adaptations to a food-rich and sulfide-
rich deep-sea environment, 167
Em amoeba, 534
Enteric bacteria, 522
Environment, 778
Enzyme activity levels, deep-living marine animals,
167
Epiphytes, 508
Epithelial conduction in colonial blood vessels, 209
Epizootic disease-complex of wild and laboratory-
maintained Hermissenda crassicornis. An, 535
Ernest Everett Just (1883-1941): a dedication, 487
Escherichia coli, 520
Estuarine crabs, see Rhithropanopeus
EVANS, ANN S., Growth and photosynthetic re-
sponses to temperature of two populations of
Zostera marina, 508
Evidence for regulation of protein synthesis at the
level of translational machinery in the sea urchin
egg, 503
Evolutionary constraint, 305
Excretion in bathypelagic mysid, 182
Exocytosis, 488, 520
Experimental studies on embryogenesis in hydro-
zoans (Trachylina and Siphonophora) with di-
rect development. 591
Eyestalk removal, 409
EYSTER, LINDA S., Infrastructure of early embryonic
shell formation in the opisthobranch gastropod
Aeolidia papillosa, 394
Factors affecting growth inhibition of enteric bacteria
by methyl «-D-glucoside, 522
FAIRFIELD, A. S., D. A. E. DOBBELAERE, AND M.
PERKINS, Characterization of the major surface
antigen of Plasmodium falciparum merozoites,
535
FAIRFIELD, A. S., see A. Flisser, 490
Fast axonal transport, 523, 527
Fast axonal transport is not affected by dimethyl
sulfoxide (DMSO) used to facilitate glycerina-
tion and/or glutaraldehyde fixation of squid ax-
ons, 527
FATH, KARL R., AND RAYMOND J. LASEK, Actin
microfilaments are a major cytoskeletal com-
ponent in squid axoplasm, 489
Fc receptor, 498
Feeding
and energetics of crab larvae, 197
in bathypelagic mysid, 182
squid, 637
Feeding structures, behavior, and microhabitat of
Echinocyamus pusillus (Echinoidea: Clypeas-
teroida), 745
FEIN, ALAN, see Richard Payne, 541
FELDMAN, SUSAN C, AND GEORGE D. PAPPAS,
Pathway tracing in the squid nervous system,
525
FENNELY, G. J., see J. Metuzals, 530
Fertilization, 503, 514
block, 5 1 3
currents, 5 1 7
in Chaetopterus, 1 1 0
Fertilization-induced ion conductances in frogs eggs,
517
Fertilization potential of eggs of the nemertean, Cer-
ebratulus. The, 516
Fiddler crab, 758
Fine structure of synapses and synaptosomes of the
squid (Loligo pealei) optic lobe, 525
FINGERMAN, MILTON, see Linda L. Vacca, 758
Fishes (Pisces), 505
Fissurella maxima, reproduction, 559
Flagellates, 241
Flight fuel utilization and flight energetics in the
migratory milkweed bug, Oncopeltus fasciatus,
532
FLISSER, A., A. S. FAIRFIELD, AND D. WIRTH, Char-
acterization of Trypanosoma brucei tubulin
genes, 490
FLISSER, A., see G. Ekapanyakul, 534, and L. D.
Sibley, 537
Fluoresceins, 524
Flourescence emission maxima, 241
FORWARD, RICHARD B., JR., AND KENNETH J.
LOHMANN, Control of egg hatching in the crab
Rhithropanopeus harrisii (Gould), 1 54
FORWARD, RICHARD B., JR., see Thomas W. Cronin,
139
FPL-55712, a leukotriene antagonist, promotes
polyspermy in sea urchins, 516
FREEMAN, GARY, Experimental studies on embryo-
genesis in hydrozoans (Trachylina and Siphon-
ophora) with direct development, 591
FREEMAN, JOHN A., TERRY L. WEST, AND JOHN
D. COSTLOW, Postlarval growth in juvenile
Rhithropanopeus harrisii, 409
Free radical kinetics, 496
Freeze-fracture, 517, 524
Freshwater, Cape Cod, 507
Frog egg fertilization, 517
FUJIWAKE HIDESHI, see Robert D. Allen, 523
Fully automated image analysis can be used to study
intramembranous particle (IMP) behavior dur-
ing development in Tetrahymena, 491
Fungus, lower marine, 429
Fusogenic protoplasts, 487
Functional and chemical characterization of squid
neurofilament polypeptides, 533
INDEX TO VOLUME 165
821
GABA receptors, presynaptic, 523
GAINER, HAROLD, see A. L. Obaid, 530, and Harish
C. Pant, 531
GALLANT, PAUL E., see Harish C. Pant, 531
Gametogenesis and reproductive periodicity of the
subtidal sea anemone Urticina lofotensis (Coe-
lenterata: Actiniaria) in California, 458
GART, SERGE, IZJA LEDERHENDLER, AND DANIEL
ALKON, An infrared macrophotographic tech-
nique for quantifying the behavioral response
to rotation of the gastropod Hermissenda cras-
sicornis, 525
GART, SERGE, see Izja Lenderhendler, 528
GASCOYNE, PETER, see Albert Szent-Gyorgyi, 496
Gastropods, 394, 543, 723
Gastrovascular cavity, 6 1 9
Genetics, 686, 699, 733
Genetic variation in the queen conch, Strombus gi-
gas, across its geographic range. Preliminary
results, 504
Geographic limits, 330
Geographic limits and local zonation: the barnacles
Semibalanus (Balanus) and Chthamalus in New
England, 330
GEORGE, ELIZABETH L., SARAH BRAY, ERIC T. Ro-
SENTHAL, AND TIM HUNT, A major maternally
encoded 41 K protein in both Spisula and Ar-
bacia binds to an anti-tubulin affinity column,
515
Giant smooth muscle cell, 49 1
Giant synapse pharmacology, 533
GILBERT, SUSAN P., see Robert D. Allen, 523
GLADFELTER, ELIZABETH H., Circulation of fluids
in the gastrovascular system of the reef coral
Acropora cervicornis, 619
GLADFELTER, ELIZABETH H., Spatial and temporal
patterns of mitosis in the cells of the axial polyp
of the reef coral Acropora cervicornis, 8 1 1
Glutaminergic agonists at the giant synapse, 533
Glutaraldehyde fixation, 527
Glycerinated axons, 527
Gnathophausia ingens, see mysid
GOLDMAN, ANNE E., see Karen M. Yokoo, 519
GOLDMAN, ROBERT D., see Karen M. Yokoo, 519,
and R. V. Zackroff, 533
Gorgonian regeneration, 778
Gossypol, 516, 517
GOULD, ROBERT M., MARTHA JACKSON, AND ICHUI
TASAKI, Phospholipid synthesis in the injected
squid giant axon, 526
GRASSLE, JUDITH P., see Kevin J. Eckelbarger, 379
and Gary E. Wagenbach, 538
Gray Reference Collection, 505
Grazing and predation as related to energy needs of
stage I zoeae of the tanner crab Chionoecetes
bairdi (Brachyura, Majidae), 197
GREER, K., see P. E. Kuwabara, 500
Growth, 507, 508
and latitude, 686
in corals, 8 1 1
in squid, 637
postlarval in crabs, 409
Growth and photosynthetic responses to temperature
of two populations of Zostera marina, 508
Growth and regeneration rates in thinly encrusting
demospongiae from temperate waters, 343
Growth responses of Zostera marina (eelgrass) to in
situ manipulations of sediment nitrogen avail-
ability, 507
GUIDITTA, ANTONIO, TIM HUNT, AND LUIGIA SAN-
TELLA, Messenger RNA in squid axoplasm, 526
H
HAGHIGHAT, NASRIN, see Rochelle S. Cohen, 525
HAINES, KATHLEEN A., see Gerald Weissmann, 503
HALL, LINDA M., A strategy to differentiate mutants
affecting voltage-sensitive sodium channels in
Drosophila, 490
HALL, ROBERT R., see George J. Skladany, 510, and
W. S. Vincent, 512
Halobacteria, 540
HALSELL, SUSAN, see Gary Lyons, 50 1
HALVORSON, H. O., A. KEYNAN, AND T. TIERNAN,
Studies on manganese oxidizing, spore forming
bacteria, 520
HAND, STEVEN C., AND GEORGE N. SOMERO, En-
