Search the history of over 505 billion pages on the Internet.
Open Library
Full text of "Malacologia"
See other formats
HARVARD UNIVERSITY
LIBRARY
OF THE
DEPARTMENT OF MOLLUSKS
IN THE
Museum of Comparative Zoology
Gift of:
VOL 41, NO. 2 1999
MALACOLOGIA
International Journal of Malacology
INTERACTIONS BETWEEN MAN AND MOLLUSCS
UNITAS MALACOLOGICA-AMERICAN MALACOLOGIGAL SOCIETY SYMPOSIUM
WASHINGTON, D.C.
26-30 JULY 1998
Publication dates
Vol.31, No. 1 29 Dec. 1989
Vol. 31, No. 2 28 May 1990
Vol.32, No. 1 30 Nov. 1990
Vol.32, No. 2 7 Jun. 1991
Vol. 33, No. 1-2 6 Sep. 1991
Vol. 34, No. 1-2 9 Sep. 1992
Vol. 35, No. 1 14 Jul. 1993
Vol. 35, No. 2 2 Dec. 1993
Vol. 36, No. 1-2 8 Jan. 1995
Vol. 37, No. 1 13 Nov. 1995
Vol.37, No. 2 8 Mar. 1996
Vol.38, No. 1-2 17 Dec. 1996
Vol.39, No. 1-2 13 May 1998
Vol. 40, No. 1-2 17 Dec. 1998
Vol.41, No. 1 22 Sep. 1999
VOL. 41, NO. 2 MALACOLOGIA 1999
CONTENTS
DANIEL L. ALKON, PLENARY ADDRESS
Molecular Principles of Associative Memory that are Conserved During the
Evolution of Species 321
A. T ABD ALLAH, S. N. THOMPSON, D. B. BORCHARDT & M. 0. A. WANAS
Biomphalaria glabrata: A Laboratory Model Illustrating the Potential of
Pulmonate Gastropods as Freshwater Biomonitors of Heavy Metal
Pollutants 345
ROLAND С ANDERSON, PAUL D. HUGHES, JENNIFER A. MATHER &
CRAIG W. STEELE
Determination of the Diet of Octopus rebescens Berry, 1953 (Cephalopoda:
Octopodidae), Through Examination of Its Beer Bottle Dens in Puget Sound 455
THERESE D. BRACKENBURY & С С. APPLETON
Structural Damage to the Foot-sole Epithelium of Bulinus africanus
Following Exposure to a Plant Molluscicide 393
JAMES T CARLTON
Molluscan Invasions in Marine and Estuarine Communities 439
GEORGE M. DAVIS
Introduction 319
GEORGE M. DAVIS, THOMAS WILKE, Yl ZHANG, XING-JIANG XU,
CHI-PING OIU, CHRISTINA SPOLSKY DONG-CHUAN OIU, YUESHENG LI,
MING-YI XIA & ZHENG FENG
SnaW- Shistosoma, Paragonimus Interactions in China: Population Ecology,
Genetic Diversity, Coevolution and Emerging Diseases 355
THOMAS K. KRISTENSEN & DAVID S. BROWN
Control of Intermediate Host Snails for Parasitic Diseases— A Threat to
Biodiversity in African Freshwaters? 379
MARIA CHRISTINA DREHER MANSUR & MARIA DA GRAÇA OLIVEIRA DA SILVA
Description of Glochidia of Five Species of Freshwater Mussels (Hyriidae:
Unionoidea) from South America 475
RICHARD NEVES
Conservation and Commerce: Management of Freshwater Mussel (Bivalvia:
Unionoidea) Resources in the United States 461
J. P. POINTIER
Invading Freshwater Gastropods: Some Conflicting Aspects for Public
Health 403
DAVID G. ROBINSON
Alien Invasions: The Effects of the Global Economy on Non-marine
Gastropod Introductions into the United States 413
TIMOTHY P YOSHINO, CHRISTINE COUSTAU, SYLVAIN MODAT &
MARIA G.CASTILLO
The Biomphalaria glabrata Embryonic (BGE) Molluscan Cell Line:
Establishment of an In Vitro Cellular Model for the Study of Snail Host-
Parasite Interactions 331
WHY NOT SUBSCRIBE TO MALACOLOGIA
ORDER FORM
Your name and address
Send U.S. $36.00 for a personal subscription (one volume) or U.S. $55.00 for an
institutional subscription. Make checks payable to "MALACOLOGIA."
Address: Malacologia
Department of Malacology
Academy of Natural Sciences
1900 Benjamin Franklin Parkway
Philadelphia, PA 19103-1195, U.S.A.
VOL 41, NO. 1 1999
MALACOLOGIA
International Journal of Malacology
Revista Internacional de Malacologia
Journal International de Malacologie
Международный Журнал Малакологии
Internationale Malakologische Zeitschrift
MALACOLOGIA
EDITOR-IN-CHIEF:
GEORGE M. DAVIS
Editorial and Subscription Offices:
Department of Malacology
The Academy of Natural Sciences of Philadelphia
1900 Benjamin Franklin Parkway
Philadelphia, Pennsylvania 19103-1195, U.S.A.
EUGENE GOAN
California Academy of Sciences
San Francisco, CA
Co-Editors:
Assistant Managing Editor:
CARYL HESTERMAN
Associate Editor:
JOHN B. BURGH
University of Michigan
Ann Arbor
CAROL JONES
Denver, CO
MALACOLOGIA is published by the INSTITUTE OF MALACOLOGY, the Sponsor Members of
which (also serving as editors) are:
RÜDIGER BIELER
President
Field Museum, Chicago
JOHN BURGH
MELBOURNE R. CARRIKER
University of Delaware, Lewes
GEORGE M. DAVIS
Secretary and Treasurer
CAROLE S. HICKMAN
President Elect
University of California, Berkeley
ALAN KOHN
University of Washington, Seattle
JAMES NYBAKKEN
Vice President
Moss Landing Marine Laboratory, California
CLYDE F. E. ROPER
Smithsonian Institution, Washington, D.C.
SHI-KUEI WU
University of Colorado Museum, Boulder
Participating Members
EDMUND GITTENBERGER
Secretary, UNITAS MALACOLOGICA
Rijksmuseum van Natuurlijke
Historie
Leiden, Netherlands
JACKIE L. VAN GOETHEM
Treasurer, UNITAS MALACOLOGICA
Koninklijk Belgisch Instituut
voor NatuuHA/etenschappen
Brüssel, Belgium
Emeritus Members
J. FRANCES ALLEN, Emérita
Environmental Protection Agency
Washington, D.C.
KENNETH J. BOSS
Museum of Comparative Zoology
Cambridge, Massachusetts
ROBERT ROBERTSON
The Academy of Natural Sciences
Philadelphia, Pennsylvania
W. D. RUSSELL-HUNTER
Easton, Maryland
Copyright © 1999 by the Institute of Malacology
ISSN: 0076-2997
1999
EDITORIAL BOARD
J.A.ALLEN
Marine Biological Station
Millport, United Kingdom
E. E. BINDER
Museum d'Histoire Naturelle
Geneve. Switzerland
A.J.CAIN
University of Liverpool
United Kingdom
P. CA LOW
University of Sheffield
United Kingdom
J. G. CARTER
University of North Carolina
Chapel Hill. U.S.A.
R. COWIE
Bishop Museum
Honolulu, HI.. U.S.A.
A. H. CLARKE, Jr.
Portland. Texas. U.S.A.
B. С CLARKE
University of Nottingham
United Kingdom
R. DILLON
College of Charleston
SC. U.S.A.
C.J. DUNCAN
University of Liverpool
United Kingdom
D.J. EERNISSE
California State University
Fullerton. U.S.A.
E. GITTENBERGER
Rijksmuseum van Natuurlijke Historie
Leiden. Netherlands
F. GIUSTI
Universita di Siena. Italy
A. N. GOLIKOV
Zoological Institute
St Petersburg. Russia
S.J.GOULD
Harvard University
Cambridge, Mass.. U.S.A.
A. V. GROSSU
Universitatea Bucuresti
Romania
J. HABE
Tokai University
Shimizu. Japan
R. HANLON
Marine Biological Laboratory
Woods Hole. Mass., U.S.A.
J. A. HENDRICKSON, Jr.
Academy of Natural Sciences
Philadelphia, PA, U.S.A.
D. M. HILLIS
University of Texas
Austin. U.S.A.
K. E. HOAGLAND
Council for Undergraduate Research
Washington, DC. U.S.A.
B. HUBENDICK
Naturhistoriska Museet
Göteborg, Sweden
S. HUNT
Lancashire
United Kingdom
R. JANSSEN
Forschungsinstitut Senckenberg.
Frankfurt am Main, Germany
R. N. KILBURN
Natal Museum
Pietermaritzburg, South Africa
M. A. KLAPPENBACH
Museo Nacional de Historia Natural
Montevideo. Uruguay
J. KNUDSEN
Zoologisk Institut Museum
Kobenhavn, Denmark
A. LUCAS
Faculte des Sciences
Brest, France
с. MEIER-BROOK
Tropenmedizinisches Institut
Tubingen. Germany
H. К. MIENIS
Hebrew University of Jerusalem
Israel
J. E. MORTON
The University
Auckland. New Zealand
J. J. MURRAY, Jr.
University of Virginia
Charlottesville. U.S.A.
R. NATARAJAN
Mahne Biological Station
Porto Novo. India
DIARMAIDO'FOIGHIL
University of Michigan
Ann Arbor. U.S.A.
J. OKLAND
University of Oslo
Norway
T. OKUTANI
University of Fisheries
Tokyo. Japan
W. L. PARAENSE
Instituto Oswalde Cruz. Rio de Janeiro
Brazil
J. J. PARODIZ
Carnegie Museum
Pittsburgh. U.S.A.
J. P. POINTIER
Ecole Pratique des Hautes Etudes
Perpignan Cedex. France
W. F. PONDER
Australian Museum
Sydney
Ol Z. Y.
Academia Sínica
Qingdao. People s Republic of China
D. G. REID
The Natural History Museum
London. United Kingdom
S. G. SEGERSTRALE
Institute of Mahne Research
Helsinki. Finland
A. STANCZYKOWSKA
Siedlce. Poland
F. STARMÜHLNER
Zoologisches Institut der Universität
Wien. Austria
Y I. STAROBOGATOV
Zoological Institute
St. Petersburg. Russia
W. STREIFE
Universite de Caen
France
J. STUARDO
Universidad de Chile
Valparaiso
S. TILLIER
Museum National d'Histoire Naturelle
Pahs. France
R. D.TURNER
Harvard University
Cambridge. Mass.. U.S.A.
J. A.M. VAN DEN BIGGELAAR
University of Utrecht
The Nethehands
N. H. VERDONK
Rijksuniversiteit
Utrecht, Netherlands
ANDERS WAREN
Swedish Museum of Natural History
Stockholm. Sweden
B. R. WILSON
Dept. Conservation and Land Management
Kallaroo. Western Australia
H.ZEISSLER
Leipzig, Germany
A. ZILCH
Forschungsinstitut Senckenberg
Frankfurt am Main. Germany
MALACOLOGIA, 1999, 41(1): 1-118
DESCRIPTIONS OF SOME OF THE GLOCHIDIA OF THE UNIONIDAE
(MOLLUSCA:BIVALVIA)
Michael A. Hoggarth^
Museum of Zoology. The Ohio State University, 1813 North High Street, Columbus, Ohio
43210, U.S. A: mhoggarth@otterbein.edu
ABSTRACT
The primary objective of this study was to describe glochidia of the family Unionidae. Glochidia
from 82 nominal taxa representing 30 genera in four subfamilies of the LJnionidae were exam-
ined. The glochidium of the European Unio elongatulus glaucinuswas found to share many char-
acters with those of the Anodontinae. Both have triangular valves and styliform hooks. Further-
more, the glochidium of Simpsonaias ambigua and U. e. glaucinus share exterior valve sculpture
and styliform hook fine structure.
The glochidia of some members of the Alasmidontini (subfamily Anodontinae), including some
species in the genera Lasmigona and Alasmidonta. and that of Pegias fabula are depressed-
pyriform, with large adductor muscles, looped exterior valve sculpture, and a double row of mi-
crostylets on the hook. These glochidia are very similar to those of Stroptiitus. The glochidia of
Arcidens and the remaining members of Alasmidonta and Lasmigona are high-pyriform, with
small adductor muscles, beaded to rosette exterior valve sculpture, and complex hooks with at
least four rows of microstylets.
The glochidia of the Ambleminae demonstrate structures also found in the primitive lampsiline
genera. The Lampsilinae are divided into four groups with the main lineage including Ptycho-
branchus. Actinonaias. Obovaria. Ligumia. Venustaconcha. Villosa and Lampsilis. Branches
from this lineage include: (1) Obliquaria. Cyprogenia and Dromus: (2) Ellipsaria. Leptodea and
Potamilus: and (3) Epioblasma.
Key words: Unionidae, glochidia, glochidial morphology, electron microscopy.
INTRODUCTION
Leeuwenhoek made the first substantial
contribution to the study of glochidia (Leeu-
wenhoek, 1695). He correctly interpreted
these tiny bivalves as young molluscs. He
also observed limited development of the lar-
vae and saw the characteristic snapping be-
havior they display when mature. It does not
appear that he ever doubted that these tiny
molluscs, which had developed within the gills
of a female, were her young. However, like so
many others during the century and a half to
follow, he was unable to provide an environ-
ment outside of the female where develop-
ment could continue. This, combined with the
observations that these tiny molluscs were al-
most identical in both Anodonta and Unio.
numbered within the gills of the larger mollusc
by the thousands, and had structures, either
real or imagined, that were quite different from
those of the larger mollusc, led some to be-
lieve that the small shelled animals were not
larvae but parasites. The tiny bivalve para-
^Also: Department of Life and Earth Science, Otterbein College, Westerville, Ohio 43081, U.S.A.
1
sites were given the name Glochidium para-
siticumby Rathke (1797).
The next major advance in the understand-
ing of the life history of the Unionacea came
with the observations of Carus and Leydig.
Carus (1 832) watched the brightly colored ova
of Potamida littoralis (Cuvier, 1 797) pass from
the oviduct to the outer gills of a female mol-
lusc. Continued observation demonstrated
that the animal known as G. parasiticum was
not a species separate from the larger mollusc
but its larval stage. Leydig's (1866) discovery
of glochidia embedded in the fins of a fish
solved the primary developmental mystery
and led the way for life history investigation.
One of the primary objectives of these early life
history investigations was to document the
changes that occur during parasitism, when
the glochidium transforms into a juvenile
(Braun, 1878; Schmidt, 1885; Schierholz,
1878, 1888; Harms, 1907a, b, c, 1908, 1909).
These studies also demonstrated that artificial
infection could be used to indicate the suscep-
tibility of fish to glochidia. Fueled by the com-
HOGGARTH
FIG. 1 . Typical life histories of two species of Unionidae. (Drawing of A. rupestris after Trautman, 1 981 , and
drawings of adult shells of P. g. grandis and L. r. luteola after Burch, 1975.)
mercial importance of North American unionid
shells in the button industry, and the dwindling
supply of those shells, the staff of the U. S. Bu-
reau of Fisheries began to use artificial infec-
tion to determine the hosts for commercially
important species (Lefevre & Curtis, 1910,
1912; Coker & Surber, 1911; Surber, 1912,
1913, 1915; Howard, 1912, 1914a, b, c). Dia-
grams of the unionid life cycle, were soon to
follow.
Figure 1 demonstrates that there are es-
sentially two very different life histories. Not
only are the larvae themselves morphologi-
cally different (triangulate, ligulate, etc.), but
so are their sites of eventual parasitism and
the way in which they gain access to their
host. The relatively large, subtriangular
glochidium of the Anodontinae bears a hook
on the ventral margin of each valve. If this
glochidium clamps down on the fin of a pass-
ing fish and pierces the fin epithelium, it may
become encapsulated by host tissue. The
smaller, subelliptical glochidium of the Lamp-
silinae lacks hooks. If this glochidium is taken
in through the mouth of its host, travels to the
gills and clamps down on a gill filament, irri-
tating the gill epithelium, it too may become
encapsulated by host tissue. Attraction de-
vices, such as the mantle flaps of Lampsilis,
facilitate the exchange of glochidium from
unionid to fish (Morrison, 1973).
The dissimilarities in these life histories re-
veal three periods when selection might be
acting upon the glochidium: (1 ) during release
from the female, (2) during initial contact with
the host, and (3) during encapsulation. The
selective advantage of glochidial structures
might be understood, therefore as they facili-
tate attraction of the host, attachment to the
host, or induction of encapsulation (Hoggarth
& Gaunt, 1988). Encapsulation provides a
stable environment where transformation can
occur, a ready source of nutrients (whether
used or not), and protection from predators
and from being washed or tumbled down-
stream (Coker & Surber, 1911; Howard &
Anson, 1922; Arey, 1932a, b; Kat, 1984). Fur-
thermore, the host functions as a dispersal
mechanism, and releases the newly trans-
formed juvenile into suitable habitats. With the
use of scanning electron microscopy, minute
glochidial structures can be examined and
used to interpret relationships among unionid
species. It is the objective of this paper to de-
GLOCHIDIAOF UNIONIDAE
scribe the glochidia of a large portion of tine
North American fauna to begin this endeavor.
MATERIALS AND METHODS
Specimens Examined
Glochidia were removed from the marsupia
of 150 female unionids representing 82 nomi-
nal taxa from 30 genera. Forty-three lots of
material were processed from specimens col-
lected or received during this study. These
specimens were deposited in the collection of
The Ohio State University Museum of Zool-
ogy (OSUM), but glochidia samples from
each specimen were retained by the author
and have MAH catalog numbers. The remain-
ing specimens were located in collections of
unionids at The Ohio State University Mu-
seum of Zoology, University of Michigan Mu-
seum of Zoology (UMMZ), University of Wis-
consin Zoological Museum (UWZY), and The
Illinois Natural History Survey (INHS).
Glochidia removed from the marsupia of fe-
male unionids were preserved (for freshly col-
lected specimens) and stored in a solution of
80% alcohol, 5% glycerin, and 15% water.
Each vial of glochidia was labeled with the
catalog number of the female from which the
sample came. Subsamples of glochidia taken
from a vial for dehydration in acetone, and
subsamples of these that were placed on a
stub for viewing with scanning electron mi-
croscopy (SEM) were also labeled with the
catalog number of the female mollusc. There-
fore, each glochidium examined can be
traced back to its maternal parent.
Procedures for Scanning
Electron Microscopy
Preserved and freshly collected glochidia
were cleaned by the removal of the glochidial
soft parts. Only glochidia that had valves gap-
ing apart (in the case of the preserved mater-
ial) or that were actively snapping their valves
(in the case of freshly collected glochidia)
were processed for the SEM.
The initial step in cleaning was to wash the
glochidia in three changes of distilled water.
Each sample was suspended in distilled water
and then allowed to settle, after which the su-
pernatant was removed by using a Pasteur
pipette. Generally, the glochidia settled to the
bottom of the vial within 10-15 sec, whereas
small pieces of the marsupium and the matrix,
within which the glochidia may be found within
the marsupium, were still suspended. These
impurities were removed with the super-
natant.
Freshly collected glochidia were cleaned
according to the method of Calloway & Turner
(1979). These glochidia initially were washed
as above, with two drops of 1 N NaOH added
to the final wash (10 ml). The glochidia were
allowed to stand in this slightly basic solution
for ten minutes and then washed in three
changes of distilled water.
Previous studies using SEM to examine
glochidia employed only freshly collected ma-
terial from which the soft parts were removed
as described above, or preserved material
from which the soft parts had not been re-
moved (Giusti, 1973; Giusti et al., 1975;
Clarke, 1981a, 1985; Rand & Wiles, 1982).
No practical method for the removal of the
preserved soft parts had been developed.
However, the procedure outlined below pre-
sents a method that was found to produce ad-
equate specimens for SEM examination. This
procedure is sufficiently flexible to allow for
differences in preservation, initial valve gape,
and amount of extraneous material in the
sample.
Following the final rinse in distilled water,
specimens were placed in 10 ml of 1% aque-
ous trypsin solution (Sorensen's Phosphate
Buffer, pH 7.00) and thoroughly mixed. This
was accomplished by drawing and expelling
the liquid in the vial into a pipette 10-15 times.
The action of the rapidly moving liquid often
dislodged the soft parts from the valves,
thereby reducing the time required in the
trypsin solution. The sample was placed in an
incubator at 37-C, visually inspected every 1 5
min, and removed when the valves began to
gape at an angle approaching 180°. The di-
gestion process was discontinued as soon as
the first valves began to open, rather than
after all valves had opened. The supernatant
was carefully removed from the sample and
the sample was washed in three changes of
distilled water. Any remaining soft parts were
mechanically dislodged by drawing a large
number of glochidia into a pipette and ex-
pelling them 20-30 times per wash.
Cleaned glochidia, whether freshly col-
lected or preserved, then were dehydrated in
an ascending gradation of acetone (10%,
30%, 50%, 70%, 90%, 95%, 100%, 100%)
and stored in Borosilicate Glass Scintillation
Vials (VWR Scientific) in the final acetone
HOGGARTH
wash. Subsamples of these glochidia were
processed for the SEM. Glochidia were air
dried on a clean glass slide and mounted on
double stick tape (3M) on 13 mm aluminum
stubs. The tape was rimmed with silver paint
and then the entire stub was placed in a vac-
uum desiccator for 24 h. Critical point drying
was found to be unnecessary since the soft
parts, which would tend to resist desiccation,
were removed. Following desiccation, the
specimens were coated with 30 nm of gold-
palladium in a Hummer VI Sputter Coater, and
then viewed in a Cambridge Stereoscan S4-
10 SEM or a Hitachi S-500 SEM at an accel-
eration voltage of 20kV.
Orientation of Specimens on Figures
The orientation of the glochidial valve,
demonstrated by the electron micrographs on
each plate, follows Hoggarth (1987). The dor-
sal aspect of the valve has been oriented to-
ward the top of the page, ventral is down. The
anterior margin is toward the right hand mar-
gin of the page and posterior is to the left.
Characters Examined
Each character examined was defined in
terms of two or more states or expressions of
that character. An index to glochidial valve
characters is found in Table 1 . Measurements
were made directly from SEM micrographs
with care taken when glochidia were placed
on the stub to ensure that some were flat and
not tilted in respect to the stub surface. Micro-
graphs of these glochidia were taken at 0- tilt.
Length - Glochidial valve length was mea-
sured as the greatest distance from anterior to
posterior margins. This measurement was
made parallel to the hinge (Fig. 2). Table 2
contains the morphometric data for each
specimen of glochidium examined.
Height - Glochidial valve height was mea-
sured as the greatest distance from dorsal to
ventral margins. This measurement was
made perpendicular to length (Fig. 2).
Hinge length - The hinge was measured in
a straight line from the points were the dorsal
margins intersect the anterior and posterior
margins, regardless of whether the hinge was
curved or straight (Fig. 2).
Hinge ligament length - In the glochidium
the hinge ligament extends the entire length
of the hinge. A portion of the ligament can be
seen when viewing the valve externally and
another portion can be viewed internally.
Herein, central ligament refers to the ex-
panded central portion of the hinge ligament
or that portion that can be viewed internally.
Posterior ligament refers to the posterior por-
tion of the hinge ligament, measured from the
posterior margin of the hinge to the posterior
margin of the central ligament. Anterior liga-
ment is the portion of the hinge ligament from
the anterior margin of the hinge to the anterior
margin of the central ligament.
Central ligament position - Central liga-
ment position was found by adding one half
the central ligament length to the length of the
posterior ligament and then dividing by the
length of the hinge. The mid-point of the cen-
tral ligament was expressed as % length of
the hinge from posterior to anterior.
Valve shape - Valve shape refers to the
outline of the shell when viewed sagittally (the
plane bisecting anterior and posterior). The
terms used to describe valve shape have
been loosely defined in the past. Valve shape
descriptions and representative electron mi-
crographs are indexed in Table 1 .
Lateral valve gape - With the valves fully
adducted some species of glochidia were
found to possess anterior and posterior valve
gape. This character was expressed as either
absent or present (Table 1 provides an index
to these and all subsequent glochidial valve
characters).
Dorsal alae - This structure was found at
the dorsal-anterior and dorsal-posterior mar-
gins of most lampsiline glochidia. They can
best be described as arch-like extensions of
the glochidial valve. This character was ex-
pressed as absent, short or long.
Microstylets - Clarke (1 981 a) proposed the
term microstylet for the larger (> 1 .0 цт long)
points on the ventral margin of the glochidial
valve. Microstylets were, (1) absent, (2) many
and unorganized, (3) arranged in one distal
row on the hook, (4) arranged in two distal
rows on the hook, or (5) arranged in many dis-
tal rows on the hook.
Micropoints -Clarke (1981a) used the term
micropoints for the small (< 1.0 |.im long)
points on the ventral margin of the glochidial
valve. Micropoints were, (1) lanceolate, aris-
ing as single attenuate points from the ventral
margin of the valve, (2) lamellate, arising as
single plate-like points, or (3) coronal, with
three to seven attenuate points arising from a
common base.
GLOCHIDIAOF UNIONIDAE
TABLE 1. Glochidial shell characters used in analysis of relationships between the species
of Unionidae. Following each character state is a reference to a figure illustrating the char-
acter state.
1. Length
a. short (46A)
b. moderate (57A)
с long (17A)
2. Height
a. short (32A)
b. moderate (58A)
с high (24A)
3. Hinge length
a. short (48C)
b. moderate (64G)
с long (4F)
4. Central ligament length
a. short (47E)
b. long (9C)
5. Central ligament position
a. posterior (7F)
b. central (50F)
6. Valve shape
a. subelliptical (36B)
b. depressed subelliptical (71 B)
с subrotund (38B)
d. subspatulate (63B)
e. fabelliform (40B)
f. subligulate (45B)
g. ligulate (51 B)
h. subtriangulate (5B)
i. lachrimiform (11 B)
j. pyriform (27B)
k. depressed pyriform (20B)
I. quadrate (23B)
7. Lateral valve gape
a. absent (64C)
b. present (50C)
8. Dorsal alae
a. absent (36A)
b. short (47A)
с long (62A)
9. Microstylets
a. absent (61 C)
b. many unorganized (15D)
с one distal row (13D)
d. two distal rows (16D)
e. many distal rows (28D)
10. Micropoints
a. lanceolate (29E)
b. lamellate (49E)
с coronal (33E)
11. Micropoint organization
a. unorganized (36H)
b. horizontal rows (71 C)
с broken vertical rows (59E)
d. complete vertical rows (45D)
12. Hook
a. absent (55D)
b. styliform (25E)
с supernumerary (70C)
d. lanceolate (50D)
13. Exterior surface sculpture
a. rough (68E)
b. beaded (15E)
с rosette (26E)
d. loose looped (9D)
e. ribbed loose looped (5C)
f. tight looped (23E)
g. vermiculate (51 E)
Micropoint organization - Micropoints were,
(1) unorganized, (2) arranged in horizontal
rows, (3) arranged in broken vertical rows, or
(4) arranged in complete vertical rows.
Hook - As pointed out by Clarke (1981a),
this term has been used indiscriminately to
apply to independently derived structures that
serve a similar function. He proposed using
the term stylet to refer to the complex hook of
the anodontine glochidia. In this study, the
glochidial hook was either, (1) absent, (2)
styliform, a V-shaped extension of the ventral
margin of the glochidium, (3) lanceolate, a re-
curved, attenuate extension at each corner of
the ventral margin of the glochidia of the
genus Potamilus, or (4) supernumerary, the
straight and sharply pointed hook found in an
area along the ventral margin of the genus
Epioblasma.
Exterior surface sculpturing - This refers to
the fine structure of the exterior surface of the
glochidial valve (viewed between 10,000 to
20,000 x). Exterior surface sculpture was, (1)
rough, (2) beaded, (3) rosette, (4) loose-
looped, (5) ribbed, loose-looped, (6) tight-
looped, or (7) vermiculate.
HOGGARTH
FIG. 2. Subelliptical glochidium of the Lampsilinae
demonstrating length (L) and height (H). The dorsal
margin of the valve is toward the top of the page,
ventral down, anterior to the right, and posterior to
the left. The adductor muscle (AM) is located in the
dorsal-anterior quadrant of the valve.
DESCRIPTIONS OF GLOCHIDIA
Subfamily Unioninae
Unio elongatulus glaucinus Porro, 1838
(Fig.3A-F)
Material Examined
MAH 2055-Rivanazzano, Pavia, Italy, ex
Coll. F. Giusti.
Description
Glochidium subtriangular, length 21 8 to 232
(am, (227 ± 6.45 ¡.im, n = 4), height 21 to 21 8
um (216 ± 4.08 цт, n = 4). Anterior and pos-
terior margins arcuate, meeting at base of
hook to form a broadly rounded ventral termi-
nus. Valve outline only slightly asymmetric,
with anterior margin slightly more produced
than posterior margin. Exterior surface of
valve malleated (large dimple-like depres-
sions) and pitted (smaller depressions gener-
ally within malleated valve surface), except
along smooth valve border. Fine sculpture of
exterior valve surface beaded (Fig. 3E). Cen-
tral ligament (seen in internal views of valve)
55 to 58 |jm (56 ± 1 .73 |im, n = 3) in length,
located about 40% from posterior to antehor.
Posterior ligament 41 to 44 цт (42 ± 1 .53
).im, n = 3) in length; anterior ligament 68 to 75
\xvr\ (72 ± 3.60 цт, n = 3) long. Dorsal margin
straight, 164 to 175 цт (171 ± 4.36 ¡xm, n =
4) in length. Styliform hook extending from
ventral terminus of each valve as a broad tri-
angular plate. Hook gradually and uniformly
tapered to a broad point distally, located about
50% from both lateral margins. Surface of
hook covered by microstylets (> 1.0 ¡am long
points on hook) and micropoints (< 1.0 |.im
long points on the hook). Most microstylets
(about 35) bluntly pointed, with some of more
proximal microstylets multifaceted, sharply
pointed. Micropoints extending from lateral
and proximal margins of styliform hook cover-
ing most of its lateral surface, except within
narrow band distally.
Remarks
The glochidium of UnIo has been described
as subtriangular with a ventral hook (Ort-
mann, 1912). Giusti (1973) stated that other
than its smaller size, this glochidium resem-
bles that of Anodonta. Ortmann (1918), how-
ever, reported the absence of a hook in Unio
caffer, but Heard & Guckert (1970) suggested
that this was the result of examining immature
specimens. They noted that Giusti (1973)
demonstrated hooks in U. e. glaucinus and
that McMichael & Hiscock (1 958) found hooks
on mature glochidia of Velesunio ambiguus,
even though the species was earlier de-
scribed as having bookless glochidia (His-
cock, 1951).
Ortmann (1912) and Heard & Guckert
(1970) placed this genus near the North
American genus Pleurobema (Ambleminae)
on the basis of similar shell morphology and
anatomy. However, Ortmann (1912) recog-
nized the similarity between the glochidia of
the Unioninae and the Anodontinae and sug-
gested that the former may have given rise to
the latter. Morhson (1955) agreed with Ort-
mann's conclusion regarding the origin of the
Anodontinae. He chose to stress develop-
mental characters (the similarity in glochidia)
over adult characters and therefore placed
the subfamily Unioninae near the Anodonti-
nae. Harms (1908, 1909) and Haas (1910)
gave length and height measurements of 290
цт for the glochidium of U. pictorum. Giusti
(1973) noted that the glochidium of U. e.
glaucinus is smaller than that of Anodonta.
but gave no measurements.
GLOCHIDIAOF UNIONIDAE
TABLE 2. Morphometric data for glochidia examined during this study. All measurements in ¡.im.
Species
Catalog #
Length
Hinge
Length
Ligament Length
ieight
Ante.
Cent.
Post.
218
172
75
56
41
216
175
168
68
57
43
210
164
173
74
55
44
218
175
352
268
112
79
77
348
273
345
268
279
115
85
79
356
277
356
279
115
88
76
352
279
118
83
78
344
272
346
270
268
115
85
68
357
279
361
273
354
267
352
262
361
264
116
91
57
350
260
114
86
60
358
274
361
280
125
96
59
289
212
293
214
288
211
290
210
118
58
34
284
209
340
264
111
88
65
257
110
88
59
345
260
111
89
60
267
114
89
64
250
106
85
59
350
253
114
84
55
345
163
350
166
52
77
37
160
48
74
38
328
237
80
101
56
324
228
74
100
54
323
233
81
98
54
320
232
78
102
52
322
231
360
250
104
92
54
355
250
360
257
111
94
52
352
250
355
259
107
90
62
353
250
350
252
367
260
368
275
363
263
100
93
70
368
288
102
115
71
366
277
370
277
358
291
351
289
98
116
75
U. e. glaucinus
A. cygnea
A. anatina
A. bering iana
A. kenneriyi
A. i m plicata
A. suborbiculata
P. g. grandis
P. g. corpulenta
P. с Cataracta
MAH 2055
MAH 2055
MAH 2055
MAH 2055
MAH 2055
MAH 2055
OSUM 20911.
OSUM 20911.
OSUM 20911.
OSUM 20911.
OSUM 20911.
OSUM 20911,
OSUM 20911,
OSUM 20911,
OSUM 20911,
OSUM 20911,
OSUM 20911,
OSUM 20912.
OSUM 20912.
OSUM 20912.
OSUM 20912.
OSUM 20912.
OSUM 20912.2
OSUM 20912.2
OSUM 3711.1
OSUM 3711.1
OSUM 3711.1
OSUM 3711.1
OSUM 3711.1
OSUM 52882.2
OSUM 52882.2
OSUM 52882.2
OSUM 52882.2
OSUM 52882.3
OSUM 52882.3
OSUM 52463.7
OSUM 52463.7
OSUM 52463.7
OSUM 13634
OSUM 13634
OSUM 13634
OSUM 13634
OSUM 13634
OSUM 38467.10
OSUM 38467.10
OSUM 38467.10
OSUM 38467.10
INHS2247
INHS2247
INHS2247
OSUM 47890
OSUM 47890
OSUM 53653
OSUM 52462.27
OSUM 52462.27
OSUM 52462.27
OSUM 52462.35
OSUM 52462.35
225
218
231
232
358
351
345
347
351
351
357
346
350
348
355
352
353
353
356
361
289
292
288
286
288
354
352
350
342
343
328
326
326
323
325
358
355
355
365
353
350
358
350
350
343
380
374
375
375
378
{continued)
HOGGARTH
TABLE 2. (Continued)
Species
Catalog #
Length
Height
Hinge
Length
Ligament Length
Ante. Cent. Post.
P. с margínala
P. doliaris
U. imbecillis
U. imbecillis
A. ferussacianus
S. ambigua
S. и. und и latus
S. и. tennesseensis
S. subvexus
A. viridis
A. heterodon
A. und и lata
OSUM 38962.6
OSUM 38962.6
OSUM 26405
OSUM 26405
MAH 435
MAH 435
OSUM 9436.2
OSUM 9436.2
OSUM 9436.2
OSUM 9436.2
OSUM 9436.2
OSUM 9436.2
OSUM 9436.2
OSUM 9436.2
UWZY 24971.1
UWZY 24971.1
UWZY 24971.1
OSUM 18275.2
OSUM 18275.2
OSUM 18275.2
MAH 989.4
MAH 989.4
MAH 989.4
MAH 989.4
UMZY 22658
UMZY 22658
UMZY 22658
UMZY 22662
UMZY 22662
UMZY 22672
OSUM 55995
OSUM 49443
OSUM 52458.4
OSUM 52458.4
OSUM 52458.4
OSUM 52458.4
OSUM 52458.4
OSUM 33381 .2
OSUM 55449
OSUM 36240
OSUM 36240
OSUM 36240
OSUM 36240
OSUM 47518
OSUM 47518
OSUM 47518
OSUM 47518
OSUM 47518
OSUM 47518
OSUM 47518
OSUM 25106.2
OSUM 25106.2
OSUM 25106.2
OSUM 25106.2
OSUM 52434.4
OSUM 52434.4
OSUM 52434.4
OSUM 52434.4
OSUM 52434.4
358
360
317
405
291
310
310
306
299
300
299
310
309
300
313
303
323
325
326
322
325
321
319
253
251
255
258
257
255
369
360
360
346
354
348
359
300
313
313
300
319
303
300
330
338
332
335
358
356
355
354
343
364
369
317
368
297
297
300
295
292
289
303
304
303
303
306
305
325
326
327
322
325
320
322
256
257
265
264
264
259
299
289
296
298
288
292
289
245
250
246
253
258
250
250
267
258
268
258
378
365
375
368
369
276
277
240
305
246
242
245
251
243
242
240
243
249
256
252
255
250
235
234
238
231
238
231
233
168
168
167
170
170
166
164
281
271
280
272
281
280
268
268
271
275
277
273
245
260
250
250
258
250
245
253
266
265
255
258
247
260
257
247
102
95
98
103
103
83
85
89
65
61
63
101
109
108
110
96
123
100
108
109
110
105
103
105
110
90
93
93
89
92
91
88
58
63
54
97
87
103
92
94
84
80
81
64
61
60
60
63
59
62
56
45
49
49
73
76
70
78
78
70
65
71
73 68
80
98
89
89
75
55
55
53
GLOCHIDIAOF UNIONIDAE
TABLE 2. (Continued)
Hinge
Ligament Length
Species
Catalog #
Length
Height
Length
Ante.