ergy metabolism pathways of hydrothermaJ vent
animals: adaptations to a food-rich and sulfide-
rich deep-sea environment, 167
HANLON, ROGER T., RAYMOND F. HIXON, AND
WILLIAM H. HULET, Survival, growth, and be-
havior of the loliginid squids Loligo plei, Loligo
pealei, and Lolligiincula brevis (Mollusca: Ce-
phalopoda) in closed sea water systems, 637
HARN, DON, see G. Ekapanyakul, 534, L. D. Sibley,
537, and Linda Swiston, 538
HAROLD, ANTONY S., see Malcolm Telford, 745
Harpacticoid copepods, 507
HARRIGAN, JUNE, see Louis Leibovitz, 535, and
D. W. Pumplin, 531
HARTMAN, JEAN M., see Susan M. Merkel, 509
Heat shock proteins, 500
Hemocytes and tanning, 758
Hemolymph acidosis, 582
Henriksen model, 507
Hermaphroditism in nudibranch, 276
Hermissenda crassicornis, 276, 525, 528, 535
Hermissenda crassicornis: a disease complex. I. The
normal animal, 528
HERNANDEZ-NICAISE, MARI-LUZ, AND GHISLAIN
NICAISE, Structure of the isolated and in situ
giant smooth muscle fibers of Mnemiopsis ley-
dii, 491
High molecular weight (380Kd) ATPase in axoplasm
of squid giant axon, 494
HIGHSTEIN, STEPHEN M., AND ROBERT BAKER,
Physiological activity of efferent vestibular neu-
rons and their action on primary afferents in
the toadnsh, 527
HILL, ELIZABETH, see Frederick I. Tsuji, 444
HILL, W. D., see R. V. Zackroff, 533
822
INDEX TO VOLUME 165
HILLER-ADAMS, PAGE, AND JAMES J. CHILDRESS,
Effects of feeding, feeding history, and food de-
privation on respiration and excretion rates of
the bathypelagic mysid Gnathophausia ingens,
182
Histochemistry, 758
Histone
genes, 501
mRNAs, 518
synthesis, 500
HIXON, RAYMOND F., see Roger T. Hanlon, 637
HODGE, ALAN J., AND W. J. ADELMAN, JR., Fast
axonal transport is not affected by dimethyl
sulfoxide (DMSO) used to facilitate glycerina-
tion and/or glutaraldehyde fixation of squid ax-
ons, 527
HODGE, ALAN J., see W. J. Adelman, Jr., 523
Horseshoe crabs, 436
Host specificity of intestinal gregarines (Protozoa,
Apicomplexa) in two sympatric species of Cap-
itclla (Polychaeta), 538
HOTANI, H., AND J. L. TRAVIS, Opposite end as-
sembly-disassembly of single microtubules, 491
HOUGHTON, R. A., see T. A. Stone, 5 1 1
HOWARTH, ROBERT W., see Susan M. Merkel, 509
HOWLETT, SARAH, JOHN MILLER, AND GILBERT
SCHULTZ, Induction of heat shock proteins in
early embryos of Arbacia punctulata, 500
Hu, S. L., AND C. Y. KAO, The pH dependence of
the tetrodotoxin-blockade of sodium channels,
528
HUFNAGEL, LINDA A., Fully automated image anal-
ysis can be used to study intramembranous par-
ticle (IMP) behavior during development in
Tt'trahymena, 49 1
HULET, WILLIAM H., see Roger T. Hanlon, 637
HUMPHREYS, TOM, see Gerald Weissmann, 503
HUNT, TIM, see Sarah Bray, 499, Richard Cornall,
513, Elizabeth L. George, 515, Antonio Giu-
ditta, 526, and Thomas Kelly, 493
Hybrid horseshoe crab LDH and MDH, 436
Hybridoma, 534
Hydra, 305
Hydrogen sulfide, influence on metabolism, 167
Hydrothermal vent animals, energy metabolism, 167
Hydrozoan experimental embryology, 591
Hyperosmotic treatment inhibits cortical granule
exocytosis in the sea urchin Lvtechinus pictus,
520
I
Identified neurons, 531
IgE monoclonal antibodies produced from mice im-
munized with irradiated cercariae of Schisto-
soma mansoni. 534
Ilyanassa obsoleta, 536, 543
Image
analysis, 491
contrast, 492
intensification, 514, 522
Immunofluorescence of Allogromia reticulopodia,
489
Immunology, echinoderm, 473
INCZE, LEWIS S., AND A. J. PAUL, Grazing and pre-
dation as related to energy needs of stage I zoeae
of the tanner crab Chionoecetes bairdi (Brachy-
ura, Majidae), 197
Induced development of sweeper tentacles on the
reef coral Agaricia agaricites: a response to direct
competition, 569
Inducer exclusion, 520
Induction of heat shock proteins in early embryos
of Arbacia punctulata, 500
Infrared macrophotographic technique for quanti-
fying the behavioral response to rotation of the
gastropod Hemussenda crassicornis, An, 525
Inhibition of a surface binding monoclonal antibody
to schistosomula of Scliistosoma mansoni by
lectins, 538
Inhibition of mitotic anaphase and cytokinesis and
reduction of spindle birefringence following
microinjection of anti-calcium transport en-
zyme IgGs into Echinaracnius parma blasto-
meres, 495
Injury, 778
Initial calcification process in shell-regenerating 7V-
gula (Archaeogastropoda), The, 265
INOUE, SHINYA, THEODORE D. INOUE, AND GOR-
DON W. ELLIS, Visualizing extremely low con-
trast images by digital enhancement of selected
portions of the image grey scale, 492
INOUE, THEODORE, see Shinya Inoue, 492
In situ
emersion of C. productus, 582
hybridization, 515
Interactions of harpacticoid copepods and photo-
synthetic microbes in the salt marsh, 507
Intermediate filaments, 519, 533
Intertidal snails, 543
Intracellular fusion between reticulopodial networks
in Allognnnia laticollari.s. 497
Intramembranous particles, 491
Intraparasitophorous vacuolar pH of Leishmania
mexicana infected macrophages, 536
Intrasexual aggression in Metridium senile, 416
Introduced species, 543
Invertebrate
ecology, 221
immunology, 473
pathology, 429, 528
In vitro transcription of histone genes in isolated
nuclei from S purpuratus, 518
Ion conductances — fertilization, 517
Iron accumulation in tunicate blood cells. I. Distri-
bution and oxidation state of iron in the blood
of Boltenia ovifera, Styela clava, and Molgula
mahnattcnsis, 1 00
Iron and phosphorus cycling in a permanently strat-
ified coastal pond, 506
Isoenzymes, 241
Isolation and characterization of tubulin clones from
Dictyostelium discoidium, 488
INDEX TO VOLUME 165
823
Isolation of cytoskeletons from Chaetopterus eggs,
514
Is there a developmental significance for mRNA lo-
calized in the cortex of Chaetopterus eggs? 5 1 8
Is there specificity in the induction of polyspermy
in sea urchins by protease inhibitors? 512
E. S. WEINBERG, Changes in histone synthesis
during Arhacia development, 500
KUZIRIAN, ALAN M., Louis LEIBOVITZ, AND DAN-
IEL L. ALKON, Hermissenda crassicornis: a dis-
ease complex. I. The normal animal, 528
KUZIRIAN, ALAN, see Louis Leibovitz, 535
JACKLET, JON W., Current clamp of photoreceptors
and pacemaker neurons in eye of Bulla, 540
JACKSON, MARTHA, see Robert M. Gould, 526
JAFFE, LAURINDA A., see Douglas Kline, 516 and
Lyanne C. Schlichter, 517
JEFFERY, WILLIAM R., An organelle complex re-
sponsible for mRNA localization in the cortex
of Chaetopterus eggs, 5 1 5
JEFFERY, WILLIAM R., Ernest Everett Just (1883-
1941): a dedication, 487
JEFFERY, WILLIAM R., see Billie J. Swalla, 518
JOSEPH-SlLVERSTEIN, JACQUELYN, AND WILLIAM
D. COHEN, Composition and function of the
cytoskeleton in "blood clam" erythrocytes, 492
JOSEPH-SILVERSTEIN, JACQUELYN, see William D.