Cent.
Post.
A. m arg i nata
MAH 277.1
340
360
233
MAH 277.1
341
362
235
107
67
61
MAH 277.1
335
372
232
100
73
59
MAH 724.1
341
367
230
P. fabula
OSUM 41308.3
386
325
210
74
73
63
OSUM 41309.1
388
319
200
69
68
62
OSUM 41309.1
385
323
208
79
73
56
A. confragosus
OSUM 52015
357
353
240
100
70
70
OSUM 52015
352
354
252
108
74
70
OSUM 52015
363
355
237
OSUM 52015
360
355
252
115
78
59
OSUM 52015
250
115
63
72
L. compressa
OSUM 23179.1
322
285
234
OSUM 23179.1
327
287
233
82
84
67
MAH 702
320
286
230
MAH 702
317
283
239
MAH 727
327
288
232
81
83
68
L. subviridis
OSUM 27131.66
370
309
253
100
88
65
OSUM 27131.66
383
318
259
OSUM 27131 .66
380
311
244
OSUM 27131.68
368
310
260
105
87
68
OSUM 27131.68
381
312
253
107
89
57
L. holstonia
OSUM 55826.6
286
294
234
109
73
52
OSUM 55826.6
291
276
235
OSUM 55826.6
281
275
221
OSUM 55826.7
290
275
228
110
63
55
OSUM 55826.7
284
291
221
L. costata
MAH 585
343
365
240
97
79
64
MAH 585
347
375
239
MAH 585
343
373
239
104
71
64
MAH 585
345
364
248
105
82
61
MAH 882.1
343
368
241
90
83
68
MAH 882.1
348
377
244
99
84
61
MAH 882.1
340
363
239
L. complánate
MAH 278.2
193
75
68
50
MAH 278.2
290
295
197
79
65
53
MAH 278.2
296
310
201
88
60
53
MAH 278.2
289
293
195
82
62
51
MAH 278.2
294
298
199
MAH 278.2
293
297
208
MAH 278.2
295
308
206
M. nervosa
OSUM 13032.66
268
347
155
OSUM 541 78
254
348
145
OSUM 54178
153
60
49
44
OSUM 541 78
262
350
147
OSUM 1986:22
260
340
145
OSUM 1986:22
155
60
52
43
OSUM 1986:22
155
63
49
43
M. boykiniana
OSUM 51107.5
245
350
150
P. dombeyana
OSUM 53273.2
223
238
132
55
46
31
OSUM 53273.2
223
240
134
OSUM 53273.2
228
259
135
OSUM 53273.2
133
58
43
32
OSUM 53273.3
224
252
134
OSUM 53273.3
230
251
132
59
42
31
OSUM 53273.3
225
243
133
OSUM 43011
223
250
136
OSUM 43011
226
243
132
OSUM 43011
231
241
134
OSUM 43011
223
243
133
51
47
35
(continued)
10
HOGGARTH
TABLE 2, (Continued
Hinge
Ligament Length
Species
Catalog #
Length
Height
Length
Ante.
Cent.
Post.
T. verrucosa
МАИ 654.1
92
101
45
MAH 654.1
46
9
32
5
MAH 654.1
88
100
40
6
31
3
MAH 654.1
43
6
33
4
MAH 654.1
90
100
46
8
33
5
MAH 654.1
85
97
44
MAH 654.1
94
101
45
MAH 654.1
44
9
30
5
MAH 654.1
93
99
44
6
34
4
Q. infucata
OSUM 48537.1
104
28
54
22
OSUM 48537.1
234
275
104
26
58
20
OSUM 48537.2
240
287
105
OSUM 48537.2
242
285
102
OSUM 48537.2
242
285
102
E. dilatata
MAH 946.9
217
225
140
56
45
39
MAH 946.9
210
219
142
56
50
36
MAH 946.9
219
219
142
MAH 946.9
219
219
147
P. fasciolaris
MAH 641
170
195
89
MAH 641
174
184
85
26
40
19
MAH 651
175
182
80
22
36
22
P. occidentalis
OSUM 45361.14
197
241
101
30
48
23
OSUM 45361.17
203
234
102
OSUM 45361.17
104
34
44
26
P. greeni
OSUM 19025.2
189
227
96
26
49
21
OSUM 19025.2
190
228
93
OSUM 19025.2
183
226
90
P. subtentum
OSUM 43156.5
195
251
85
25
36
24
OSUM 43156.5
181
239
90
OSUM 43156.5
191
251
85
OSUM 43156.5
83
26
36
21
OSUM 43156.5
190
236
82
OSUM 43156.5
194
239
83
25
38
20
0. reflexa
OSUM 54361.1
219
221
119
OSUM 54361.1
213
219
127
44
52
31
OSUM 54361.1
219
206
121
37
54
30
С stegaria
OSUM 6298.21
204
164
113
OSUM 6298.21
206
167
116
46
40
30
OSUM 6298.21
208
170
120
OSUM 6298.21
115
43
39
33
OSUM 6298.21
114
47
35
32
С aberti
OSUM 48067
200
145
128
50
29
49
OSUM 48067
207
155
136
OSUM 48067
200
143
125
48
31
46
OSUM 48067
211
156
128
51
36
41
OSUM 48067
218
161
132
OSUM 48067
211
161
131
60
28
43
OSUM 48067
136
61
32
43
D. dramas
OSUM 23209
171
76
39
56
OSUM 20407.1
230
120
182
OSUM 20407.1
230
114
182
82
39
61
OSUM 20407.1
222
118
169
78
40
51
OSUM 20407.1
221
120
160
OSUM 20407.1
219
120
182
A. pectorosa
OSUM 24337
253
270
139
46
62
31
OSUM 48748.3
248
270
151
52
62
37
OSUM 48748.3
244
260
144
47
58
39
A. 1. cari nata
MAH 842.1
220
243
125
38
57
30
0. retusa
UUMZ UNCAT.
233
278
115
UUMZ UNCAT.
218
272
117
UUMZ UNCAT.
119
38
45
36
GLOCHIDIAOF UNIONIDAE
11
TABLE 2. {Continued)
Species
Catalog #
Length
Height
Hinge
Length
Ligament Length
Ante. Cent. Post.
O. olivaria
O. subrotunda
O. jacksoniana
O. unicolor
E. lineolata
L. frag i I is
L. ochracea
P. otiiensis
P. amphiciiaena
P. alatus
P. purpuratus
L. recta
OSUM 51282.2
OSUM 51282.2
OSUM 51282.2
OSUM 51282.2
OSUM 51282.2
MAN 659.2
MAH 659.2
MAH 659.2
MAH 805.1
MAH 805.1
MAH 805.1
OSUM 50233.8
OSUM 50233.8
OSUM 50233.8
OSUM 50233.8
OSUM 33158
OSUM 33158
OSUM 33158
OSUM 47696.6
OSUM 47696.6
OSUM 47696.6
1984
1984
1984
1984
1984
1984
1984
OSUM:
OSUM:
OSUM:
OSUM:
OSUM:
OSUM:
OSUM:
MAH 626.1
MAH 626.1
MAH 626.1
MAH 896
MAH 896
MAH 896
MAH 896
MAH 896
MAH 896
OSUM 54521
OSUM 54521
OSUM 54521
OSUM 54521
OSUM 54521
OSUM 33163.13
OSUM 33163.13
OSUM 55465
OSUM 55465
OSUM 55465
OSUM 55465
OSUM 55465
OSUM 15738.2
OSUM 15738.2
OSUM 15738.2
OSUM 15738.2
OSUM
OSUM
OSUM
OSUM
OSUM
OSUM
1984:2
1984:2
1984:2
1984:2
1984:2
1984:2
202
198
206
174
180
178
180
174
175
183
184
187
172
174
180
168
168
240
229
240
235
245
233
241
73
72
72
246
246
241
241
241
126
122
120
113
111
206
208
227
223
217
200
195
190
205
219
207
211
254
261
258
197
210
205
210
199
243
230
230
240
222
224
229
218
234
324
319
325
310
325
319
325
82
80
83
289
293
289
291
291
182
187
175
171
170
371
371
386
386
375
356
347
348
257
263
257
265
107
106
105
101
110
94
89
95
92
88
85
89
95
100
97
85
88
96
93
89
88
93
88
91
87
96
91
89
35
35
30
109
110
102
105
105
110
42
45
50
42
47
41
40
104
96
109
96
106
106
103
100
109
105
112
115
110
105
110
32
44
29
41
34
26
43
40
27
29
40
25
24
43
21
26
42
21
31
14
15
11
41
45
41
11
21
20
40
35
25
41
32
23
32
33
28
32
32
24
23
50
20
27
37
24
29
35
23
23
51
22
10
11
11
8
32
47
30
36
42
32
34
43
28
37
39
34
13
16
13
14
10
35
42
19
38
42
29
37
41
28
37
37
29
34
30
[continued)
12
HOGGA
RTH
TABLE 2. (Continued)
Hinge
Ligament Length
Species
Catalog #
Length
Height
Length
Ante.
Cent.
Post.
V. ellipsiformis
MAH 947.2
230
280
110
39
40
31
MAH 947.2
102
34
40
28
MAH 947.2
226
287
105
MAH 947.2
226
287
103
MAH 947.2
104
34
38
32
MAH 947.2
223
284
102
V. trabalis
OSUM 9516.49
213
278
94
OSUM 9516.49
211
277
95
36
31
28
OSUM 9516.49
99
36
36
27
OSUM 9516.49
214
279
96
40
31
25
OSUM 9516.49
216
280
97
V. perpurpurea
OSUM 16262
165
241
88
V. villosa
OSUM 45943
246
300
114
OSUM 45943
245
298
105
OSUM 45943
246
296
110
35
42
33
OSUM 45947
250
308
116
OSUM 45947
248
304
108
OSUM 45947
240
307
110
OSUM 45947
241
308
113
43
37
33
V. vibex
OSUM 54631
230
298
117
OSUM 24124
231
302
120
OSUM 24124
232
304
102
29
44
29
OSUM 24124
231
302
117
OSUM 24124
239
299
105
32
45
28
OSUM 48628
231
300
109
OSUM 48628
225
297
105
OSUM 48628
224
299
103
V. i. iris
MAH 641.1
230
289
112
MAH 641.1
217
289
113
38
42
33
MAH 641.1
230
300
112
MAH 641.1
232
303
115
40
41
34
MAH 641.1
228
305
115
OSUM 55828.3
220
293
113
OSUM 55828.3
227
300
114
39
42
33
OSUM 55828.3
217
290
107
L t. teres
OSUM 36409
189
256
103
OSUM 36409
113
54
27
32
OSUM 36409
189
255
100
OSUM 36409
190
255
107
OSUM 36409
191
256
108
OSUM 36409
115
49
34
32
OSUM 36409
106
42
34
30
OSUM 36531.1
190
258
106
OSUM 36531.1
107
42
33
32
OSUM 51669.2
190
259
115
OSUM 51669.2
116
49
37
30
OSUM 51669.2
192
261
115
OSUM 51669.2
193
265
112
41
39
32
OSUM 51669.2
194
260
112
L. t. anodontoides
OSUM 35612
197
255
105
OSUM 35612
206
249
115
46
36
33
OSUM 35612
207
254
110
OSUM 35612
107
39
36
32
OSUM 41762.2
115
47
38
30
OSUM 41762.2
187
249
105
OSUM 41762.2
199
254
110
OSUM 41 762.2
204
249
118
50
37
31
OSUM 41762.2
194
249
118
46
42
30
GLOCHIDIA OF UNIONIDAE
13
TABLE 2. {Continued)
Species
Catalog #
Length
Height
Hinge
Length
Ligament Length
Ante. Cent. Post.
L. r. radíala
L. r. luteals
L. r. luteola
L. abrupta
L. higginsi
L. ovata
L. ornata
L. satura
L. ventricosa
MAH 897,1
MAH 897.1
MAH 897.1
MAH 897.1
MAH 321.6
MAH 321 .6
MAH 587.1
MAH 587.1
MAH 587.1
MAH 587.1
MAH 587.1
MAH 628.1
MAH 628.1
MAH 727.2
MAH 727.2
MAH 727.2
OSUM 13303
OSUM 13303
OSUM 13303
OSUM 13303
OSUM 13303
OSUM 38841
OSUM 38841
OSUM 49024.1
OSUM 49024.1
OSUM 49024.1
OSUM 49024.1
OSUM 43164.1
OSUM 43164.1
OSUM 43164.1
OSUM 43164.1
OSUM 54661
OSUM 54661
OSUM 54661
OSUM 38973.51
OSUM 38973.51
OSUM 38973.51
OSUM 38973.52
MAH 846.1
MAH 846.1
MAH 846.1
MAH 846.1
MAH 954.1
MAH 954.1
MAH 954.1
MAH 954.2
255
260
250
255
230
232
227
235
230
232
232
230
208
214
207
210
214
208
214
217
232
232
203
204
198
220
223
255
253
254
245
250
240
255
311
295
304
288
295
280
290
280
280
280
285
251
252
257
259
252
251
257
254
271
276
258
257
260
269
268
292
287
291
275
274
274
286
128
129
122
121
110
112
109
118
107
120
119
113
115
110
120
123
96
112
98
111
106
96
97
108
111
108
118
113
116
116
119
98
100
96
117
115
116
113
111
109
118
111
117
108
106
105
50
39
46
46
46
38
32
46
47
29
49
45
27
37
40
33
33
40
40
40
42
45
32
37
42
36
40
41
39
27
38
39
31
40
47
31
39
43
34
40
44
35
38
33
29
42
40
35
43
40
33
45
41
33
38
40
33
45
39
33
Subfamily Anodontinae
Anodonta cygnea (Linnaeus, 1758)
(Fig. 4A-F)
Material Examined
OSUM 20911.1 -"Maubroux (Genval) petit
etang près du lac", Belgium [étang = pool,
pond], 24 January 1949, W. Adam.
Description
Glochidium subtriangular, length 345 to 358
цт (351 ± 4.33 цт, n = 9), height 344 to 357
\xm (351 ± 4.41 [im, n = 9). Anterior margin
rounded, slightly more produced than poste-
rior margin, with maximum inflation at approx-
imately 50-60% from dorsal to ventral. Poste-
rior margin gently curved throughout its length,
14
HOGGARTH
FIG. 3. Glochidium of Unió elongatulus glaucinus. MAH 2055: A. exterior valve, bar length = 30 цт; В. inte-
rior valve, bar length = 30 |im; С styliform hook, bar length = 10 цт: D. styliform hook, bar length = 10 |дт;
E. exterior valve sculpture, bar length = 1 цт; F. hinge, bar length = 25 |.im.
producing a moderately asymmetric valve out-
line. Ventral terminus narrowly pointed, lo-
cated about 40% from posterior to anterior. Ex-
terior surface of valve malleated and pitted,
except in a narrow band along lateral margins.
Pit density reduced in the area of adductor
muscle scar (Fig. 4B, C). Loose-looped sculp-
turing covering exterior surface of valves (Fig.
4E). Central ligament 79 to 88 |.im (84 ± 3.27
j.im, n = 5) in length, centered about 43% from
posterior to anterior. Posterior ligament 68 to
79 \.ivn (73 ± 5.22 |.im, n = 5) in length; anterior
ligament 112 to 118|.im(115 ± 2.12мт,п = 5)
long. Hinge straight, 268 to 279 цт (274 ±
GLOCHIDIAOF UNIONIDAE
15
|HHK ' 'ЖУК
^- ■
■ *, '..jk
i *-
.^.„-я^,^^ , ,„.„^_.^,, ^„^
^ ' *.
•>. fc - *
s
FIG. 4. Glochidium of Anodonta cygnea. OSUM 20911.1; A. exterior valve, bar length = 55 jam; B. interior
valve, bar length = 50 |am; С interior valve pitting, bar length = 5 (im; D. styliform hook, bar length = 20 цт;
E. exterior valve sculpture, bar length = 1 цт; F. hinge, bar length = 40 \xm.
3.35 ).im, n = 1 1 ) in length. A styliform hook ex-
tends from ventral terminus as a broad trian-
gular plate. Lateral hook margins concave,
rapidly narrowed, producing a sharp point.
Microstylets lanceolate, sharply pointed, ar-
ranged in four row/s near ventral terminus, re-
duced to single row distally. Micropoints cov-
ering ventral terminus and lateral surfaces of
the hook, leaving narrow unsculptured band
along distal edge of the hook.
16
HOGGARTH
Remarks
The glochidium of A. cygnea can be distin-
guished from that of Unio by its shape, hook
structure, exterior valve sculpturing and size.
Length and height measurements of 350 дт
have been given for this glochidium by Harms
(1909), Haas (1910), and Ortmann (1912). It
lias been figured by Ortmann (1912: pi. 19,
fig. 2) and Wood (1974: figs. 3, 4).
Anodonta anatina (Linnaeus, 1758)
(Fig. 5A-G)
Material Examined
OSUM 20912.1, 20912.2 -"lxelles,etang",
Belgium [etang = pool, pond], 16 March 1950,
W. Adam.
Description
Glochidium subtriangular, moderately
asymmetric, length 350 to 361 |дт (357 ±
3.91 |.im, n = 7), height 348 to 361 |im (354 ±
4.46 цт, n == 7). Anterior margin broadly
curved near the dorsal margin, more moder-
ately curved ventrally. Posterior margin gently
and evenly curved throughout its length. Mal-
leations and pits covering exterior surface of
valve, except along the valve margin. Interior
surface uniformly pitted, without a noticeable
adductor muscle scar. Exterior valve sculptur-
ing ribbed loose-looped, consisting of raised
loops resthcted to bands that run parallel to
the dorsoventral axis of valve, separated by
unsculptured bands (Fig. 5C). Larval thread
present (Fig. 5B, E, F). Central ligament 86 to
96 |im (91 ± 5.00 |.im, n = 3) in length, cen-
tered about 43% from posterior to anterior.
Posterior ligament 57 to 60 (.im (59 ± 5.51
|im, n = 3) long; anterior ligament 114 to 125
|im (118 ± 5.86 |im, n = 3) long. Hinge
straight, with a length of 260 to 280 yim (269
± 6.37 цт, n = 7). Styliform hook with bicon-
cave lateral margins, forming sharp distal
point. Microstylets (about 15) lanceolate (Fig.
5G), sharply pointed, reduced to a single row
distally. Micropoints numerous, covering ven-
tral terminus and lateral surface of hook. Un-
sculptured distal hook margin narrow. Hook
located about 40% from posterior to anterior.