Cohen, 489
K
KAO, C. Y., see S. L. Hu, 528
KAPLAN, SAUL W., Intrasexual aggression in Me-
tndium senile, 416
Karyology of Teredo utriculus (Gmelin) (Mollusca,
Pelecypoda), The, 450
KASS, LEONARD, JANICE, L. PELLETIER, GEORGE
H. RENNINGER, AND ROBERT B. BARLOW, JR.,
cAMP: a possible intracellular transmitter of
circadian rhythms in Limulus photoreceptors,
540
KASS, LEONARD, see Robert B. Barlow, Jr., 539, and
Jennifer J. Marler, 541
K+ channels, 524
KELLY, THOMAS, JOEL L. ROSENBAUM, AND TIM
HUNT, Two-dimensional gel analysis of sea ur-
chin ciliary tubulins, 493
KEMPF, STEPHEN C., see Louise R. Bickell, 1 19
KEYNAN, A., see H. O. Halvorson, 520
KJEHART, D. P., see A. Eisen, 514
Kinetoplast DNA, 537
KING, KENNETH R., see Richard L. Miller, 419
KLINE, DOUGLAS, AND LAURINDA A. JAFFE, The
fertilization potential of eggs of the nemertean,
Cerebratulus, 516
Ko, A., see G. Ekapanyakul, 534, and Linda Swiston,
538
KOIDE, S. S., see Eimei Sato, 516, and S. J. Segal,
517
KORNBERG, H. L., see E. B. Ackerman, 520, and
D. F. Sutherland, 522
KRAKOW, J., see L. D. Sibley, 537
KUSTIN, KENNETH, see Maria I. Agudelo, 100
KUWABARA, P. E., K. GREER, S. MAEKAWA, AND
Laboratory maintenance, 637
Lactate dehydrogenase of hybrid horseshoe crab em-
bryos, 436
Lactoperoxidase-tubulin interaction in ciliary mem-
branes, 496
Laguna Figueroa, 241
LANDFEAR, SCOTT, see Claire Wyman, 539
LANDFEAR, SCOTT, Structure and expression of tu-
bulin genes in the protozoan parasite Leish-
mania enriettii, 493
LANGFORD, GEORGE M., Enhancement of the ap-
pearance of lateral projections on negatively
stained microtubules after glutaraldehyde —
tannic acid fixation, 493
LANGFORD, GEORGE M., see William R. Eckberg,
514
Larvaceans, 419
Larval and metamorphic morphogenesis in the nu-
dibranch Melibe leonina (Mollusca: Opistho-
branchia), 1 19
Larval crabs, 139
Larval ecology, 221
Larval settlement and metamorphosis otAlcyonium,
286
LASEK, RAYMOND J., see Karl R. Path, 489
Lateral eye, 541
Latitude, 686, 699
Latitude and intraspecific growth, 699
Latitudinal compensation hypothesis: growth data
and a model of latitudinal growth differentiation
based upon energy budgets. I. Interspecific
comparison ofOphryotrocha (Polychaeta: Dor-
villeidae). The, 686
Latitudinal compensation hypothesis: growth data
and a model of latitudinal growth differentiation
based upon energy budgets. II. Intraspecific
comparisons between subspecies of Ophryotro-
chapuerilis (Polychaeta: Dorvilleidae), The, 699
LATZ, MICHAEL I., see Edith A. Widder, 791
Lectin, 525, 538
LEIBOVITZ, Louis, ALAN KUZIRIAN, JUNE HAR-
RIGAN, EDWARD F. SCHOTT, IZJA LEDER-
HENDLER, AND DANIEL L. ALKON, An epizootic
disease-complex of wild and laboratory-main-
tained Hermissenda crassicornis, 535
LEIBOVITZ, Louis, see Alan M. Kuzirian, 528
Leishmania. 534, 536, 539
identification, 537
tubulin genes, 493
LENDERHENDLER, IZJA, see Serge Gart, 525
LENDERHENDLER, IZJA, SERGE GART, AND DANIEL
L. ALKON, Associative learning in Hermissenda
crassicornis (Gastropoda): evidence that light
824
INDEX TO VOLUME 165
(the CS) takes on characteristics of rotation (the
UCS), 528
Lens, 499
Leucine aminopeptidase, 241
Leukotriene, 503, 516
Leukotriene B4 promotes the calcium-dependent ag-
gregation of marine sponge cells, 503
LEVIN, JACK, see Peter B. Armstrong, 488
LEVINTON, JEFFERY S., AND ROSEMARY K. MON-
AHAN, The latitudinal compensation hypothesis:
growth data and a model of latitudinal growth
differentiation based upon energy budgets. II.
Intraspecific comparisons between subspecies
of Ophryotrocha puerilis (Polychaeta: Dorvil-
leidae), 699
LEVINTON, JEFFREY S., The latitudinal compensa-
tion hypothesis: growth data and a model of
latitudinal growth differentiation based upon
energy budgets. I. Interspecific comparison of
Ophryotrocha (Polychaeta: Dorvilleidae), 686
LEWENSTEIN, LISA A., Propagating calcium spikes
in identified cells in the supraesophageal gan-
glion of the giant barnacle, Balanus nubilus,
529
LIEBMAN, MATTHEW, Trematode infection in lly-
anassa obsoleta: dependence on size and sex of
the host and effect on chemotaxis, 536
Life history, Octocorallia, 353
Limulus, 488, 495, 539
photoreceptors, 540, 541
Littorina littorea, 543
LLINAS, R., M. SUGIMORI, AND J. M. BOWER, Vi-
sualization of depolarization-evoked presynaptic
calcium entry and voltage dependence of trans-
mitter release in squid giant synapse, 529
Lobster, 532
Localization of calcium transients in the presynaptic
terminals of a barnacle photoreceptor detected
using Arsenazo III, 542
LOHMANN, KENNETH J., see Richard B. Forward,
Jr., 154
Loligo, 637
Lower marine fungus associated with black line dis-
ease in star corals (Montastrea annularis, E. &
S.), 429
LOYA, Y., see. Y. Benayahu, 353
Lymphocyte mitogen, 536
Lyngbya aestuarii. 521
LYONS, GARY, SUSAN HALSELL, AND ROB MAXSON,
Accumulation of late H2b histone mRNA in
sea urchin embryogenesis, 501
Lysosome, 534
Lysosomal pH, 536
M
MACKIE, G. O., AND C. L. SINGLA, Coordination
of compound ascidians by epithelial conduction
in the colonial blood vessels, 209
Macrophage, 536
r^AEKAWA, S., see P. E. Kuwabara, 500
MAJORCA, A., see. R. Vitturi, 450
Major maternally encoded 4 1 K protein in both Spi-
sula andArbacia binds to an anti-tubulin affinity
column. A, 515
Malaria, 536
Malate dehydrogenase of hybrid horseshoe crab em-
bryos, 436
Mammalian nerve terminals, 530
MANCINI, VIVIAN, see Robert B. Barlow, Jr., 539
Manganese-oxidizing bacteria, 24 1
MAPS (microtubule-associated proteins), 492
Marginal band function, 492
Marginal band function in the dogfish erythrocyte,
489
MARGULIS, LYNN, see Laurie K. Read, 241
Marine bioluminescence spectra measured with an
optical multichannel detection system, 791
Marine molluscan genomes contain sequences ho-
mologous to the octopine synthase gene of
Agrobacterium tumefaciens, 498
MARQUIS, SALLY, see Jennifer Dungan, 507
MARLER, JENNIFER J., ROBERT B. BARLOW, JR.,
LESLIE EISELE, AND LEONARD KASS, Photo-
receptors add at tht anterior edge of Limulus
lateral eye, 541
Maternal mRNA, 499
Mating and egg mass production in the aeolid nu-
dibranch Hermissenda crassicornis (Gastro-
poda: Opisthobranchia), 276
MATSUO, N., see Eimei Sato, 516
MATTESON, D. R., Voltage clamp studies of dispersed
toadfish pancreatic islet cells, 494
Maturation of sea urchin and Chaetopterus oocytes
results in a change in the pattern of protein
synthesis, 519
MAXSON, ROB, see Gary Lyons, 501
MAZZELLA, L., W. C. DENNISON, AND R. S. AL-
BERTE, Photosynthetic activity of Zostera ma-
rina L. epiphytes in relation to light regime and
substratum, 508
MCLAUGHLIN, JANE, see Albert Szent-Gyorgyi, 496
McLEOD, GUY C., see Maria I. Agudelo, 100
McMAHON, BRIAN R., see Peter L. deFur, 582
Mechanical stimulation of bioluminescence in dilute
suspensions of dinoflagellates, 522
Mechanosensitivity, 496
Meiobenthos, 507
Melampus bidentatus, 510, 511
Melanin, 758
Melibe leonina, 1 1 9
MELILLO, J. M., see T. A. Stone, 51 1
Memory, 473
Membrane potential, 540
Membranes, 496
Merccnaria mercenaria, 509
MERKEL, SUSAN M., JEAN M. HARTMAN, AND
ROBERT W. HOWARTH, Sulfate reduction fol-
lowing marsh grass die-back, 509
Merozoite surface glycoprotein, 535
Messenger RNA in squid axoplasm, 526
Metabolism of a bathypelagic mysid, 182
Metamorphosis
of Melibe. 119
of soft-coral larvae, 286
INDEX TO VOLUME 165
825
Methanogenesis, 505
Metridium, intrasexual aggression, 416
METUZALS, J., D. F. CLAPIN, P. A. M. EAGLES, AND
G. J. FENNELY, Ordered assemblies of neuro-
filament proteins isolated from squid giant axon,
530
Microbial selection in an artificial ecosystem, 512
Microhabitat of Echinocyamus, 745
Microscope, 492
Microtubule, 491, 493, 494
MILLER, JOHN, see Sarah Hewlett, 500
MILLER, RICHARD L., AND KENNETH R. KING,
Sperm chemotaxis in Oikopleura dioica Fol,
1872 (Urochordata: Larvacea), 419
MILLS, SUSAN W., see Gary E. Wagenbach, 538
MITCHELL, TRICIA A., AND COLLEEN M. CAVA-
NAUGH, Numbers of symbiotic bacteria in the
gill tissue of the bivalve Solemva velum Say,
521
Mitochondria! and spherosomal movement along a
filamentous network in the marine slime mold
Gymnophrydium marinum. 498
Mitogenic activity of extracts and supernates from
Plasmodium falcipamm. 536
Mitosis, 495, 811
Mitotic index in coral tissues, 81 1
MITTON, JEFFRY, B., see Carl J. Berg, Jr., 504
MIYAMOTO, DAVID M., A video time lapse study
of cell behavior during notochord morphogen-
esis in ascidian embryos, 501
Model system, 510
Molgula manhattensis. see tunicates
Mollusc reproduction, 559
Molting cycle, 409, 758
Molting in Crustacea, 32 1
Monoclonal antibody, 535, 538
Mooi, RICH, see Malcolm Telford, 745
MOON, RANDALL T., see Billie J. Swalla, 518
Morphogenesis in a nudibranch (Melibe), 119
Morphology and genetics of rejection reactions be-
tween oozooids from the tunicate Botryllus
schlosseri. 733
MORROW, LAURA L., see Mary Ann Rankin, 532
MOSS, R., R. SCHUEL, AND H. SCHUEL, FPL-557 1 2,
a leukotriene antagonist, promotes polyspermy
in sea urchins, 516
Motility, 523
MUEHLSTEIN, LISA, AND RICHARD W. CASTEN-
HOLZ, Sheath pigment formation in a blue-green
alga, Lvngbva aestuarii, as an adaptation to
high light, 52 1
Multiple fission, 241
Mutants of Escherichia coll affected in "inducer ex-
clusion," 520
Mysid, bathypelagic, 182
N
NACHTIGALL, MARGARET, see Frederic Blum, 513
NAGASHIMA, LAUREN S., see Virginia L. Scofield,
733
NAKAMURA, SHOGO, see Sidney L. Tamm, 497
Nerve conduction, 526
Neurofilament, 530, 531, 533
Neurogenesis in Melibe. 1 19
Neuronal cytoskeleton, 533
Neurotubules, 493
New strain of Paraletramitus jugosus from Laguna
Figueroa, Baja California, Mexico, A, 241
NICAISE, GHISLAIN, see Mari-Luz Hernandez-Ni-
caise, 491
NICCHITTA, C. V., AND W. R. ELLINGTON, Energy
metabolism during air exposure and recovery
in the high intertidal bivalve mollusc Geukensia
demissa granossisima and the subtidal bivalve
mollusc Modiolus sqiiamosus, 708
Noradrenalin, 758
NOLL, KENNETH M., AND RICHARD W. CASTEN-
HOLZ, Diel vertical movements of bacteria in
intertidal streams of Sippewissett Marsh, 509
Notochord, 501
Nudibranchia, 119, 276, 528, 535
Numbers of symbiotic bacteria in the gill tissue of
the bivalve Solemya velum Say, 521
o
OBAID, A. L., H. GAINER, AND B. M. SALZBERG,
Optical recording of action potentials from
mammalian nerve terminals in situ. 530
OBAR, ROBERT, see Laurie K. Read, 241
Octocoral, 353, 778
Octopine synthase, 498
Ocular lens aging in the skate, 499
Odontosyllis, 444
Oikopleura. 419
OLSON, RICHARD RANDOLPH, Ascidian-Prochloron
symbiosis: the role of larval photoadaptations
in midday larval release and settlement, 221
O'MELIA, ANNE F., Rates of 5S RNA and tRNA
synthesis in sea urchin embryos animalized by
Evans Blue, 502
Ommatidia, 541
Oncopeltus Jasciatus, 532
On the evolutionary constraint surface of hydra, 305
Oocytes, 379
Oogenesis, 379
Ooplasmic segregation, 379
Opposite end assembly-disassembly of single micro-
tubules, 491
Opisthobranchia, 1 19
Optical recording of action potentials from mam-
malian nerve terminals in situ, 530
Optic lobe, 525