Remarks
This glochidium is distinguished from that of
A. cygnea by its unique exterior valve sculp-
turing. Giusti (1973) and Giusti et al. (1975)
were the first to demonstrate this unusual
sculptuhng. They refer to their species as A.
cygnea {Anodonta piscinalis). Anodonta pisci-
nalis was placed in the synonymy of A.
cygnea by Simpson (1900, 1914) and Ort-
mann (1912). Anodonta anatina has also
been synonymized with A. cygnea by Ort-
mann (1912), although it was treated as a
variant of A. cygnea by Simpson (1900,
1914).
There is little question that there are two
species of European Anodonta based on
glochidial characters. These species have
many characters in common, but their exterior
valve sculpturing is sufficiently different to
separate them. Glochidia of A. anatina were
figured by Flemming (1875: pi. 3, fig. 11),
Schierholz (1888: pi. 2, fig. 29), Ortmann
(1 91 2: pi . 1 9, fig. 3, as A. complanata), Giusti
(1973: figs. 13-25, as A. cygnea), and Giusti
et al. (1 975: figs. 10-21, as A. cygnea).
Anodonta beringiana Middendorff, 1851
(Fig. 6A-F)
Material Examined
OSUM 3711.1 - Outlet of Peper Lake,
Kenai Peninsula, Alaska, 13 August 1957, R.
Rausch.
Description
Glochidium subtriangular, moderately
asymmetric, length 286 to 292 ¡.im (289 ±
2.28 |.im, n - 5), height 284 to 293 |.im (290 ±
3.42 |.im, n = 5). Anterior and posterior mar-
gins about equal, more or less gently curved,
with anterior margin slightly more produced,
especially in the dorsal half of valve. Valve
outline only slightly asymmetric. Exterior sur-
face of valve malleated and pitted. Pits uni-
formly distributed, except along the valve
margin, where they are absent. Loose-looped
sculpture covering extehor surface of valve.
Larval thread present (Fig. 6A, C). Hinge
straight, length 209 to 214 цт (211 ± 2.35
(.im, n = 5). Posterior, central and anterior lig-
aments of a single specimen, 34 цт, 58 цт
and 1 1 8 |jm, respectively. Central ligament far
posterior, at 31% from posterior to anterior.
Styliform hook ahsing as a broad biconcave
triangular plate. Microstylets robust, multifac-
eted few in number (about 15). Micropoints
bluntly lanceolate, on ventral rim of valve,
along lateral edges of the microstylets and for
GLOCHIDIAOF UNIONIDAE
17
FIG. 5. Glochldium of Anodonta anatina: A. exterior valve, OSUM 20912.2, bar length = 55 цгл; В. interior
valve, OSUM 20912.2, bar length = 50 |am; С exterior valve sculpture, OSUM 20912.1, bar length = 1 цт;
D. styliform hook, OSUM 20912.2, bar length = 20 цт; E. larval thread, OSUM 20912.2, bar length = 25 цгл;
F. larval thread, OSUM 20912.2, bar length = 15 ¡am; G. microstylets, OSUM 20912.2, bar length = 5 цт.
HOGGARTH
FIG. 6. Glochidium of Anodonta beringiana, OSUM 3711.1: A. exterior valve, bar length = 45 цт; В. interior
valve, bar length = 50 |.im; С ventral valve view, bar length = 40 |.im; D. styliform hook, bar length = 10 |.im;
E. styliform hook, bar length = 10 \хт; F. styliform hook, bar length = 10 pm.
GLOCHIDIAOF UNIONIDAE
19
a short distance on lateral, surfaces of hook.
Unsculptured distal hook margin, wide. Hook
located approximately 41% from posterior to
anterior.
Remarks
This species was allied, on the basis of
adult shell morphology, to A. cygnea by Simp-
son (1914). The shape of this glochidium also
allies A. beringiana with A. cygnea. The glo-
chidium of A. beringiana can be distinguished
from that of A. cygnea by its smaller overall
size and hook structure. Inaba (1941) gave
296 |.im for the length and height of this glo-
chidium, and Cope (1959) reported length
and height measurement of 275 |im x 300 [im.
Anodonta lienneriyi Lea, 1 860
(Fig. 7A-F)
Material Examined
OSUM 52882.2, 52882.3 - "Griffin Creek,"
Snohomish Co., Washington, January 1972,
L. Gilbertson.
Description
Glochidium subtriangular, moderately
asymmetric, length 350 to 354 ).im (352 ±
2.08 |Lim, n - 3), height 340 to 350 |.im (344 ±
5.13 цт, n = 3). Anterior margin rounded, ta-
pering to meet a slightly curved posterior mar-
gin at ventral terminus. Ventral terminus nar-
rowly pointed, located about 40% from
posterior to anterior. Malleations and pits uni-
formly distributed on valve, except along
valve margin, where dorsally converging lon-
gitudinal ridges occur (Fig. 7C). Exterior valve
sculpturing intermediate between beaded and
loose-looped (Fig. 7E), resembling short
lengths of strung beads closely packed on
surface of valve. Hinge straight, 250 to 267
|.im (259 ± 7.31 ¡am, n = 6) in length. Central
ligament 84 to 89 \xvc\ (87 ± 2.07 цт, n = 6)
long, centered about 40% from posterior mar-
gin. Posterior ligament 55 to 65 цт (60 ± 3.25
цт, n = 6) in length; anterior ligament 106 to
114 |.im (111 ± 2.94 lam, n = 6) long. Styliform
hook sharply pointed, with fewer than 20 mi-
crostylets (Fig. 7B, D). Microstylets lanceo-
late, sharply pointed, arranged in four proxi-
mal rows, reduced to a single row distally. Two
to three microstylets, and as many as seven
micropoints, forming a cluster near tip of
hook. Additional micropoints covering lateral
surfaces of hook and along lateral borders of
microstylets, leaving narrow unsculptured dis-
tal hook edge.
Remarks
This species was also allied to A. cygnea by
Simpson (1914) due to similarity in adult shell
morphology. The shape of this glochidium is
also similar to that of A. cygnea, although this
species can be distinguished by its exterior
surface sculpturing and the cluster of mi-
crostylets and micropoints near the point of
the hook. This aspect of the styliform hook re-
sembles that of Pyganodon Cataracta and re-
lated species east of the Rocky Mountains.
Anodonta implicata Say, 1 829
(Fig. 8A-F)
Material Examined
OSUM 52463.7 - Great Herring Pond, S
shore by the Herring River, 0.8 mi. N of
Bournedale, 1.6 mi. WNW of Sagamore,
Barnstable Co., Massachusetts, 4 October
1982, D. H. Stansbery & K. E. Wright.
Description
Glochidium lachrimiform, asymmetric,
length 342 to 343 цт (343 ± 0.71 цт, n = 2),
height 345 to 350 |.im (348 ± 3.54 цт, n = 2).
Posterior margin gently curved, anterior mar-
gin subrotund. Maximum inflation of anterior
margin at about 50% from dorsal to ventral,
broadly rounded ventral terminus occuring
about 35% from posterior to anterior. Exterior
valve surface finely malleated and pitted, ex-
cept at valve margin. Pits uniformly distributed
in the malleated surface, exterior valve sur-
face with sculpturing intermediate between
beaded and loose-looped (Fig. 8F). Hinge
straight, 160 to 166 цт (163 ± 3.06 цт, n -
3) long. Central ligament 74 to 77 \лт (76 ±
2.12 |im, n = 2) in length, centered about 46%
from posterior border of hinge. Posterior liga-
ment 37 to 38 цт (38 ± 0.71 цт, n = 2) long;
anterior ligament 48 to 52 цт (50 ± 2.36 ¡лт,
n = 2) long. Hook styliform, broadly triangular,
gradually tapered to blunt point (Fig. 8C). Mi-
crostylets lanceolate, gradually increasing in
size toward center, arranged in four rows near
proximal border of hook. Micropoints numer-
ous, covering lateral surfaces of hook. Un-
sculptured distal hook margin, narrow.
20
HOGGARTH
FIG. 7. Glochidium of Anodonta kennerlyi:A. exterior valve, OSUM 52882.2, bar length = 50 цт; В. interior
valve, OSUM 52882.2, bar length = 50 |.im; С lateral view, OSUM 52882.3, bar length = 55 цт: D. styliform
hook, OSUM 52882.2, bar length = 1 5 цгл; E. exterior valve sculpture. OSUM 52882.3. bar length = Uim; F.
hinge, OSUM 52882.3, bar length = 35 цт.
GLOCHIDIAOF UNIONIDAE
21
FIG. 8. Glochidium of Anodonta implicata. OSUM 52463.7; A. exterior valve, bar length = 50 |im; B. interior
valve, bar length = 45 |im; С styliform hook, bar length = 1 ¡xm; D. valve pitting, bar length = 1 5 |.im; E. hinge,
bar length = 25 цт; F. exterior valve sculpture, bar length = 2 цгл.
22
HOGGARTH
Remarks
Johnson (1946) describes this glochidium
as, "typical of the genus Anodonta," even
though his figure shows the short hinge line
and greatly inflated anterior margin, both char-
acters far from typical for this genus. Rand &
Wiles (1 982) have also failed to recognize that
this species can be distinguished from all other
glochidia simply by the tear-drop outline of the
valve. This glochidium is further distinguished
by its hook structure and exterior valve sculp-
ture. The glochidium of this species is figured
by Johnson (1 946: pi. 1 6, fig. 3), Wiles (1 975:
figs. 1 , 2, as A. ¡mplicataand A. Cataracta), and
Rand & Wiles (1982: figs. 5-8). Rand & Wiles
gave length and height measurements of 345
|im X 345 |.im.
Anodonta suborbiculata Say, 1831
(Fig. 9A-F)
Material Examined
OSUM 13634 - Black River at U.S. Rt. 67
bridge W of Hendrickson, Butler Co., Mis-
souri, 14 October 1964, С В. Stein.
Description
Glochidium subtriangular, length 323 to 328
|im (325 ± 2.08 \хт, n = 5), height 320 to 328
).im (323 ± 3.46 [ivn, n = 5). Dorsal margin
straight, 231 to 237 ¡.im (232 ± 2.41 цт, n =
5) long. Posterior margin gently curved; an-
terior margin, broadly curved. Maximum an-
terior inflation at about 50% from dorsal to
ventral. Ventral terminus, narrowly pointed, lo-
cated about 42% from posterior to anterior.
Exterior surface malleated, pitted (Fig. 9E),
except along the edge of valve (Fig. 9C); fine
structure of exterior surface consisting of fine
non-overlapping lines referred to here as ver-
miculate sculpturing (Fig. 9F). Central liga-
ment 98 to 1 02 цт (1 00 ± 1 .71 цт, n = 4) in
length, centered about 45% from posterior to
anterior (Fig. 9B). Posterior ligament 52 to 56
|im (54 ± 1.83 [xvr\, n = 4) long; anterior liga-
ment, 74 to 81 fim (78 ± 2.99 дт, n = 4) long.
Hook styliform, with about 40 lanceolate mi-
crostylets and many micropoints. Unsculp-
tured distal margin of hook narrow.
Remarks
This glochidium can be distinguished by its
exterior valve sculpturing and hook. It resem-
bles A. cygnea in shape but it has no other
characters to tie it to that species. Surber
(1915) stated that in, "general outline [the
glochidium of] suborbiculata closely resem-
bles Anodonta grandis but may be distin-
guished by its smaller size". He gave mea-
surements of 325 цт X 320 ¡am for length and
height. However, the glochidia of these spe-
cies are only superficially similar, and a close
relationship between this species and any
other member of the genus is not supported by
glochidial characters. The glochidium of A.
suborbiculata is figured by Surber (1 91 5: pi. 1 ,
fig. 1 ) and Utterback (191 5-1 91 6: fig. 7).
Pyganodon grandis grandis (Say, 1 829)
(Fig. 10A-F)
Material Examined
P. g. grandis: OSUM 38467.10 - Miami
River, R.Mi. 82.4, at 1-75 bridge at Dayton, just
above mouth of Mad River, Harrison Twp.,
Montgomery Co., Ohio, 6 February 1976, D.
H. Stansbery. MAH 668-Olentangy River
below Fifth Ave. bridge near Ohio State Uni-
versity main campus, Columbus, Franklin Co.,
Ohio, 30 September 1984, K. Wright & K. Gal-
lant. INHS 2247-Kankakee River at Kanka-
kee, below hydroelectric plant, Kankakee Co.,
Illinois. 11 October 1985, J. M. Kasprowicz. P.
g. corpulenta: OSUM 47890-Stonelick Creek
at Stonelick Reservoir, 1.1 mi. SW of Edenton,
2.8 mi. N of Newtonville, Wayne Twp., Cler-
mont Co., Ohio, 1 October 1978, D. H. Stans-
bery & K. G. Borror. OSUM 53653-Ohio River
bank, R.Mi. 442.8-443.0, 0.3-0.4 mi. NW of
Moscow, 2.2-2.4 mi. S of Point Pleasant, 6.9
mi. SE of New Richmond, Clermont Co., Ohio,
22 October 1984, K. E. Wright et al.
Description
Glochidium subtriangular, asymmetric,
length 350 to 365 f.im (356 ± 5.56 цт, n = 7),
height 350 to 360 |.im (355 ± 3.10 цт, n = 7).
Posterior margin gently curved throughout its
length. Anterior margin broadly curved to its
point of maximum inflation at about 70% from
dorsal to ventral. Ventral terminus broadly
rounded, located about 40% from posterior to
anterior. Valve surface malleated and pitted,
except along its margin, where dorsally con-
verging longitudinal ridges are found (Fig.
IOC). Adductor muscle scar not evident.
Coarse loose-looped exterior valve sculpture
covering surface of valve (Fig. 10F). Hinge
GLOCHIDIAOF UNIONIDAE
23
FIG. 9. Glochidium of Anodonta suborbiculata. OSUM 13634; A. exterior valve, bar length = 40 цт; В. inte-
rior valve, bar length = 40 |дт; С. lateral view, bar length = 40 |am: D. styliform hook, bar length = 10 |im; E.
valve pitting, bar length = 10 |am; F. exterior valve sculpture, bar length = 2 ¡am.
24
HOGGARTH
FIG. 10. Glochidium of Pyganodon д. grandis. OSUM 38467.10: A. exterior valve, bar length = 55 (.im: B. in-
terior valve, bar length = 55 |.im; С lateral view, bar length = 55 |.im: D. styliform hook, bar length = 15 |.im;
E. interior valve pitting, bar length = 15 цт; F. exterior valve sculpture, bar length = 1 цт.
GLOCHIDIAOF UNIONIDAE
25
straight, 250 to 259 цт (253 ± 3.20 цт, n =
7) in length. Central ligament 90 to 94 ¡.im (92
± 2.08 |.im, n = 3) long, centered between
41 -42% from posterior margin. Posterior liga-
ment 54 to 62 цт (57 ± 4.04 |.im, n = 3) long;
anterior ligament 104 to 111 (.im (107 ± 3.61
j.im, n = 3) in length. Hook styliform, with about
20 microstylets. Microstylets arranged in four
proximal rows, reduced to a single row distally
(Fig. 10B, D). Five to six sharply pointed mi-
crostylets forming a cluster near tip of hook.
Micropoints few in number, found only along
ventral rim of valve and along borders of mi-
crostylets. Unsculptured distal hook margin
very wide.
The glochidium of P. g. corpulenta is nearly
identical to that of P. g. grandis, with a length
of 343 to 350 цт (348 ± 5.13 цт, n - 3), a
height of 363 to 368 цт (366 ± 3.54 цт, n =
3), and a hinge length of 260 to 275 цт (266
± 7.94 j.im, n = 3). The posterior, central and
anterior ligaments of a single specimen were
70 i-im, 93 цт and 100 (.im in length, respec-
tively. Surber figured this glochidium (1912:
fig. 4; 1913: fig. 1 ) and gave length and height
measurements of 350 цт.
Remarks
Ortmann (1912) gave 360 цт x 370 цт for
length and height of the glochidium of P. g.
grandis, and Tucker (1928) reported the fol-
lowing ranges: length, 350-398 цт; height,
343-390 цт. However, Surber gave measure-
ments of 410 |im X 420 |.im. This extremely
large range in size led Tucker (1928) to sug-
gest that this was the result of examining the
glochidia of a far ranging, variable species.
Her material, Ortmann's and mine came from
small streams, whereas Surber's material
probably came from the Mississippi River and
probably represents P. g. gigantia. The glo-
chidium of P. g. grandis was figured by Lea
(1858: pi. 5, figs. 32, 33, 34, as A. lewisii. A.
ovata and A. decora = P. g. grandis, fide Ort-
mann, 1919) and Surber (1912: pi. 3, fig. 45).
Lea's figures are slightly different from each
other, and none show the correct outline.
Surber's figure demonstrates the asymmetri-
cal valves of this glochidium.
Pyganodon Cataracta Cataracta (Say, 1817)
(Fig. 11A-E)
Material Examined
P с Cataracta; OSUU 52462.27, 52462.35
- Great Herring Pond, S Shore by the Herring
River, 0.8 mi. N of Bournedale [1.6 mi. WNW
of Sagamore], Barnstable Co., Massachu-
setts, 4 October 1984, D. H. Stansbery & K.
Wright. P с marginata - OS UM 38962.6
"Black Moshannon Lake" [near or part of
Black Moshannon Creek?], Snowshoe Twp.,
Centre Co., Pennsylvania, September 1975,
D. S. Gussman.
Description
Glochidium subtriangular, asymmetric,
length 374 to 380 цт (376 ± 2.61 |im, n = 5),
height 351 to 370 цт (363 ± 7.40 цт, n = 5).
Dorsal margin straight; posterior margin gen-
tly, evenly curved; anterior margin broadly
curved to its maximum inflation at about 70%
from dorsal to ventral. Ventral terminus
broadly pointed, about 43% from posterior to
anterior. Malleations and pits uniformly dis-
tributed on the valve, except along valve mar-
gin. Coarse loose-looped sculpture covering
exterior surface of valve (Fig. 11 D). Hinge
straight, 277 to 291 |im (284 ± 5.92 цт, n =
5) long. Central ligament 98 to 1 02 |.im (1 00 ±
2.83 }.im, n = 2) in length, centered about 42%
from posterior to anterior. Posterior ligament
71 to 75 |im (73 ± 2.83 (лт, n = 2) in length;
anterior ligament, 115 to 116 цт (116 ± 0.71
|im, n = 2) long. Hook styliform, with about 20
lanceolate microstylets (Fig. 11 E). Micro-
stylets arranged in four rows near ventral ter-
minus, reduced to a single row distally, clus-
tered near tip of hook (four to six per hook).