Ordered assemblies of neurofilament proteins isolated
from squid giant axon, 530
Organelle complex responsible for mRNA localiza-
tion in the cortex of Chaetopterus eggs, An, 5 1 5
Organelle transport, 498
ORR, KATHERINE S., see Carl J. Berg, Jr., 504
Osmotic, 520
826
INDEX TO VOLUME 165
Ovarian
ecdysteroids, 321
follicle cells, 379
PALASZEWSKI, PAUL P., AND BARBARA C. BOYER,
Reproduction in Haploplana and Stylochus:
developmental and cytoskeletal research pos-
sibilities, 502
Pancreatic islet cells, 494
PANT, HARISH C., PAUL E. GALLANT, ROCHELLE
S. COHEN, AND HAROLD GAINER, A relatively
stable lOOKd protein is derived from the Ca2+-
dependent proteolysis of neurofilament proteins
in the squid axoplasm, 531
PAPPAS, GEORGE D., see Rochelle S. Cohen, 525,
and Susan C. Feldman, 525
Parasite antigens, 537
Parasites, 538
Parasitism, 536
Paratetramitiis jugosus, 241
Paternal forms of LDH and MDH, 436
Particle velocity distribution, 523
Pathway tracing in the squid nervous system, 525
PAUL, A. J., see Lewis S. Incze, 197
PAXHIA, TERESA, see Seymour Zigman, 499
PAYNE, RICHARD, AND ALAN FEIN, The effects of
intracellular calcium/EGTA on the photoac-
tivation of Limulus ventral photoreceptors, 541
PELLETIER, JANICE L., see Robert B. Barlow, Jr.,
539, and Leonard Kass, 540
PEP-phosphotransferase system, 522
PERCY, A., Mitogenic activity of extracts and su-
pernates from Plasmodium falciparum, 536
PERKJNS M., see A. S. Fairfield, 535
PETERSON, BRUCE, see Jennifer Dungan, 507
PETHIG. RONALD, see Albert Szent-Gyorgyi, 496
Phagocytosis and intralysosomal killing of Leish-
mania mexicana by Entamoeba histolytica, 534
pH dependence of the tetrodotoxin-blockade of so-
dium channels. The, 528
Phenolic acid, 510
Phosphate removal, 510
Phospholipid synthesis in the injected squid giant
axon, 526
Phosphorous cycling, 506
Photoactivation kinetics, 541
Photoadaptation, 221
Photographic (infrared analysis), 525
Photophore, 791
Photoreceptors, 540, 542
Photoreceptors add at the anterior edge of Limulus
lateral eye, 541
Photosensory excitation, 540
Photosynthesis, 508
Photosynthetic activity of Zostera marina L. epi-
phytes in relation to light regime and substra-
tum, 508
Photosynthetic microbes, 507
Physiological activity of efferent vestibular neurons
and their action on primary afferents in the
toadfish, 527
Pigment, 521
Plasmodium falciparum. 535
Plasmodium mitogen, 536
POHL, CHARLES, see Jay Shiro Tashiro, 5 1 1
Pollination, submarine, 504
Polychaeta, 379, 686, 699
Polyclads, 502
Polyspermic fertilization, 513
Polyspermy, 512, 516
Population crosses, 506
Population dynamics, 538
Population ecology of the Caribbean bivalve Asaphis
deflorata (Linne, 1758), 504
Postlarval growth in juvenile Rhithropanopeus har-
risii. 409
Posttraslational modification, 493
PRATT, M. M., High molecular weight (380Kd)
ATPase in axoplasm of squid giant axon, 494
Preliminary evidence indicating the existance of in-
termediate filament-like proteins in unfertilized
eggs of the surf clam, Spisula solidissima, 519
Presynaptic action of baclofen, a GABA analog, at
the crayfish neuromuscular juction, 523
Prochloron, symbiosis with Ascidian, 221
Promitosis, 241
Propagating calcium spikes in identified cells in the
supraesophageal ganglion of the giant barnacle,
Balanus nubi/us, 529
Propionyl esterase, 241
Protease inhibitors, 5 1 2
Protein
microtubule-associated, 493
synthesis, 503, 513, 519
Proteinase inhibitor, 488, 495
Proteinase inhibitor released from the Limulus ame-
bocyte during exocytosis. A, 488
Pseudostereoscopy allows direct visualization of the
velocity distribution of particles undergoing fast
axonal transport, 523
PUMPLIN, D. W., AND J. HARRIGAN, Some mor-
phological observations on the giant synapse of
immature squid, 531
QUIGLEY, JAMES P. AND PETER B. ARMSTRONG,
Characterization and isolation of a homologue
of alpha-2-macroglobulin from the plasma of
the horseshoe crab Limulus, 495
QUIGLEY, JAMES P., see Peter B. Armstrong, 488
R
RAMOS-FLORES, TALIA, Lower marine fungus as-
sociated with black line disease in star corals
(Montastrea annularis, E. & S.), 429
RANKIN, MARY ANN, AND LAURA L. MORROW,
Right fuel utilization and flight energetics in
INDEX TO VOLUME 165
827
the migratory milkweed bug, Oncopeltus fas-
ciatus, 532
Rates of 5S RNA and tRNA synthesis in sea urchin
embryos animalized by Evans Blue, 502
READ, LAURIE K., LYNN MARGULIS, JOHN STOLZ,
ROBERT OBAR, AND THOMAS K. SAWYER, A
new strain of Paratetramitus jugosus from La-
guna Figueroa, Baja California, Mexico, 241
Receptor aggregation, 498
Recognition, 502, 569
Red Sea, soft coral, 353
Reef coral, 619
REED-MILLER, CHARLENE, The initial calcification
process in shell-regenerating Tegida (Archaeo-
gastropoda), 265
REED-MILLER, CHARLENE, Scanning electron mi-
croscopy of the regenerated shell of the marine
archaeogastropod, Tegula. 723
REESE, TOM S., see Donald C. Chang, 524
Regenerated Tegula shell, 723
Regeneration of injuries among Jamaican gorgoni-
ans: the roles of colony physiology and envi-
ronment, 778
Regeneration, sponge, 343
Rejection reaction, 733
Relatively stable lOOKj protein is derived from the
Ca2+-dependent proteolysis of neurofilament
proteins in the squid axoplasm. A, 531
Remote sensing, 5 1 1
RENNINGER, GEORGE H., see Leonard Kass, 540
Repetitive cycles of bioluminescence and spawning
in the polychaete, Odontosyllis phosphorea, 444
Reproduction, 458, 502, 559
Reproduction in Haploplana and Stylochus: devel-
opmental and cytoskeletal research possibilities,
502
Reproductive behavior in nudibranch, 276
Reproductive periodicity, 458
Reproductive phase, 778
Respiration in bathypelagic mysid, 182
Restriction mapping, 490
Reticulopodial network, 497
Retinal growth, 541
REVELAS, EUGENE C., Vertical movements of the
hard clam, Mercenaria mercenaria. in response
to changes in barometric pressure, 509
REYNOLDS, G. T., AND ALAN J. WALTON, Me-
chanical stimulation of bioluminescence in di-
lute suspensions of dinoflagellates, 522
REYNOLDS, G. T., see A. Eisen, 514
Rhithropanopeus, 1 39, 1 54, 409
Rhythms in Rhithropanopeus, \ 39, 1 54
RICE, ROBERT V., see Stanley W. Watson, 498
RICH, ABBY M., see Gerald Weissmann, 503
RIETSMA, CAROL S., Effect of age and quality of
detritus on growth of the salt marsh snail, Me-
lampus bidentatus, 510
Rise of free intracellular Ca2+ in mouse macrophage
associated with -y2b/-yl Fc receptor-ligand in-
teraction, 498
RIVAS, L., AND K.-P. CHANG, Intraparasitophorous
vacuolar pH of Leishmania mexicana infected
macrophages, 536
RNA
accumulation, 501
mRNA, 515, 518
rRNA gene cluster, 488
ROBINSON, WILLIAM E., see Maria I. Agudelo. 100
Rock intertidal, 330
Role of freshwater wetlands in the ontogeny of a
New England saltmarsh. The, 512
Roles of hemocytes in tanning during the molting
cycle: a histochemical study of the fiddler crab,
Uca pugilator. The, 758
ROMERO, GIULLERMO, YARA CSEK.O, AND DYANN
WIRTH, Detection of Leishmania kinetoplast
DNA using biotinylated DNA probes, 537
ROSENBAUM, JOEL L., see Thomas Kelly, 493
ROSENTHAL, ERIC T., see Elizabeth L. George, 515
Ross, WILLIAM N., AND N. STOCKBRIDGE, Local-
ization of calcium transients in the presynaptic
terminals of a barnacle photoreceptor detected
using Arsenazo III, 542
RUTOWSKI, RONALD L., Mating and egg mass pro-
duction in the aeolid nudibranch Hermissenda
crassicornis (Gastropoda: Opisthobranchia), 276
Salt marsh. 512, 543
Salt-tolerant amoebae, 241
SALZBERG, B. M., see A. L. Obaid, 530
SANTELLA, LUIGIA, see Antonio Giuditta, 526
Satellite fusion, 497
SATO, EIMEI, N. MATSUO, M. H. BURGOS, S. S.