Micropoints limited to proximal half of hook,
except along lateral margins of microstylets,
where they form a single row, leaving wide un-
sculptured distal hook margin.
The glochidium of P c. marginata had a
length of 358 to 360 цт (359 ± 1 .41 цт, n =
2), a height of 364 to 369 цт (367 ± 3.54 |im,
n = 2), and a hinge length of 276 to 277 ¡im
(277 ± 0.71 цт, n = 2). The posterior, central
and anterior ligaments of a single specimen
were 64 цт, 1 1 цт and 1 02 цт in length, re-
spectively.
Remarks
Ortmann (1912) believed this species was
the eastern representative of the P grandis
complex. The similarities between the glo-
chidia of these two species support the view
of close relationship; however, the glochidium
of P. Cataracta can be distinguished from
that of P grandis by its longer central liga-
ment. The glochidia of both P с Cataracta and
P. с marginata have longer anterior ligaments
26
HOGGARTH
FIG. 11. Glochidium of Pyganodon с Cataracta: A. exterior valve, OSUM 52462.27, bar length = 55 |.im; B.
interior valve, OSUM 52462.35, bar length = 55 цгл: С. hinge, OSUM 52462.27, bar length = 40 |.im; D. ex-
terior valve sculpture, OSUM 52462.27, bar length = 2 |.im; E. styliform hook, OSUM 52462.27, bar length =
20|am.
than any species in the Anodontinae. Rand &
Wiles (1982) gave length and height mea-
surement of 382 цт X 383 |.im, whereas Ort-
mann (1912) reported 360 |.im x 370 цт
(identical to his figures for P. g. grandis). This
glochidium is figured by Lefevre & Curtis
(1910: fig. C; 1912: fig. 1С), Calloway &
Turner (1979: pi. 3, figs. 1, 3, 5, 7, 8), Wiles
(1975: figs. 3, 7), and Rand & Wiles (1982:
figs. 1-4).
Pyganodon doliaris (Lea, 1863)
(Fig. 12A-D)
Material Examined
OSUM 26405 - Potomac River along W
shore Theodore Roosevelt Island, W. edge
Washington D.C., just E of Arlington, Va., Dis-
thct of Columbia. 22 February 1970, K. Hef-
felfinger Olson.
GLOCHIDIAOF UNIONIDAE
27
FIG. 12. Glochidium of Pyganodon doliaris. OSUM 26405; A. exterior valve, bar length = 55 jam; B. interior
valve, bar length = 55 |.im; С exterior valve sculpture, bar length = 1 цт; D. styliform hook, bar length = 20
|im.
Description
The glochidium of this species is essentially
identical to that of P. g. grandis and P. с
Cataracta. Its shape and hinge structure, ex-
terior valve sculpture (Fig. 12C) and hook
structure (Fig. 12D) place it firmly with these
species. One glochidium measured 31 7 цт x
317 цт X 240 |.im (length x height x hinge
length) (Fig. 12B), while another measured
405 |im x 368 urn x ЯП.Я nm (F\a. 12A). Both
glochidia were removed from the marsupium
of the same female.
Remarks
With such a limited amount of material and
such great variability in size, it would not be
appropriate to suggest limits. However, the
larger measurements seem out of range for
glochidia in this group, with the exception of P.
g. gigantia.
28
HOGGARTH
Utterbackia imbecillis (Say, 1 829)
(Fig. 13A-F)
Material Examined
OSUM 9436.2 - Lake Erie off East Sister Is-
land, Ontario, Canada, 13 July 1960, D.
Mount. OSUM 34463 - Clear Fork Mohican
River at Clear Fork State Park Campground
B, 5 mi. NNW of Jelloway, Hanover Twp., Ash-
land Co., Ohio, 13 July 1973, E. W. Tittsler.
MAH 435 - Hellbranch Run at Bausch Rd.
(Co. Rt. 1 1 ) bridge, 0.7 mi. S of Galloway, 1 0.0
mi. SW of Columbus, Prairie Twp., Franklin
Co., Ohio, 15 June 1983, M. A. Hoggarth.
UWZY 24971.1 -Baraboo River, Til N, R6E,
Sec. 1, NW 1/4 of NE 1/4, 25 meters down-
stream of dam at Baraboo Sauk Co., Wiscon-
sin, 12 October 1984, D. J. Heath.
Description
Glochidium subtriangular, length 291 to 313
).im (304 ± 6.41 |.im, n = 1 2), height 289 to 306
цт (300 ± 6.11 |.im, n = 12). Dorsal margin
straight 240 to 256 цт (246 ± 5.97 )im, n =
13) long. Anterior and posterior margins
slightly and evenly curved. Anterior margin
only slightly more produced than posterior
margin, producing a slightly asymmetric valve
outline. Surface of valve weakly malleated
and pitted, except along smooth valve margin.
Surface of valve covered with a uniform
looped pattern, referred to here as tight-
looped sculpture (Fig. 13F). Central ligament
85 to 93 (.im (91 ± 3.50 цт, n = 4) long, cen-
tered about 42% from posterior border of
hinge. Anterior ligament 95 to 103 цт (100 ±
3.79 jim, n = 4) in length; posterior ligament
60 to 63 |.im (61 ± 1 .50 |.im, n = 4) long. Hook
styliform, arising as a broad triangular plate
from ventral terminus. Microstylets lanceolate
near distal margin of hook, multifaceted near
proximal border (Fig. 13C, D), arranged in
three to four rows near ventral terminus, re-
duced to a single row distally. A cluster of mi-
crostylets and micropoints near point of hook
present in some material (Fig. 13C); however,
most micropoints cover proximal half of lateral
surface of hook, leaving wide unsculptured
distal margin. Ventral terminus located about
45% from posterior to anterior.
Remarks
This glochidium can be distinguished from
others in the subfamily by its exterior valve
sculpture, its rather small size, and its rela-
tively long central ligament. It resembles A.
suborbiculata in these last two characters, but
differs in valve sculpture and hook structure.
The outline of this glochidium is like that of A.
cygnea, whereas the hook allies this species
to P. grandis. Ortmann (1912) gave length
and height measurements of 300 цт x 310
).im (for A. imbecillis) and 290 ¡.im x 300 цт
(for /A. henryana. = U. imbecillis, fide Johnson,
1 970). Surber (1912) gave 31 ¡im x 290 |im,
and Tucker (1 927) reported 290 |am x 300 ¡im.
This glochidium was figured by Lea (1858: pi.
5, fig. 36), Ortmann (1911: pi. 89, fig. 13),
Surber (1912: pi. 1 , fig. 2), and Tucker (1927:
pi. 10, figs. 1,2).
Anodontoides ferussacianus (Lea, 1834)
(Fig. 14A-E)
Material Examined
OSUM 18275.2 - South Branch Phelps
Creek at Rt. 322 bridge, 1.5 mi. E of Hunts-
burg, 5.5 mi. NE of Middlefield, Huntsburg
Twp., Geauga Co., Ohio, 23 March 1966, R.
E. Jezerinac. MAH 989.4 - Silver Creek, R.M.
4.6, at St. Rt. 15 bridge, 1.5 mi. N of Pioneer,
8.4 mi. NNE of Montpelier, T9S, R2W, Sec.
8/9, Madison Twp., Williams Co., Ohio, 2 Oc-
tober 1986, M. A. Hoggarth.
Description
Glochidium subtriangular, only slightly
asymmetric, length 319 to 326 цт (323 ±
3.20 цт, n = 7), height 320 to 327 цт (324 ±
2.64 ).im, n = 7). Hinge 231 to 238 цт (234
± 2.93 |.im, n = 7) in length. Dorsal margin
straight. Posterior and anterior margins about
equally curved to a bluntly rounded ventral
terminus. Exterior surface malleated and pit-
ted, except along lateral margins and within a
circular dorsomedial area or umbo (Fig. 14A,
C). Exterior surface of valve covered by
loose-looped sculpture (Fig. 14E). Central lig-
ament 88 to 92|.im (91 ± 2.31 |.im, n = 3) long,
centered about 42% from posterior to anterior.
Anterior ligament 83 to 89 pm (86 ± 3.06 цт,
n = 3) in length; posterior ligament 56 to 62
(.im (59 ± 3.06 ^m, n = 3) long. Hook styliform,
with two rows of microstylets extending dis-
tally for at least three quarters of its length, lo-
cated about 50% from posterior to anterior. A
cluster of microstylets and micropoints occur-
ring at distal end of hook. Micropoints located
GLOCHIDIAOF UNIONIDAE
29
FIG. 13. Glochidium of Utterbackia imbecillis: A. exterior valve, UWZY 24971 .1 , bar length = 50 ¡.im; B. inte-
rior valve, OSUM 9436.2, bar length = 60 |am; С styliform hook, UWZY 24971 .1 , bar length = 1 5 |am; D. styli-
form hook, MAH 435, bar length = 1 5 ).im; E. hinge, UWZY 24971 .1 , bar length = 2 |.im; F. exterior valve sculp-
ture, UWZY 24971 ,1 , bar length = 2 цт.
30
HOGGARTH
FIG. 14. Glochidium of Anodontoides ferussacianus:A. exterior valve. MAH 989.4, bar length = 50 цт; В. in-
terior valve, MAH 989.4, bar length = 50).im;C. hinge ligament. OSUM 18275.2, bar length = 30цт: D. styli-
form hook, MAH 989.4, bar length = 25 |дт; E. exterior valve sculpture, MAH 989.4, bar length = 1 цт.
GLOCHIDIAOF UNIONIDAE
31
on ventral rim of valve and on proximal sur-
faces of hook, leaving wide unsculptured dis-
tal hook margin.
Remarks
The glochidium of Anodontoides ferussa-
cianus can be distinguished from above-de-
scribed members of the Anodontinae by its
double row of microstylets distally and its
symmetry. The hook of this glochidium re-
sembles that of Strophitus, but the outline of
its valve is essentially like that of Anodonta.
Ortmann (1912) and Surber (1912) gave the
following size range for the glochidium of A.
ferussacianus: 320-330)дт x 320-330 цт.
This glochidium has been figured by Lea
(1858: pi. 5, fig. 35), Ortmann (1911: pi. 89,
fig. 12), and Surber (1912: pi. 13, fig. 43). Lea
figured this glochidium without hooks, but Ort-
mann (1919) suggests that Lea's specimens
were immature.
Simpsonaias ambigua (Say, 1 825)
(Fig. 15A-F)
Material Examined
UWZY 22658, 22662, 22672 - Wisconsin
River, T8N, RIB, Sec. 5, 9.5 mi. S. of Rich-
land Center, Richland Co., Wisconsin, 1 6 July
1984, D. J. Heath. OSUM 55995 - Wisconsin
River, T8N, RÍE, Sec. 5, N side of river, 100
meters upstream of public boat landing, 9.5
mi. S. of Richland Center, Richland Co., Wis-
consin, 20 April 1 985, D. J. Heath.
Description
Glochidium ovate subtriangular, slightly
asymmetric, length 250 to 258 |im (255 ±
3.39 (im, n = 6), height 256 to 265 (im (261 ±
3.43 (im, n = 6). Dorsal margin straight, 164
to 170 (im (168 ± 2.22 (im, n = 7) in length.
Posterior margin gently and evenly curved.
Anterior margin slightly more rounded than
posterior margin. Maximum -anterior inflation
occurring near 60% from dorsal to ventral.
Ventral terminus bluntly rounded, about 45%
from posterior to anterior. Exterior surface of
valve finely malleated, uniformly pitted, ex-
cept along smooth valve margin and umbo.
Exterior surface sculpture beaded (Fig. 15E).
Central ligament 54 to 63 (.im (58 ± 4.51 (im,
n = 3) long, centered about 45% from poste-
rior to anterior. Posterior ligament 45 to 49 (im
(48 ± 2.31 (.im, n = 3) long; anterior ligament
61 to 65 (im (63 ± 2.00 (Lim, n = 3) in length.
Hook styliform, arising as a broad biconcave
triangular plate from ventral terminus, taper-
ing to a sharp point. Micropoints grading more
centrally located microstylets. Proximal mi-
crostylets multifaceted, sharply pointed, ar-
ranged four to five abreast. Distal microstylets
and micropoints lanceolate. Micropoints lo-
cated on ventral terminus and lateral surfaces
of hook, leaving narrow unsculptured distal
hook margin.
Remarks
The glochidium of S. ambigua has only
rarely been available for description. Lea
(1858) described the glochidium of Margari-
tanaliildrettiiana{^ S. ambigua, f /сУе Simpson,
1900) as subrotund, with a straight or slightly
incurved dorsal line and lacking hooks. He
suggested, however, that hooks might be
present in more mature specimens. Howard
(1915, 1951) described the glochidium as
triangular, with well-developed hooks, and
Clarke (1985) added that the glochidium is
slightly asymmetric with malleated surfaces.
Clarke's description, however, is of Howard's
figure rather than of glochidia he examined.
This glochidium can be distinguished from
most others by its shape, and from that of the
other Anodontinae by its size, hook structure
and exterior valve sculpturing. In regards to
valve symmetry, size, hook structure, hinge
structure and exterior valve sculpture, this
glochidium is reminiscent of Unio. The
glochidium of S. ambigua was figured by Lea
(1 858: pi. 5, fig. 31 ), and Howard (1 951 : figs,
4a, b). Lea's figure is much too round and
lacks hooks. Howard's figures show the out-
line of this glochidium correctly, but the draw-
ings lack microstylets on the styliform hook.
Howard's figures are reprinted in Clarke
(1985: fig. 20d).
Strophitus undulatus undulatus (Say, 1817)
(Fig. 16A-F)
Material Examined
S. u. undulatus -OSUM 49443 Mississippi
River, R.Mi. 635.0, East Channel across from
Prairie du Chien, N tip of island above U.S. Rt.
1 8 bridge, Crawford Co., Wisconsin, 21 March
32
HOGGARTH
FIG. 15. Glochidium of Simpsonaias ambigua: A. exterior valve, UWZY 22658, bar length = 40 |.im; B. inte-
rior valve, UWZY 22662, bar length = 40 цт; С, lateral view, UWZY 22662, bar length = 45 цт: D. styliform
hook, UWZY 22658, bar length = 10 цт; E. exterior valve sculpture, OSUM 55995, bar length = 2 цт; F.
hinge, UWZY 22672, bar length = 20 pm.
GLOCHIDIAOF UNIONIDAE
33
FIG. 16. Glochidium of Strophitus u. undulatus:A. exterior valve, OSUM 49443, bar length = 50 цт; В. inte-
rior valve, OSUM 52458.4, bar length = 50 urn; С lateral view, OSUM 52458.4, bar length = 70 |im; D. styli-
form hook, OSUM 52458.4, bar length = 20 urn: E. exterior valve sculpture, OSUM 49443, bar length = 1 |дт;
F. microstylets, OSUM 52458.4, bar length = 10 цт.
34
HOGGARTH
1981, M. E. Havlik et al.. OSUM 52458.4 -
Ashuelot River 0.4 mi. S of Surrey Mountain
Dam, 4.3 mi. NNW of Keene, Cheshire Co.,
New Hampshire, 22 August 1 982, K. Wright &
J. LeBlanc. MAH 792.1 - Fish Creek 0.4 mi.
above its mouth at Co. Rt. 49 bridge, 1.1 mi. N
of Edgerton, 10.4 mi. W of Bryan, St. Joseph
Twp., Williams Co., Ohio, 29 October 1 986, D.
H. Stansbery et al. S. и. tennesseensis -
OSUM 33381.2 Laurel Creek 0.4 mi. N of
Bradford along Va. Rt. 91, 6.3 mi. NE of
Saltville, Rich Valley District, Smyth Co., Vir-
ginia, 29 September 1971, D. H. Stansbery &
W. J. Clench. OSUM 55449 - Clinch River,
R.Mi. 270.6-270.9, 0.7-1 .0 mi. SW of Cleve-
land, 1.5-1.8 mi. NE of Carbo, Russell Co.,
Virginia, 3 October 1985, G. T. Waiters.
Description
Glochidium depressed pyriform, asymmet-
ric, length 360 to 369|.im (363 ± 5.20 ,um, n -
3), height 289 to 299 |.im (295 ± 5.29 цт, n =
3). Dorsal margin straight, 271 to 281 цт (278
± 3.78 |jm, n = 6) in length. Posterior margin
broadly arcuate. Anterior margin almost
round, meeting posterior margin at a slightly
rounded and ventrally produced, nipple-like,
ventral terminus. Ventral terminus located
about 44% from posterior to anterior. Exterior
valve surface malleated and pitted. Pits uni-
formly distributed throughout valve, except
along valve margin and at umbo. Coarse
loose-looped sculpture covering exterior sur-
face of valves (Fig. 16E). Central ligament 87
to 103 цт (95 ± 6.55 цт, n = 4) in length, lo-
cated about 44% from posterior margin. Pos-
terior ligament 70 to 78 |.im (74 ± 3.30 цт, n
= 4) long; anterior ligament 101 to 110 цт
(107 ± 4.08 |дт, n = 4) long. Hook styliform,
covered with about 30 microstylets and nu-
merous micropoints. Proximal microstylets
bluntly pointed. Distal microstylets, multifac-
eted, sharply pointed (Fig. 16F). Two rows of
microstylets extend distally, and a few mi-
crostylets and micropoints forming a cluster
near sharply pointed terminus of hook. Un-
sculptured distal hook margin narrow.
The glochidium of S. u. tennesseensis is
identical to that described above. One glo-
chidium had the following dimensions: length,
346 |.im; height, 298 (.im; hinge length, 268
jjm; central ligament length, 94 |дт; posterior
ligament length, 78 ¡im; anterior ligament
length, 96 jjm.
Remarks
The glochidium of Strophitus is depressed-
pyriform and possesses a styliform hook at
the ventral terminus of each valve. Lea (1 858)
described the glochidia of Strophitus edentula
(= S. undulatus, ^/de Johnson, 1970) and S.
undulata as subthangular, with a long straight
dorsal margin, inflated side margins (Lea er-
roneously viewed all glochidia as symmetrical
about the dorsoventral axis), and a large hook
with four rows of "granules" proximally, re-
duced to two rows distally. Other than shape.
Lea's description is surprisingly accurate.
The glochidium of S. u. undulatus is unlike
any so far described. Its shape, hook structure
and coarse looped sculpture will distinguish it
from other species examined. Glochidia of S.
u. undulatus have been figured by Lea (1858:
pi. 5, fig. 37, as S. edentula, pi. 5, fig. 38, as S.
undulatus), and Surber (1912: pi. 1, fig. 3, as
S. edentula). Surber gave length and height
measurements of 350 цт x 285 |jm, and Ort-
mann (1 91 2) gave 360 цт x 300 |Lim.