KOIDE, AND S. J. SEGAL, Binding of 14C-gos-
sypol by Arbacia sperm, 516
SAWYER, THOMAS K., see Laurie K. Read, 241
Scanning electron microscopy of the regenerated shell
of the marine archaeogastropod, Tegula. 723
SCHEN, CATHY R., see Barbara E. Ehrlich. 540
Schistosoma mansoni, 534, 537, 538
SCHOTT, EDWARD F., see Louis Leibovitz, 535
SCHLICHTER, LYANNE C., AND LAURJNDA A. JAFFE,
Fertilization-induced ion conductances in frog
eggs, 517
SCHNAPP, BRUCE J., see Stanley W. Watson, 498
SCHNEIDER, E. GAYLE, Cell-cell recognition and
adhesion during embrvogenesis in the sea ur-
chin, 502
SCHUEL, H., see M. C. Alliegro, 512, and R. Moss,
516
SCHUEL, R., see R. Moss, 516
SCHULTZ, GILBERT, see Sarah Hewlett, 500
Scleractinian corals, 569
SCOFIELD, VIRGINIA L., AND LAUREN S. NAGASH-
IMA, Morphology and genetics of rejection re-
actions between oozooids from the tunicate Bo-
tryllus schlosseri, 733
Sea anemone, see urticina
Sea urchin, 502, 519
ciliary tubulin, 493
828
INDEX TO VOLUME 165
eggs, 503, 520
embryo, 502
fertilization, 512, 516
SEBENS, KENNETH, P., Settlement and metamor-
phosis of a temperate soft-coral larva (Alcyonium
siderium Verrill): induction by crustose algae,
286
Sediment nitrogen availability, 507
SEGAL, S. J., M. BURGOS, AND S. S. KOIDE, Ultra-
structural changes characteristic of Arbacia
sperm exposed to gossypol, 5 1 7
SEGAL, S. J., see Eimei Sato, 516
SEKIGUCHI, KIOCHI, see Hiroaki Sugita, 436
Semibalanus, 330
Sensitivity of freshwaters of Cape Cod, Massachusetts
to acid precipitation. The, 507
Settlement and metamorphosis of a temperate soft-
coral larva (Alcyonium siderium Verrill): in-
duction by crustose algae, 286
Sexual dimorphism and reproductive behavior in
Almyracuma proximolculi (Crustacea: Cuma-
cea): the effect of habitat, 370
Sheath pigment formation in a blue-green alga,
Lvngbva aestuarii, as an adaptation to high light,
521
Shell gland, 394
Shell regeneration in Tegula, 265
Shell ultrastructure, 723
SHETTkES, BREWER, see Matthew Winkler, 503
SHUPE, KATHLEEN, AND ERIC WEINBERG, In vitro
transcription of histone genes in isolated nuclei
from 5. purpuratus, 518
SIBLEY, L. D., J. KRAKOW, A. FLISSER, AND D.
HARN, Antigens on both mechanical and lung
stage schistosomula of Schistosoma mansoni
are masked by host molecules, 537
Sibling species, Capilella, 379
SILVER, ROBERT B., Inhibition of mitotic anaphase
and cytokinesis and reduction of spindle bire-
fringence following microinjection of anti-cal-
cium transport enzyme IgGs into Echinaracnius
parma blastomeres, 495
SINGLA, C. L., see G. O. Mackie, 209
Single amino acids stimulate lobster (Homarns
americanus) behavior against ambient and
modified amino acid backgrounds, 532
Siphonphora, 591
Skate, 499
SKINNER, DOROTHY M., see Cynthia Soumoff, 321
SKLADANY, GEORGE J., BRIAN A. WRENN, AND
ROBERT R. HALL, The design and construction
of a benchtop reactor to model and anaerobic/
oxic wastewater treatment system, 510
Slime mold, 498
SLOBODKIN, L. B., AND KENNETH DUNN, On the
evolutionary constraint surface of hydra, 305
Slow rearrangements of membrane bound, haloge-
nated fluoresceins produce altered K+ currents
in squid axon, 524
SMITH, STEPHEN J., GEORGE J. AUGUSTINE, AND
MILTON P. CHARLTON, Calcium transients in
voltage clamped presynaptic terminals, 532
Snails, see Tegula
Sodium channel, 490, 528
Some morphological observations on the giant syn-
apse of immature squid, 531
SOMERO, GEORGE N., see Steven C. Hand, 167
SOUMOFF, CYNTHIA, AND DOROTHY M. SKJNNER,
Ecdysteroid tilers during the molt cycle of the
blue crab resemble those of other Crustacea,
321
SPALDING, MARILYN, AND JELLA ATEMA, Single
amino acids stimulate lobster (Homarus amer-
icanus) behavior against ambient and modified
amino acid backgrounds, 532
Spartina altemqflora, 509
Spatial and temporal patterns of mitosis in the cells
of the axial polyp of the reef coral Acropora
cervicornis. 8 1 1
Spawning cycles in Odontosyllis, 444
Spectra, 791
Speciation in the brine shrimp A rtemia. cross-breed-
ing between sexual Mediterranean populations,
506
Spectroscopy, 791
Sperm chemotaxis in Oikopleura dioica Fol, 1872
(Urochordata: Larvacea), 419
Sperm motility, 516
Spisula
early development, 515
oocyte, 5 1 9
Sponge cell aggregation, 503
Sponge growth and regeneration, 343
SPUDICH, JOHN L., see Barbara E. Ehrlich, 540
Squid
axon, 523, 524, 526
axoplasm, 494, 526, 531
behavior, 637
development, 531
growth, 637
maintenance, 637
nervous system, 525
synapse, 529
5S RNA and tRNA synthesis, 502
STANLEY, E. F., Depolarizing and desensitizing ac-
tions of glutaminergic and cholinergic agonists
at the squid giant synapse, 533
Starvation, effects of in bathypelagic mysid, 1 82
Stellate ganglion, 531
STEPHENS, R. E., Lactoperoxidase-tubulin interac-
tion in ciliary membranes, 496
STOCKBRIDGE, N., see William N. Ross, 542
STOLZ, JOHN, see Laurie K. Read, 241
STOMMEL, ELIJAH W., Calcium activated channels
in the mechanically sensitive abfrontal ciliated
cells of Mytilus gill, 496
STONE, T. A., R. A. HOUGHTON, J. M. MELILLO,
AND G. M. WOODWELL, Deforestation in the
Amazon Basin measured by satellite: a release
of CO2 to the atmosphere, 5 1 1
INDEX TO VOLUME 165
829
Strategy to differentiate mutants affecting voltage-
sensitive sodium channels in Drosophila, A, 490
Strombus gigas, 504
Strongylocentrotus purpuratus, 473, 501, 518
Structure and expression of tubulin genes in the pro-
tozoan parasite Leishmania enriettii, 493
Structure of the isolated and in situ giant smooth
muscle fibers of Mnemiopsis leydii, 491
Structure of the squid axon membrane as seen after
freeze-fracture, 524
Structure of tubulin RNA from Leishmania enriettii,
539
Studies of ctyotoxic free radicals produced by some
methoxy-quinones plus ascorbate in the pres-
ence of Ehrlich ascites cells, 496
Studies on manganese oxidizing, spore forming bac-
teria, 520
Studies of the isolation and calcium-induced fusion
of fusogenic wild carrot protoplasts, 487
Styela clava, see tunicates
Subtidal sponge growth, 343
SUGITA, HIROAKI, AND KoiCHi SEKIGUCHI, The de-
velopmental appearance of paternal forms of
lactate dehydrogenase and malate dehydroge-
nase in hybrid horseshoe crabs, 436
SUGIMORI, M., see R. Llinas, 529
Sulfate reduction, 505, 509
Sulfate reduction following marsh grass die-back,