Strophitus subvexus (Conrad, 1834)
(Fig. 17A-F)
Material Examined
OSUM 36240 - Buttahatchie River about
0.5 mi. above its mouth, 12 mi. NNW of
Columbus, T16S, R19W, Lowndes Co., Mis-
sissippi, 4 October 1974, R. Grace et al.
Description
Glochidium depressed pyriform, asymmet-
ric, length 348 to 359 цт (354 ± 5.51 цт, n =
3), height 288 to 292 ,um (290 ± 2.08 цт, n =
3). Dorsal margin straight, 271 to 277 цт (274
± 2.50 jam, n = 4) in length. Posterior margin
broadly curved; anterior margin rounded. Lat-
eral margins meeting at a narrowly rounded,
nipple-like ventral terminus located about
50% from posterior to anterior. Surface of
valve coarsely malleated, uniformly pitted, ex-
cept along valve margin and at umbo (Fig.
17B, C). Coarse loose-looped sculpture cov-
ering the exterior surface of valves (Fig. 1 7E).
Central ligament about 84 |Lim in length, cen-
tered about 40% from posterior to anterior.
Posterior ligament about 70 цт long; anterior
ligament about 123 цт long. Hook styliform,
with proximal microstylets bluntly pointed dis-
GLOCHIDIAOF UNIONIDAE
35
FIG. 17. Glochidium of Strophitus subvexus. OSUM 36240: A. exterior valve, bar length = 50 |.im; B. interior
valve, bar length = 50 |.im; С exterior valve, bar length = 50 цт; D. styliform hook, bar length = 20 цт; E. ex-
terior valve sculpture, bar length = 2 |.im; F. hinge, bar length = 35 ).im.
36
HOGGARTH
tal microstylets lanceolate, multifaceted and
sharply pointed. Microstylets arranged in a
double row distally, forming a cluster near
point of hook. Micropoints located on ventral
valve rim and on lateral surfaces of hook,
leaving narrow unsculptured distal hook mar-
gin.
Remarks
The glochidium of S. subvexus resembles
that of S. u. undulatus, except the former has
a roundish outline and a more centrally posi-
tioned ventral terminus. The near symmetrical
outline of this glochidium is overemphasized
slightly by Lea (1874: PI. 21, fig. 15, as Mar-
gan tana spillmanii. = S. subvexus, f/de John-
son, 1967), but his figure correctly shows the
broadly curving margins.
Alasmidonta wr/d/s (Rafinesque, 1820)
(Fig. 18A-F)
Material Examined
OSUM 47518 - Horse Lick Creek, 0.3 mi.
below mouth of Raccoon Creek at Dango, 7.6
mi. SW of Mckee, Jackson Co., Kentucky, 28
February 1980, S. Call et al.
Description
Glochidium depressed pyriform, only
slightly asymmetric, length of 300 to 319 \.^m
(307 ± 7.59 |im, n - 7), height 245 to 260 цт
(251 ± 4.35 ¡am, n = 7). Dorsal margin
straight, 245 to 258 |.im (250 ± 3.93 ,um, n =
7) long. Posterior margin gently and evenly
curved; anterior margin only slightly more pro-
duced than posterior margin. Ventral terminus
roundly pedicellate, about 45% from posterior
to anterior. Exterior surface of valve finely
malleated and pitted, except along valve mar-
gin (Fig. 18C) and at umbo. Pit density re-
duced in area of adductor muscle scar (Fig.
1 8F), and loose-looped sculpture covering the
exterior surface of valve (Fig. 18E). Central
ligament 73 to 81 цт (76 ± 4.1 6 |.im, n = 3) in
length, centered about 42% from posterior to
anterior. Postehor ligament 65 to 71 цт (68 ±
4.04 цт, n = 3) long; anterior ligament 100 to
109|.im(105 ± 4.93 |.im,n = 3) in length. Hook
styliform, broadly connected to ventral termi-
nus and covered with about 30 microstylets
and numerous micropoints. Microstylets
lanceolate, arranged in three proximal rows
and reduced to two distal rows. Six to eight
microstylets and micropoints forming a cluster
near point of hook. Micropoints extend over
nipple-like ventral terminus beyond edge of
ventral margin of valve, leaving wide unsculp-
tured distal hook margin.
Remarks
Clarke (1981a) described many of the
glochidia of the Alasmidontini using SEM, in-
cluding this species. Our descriptions of this
glochidium are almost identical, except for the
location of the hook. Actually, we agree here
as well, but Clarke confused the anterior-pos-
terior orientation of the glochidium. He stated,
"apices are located slightly anterior of center
(about 47%)." Actually the apices (= ventral
termini) are located about that distance from
the posterior margin.
Ortmann (1912) reported length and height
measurements of 300 цт x 250 |am, and
Surber (1912) gave 300 |am x 255 цт for this
glochidium. Clarke (1 981 a) gave the following
ranges; length 286-292 цт, height 232-235
|лт, and hinge length 205 (.im. The figure for
hinge length is probably a typographical error
and should read 250 цт (approximate hinge
length taken from his fig. 6b, c).
This glochidium is distinguished from that of
Strophitus by its finer exterior valve sculpture
and its broadly connected hook. This glochid-
ium is figured by Lea (1858; pi. 5, fig. 30, as
Margaritana deltoidea. = A. viridis, fide Simp-
son, 1900), Surber (1912: pi. 1,fig. 1, as Л /as-
midonta calceola. = A. viridis, fide Clarke,
1 981 a), Ortmann (1 91 2: pi. 1 9, fig. 4, as Alas-
midonta minor = A. viridis, fide Clarke,
1981a), Clarke (1981a: fig. 6), and Zaie &
Neves (1982: fig. 1 , as A. minor).
Alasmidonta heterodon (Lea, 1829)
(Fig. 19A-F)
Material Examined
OSUM 25106.2 - Canoe River at old New-
land St. bridge, 2.45 mi. NNE of Norton, Bris-
tol Co., Massachusetts, 2 June 1 969, H. D. At-
hearn.
Description
Glochidium depressed-pyrlform, symmetric
when immature (Fig. 18A). becoming asym-
metric with the development of hook (Fig.
GLOCHIDIAOF UNIONIDAE
37
— «-««-^ШЩ^
1Ш^^
eWnîlffraSRI«»
w
rW^
^ЩЯЁ^Ш-
i
'M
J
^
D
FIG. 18. Glochidium of Alasmidonta viridis, OSUM 47518; A. exterior valve, bar length = 45 цт; В. interior
valve, bar length = 45 lam; С lateral view, bar length = 45 ¡.im; D. styliform hook, bar length = 15 |дт; E. ex-
terior valve sculpture, bar length = 2 (.im: F, interior valve pitting, bar length = 10 цт.
FIG. 19. Glochidium of Alasmidonta heterodon. OSUM 25106.2; A. exterior valve of immature glochidium.
bar lengthi = 45 |am; B. interior valve, bar length = 70 цт; С. exterior valve, bar length = 45 цт; D. interior
valve, bar lengtfi = 60 цт; E. styliform hook, bar length = 10 |.im: F. exterior valve sculpture, bar length =
1 \im.
GLOCHIDIAOF UNIONIDAE
39
18B, С, D), length 330 to 338 цт (334 ± 3.65
цт, n = 4), height 258 to 268 цт (265 ± 4.86
|am, n = 4). Dorsal margin straight, 253 to
266 цт (261 ± 5.94 цт, n = 4) in length. Pos-
terior margin slightly and evenly curved; ante-
rior margin more broadly curved, especially
near dorsal margin. Point of maximum lateral
inflation at between 40-50% from dorsal to
ventral. Ventral terminus curved, slightly out-
wardly produced. Exterior valve surface finely
malleated and pitted, except along its margins
and at umbo. Interior pitting not uniform; pit
density reduced in adductor muscle scar.
Loose-looped sculpture occurring on exterior
surface of valves (Fig. 19F). Central ligament
about 80 |.im long, centered about 42% from
posterior to anterior. Posterior ligament about
75 цт long; anterior ligament about 110 цт
long. A styliform hook extending dorsally from
ventral terminus of each valve. Hook covered
with about 40 microstylets and many micro-
points, located about 42% from posterior to
anterior. Microstylets arranged in four to six
proximal rows and reduced to two rows dis-
tally. Micropoints extending over edge of ven-
tral terminus, along ventral margin of valve,
and for a short distance on lateral surfaces of
hook, leaving wide unsculptured distal hook
margin.
Remarks
Clarke's (1981a) and my descriptions of
this glochidium are, not surprisingly, very sim-
ilar. We both examined glochidia removed the
same adult female. He reported length,
height, and hinge length measurements of
325 ,um X 255 цт x 267 цт, respectively. All
of these are within the ranges found during
this study. This glochidium is very similar to
that of A. viridis, but differs from that species
in being more inflated toward the anterior end
and in having a less broadly attached styli-
form hook. This glochidium has been figured
by Clarke (1981a; fig. 9).
Alasmidonta undulata (Say, 1817)
(Fig. 20A-F)
Material Examined
OSUM 52434.4 - Merrimack River just
below Sewalls Fall's Dam, 2.7 mi. SE of Pe-
nacook, 3.7 mi. NNW of Concord, Concord
Twp., Merrimack Co.. New Hampshire, 31 Oc-
tober 1982, K. E.Wright.
Description
Glochidium pyriform, much higher than
long, asymmetric, length 343 to 358 ¡.im (353
± 5.89 |.im, n = 5), height 365 to 378 |am (371
± 5.13 цт, n ^ 5). Dorsal margin straight, 247
to 260 |.im (254 ± 5.59 (.im, n = 5) in length.
Posterior margin outwardly curved dorsally,
becoming straight to slightly incurved before
ventral terminus. Anterior margin broadly
curved. Ventral terminus slightly incurved.
Maximum anterior and posterior inflation at
about 30% from dorsal to ventral, with ventral
terminus rounded, located about 40% from
posterior to anterior. Valve surface coarsely
malleated, uniformly pitted, except at the
umbo and along valve margin, where pits and
malleations are absent. Exterior surface cov-
ered with sparse rosette sculpture, separated
by areas of beaded sculpture (Fig. 20E). Lar-
val thread present (Fig. 20D). Central liga-
ment 89 to 98 |.im (92 ± 5.19 цт, n = 3) in
length, centered about 40% from posterior to
anterior. Posterior ligament 53 to 55 цт (54 ±
1.15 jjm, n = 3) in length; anterior ligament
103 to 105 |im (104 ± 1.00 цт, n = 3) long.
Hook styliform, covered with numerous (about
120) microstylets and micropoints. Micro-
stylets pyramidal, multifaceted, arranged in
about six proximal rows, reduced to about
four rows distally. Micropoints present along
ventral valve margin, at lateral margins of mi-
crostylets and for a short distance down lat-
eral surfaces of hook, leaving wide unsculp-
tured distal hook margin.
Remarks
The glochidium of A. undulata can be dis-
tinguished by its large size, pyriform valve
shape, complex hook and unique exterior
valve sculpture. Ortmann (1912) described
this glochidium as, "moderately large, higher
than long, with strong hooks. Length 0.34
mm; height 0.36 mm." Clarke (1981a) re-
ported length and height measurements of
310 i-im X 370 i^m and figured the glochidium
(his fig. 13). This glochidium also is figured by
Wiles (1975; fig. 6, as Anodonta Cataracta).
Alasmidonta marginata Say, 1818
(Fig.21A-l)
Material Examined
M AH 277.1 - Big Darby Creek at and above
McLean Mill Rd. bridge, 0.2 mi. SW of Fox, 4.7
40
HOGGARTH
FIG. 20. Glochidium of Alasmidonta undulata. OSUM 52434.4; A. exterior valve, bar length = 60 цт: В. in-
terior valve, bar length = 55 цт; С. styliform hook, bar length = 20 ¡.im; D. interior valve, bar length = 50 цт;
E. exterior valve sculpture, bar length = 1 цт; F. micropoints, bar length = 10 |.im.
GLOCHIDIAOF UNIONIDAE
41
FIG. 21 . Glochidium of Alasmidonta marginata: A. exterior valve, MAH 724.1 , bar length = 55 цт; В. interior
valve, MAH 724.1 , bar length = 45 (.im; С exterior valve sculpture, MAH 724.1 , bar length = 1 |im; D. exte-
rior valve sculpture, MAH 277.1 , bar length = 1 цт; E. styliform hook, MAH 277.1 , bar length = 25 |am; F. hair
cell, MAH 724.1 , bar length = 5 ¡am; G. adductor rnuscle, MAH 724.1 , bar length = 5 ¡jm: H. mantle cells. MAH
724.1, bar length = 10 |im; I. larval thread, MAH 724.1, bar length = 10 цт.
42
HOGGARTH
mi. NW of Circleville, Jackson Twp., Pickaway
Co., Ohio, 1 October 1982, IVl. A. Hoggartln et
a!.. MAH 724.1 - Fisti Creek above and below
Edon Rd. bridge, 1 .9 mi. NW of Edgerton, St.
Josepfi Twp., Williams Co., Ohio, 2 October
1985, M. A. Hoggarth & D. Rice.
Description
Glochidium pyriform, higher than long,
asymmetric, length 335 to 341 ).im (339 ±
2.87 (.im, n = 4), height 360 to 372 (im (365 ±
5.74 [im, n = 4). Dorsal margin straight, 230 to
235 |.im (233 ± 0.96 |im, n - 4) in length. Pos-
terior margin broadly curved (especially dor-
sally), with maximum inflation at about 30%
from dorsal to ventral. Anterior margin more
broadly curved than posterior margin. Maxi-
mum anterior inflation at about 40% from dor-
sal margin. Lateral margins slightly incurved
ventrally, producing a narrowly rounded ven-
tral terminus. Ventral terminus about 40%
from posterior to anterior. Exterior surface of
valve coarsely malleated and densely pitted,
except at valve margins and at umbo. Exterior
surface covered with beaded sculpture (Fig.
21 D), although fine loose-looped sculpture
present near umbo of one glochidium (Fig.
21 C). Larval thread present (Fig. 21 1). Central
ligament 67 to 73 |лт (70 ± 3.54 цт, n = 2) in
length, centered about 40% from posterior to
anterior. Posterior ligament 59 to 61 цт (60 ±
1 .41 цт, n = 2) long; anterior ligament 1 00 to
1 07 |.im (1 04 ± 4.95 цт, n - 2) in length. Hook
styliform, with many microstylets (about 120)
and many micropoints. As in A. undulata, dis-
tal margin of hook parallel to ventral margin,
except at its center, where it becomes
strongly curved to a sharp distal point. Mi-
crostylets pyramidal, multifaceted, arranged
in six to seven proximal rows, reduced to four
rows distally. Micropoints occur along rim of
ventral terminus, at lateral margins of mi-
crostylets and for a short distance down lat-
eral surfaces of hook. Distal hook margin with
wide unsculptured band.
Sensory hair cells (Fig. 21 F) have been de-
scribed using light microscopy (Lillie, 1895;
Wood, 1974), SEM (Giusti et al., 1975; Rand
& Wiles, 1982) and transmission electron mi-
croscopy (Zs.-Nagy & Labos, 1969). It is gen-
erally thought that during attachment to the
host, the tissues of the host push down on the
hair cells and that the response to this stimu-
lus is prolonged muscle contraction. It ap-
pears that when the hairs are bent, their
movement stimulates a ring of tissue, possibly
composed of nerve tissue that encompass the
base of the cell (at arrow).
The adductor muscle (Fig. 21 G) is com-
posed of long cells of contractile elements
that attach to the crystalline matrix of the
valves. The adductor muscle scar is often vis-
ible as a rough elliptical area or by its reduced
number of pits. There is no evidence that the
muscle cells actually insert within these pits.
The larval mantle cells are five to seven sided
(Fig. 21 H) and do not appear to be pitted, as
suggested by Rand & Wiles (1982). They
suggested that the larval mantle was pitted to
correspond to the pits in the valve and that
this might facilitate gas exchange, nutrient up-
take, or waste elimination. This was not ob-
served, and therefore it is suggested that the
pits are simply a result of the absence of
stress traversing the body of the shell. Where
stress is transferred from ventral margin to
hinge (the lateral margins), pitting is absent
and ridges (providing additional strength)
occur. The pits may serve no function and
may simply be a consequence of glochidial
valve morphology (Hoggarth & Gaunt, 1988).
Remarks
This glochidium can be distinguished from
that of A. undulata by differences in exterior
valve sculpturing. Otherwise they are very
similar. Lea (1858) described this glochidium
as subthangular, with a long, straight dorsal
line and inflated side margins. He described
the hook as terminating in an arrowhead
point. Clarke (1 981 a) also described the point
of the hook as arrowhead-like. The arrowhead
effect is probably due to the collapse of the
lateral surfaces of the hook and is therefore
an artifact of drying. This glochidium has a
wide size range: (length x height) 330 цт x
360 |Lim (Ortmann, 1912); 350 |im x 380 |im
(Surber, 1912); 300 цт x 350 |nm (Utterback,
1915-1916); 341 |.im x 346 |.im (Clarke,
1981a) and is figured by Lea (1858:pl. 5, fig.
27), Surber (1912: pi. 3, fig. 42, as A. truncata.
= A. marginata, fide Stansbery et al., 1985),
and Clarke (1981a: fig. 20).
Pegias fabula (Lea, 1838)
(Fig. 22A-F)
Material Examined
OSUM 41 308.3 - Little South Fork Cumber-
land River at Freedom Church Ford. 2.0 mi.
ENE of Ritner, 14.3 mi. E of Monticello, Wayne
FIG. 22. Glochidium of Pegias fabula: A. exterior vaive, OSUM 41 308.3, bar length = 55 цт; В. interior valve,
OSUM 41 309.1 . bar length = 55 \xm; С exterior valve. OSUM 41 308.3, bar length = 55 |дт; D. lateral view,
OSUM 41308.3, bar length = 55 цгл; E. exterior valve sculpture, OSUM 41308.3, bar length = 2 |im; F. styii-
form hook, OSUM 41309.1, bar length = 20 urn.
44
HOGGARTH
Co., Kentucky, 22 October 1977, W. & L.
Starnes. OSUM 41309.1 - Little South Fork
Cumberland River at Freedom Church Ford,
2.0 mi. ENE of Ritner, 14.3 mi. E of Monticello,
Wayne Co., Kentucky, 22 October 1977, A.
Bogan.
Description
Glochidium oval quadrate to roundly trape-
zoidal, length 385 to 388 цт (386 ± 1 .53 |am,
n = 3), height 319 to 325 цт (322 ± 3.06 |am,
n = 3). Dorsal margin straight, 200 to 210 |am
(205 ± 5.03 цт, n = 3) long. Posterior margin
roundly arcuate. Anterior margin inflated dor-
sally, slightly incurved near ventral terminus.