509
Sulfur-oxidizing chemoautotrophic bacteria, 521
Superoxide dismutase biomimetic compounds pre-
vent fertilization in Arbacia punctulala eggs,
513
Surface brooding in the Red Sea soft coral Parery-
thropodiumfulvumfulvum(FoTska\, 1775), 353
Survival, growth, and behavior of the loliginid squids
Loligo plei. Loligo pealei, and Lolliguncula
brevis (Mollusca: Cephalopoda) in closed sea
water systems, 637
SUTHERLAND, D. F., AND H. L. KORNBERG, Factors
affecting growth inhibition of enteric bacteria
by methyl a-D-glucoside, 522
SWALLA, BlLLIE J., RANDALL T. MOON, AND WlL-
LIAM R. JEFFERY, Is there a developmental sig-
nificance for mRNA localized in the cortex of
Chaetopterus eggs? 5 1 8
Swarming in polychaetes, 444
Sweeper tentacle, 569
SWISTON, LINDA, ALBERT Ko, AND DON HARN,
Inhibition of a surface binding monoclonal an-
tibody to schistosomula of Schistosoma man-
soni by lectins, 538
Symbiosis, Ascidian-Proc/z/o/wz, 221
Synapses, 525
Synaptic transmission, 532
Synaptosomes, 525
SZENT-GYORGYl, ALBERT, PETER GASCOYNE,
RONALD PETHIG, AND JANE MCLAUGHLIN,
Studies of ctyotoxic free radicals produced by
some methoxy-quinones plus ascorbate in the
presence of Ehrlich ascites cells, 496
Tachypleus, see horseshoe crabs
TAMM, SIDNEY L., AND SHOGO NAKAMURA, ATP-
reactivated models of ctenophore comb plates,
497
Tanner crab, energy needs of zoeae, 197
Tanning, hormone, 758
TASAKJ, ICHLJI, see Donald C. Chang, 524 and Robert
M. Gould, 526
TASHIRO, JAY SHIRO, MARK WILTSHIRE, AND
CHARLES POHL, Age of first reproduction in
Melampus bidentatus: the effects of overwin-
tering degrowth and repair, 5 1 1
Tegula. 265, 723
TELFORD, MALCOLM, ANTONY S. HAROLD, AND
RICH Mooi, Feeding structures, behavior, and
microhabitat of Echinocyamus pusillus (Echi-
noidea: Clypeasteroida), 745
Temperature, 508
Teredo, karyology, 450
Terminal anecdysis, 321
TESORIERI, ITALO, see Marta Bretos, 559
Tetrahvmena, 491
Tetrodotoxin, 490, 528
TIERNAN, T., see H. O. Halvorson, 520
Ti plasmid, 498
Toadfish, 494
Trachylina, 591
Transcriptional regulation, 518
Translational control, 515
Transport of vesicles along filaments dissociated from
squid axoplasm, 523
TRAVIS, J. L., AND V. E. CENTONZE, Intracellular
fusion between reticulopodial networks in Al-
logromia laticollaris, 497
TRAVIS, J. L., see V. E. Centonze, 489 and H. Hotani,
491
Treadmill, 49 1
TREGGOR, JOSEF P., The role of freshwater wetlands
in the ontogeny of a New England saltmarsh,
512
Trematode infection in Ilyanassa obsoleta: depen-
dence on size and sex of the host and effect on
chemotaxis, 536
TROLL, WALTER, see Frederic Blum, 5 1 3
Tropical deforestation, 5 1 1
Trypanosoma brucei, 490
TSUJI, FREDERICK L, AND ELIZABETH HILL, Re-
petitive cycles of bioluminescence and spawning
in the polychaete Odontosyllis phosphorea, 444
Tubulin, 489, 496
genes, 490, 493
RNA, 539
Tunicates, 100, 733, see ascidians
Two-dimensional gel analysis of sea urchin ciliary
tubulins, 493
TYTELL, M., see R. V. Zackroff, 533
830
INDEX TO VOLUME 165
u
Uca pugilator, 758
Ultrastructural changes characteristic of Arbacia
sperm exposed to gossypol, 5 1 7
Ultrastructure, 491
Unconditioned behavioral response to learning, 525
Undulipodia, 241
Urticina lofotensis. 458
Urochordata, 419
Ultrastructural differences in the eggs and ovarian
follicle cells of Capitella (Polychaeta) sibling
species, 379
Ultrastructure of early embryonic shell formation in
the opisthobranch gastropod Aeolidia papillosa.
394
VACCA, LINDA L., AND MILTON FINGERMAN, The
roles of hemocytes in tanning during the molting
cycle: a histochemical study of the fiddler crab,
Uca pugilator, 758
Vahlkampfid amoebae, 241
VALIELA, IVAN, see Nina Caraco, 506
Vascular system of ascidians, 209
Veratridine, 490
Vertical migration rhythms of newly hatched larvae
of the estuarine crab, Rhithropanopeus harrisii,
139
Vertical movements of the hard clam, Mercenaria
mercenaria, in response to changes in baro-
metric pressure, 509
Vestibular system, 527
Video, 492
Video time lapse study of cell behavior during no-
tochord morphogenesis in ascidian embryos. A,
501
VINCENT, W. S., AND ROBERT M. HALL, Microbial
selection in an artificial ecosystem, 5 1 2
Vision in Limulus mating behavior: tests for detection
and form discrimination, 539
Visualization of depolarization-evoked presynaptic
calcium entry and voltage dependence of trans-
mitter release in squid giant synapse, 529
Visualizing extremely low contrast images by digital
enhancement of selected portions of the image
grey scale, 492
VITTURI, R., A. MAIORCA, AND E. CATALANO, The
karyology of Teredo ittriciilus (Gmelin) (Mol-
lusca, Pelecypoda), 450
Voltage
clamp, 532
dependence, 529
sensitive dyes, 530
Voltage clamp studies of dispersed toadfish pancreatic
islet cells, 494
VON BRUNN, ALBRECHT, RONALD A. CONLON, AND
M. M. WINK.LER, Maturation of sea urchin and
Chaetopterus oocytes results in a change in the
pattern of protein synthesis, 519
VOSSHALL, LESLIE B., see Gerald Weissmann, 503
W
WAGENBACH, GARY E., JUDITH P. GRASSLE, AND
SUSAN W. MILLS, Host specificity of intestinal
gregarines (Protozoa, Apicomplexa) in two
sympatric species of Capitella (Polychaeta), 538
WAHLE, CHARLES M., Regeneration of injuries
among Jamaican gorgonians: the roles of colony
physiology and environment, 778
WALTON, ALAN J., see G. T. Reynolds, 522
WALDRON, WILLIAM, see Seymour Zigman, 499
WANG, ROBERT T., see Maria I. Agudelo, 100
WARD ERIC R., AND WAYNE M. BARNES, Marine
molluscan genomes contain sequences homol-
ogous to the octopine synthase gene of Agro-
bacterium tumefaciens, 498
Wastewater treatment, 510, 512
WATSON, STANLEY W., BRUCE J. SCHNAPP, AND
ROBERT V. RICE, Mitochondria! and sphero-
somal movement along a filamentous network
in the marine slime mold Gymnophrydium
marinutn, 498
WEDI, STEVEN E., AND DAPHNE FAUTIN DUNN,
Gametogenesis and reproductive periodicity of
the subtidal sea anemone Urticina lofotensis
(Coelenterata: Actiniaria) in California, 458
WEINBERG, ERIC, see Kathleen Shupe, 518
WEINBERG, E. S., see P. E. Kuwabara, 500
WEISSMANN, GERALD, CATHLEEN ANDERSON,
LESLIE B. VOSSHALL, ABBY M. RICH, KATH-
LEEN A. HAINES, TOM HUMPHREYS, AND PHI-
LIP DUNHAM, Leukotriene B4 promotes the
calcium-dependent aggregation of marine
sponge cells, 503
WEST, TERRY L., see John A. Freeman, 409