Ventral terminus slightly rounded, about 35%
from posterior to anterior. Exterior surface pit-
ted but not malleated, smooth in area of ad-
ductor muscle scar and at margin of valve.
Adductor muscle scar very large (Fig. 22B-D).
Exterior sculpturing tight-looped (Fig. 22E).
Central ligament 68 to 73 цт (71 ± 2.89 |jm,
n = 3) long, centered about 45% from poste-
rior to anterior. Posterior ligament 56 to 63 цт
(60 ± 3.51 цт, n = 3) in length; anterior liga-
ment 69 to 79 ).im (74 ± 5.03 |.im, n = 3) long.
Hook, styliform, sharply pointed, broadly con-
nected to ventral valve margin. Hook with
about 75 lanceolate microstylets arranged
three abreast near proximal end of hook and
in a double row distally. About 15 microstylets
forming a cluster near distal end of hook. Mi-
cropoints limited to ventral rim of valve and
along margins of microstylets, leaving very
wide unsculptured distal hook margin.
Remarks
The glochidium of P. fabula cannot be con-
fused with that of any other species. Its
quadrate shape, broadly connected styliform
hook, tight looped exterior valve sculpturing
and extremely large adductor muscle scar
distinguish it. Clarke (1981a) figured this
glochidium (his fig. 3) and provided length and
height measurements for two specimens: 354
jim X 309 цт and 380 цт x 310 yim.
Arcidens confragosus (Say, 1 829)
(Fig. 23A-G)
Material Examined
OSUM 52015 - Green River at Glenmore
below lock 5 dam, 12 mi. N of Bowling Green,
Warren Co., Kentucky, 2 November 1977,
D. H. Stansbery et al.
Description
Glochidium pyriform, about as long as high,
asymmetric, length 352 to 363 цт (359 ±
5.32 |.im, n - 4), height 353 to 355 |am (354 ±
1 .41 |im, n = 2). Dorsal margin straight, 237 to
252 |.im (246 ± 7. 1 6 цт, n = 5) in length. Pos-
terior margin produced dorsally, incurved ven-
trally, with its maximum inflation between
30-40% from dorsal to ventral. Anterior mar-
gin broadly rounded dorsally, incurved just be-
fore ventral terminus, maximum inflation at
about 50% from dorsal margin. Ventral termi-
nus narrowly rounded, located about 40%
from posterior to anterior. Exterior surface
coarsely malleated and densely pitted, except
at umbo and along valve margin. Dense
rosette sculpturing covering exterior surface
of valve (Fig. 23E). Central ligament 63 to
78 |im (71 ± 6.40 |лт, n = 4) long, centered
about 42% from posterior to anterior. Poste-
rior ligament 59 to 72 |.im (68 ± 5.91 ¡am, n =
4) long; anterior ligament 100 to 115 цт (110
± 7.14 цт, n = 4) long. Styliform hook ex-
tending from ventral terminus as a very
strongly biconcave triangular plate with about
80 pyramidal microstylets arranged in about
five rows (Fig. 23D, F). Number of rows of
microstylets same from proximal to distal
ends of hook. Micropoints occur on rim of
valve, at ventral terminus and along mi-
crostylet border, leaving wide unsculptured
distal hook margin.
Remarks
The glochidium of A. confragosus is pyri-
form and about as long as high. Anterior and
posterior margins are greatly inflated dorsally
and slightly incurved ventrally. The ventral ter-
minus is narrowly rounded, and the number of
rows of microstylets remain constant from
proximal to distal ends of the styliform hook.
This glochidium is similar to that of some
members of the genera Alasmidonta and Las-
migona but can be distinguished by its equal
height and length, its exterior valve sculpture,
and the arrangement of microstylets on the
hook. Length and height measurements are
given by Surber (1912: pi. 1, fig. 5) 355 цт x
350 цт, and Clarke (1981a: fig. 31) 359 pm x
360 i.im, both of whom also figure the glochid-
ium.
GLOCHIDIAOF UNIONIDAE
45
i
FIG. 23. Glochidium of Arcidens confragosus. OSUM 52015; A. exterior valve, bar length = 65 цт; В. inte-
rior valve, bar length = 65 (.im; С styliform hook, bar length = 25 i-im; D. styliform hook, bar length = 5 цт; E.
exterior valve sculpture, bar length = 1 цт; F. microstylets, bar length = 5 |.im; g. hinge, bar length = 40 |im.
Lasmigona compressa (Lea, 1829)
(Fig. 24A-F)
Material Examined
OSUM 23179.1 - Little Darby Creek above
Rosedale-Plain City Rd. bridge, 2.8 mi. E of
Rosedale, Pike Twp., Madison Co., Ohio, 20
October 1969, С В. Stein et at.. MAH 702 -
Big Darby Creek below access point within
Battelle-Darby Metro Park, 0.6 mi. S of
Georgesville, 3.5 mi. SW of Galloway, Pleas-
46
HOGGARTH
Ki
FIG. 24. Glochidium of Lasmigona compressa: A. exterior valve. OSUM 23179.1, bar length = 45|.im: B. in-
terior valve, OSUM 23179.1, bar length = 45 цт; С lateral view, MAH 702, bar length = 70цт; D. styliform
hook, MAH 702, bar length = 30 цт; E. extehor valve sculpture, MAH 702, bar length = 1 |.im; F. microstylets,
MAH 702, bar length = 10 цт.
GLOCHIDIAOF UNIONIDAE
47
ant Twp., Franklin Co., Ohio, 9 September
1985, H. T. Albin. MAH 727 - Fish Creek
above and below Edon Rd. bridge, 1.9 гл1.
NW of Edgerton, St. Joseph Twp., Williams
Co., Ohio, 2 October 1985, M. A. Hoggarth &
D. Rice.
Description
Glochidium depressed pyriform, longer
than high, strongly asymmetric, length 317 to
327 |iim (323 ± 4.39 цт, n = 5), height 283 to
288 цт (286 ± 2.51 цт, n = 5). Dorsal mar-
gin straight, 230 to 239 (.im (234 ± 3.36 цт, n
= 5) long. Posterior margin strongly curved;
anterior margin greatly inflated dorsally, be-
coming more gently curved ventrally. Poste-
rior and anterior margins joining at a gently
rounded, nipple-like ventral terminus. Exterior
valve surface coarsely malleated and pitted
throughout, except at umbo and at margin of
valve (Fig. 24C). Loose-looped sculpture cov-
ering exterior surface of valve (Fig. 24E).
Central ligament 83 to 84 цт (84 ± 0.71 ¡um,
n = 2) in length, centered about 46% from
posterior margin. Posterior ligament 67 to 68
|nm (68 ± 0.71 |im, n = 2) long; anterior liga-
ment 81 to 82 )im (82 ± 0.71 ¡.im, n - 2) in
length. Styliform hook armed with about 25
stout microstylets arrange in three proximal
rows reduced to two widely off-set distal rows,
extending from ventral terminus and located
approximately 45% from posterior to anterior.
Micropoints extend over edge of valve at ven-
tral terminus and along valve margin, leaving
very wide unsculptured distal hook margin.
Remarks
Surber (1912), Ortmann (1912), Tompa
(1979), and Clarke (1985) gave length and
height measurements for this glochidium; 353
|im X 31 3 цт, 340 цт x 280 ^im, 320 цт x 260
|im, and 344 ¡.im x 275 |.im. This glochidium
can be distinguished by its widely offset dou-
ble row of microstylets and its wide unsculp-
tured distal hook margin. Clarke (1985) de-
scribed the hook of this species as having a
single distal row of microstylets; however, his
micrographs show only collapsed hooks that
are very difficult to interpret. This glochidium
was figured by Lea (1 858: pi. 5, fig. 23, as Unio
pressus, = L. compressa, f/de Simpson 1900),
Ortmann (1911: pi. 89, fig. 10), Surber (1912:
pi. 3, fig. 44), and Clarke (1985: fig. 13).
Lasmigona subviridis (Conrad, 1835)
(Fig. 25A-D)
Material Examined
OSUM 271 31 .66, 271 31 .68 - Little River at
U.S. Rt. 221 bridge, at Woods Store, 5.3 mi.
NE of Floyd, 33.7 mi. SW of Roanoke, Floyd
Co., Virginia, 3 October 1 970, D. H. Stansbery
& W.J. Clench.
Description
Glochidium depressed pyriform, 368 to 383
цт (376 ± 5.81 цт, n = 5) in length, 309 to
31 8 |.im (31 2 ± 3.71 цт, n = 5) in height. Pos-
terior margin, strongly curved; anterior margin
broadly rounded. Point of maximum posterior
inflation at about 30% from dorsal to ventral;
point of maximum anterior inflation at about
40-50% from dorsal margin. Dorsal margin
straight, 245 to 264 |.im (254 ± 6.82 цт, n =
5) in length. Exterior valve surface coarsely
malleated and uniformly pitted, except within
a narrow marginal band. Ventral terminus
broadly rounded, not produced or nipple-like
as in the other members of the genus, located
about 40% from posterior to anterior. Exterior
surface sculpturing loose-looped (Fig. 25C).
Central ligament 87 to 89 цт (88 ± 1 .00 цт,
n = 3) in length, centered about 40% from
posterior to anterior. Posterior ligament 57 to
68 цт (63 ± 5.51 |im, n= 3) long; anterior lig-
ament 100 to 107 |im (104 ± 3.61 |.im, n = 3)
long. Styliform hook armed with about 25 mi-
crostylets and numerous micropoints. Mi-
crostylets lanceolate, about five to six abreast
proximally, arranged in a double off-set row
distally. Micropoints located on ventral margin
of valve and on the lateral surfaces of hook
but not over the edge of valve at the ventral
terminus (Clarke 1 985, Fig. 1 6b), leaving wide
unsculptured distal hook margin.
Remarks
This glochidium can be distinguished by its
more broadly curved lateral margins and the
absence of micropoints extending onto the
valve at the ventral terminus. Ortmann (1912)
gave 360 |im x 300 цт for the length and
height of this glochidium, and Clarke (1985)
gave 350-372 |am x 285-303 цт. This
glochidium was figured by Lea (1874: pi. 21,
fig. 14, as Unio tappanianus, = L. subviridis,
fide Ortmann & Walker, 1922) and Clarke
48
HOGGARTH
FIG. 25. Glochidium of Lasmigona subviridis: A. exterior valve. OSUM 27131 .68, bar length = 55 цт: В. in-
terior valve, OSUM 27131 .68, bar length = 80 цт: С. extehor valve sculpture, OSUM 27131 .66, bar length
= 1 jam; D. styliform hook, OSUM 271 31 .68, bar length = 1 5 цт.
(1 985: fig. 1 6). Lea's figure does not show the
morphologically depression of the glochidium.
Tazewell, Tazewell Co., Virginia, 13 October
1985, D. H. Stansbery.
Lasmigona holstonia (Lea, 1838)
(Fig. 26A-D)
Material Examined
OSUM 55826.6, 55826.7 - South Fork
Clinch River at St. Rt. 61 bridge, E edge of
Description
Glochidium subtriangular, length and height
about equal. 281 to 291 |.im (286 ± 3.65 |.im, n
= 5) in length, 275 to 294 |.im (282 ± 7.76 pm.
n = 5) in height. Dorsal margin straight, 221 to
235 цт (228 ± 5.32 |.im, n = 5) long. Posterior
GLOCHIDIAOF UNIONIDAE
49
FIG. 26. Glochidium of Lasmigona holstonia. OSUM 55826.6; A. exterior valve, bar length = 45 цт; В. inte-
rior valve, bar length = 40 цт; С. exterior valve sculpture, bar length = 1 ¡.im; D. styliform hook, bar length =
10 цт.
margin slightly and evenly curved, with its
point of maximum inflation about 20-30%
from dorsal margin. Anterior margin broadly
curved, with a tendency to be incurved before
ventral terminus, with its maximum inflation at
about 40-50% from dorsal to ventral. Ventral
terminus narrowly rounded, located about
40% from posterior to anterior. Exterior sur-
face finely malleated, uniformly pitted, except
along valve margin. Tight-looped exterior
valve sculpture covering surface of valve (Fig.
26C). Clarke (1985) reported a central liga-
ment length of 65 |.im centered 35% from the
anterior margin (actually the posterior margin).
A central ligament length of 63 to 73 |jm (68 ±
7.07 |лт, n = 2) was found during this study,
with a midpoint about 38% form posterior to
anterior. Posterior and anterior ligaments 52 to
50
HOGGARTH
55 |.im (54 ± 2.12 |.im, n = 2), and 109 to 110
цт (110 ± 0.71 |.im, n = 2) long, respectively.
Hook styliform, with about 20 microstylets and
many micropoints. Micropoints located on
ventral rim and lateral surfaces of hook, leav-
ing narrow unsculptured distal hook margin.
Microstylets lanceolate, arranged in a double
offset rows distally.
Remarks
The shape of this glochidium resembles
that of Anodonta. except for the slightly in-
curved margins prior to ventral terminus. The
hook structure also differs from that of An-
odonta. These two characters, as well as the
far posterior position of the hook, the broadly
rounded anterior margin and looped exterior
valve sculpture ally this glochidium with some
members of the genera Alasmidonta and Las-
migona. It is easily distinguished from these,
however, by its shape and the tightness of its
looped sculpture.
Lasmigona costata (Rafinesque, 1820)
(Fig. 27A-E)
Material Examined
MAH 279.1 - Big Darby Creek at and above
McLean Mill Rd. bridge, 0.2 mi. SW of Fox,
4.7 mi. NW of Circleville, Jackson Twp., Pick-
away Co., Ohio, 1 October 1982, M. A. Hog-
garth et al. MAH 585 - Big Darby Creek at ac-
cess, 0.9 mi. N of Harrisburg, 1.7 mi. NW of
Orient, Pleasant Twp., Franklin Co., Ohio, 27
September 1983, M. A. Hoggarth. MAH 882.1
- Fish Creek at bridge 0.7 mi. W of Arctic, 3.8
mi. NE of Butler, Sec. 20/29, Troy Twp.,
Dekalb Co., Indiana, 30 October 1985, D. H.
Stansbery et al.
Description
Glochidium pyriform, asymmetric, length
340 to 348 |im (344 ± 2.73 |im, n = 7), height
363 to 377 ¡im (369 ± 5.68 |im, n = 7). Dorsal
margin 239 to 245 цт (241 ± 2.23 цт, n - 7)
long. Posterior and anterior margins broadly
rounded dorsally, slightly incurved ventrally.
Maximum inflation of posterior margin at
about 30% from dorsal to ventral; maximum
inflation of anterior margin at about 40% from
dorsal margin. Exterior valve surface coarsely
malleated and densely pitted, except at valve
margins and at umbo. Exterior surface cov-
ered with densely beaded sculpture (Fig.
27C). Central ligament 78 to 84 ¡.im (80 ±
2.49 Mm, n = 5) long, centered at about 45%
from posterior to anterior (Fig. 27B). Posterior
ligament 61 to 68 цт (64 ± 2.77 |.im, n = 5) in
length; anterior ligament 90 to 105 цт (99 ±
7.05 |.im, n = 5) long. Hook styliform, arising
from ventral terminus as a broadly incurved
triangular plate. Microstylets lanceolate, mul-
tifaceted, arranged in about seven proximal
rows, reduced to five distal rows and number-
ing about 100. Micropoints on proximal border
of hook but ending abruptly at ventral rim of
valve, not extending onto exterior valve sur-
face (Fig. 27D) nor very far onto lateral sur-
face of the hook, leaving wide unsculptured
distal hook margin.
Remarks
The glochidium of L. costata can be distin-
guished by its distinctly pear-shaped outline,
its exterior valve sculpture, and hook struc-
ture. Glochidia of this species were figured by
Lea (1858: pi. 5, fig. 26, as Margaritana ru-
gosa, = L. costata. fide Ortmann & Walker,
1922), Lefevre& Curtis (1910: fig. B, 1912:fig.
1 B), Surber (1912: pi. 1 , fig. 7), Arey (1 924: pi.
1, fig. 2), and Clarke (1985: fig. 5). This glo-
chidium varies greatly in size: (length x height)
Lea, 368 цт x 400 цт; Surber, 385 jam x 390
jjm; Lefevre & Curtis, 350 ¡am x 390 |jm; Ort-
mann, 340 |im X 370 цт ; Clarke, 333 цт x
364 цт.
Lasmigona complanata (Barnes, 1823)
(Fig. 28A-F)
Material Examined
MAH 278.2 - Big Darby Creek at and above
McLean Mill Rd. bridge, 0.2 mi. SW of Fox,
4.7 mi. NW of Circleville, Jackson Twp., Pick-
away Co., Ohio, 1 October 1982, M. A. Hog-
garth et al.
Description
Glochidium pyriform. almost symmetrical,
length 289 to 296 |.im (293 ± 2.90 цт, n = 6),
height 293 to 310 |.im (300 ± 6.90 цт, n = 6).
Dorsal margin straight, 193 to 208 i^im (200 ±
4.69 цт, n = 7) in length. Maximum inflation
of anterior and posterior margins at about
40% from dorsal to ventral. Exterior surface
coarsely malleated, densely pitted except
along valve margin. Umbo malleated but not
GLOCHIDIAOF UNIONIDAE
51
FIG. 27. Glochidium of Lasmigona costata:A. exterior valve, MAH 585, bar length = 50 |am; B. interior valve,
MAH 279.1 , bar length = 50 цт; С. exterior valve sculpture, MAH 585, bar length = 1 ¡im; D. styliform hook,
MAH 585, bar length = 20 цт; E. styliform hook, MAH 585, bar length = 20 цт.
pitted. Dense rosette sculpture covering exte-
rior surface of valves (Fig. 28E, F). Central lig-
ament 60 to 68 (.im (64 ± 3.32 ¡.im, n = 4) long,
centered about 42% from posterior to anterior.
Posterior ligament 50 to 53 цт (52 ± 1 .29 |лт,
n = 4) in length; anterior ligament 75 to 88 цт
(81 ± 5.48|.im,n = 4)long. Hook styliform, very
similar to that of L. costata. Microstylets (about
100) lanceolate, multifaceted, arranged in six
proximal rows reduced to four rows distally. Mi-
cropoints restricted to the proximal margin of
hook, leaving wide unsculptured distal hook
margin. Ventral terminus located about 40%
from posterior to anterior.
Remarks
This glochidium can be distinguished by its
nearly equal length and height, few micro-
points and exterior valve sculpture. It is figured
by Lea (1858: pi. 5, fig. 29), Lefevre & Curtis
(1910:fig.A, 1912:fig. 1A), Ortmann (1911 :pl.