WETHEY, DAVID S., Geographic limits and local
zonation: the barnacles Semibalanus (Balanus)
and Chthamalus in New England, 330
What makes cyclin cycle? 513
Wheat germ agglutinin, 525
WIDDER, EDITH A., MICHAEL I. LATZ, AND JAMES
F. CASE, Marine bioluminescence spectra mea-
sured with an optical multichannel detection
system, 791
WIELAND, S., see A. Eisen, 514
WILTSHIRE, MARK, see Jay Shiro Tashiro, 51 1
WINK.LER, MATTHEW, AND BREWER SHETTLES, Ev-
idence for regulation of protein synthesis at the
level of translational machinery in the sea urchin
egg, 503
WINKLER, M. M., see Albrecht Von Brunn, 519
WIRTH, D., see A. Flisser, 490, and Giullermo Ro-
mero, 537
WOLFE, RALPH S., see Joseph N. Boyer, 505
WOODWELL, G. M., see T. A. Stone, 51 1
WRENN, BRIAN A., see George J. Skladany, 510
INDEX TO VOLUME 165
831
WYMAN, CLAIRE, AND SCOTT LANDFEAR, Structure
of tubulin RNA from Leishmania enriettii, 539
noderm immunology: bacterial clearance by the
sea urchin Strongylocentrotus purpuratus, 473
YOKOO, KAREN M, ANNE E. GOLDMAN, AND
ROBERT D. GOLDMAN, Preliminary evidence
indicating the existance of intermediate fila-
ment-like proteins in unfertilized eggs of the
surf clam, Spisula solidissima, 519
YOUNG, JOHN DiNG-E, Rise of free intracellularCa2+
in mouse macrophage associated with 72b/7l
Fc receptor-ligand interaction, 498
Yui, MARY A., AND CHRISTOPHER J. BAYNE, Echi-
ZACK.ROFF, R. V., W. D. HILL, M. TYTELL, AND
R. D. GOLDMAN, Functional and chemical
characterization of squid neurofilament poly-
peptides, 533
ZIGMAN, SEYMOUR, TERESA PAXHIA, BLENDA AN-
TONELLIS, AND WILLIAM WALDRON, Ocular
lens aging in the skate, 499
ZIMMERBERG, JOSHUA, Hyperosmotic treatment
inhibits cortical granule exocytosis in the sea
urchin Lytechinus pictus, 520
Zoslera marina, 504, 507, 508
Continued from Cover Two
current addresses, acknowledgments or contribution 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.
4. A condensed title or running head of no more than 35 letters and spaces should appear at the top
of the title page.
5. Literature cited. In the text, literature should be cited by the Harvard system, with papers by
more than two authors cited as Jones el al., 1980. Personal communications and material 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 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 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 PERIODICALS, may find
a booklet published by the Biological Council of the U.K. (obtainable from the Institute of Biology, 4 1
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 /. 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 Visindafjelags Islendinga without abbreviation.
F. All single word 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 preferred forms rather than WORLD LIST or
USASI usage (e.g. Nature, Science, Evolution NOT Nature, Land., Science, N.Y.; Evolution, Lancaster,
Pa.)
6. Reprints, charges. The Biological Bulletin has no page charges. However, authors will be requested
to help pay printing charges of manuscripts that are unusually costly due to length or numbers of tables,
figures, or formulae. Reprints may be ordered at time of publication and normally will be delivered about
two to three months after the issue date. Authors (or delegates or 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).
CONTENTS
BRENCHLEY, G. A., AND J. T. CARLTON
Competitive displacement of native mud snails by introduced periwinkles
in the New England intertidal zone 543
BRETOS, MARTA, ITALO TESORIERI, AND Luis ALVAREZ
The biology of Fissurella maxima Sowerby (Mollusca: Archaeogastro-
poda) in northern Chile. 2. Notes on its reproduction 559
CHORNESKY, ELIZABETH A.
Induced development of sweeper tentacles on the reef coral Agaricia
agaricites: a response to direct competition 569
DEFUR, PETER L., BRIAN R. MCMAHON, AND CHARLES E. BOOTH
Analysis of hemolymph oxygen levels and acid-base status during emer-
sion 'in situ' in the red rock crab, Cancer product us 582
FREEMAN, GARY
Experimental studies on embryogenesis in hydrozoans (Trachylina and
Siphonophora) with direct development 591
GLADFELTER, ELIZABETH H.
Circulation of fluids in the gastro vascular system of the reef coral Acropora
cervicornis 619
HANLON, ROGER T., RAYMOND F. HIXON, AND WILLIAM H. HULET
Survival, growth, and behavior of the loliginid squids Loligo plei, Loligo
pealei, and Lolliguncula brevis (Mollusca: Cephalopoda) in closed sea
water systems 637
LEVINTON, JEFFREY S.
The latitudinal compensation hypothesis: growth data and a model of
latitudinal growth differentiation based upon energy budgets. I. Inter-
specific comparison of Ophryotrocha (Polychaeta: Dorvilleidae) 686
LEVINTON, JEFFREY S., AND ROSEMARY K. MONAHAN
The latitudinal compensation hypothesis: growth data and a model of
latitudinal growth differentiation based upon energy budgets. II. Intra-
specific comparisons between subspecies of Ophryotrocha puerilis (Po-
lychaeta: Dorvilleidae) 699
NlCCHITTA, C. V., AND W. R. ELLINGTON
Energy metabolism during air exposure and recovery in the high intertidal
bivalve mollusc Geukensia demissa granosissima and the subtidal bivalve
mollusc Modiolus squamosus 708
REED-MILLER, CHARLENE
Scanning electron microscopy of the regenerated shell of the marine
archaeogastropod, Tegula 723
SCOFIELD, VIRGINIA L., AND LAUREN S. NAGASHIMA
Morphology and genetics of rejection reactions between oozooids from
the tunicate Botryllus schlosseri 733
TELFORD, MALCOLM, ANTHONY S. HAROLD, AND RICH Mooi
Feeding structures, behavior, and microhabitat of Echinocyamus pusillus
(Echinoidea: Clypeasteroida) 745
VACCA, LINDA L., AND MILTON FINGERMAN
The roles of hemocytes in tanning during the molting cycle: a histo-
chemical study of the fiddler crab, Uca pugilator 758
WAHLE, CHARLES M.
Regeneration of injuries among Jamaican gorgonians: the roles of colony
physiology and environment 778
WIDDER, EDITH A., MICHAEL I. LATZ, AND JAMES F. CASE
Marine bioluminescence spectra measured with an optical multichannel
detection system 791
Short Report
GLADFELTER, ELIZABETH H.
Spatial and temporal patterns of mitosis in the cells of the axial polyp
of the reef coral Acropora cervicornis 811
INDEX TO VOLUME 165 816
100 -.M.
&, :
, 'k-
MBL WHOI LIBRARY
1B^^ o