52
HOGGARTH
FIG. 28. Glochidium of Lasmigona complanata, MAH 278.2; A. exterior valve, bar length = 45 |лт; В. interior
valve, bar length = 45 цт; С. styliform hook, bar length = 15 |um; D. styliform hook, bar length = 20 цгл; E.
exterior valve sculpture, bar length = 1 цт; F. exterior valve sculpture, bar length = 2 цт.
GLOCHIDIAOF UNIONIDAE
53
89, fig. 11), Surber (1912: pi. 1, fig. 6), Arey
(1 921 : pi. 1 , fig. 1,2), and Clarke (1 985: fig. 8).
Measurements given for this glochidium are:
290 цгл X 300 |.im (Lefevre & Curtis, 1 91 0), 340
(im X 340 ц m (Ortmann, 1911), 310 цт x 320
|лт (Surber, 1912), and 337 j.im x 337 (.im
(Clarke, 1985). This glochidium appears to
vary a great deal in size, but its relative di-
mensions remain fairly constant (i.e., length =
height).
Subfamily Ambleminae
Magalonaias nervosa (Rafinesque, 1820)
(Fig. 29A-G)
Material Examined
OSUM - 13032.66 St. Francis River Bay
halfway between Wynne and Perkin, 1 .0 mi. S
of Rt. 64 bridge, Cross Co., Arkansas, 25 Oc-
tober 1964, С В. Stein. OSUM 54178 - Mis-
sissippi River, R.MI. 299.8-301 .1 , 7.8-9.0 mi.
SE of Hannibal (MO), Pike/Ralls Co., Illi-
nois/Missouri, 16 October 1979, R. B. Lewis
et al. OSUM 178-Licking River, 1.6 mi. E of
Butler, immediately above mouth of Flour
Creek, 22 mi. SE of Cincinnati, Pendleton
Co., Kentucky, 31 October 1986, D. H. Stans-
bery et al. OSUM - 1986:22 Green River at
Glenmore, below Lock 5 Dam, 12.0 mi. N of
Bowling Green, Warren Co., Kentucky, 1 No-
vember 1986, D. H. Stansbery et al.
Description
Glochidium subelliptical, length 254 to 268
¡.im (261 ± 6.58 (.im, n = 4), height 340 to 350
(.im (346 ± 4.78 (дт, n = 4). Dorsal margin
straight, 145 to 155 (.im (150 ± 3.27 (.im, n =
6) in length. Lateral margins gently curved but
unequal. Maximum inflation of posterior mar-
gin at about 70% from dorsal to ventral; max-
imum inflation of anterior margin at about 40%
from dorsal margin. Ventral margin narrowly
rounded. Tight-looped sculpture covering ex-
terior surface of valve (Fig. 29F). Coiled larval
thread present (Figs. 29C, D). Central liga-
ment 49 to 52 ¡.im (50 ± 1 .53 (.im, n - 3) long,
centered about 44% from posterior to anterior.
Anterior ligament 60 to 63 ¡.im (61 ± 2.00 (.im,
n = 3) long, posterior ligament 43 to 44 (.tm (43
± 0.58 (.im, n = 3) in length. Lanceolate mi-
cropoints occurring in broken vertical rows on
a narrow ventral flange, and along rim of ven-
tral margin of valve, covering most of ventral
flange, leaving very narrow unsculptured dis-
tal flange margin.
Remarks
Surber (1915) stated, "notwithstanding its
great variation in size, and even outline, this
species cannot be readily confused with any
other, even though the larval gland may have
been absorbed. . ." Surber (1912), Howard
(1914c), and Surber (1915) gave length and
height measurements of 260 ¡.im x 340 (.im,
250-260 (.im X 316-340 (.im, and 250-280
(.im X 300-380 jLim (as Quadrula heros = M.
nervosa, fide Stansbery et al., 1985). This
glochidium has been figured by Lea (1858: pi.
5, fig. 3, as Unio multiplicatus. = M. nervosa
fide, Stansbery et al., 1985), Surber (1912: pi.
2, fig. 32), Howard (1914c: pi. 3, fig. 21; pi. 5,
fig. 35), Surber (1915: pi. 1, fig. 10), and Ut-
terback (1915-1916: fig. 3a, b). Surber's,
Howard's, and Utterback's figures agree with
mine, whereas Lea's does not.
Megalonaias boykiniana (Lea, 1840)
(Fig. 30A-D)
Material Examined
OSUM 511 07.5 - Apalachicola River below
U.S. Rt. 90 bridge, 1.0 mi. W of Chatta-
hoochee, 17.8 mi. WNW of Quincy, T4N,
R6W, Sec. 32, Gadsden Co., Florida, 29 Oc-
tober 1981, D. H. Stansbery et al.
Description
Glochidium subelliptical, with a straight dor-
sal margin, a narrowly rounded ventral mar-
gin, and gently but unequally curved lateral
margins. A single specimen gave the following
measurements for length, height, and hinge
length: 245 (im x 350 jim x 150 (.im. Tight-
looped sculpture covering exterior surface
of valve (Fig. 30C), lanceolate micropoints
cover a narrow ventral flange (Fig. 30D), and
a larval thread is coiled around adductor mus-
cle, not supercoiled as in /W. nervosa (Fig.
ЗОВ)
Remarks
The glochidium of this species can be dis-
tinguished from that of M. nervosa by its larval
thread and from all other glochidia by its di-
mensions and outline. No published figure of
this glochidium was found.
54
HOGGARTH
FIG. 29. Glochidium of Megalonalas nervosa: A. exterior valve, OSUM 54178, bar length = 50 цт; В. inte-
rior valve, OSUM 178, bar length = 50 цт; С. larval thread, OSUM 54178, bar length = 25 цт; D. larval
thread, OSUM 54178, bar length = 5 цт; E. micropoints, OSUM 54178, bar length = 5 ¡.im: F. exterior valve
sculpture, OSUM: 1986:22, bar length = 2 |дт; G. hinge, OSUM 178, bar length = 20 цт.
Plectomerus dombeyana
(Valenciennes, 1827)
(Fig. 31A-H)
Material Examined
OSUM 42011 - Black Warrior River at Hall
Shoals, below Eutaw Dam, 5.8 mi. SE of
Eutaw, Sec. 25, T21N, R2E, Green Co., Al-
abama, 28-30 July 1 975, J. D. Williams et al..
OSUM 53273.2, 53273.3 - Calcasieu River at
Novel's Bluff, 2.0 mi. NE of Indian Village, 6.9
mi. WSW of Kinder, 22.0 mi. NE of Lake
Charles, Allen Pahsh, Louisiana, 25 July
1982, D. H. Stansbery & M. A. Hoggarth.
Description
Glochidium subelliptical, length 223 to 231
|.im (226 ± 2.99 цт. n = 1 0). height 238 to 259
цт (246 ± 7.07 цт, n = 10). Dorsal margin
straight, 130 to 135 |.im (133 ± 1.36 цт, n =
GLOCHIDIAOF UNIONIDAE
55
FIG. 30. Glochidium of Megalonaias boykiniana, OSUM 51107.5; A. exterior valve, bar length = 50 |.im; B.
larval thread, bar length = 25 jam; C. exterior valve sculpture, bar length = 1 |im; D. micropoints, bar length
= 5 [ivn.
11) long. Lateral nnargins gently and equally
curved, symmetrical. Ventral margin semicir-
cular. Valve surface finely malleated, with
many pits (Fig. 31 E). Loose-looped sculpture
covers exterior valve surface (Fig. 31C,H).
Figure 31 С demonstrates that the larval valve
has an exterior membrane that is at least
analogous, if not homologous, with the pe-
riostracum of the adult. It is also evident from
this micrograph that the exterior valve sculp-
ture occurs within this membrane. Central lig-
ament 42 to 47 |im (44 ± 2.06 fim, n = 4) long,
centered about 40% from posterior to anterior.
Anterior ligament 51 to 59 |лт (56 ± 3.59 ¡am.
n = 4) long; posterior ligament 31 to 35 ¡am (32
± 1.89 цт, n = 4) in length. Lanceolate mi-
cropoints in broken rows on rim of ventral
margin of valve and on a narrow ventral
flange. Surface of ventral flange mostly cov-
ered with micropoints, leaving narrow un-
sculptured distal flange margin.
Remarks
This glochidium is distinguished from oth-
ers examined by its shape and dimensions.
No published figure of this glochidium was
found.
56
HOGGARTH
FIG. 31. Glochidium of Plectomerus dombeyana: A. exterior valve, OSUM 53273.3, bar length = 35 цт: В,
interior valve, OSUM 53273.3, bar length = 35 цт; С. exterior valve sculpture and torn exterior valve mem-
brane, OSUM 53273.3, bar length = 2цт; D. hinge, OSUM 53273.2, bar length = 25 мт; E. interior valve
pitting, OSUM 42011, bar length = 10|.im; F. micropoints, OSUM 53273.3, bar length = 5 ^m: G. micropoints,
OSUM 42011 , bar length = 3 цт; H, exterior valve sculpture, OSUM 42011 , bar length = 1 pm.
GLOCHIDIAOF UNIONIDAE
57
FIG. 32. Glochidium of Tritogonia verrucosa, MAH 654.1; A. exterior valve, bar length = 15 цт; В. interior
valve, bar length = 15 цгл; С. micropoints, bar length = 5 цт; D. hinge, bar length = 7 \im; E. exterior valve
sculpture, bar length = 2 цт.
Tritogonia verrucosa (Rafinesque, 1820)
(Fig. 32A-E)
Material Examined
MAH 654.1 - Big Darby Creek at Scioto-
Darby (Mt. 31ег11пд-Сотглегс1а1 Pt.) Rd.
bridge, 3.4 mi. S of Orient, 1 5.3 mi. NW of Cir-
cleville, Scioto/Darby Twp., Pickaway Co.,
Ohio, 18 May 1984, M. A. Hoggarth & G. T.
Watte rs.
Description
Glochidium subelliptical to subrotund,
length 85 to 94 ).im (90 ± 3.38 |jm, n = 6),
height 97 to 101 цт (100 ± 1.75 цт, n = 6).
Dorsal margin slightly curved, 43 to 46 |im (44
±1.11 jim, n = 7) long. Lateral margins gen-
tly and equally curving throughout their
lengths. Ventral margin semicircular. Exterior
surface rough, sparsely pitted (Fig. 32E).
Central ligament 30 to 34 |am (32 ± 1 .47 цт,
58
HOGGARTH
n = 6) long, centered about 45% from poste-
rior to anterior. Anterior ligament 6 to 9 |.im (8
± 1 .51 |.im, n = 6) long; posterior ligament 3 to
5 цт (4 ± 0.82 цт, n = 6) in length; central lig-
ament comprising about 75% of total hiinge
length. Micropoints extremely small, almost
undetected even at high magnification, unor-
ganized (Fig. 32C), on rim of ventral valve
margin and on exterior surface of valve. Ven-
tral flange not observed.
Remarks
This glochidium is figured by Surber (1912;
pi. 2, fig. 31 , as T. tuberculata, = T. verrucosa,
fide Ortmann, 1919), who gave length and
height measurements of 85 цт x 90 цт. It can
be distinguished from all other glochidia by its
very small size and the outline of its valve.
Quincuncina ¡nfucata (Conrad, 1834)
(Fig. 33A-G)
Material Examined
OSUM 48537.1, 48537.2 - Suwannee
River at Fl. Rt. 51 bridge, 8.4 mi. SSE of
Jasper, Sec. 17, Hamilton/Suwannee Co.,
Florida, 14 May 1978, W. J. Clench et al.
Description
Glochidium subelliptical, with a short hinge
line, equally curved lateral margins, a broadly
curved ventral margin. Glochidium 234 to 242
|im (240 ± 4.00 цт, n = 4) in length, 275 to 287
[im (283 ± 5.68 цт, n - 4) in height. Dorsal
margin straight, 102 to 105 цт (103 ± 1.34
|im, n = 5) in length. Valve surface densely pit-
ted, except along valve margin, and rough
sculpture covering exterior valve surface (Fig.
33F). Central ligament 54 to 58 цт (56 ± 2.83
|im, n = 2) long, centered about 48% from pos-
terior to anterior. Anterior ligament 27 to 28 ¡.im
(28 ± 0.71 jam, n = 2) long, posterior ligament
20 to 22 f.im (21 ± 1 .54 |.im, n - 2) in length. Mi-
cropoints coronal, with fused bases and lance-
olate points, not extending onto ventral rim or
on ventral margin of valve, covering about
90% of ventral flange, leaving narrow un-
sculptured distal flange margin.
Remarks
This glochidium will not be confused with
that of any other species examined. The most
striking feature is its unusual micropoint struc-
ture. The more proximal micropoints resem-
ble crowns. The bases of the micropoints are
fused with their points extending outward. The
number of points in each "crown" range from
seven along the proximal margin of the flange
to two points distally. The furthest micropoints
on the flange are simple lanceolate points. No
published figure of this glochidium was found.
Elliptio dilatata (Rafinesque, 1820)
(Fig. 34A-F)
Material Examined
MAH 946.9 - Kalamazoo River above St.
Rt. 60 bridge, 3.0 mi. WSW of Spring Arbor,
12.0 mi. WSW of Jackson, Jackson Co.,
Michigan, 13 May 1986, M. A. Hoggarth.
Description
Glochidium subelliptical, length 210 to 219
цт (216 ± 4.27 цт, n = 4), height 219 to 225
(.im (221 ± 3.00 (im, n = 4). Dorsal margin
straight, 140 to 147 [xm (143 ± 2.99 }im, n =
4) in length. Ventral margin semicircular. Lat-
eral margins subequal, with anterior margin
slightly more produced than posterior margin.
Valve pitting eliminated in region of adductor
muscle scar (Fig. 34D) and sparse throughout
remainder of valve. Loose-looped sculpture
covering exterior surface of valve (Fig. 34C).
Central ligament 45 to 50 цт (48 ± 3.54 цт,
n = 2) long, centered about 43% from poste-
rior to anterior. Anterior ligament about 56 цт
long; posterior ligament 36 to 39 цт (38 ±
2.12 цт, n = 2) in length. Micropoints lanceo-
late, located on ventral rim of valve and
on narrow ventral flange, arranged in broken
vertical rows, covering most of the area of
flange. Unsculptured distal flange margin nar-
row.
Remarks
This glochidium is figured by Lea (1 874: pi.
21, fig. 10, as Unio gibbosus, = E. dilatata.
We Ortmann & Walker, 1922), Lefevre & Cur-
tis (1910; fig. N; 191 2; fig. 10, as U. gibbosus),
Ortmann (1911 ; pi. 89, fig. 7, as E. gibbosus),
and Surber (1912: pi. 2, fig. 38, as U. gibbo-
sus). Lefevre & Curtis (1910, 1912) gave
length and height measurements of 220 цт x
190 МГП for this glochidium, Surber (1912)
gave 200 цт x 21 5 ¡.im, Ortmann (1912) gave
200 цт x 220 yivn, and Ortmann (1919, as E.
cuprous. = E. dilatata, fide Johnson, 1970)
gave 200 |.im x 200 ).im.
GLOCHIDIAOF UNIONIDAE
59
FIG. 33. Glochidium of Quincuncina infucata:A. exterior valve. OSUM 48537.2, bar length = 40 .urn; B. inte-
rior valve, OSUM 48537.2, bar length = 40 цт; С. micropoints, OSUM 48537.2, bar length = 5 цт; D. mi-
cropoints. OSUM 48537.2, bar length = 5 цт; E. micropoints, OSUM 48537.1 , bar length = 2 цт; F. exterior
valve sculpture. OSUM 48537.1, bar length = 2|лт; G. hinge, OSUM 48537.1, bar length = 15 цт.
Subfamily Lampsilinae
Ptychobranchus fasciolaris
(Rafinesque. 1820)
(Fig. 35A-E)
Material Examined
MAH 640.1 - Little Darby Creek at Co.
Rt. 131 bridge (Grewell Rd.), 1.8 mi. E of
Plumwood, 7.7 mi. NW of West Jefferson,
Monroe Twp., Madison Co., Ohio, 11 May
1984, M. A. Hoggartli & G. T. Watters. MAH
651 - Little Darby Creek at Little Darby Rd.
access, 2.7 mi. SE of Plumwood, 6.4 mi. NW
of West Jefferson, Monroe Twp., Madison
Co., Ohio, 17 May 1984, M. A. Hoggarth & G.
T. Watters.
HOGGARTH
FIG. 34. Glochidium of Elliptio dilatata. MAH 946.9: A. exterior valve, bar length = 35 ,um: B. interior valve,
bar length = 35 цт; С. exterior valve, bar length = 1 \.ivr\: D. interior valve, bar length = 1 |.im; E. micropoints,
bar length = 5 |jm; F. hinge, bar length = 20 цт.
Description
Glochidium subelliptical, higlier than long,
length 1 70 to 1 75 цт (1 73 ± 2.89 цт, n = 3),
height 182 to 195 цт (187 ± 6.81 цт, n = 3).
Dorsal margin slightly curved, 80 to 89 )Lim (83
± 4.93 |лт, n = 3) in length. Anterior and pos-
terior margins equally curved to a maximum
inflation at about 60% from dorsal to ventral;
ventral margin broadly rounded. Surface of
the valve smooth, with only a few pits. Dorsal
alae absent. Loose-looped sculpture covering
exterior surface of valve (Fig. 35E). Central
ligament 36 to 40 мт (38 ± 2.83 цт, n = 2)
GLOCHIDIAOF UNIONIDAE
61
FIG. 35. Glochidium of Ptychobranchus fasciolaris: A. exterior valve, MAH 640.1 , bar length = 25 цгл; В. in-
terior valve, MAH 640.1 , bar length = 25 цгл; С. lateral view, MAH 640.1 , bar length = 35 |am; D. ventral valve
edge, MAH 651 , bar length = 5 цт; E. exterior valve sculpture, MAH 651 , bar length = 1 |am.
long, centered about 49% from posterior to
anterior. Posterior ligament 1 9 to 22 цт (21 ±
2.12 цт, n = 2) long; anterior ligament 22 to
26 цт (24 ± 2.83 |.im, n = 2) in length. Micro-
points very small, few in number, located on
ventral margin of valve rather than on broad
ventral flange.
Remarks
This glochidium can be distinguished by its
small size, central ligament position, and its
very simple ventral margin. It was figured by
Lea (1858: pi. 5, fig. 12, as Unio phaseolus, =
P. fasciolaris, fide Ortmann & Walker, 1922)
62
HOGGARTH
and Ortmann (1911: pi. 89, fig. 14, as P.
phaseolus). Ortmann (1912) gave 170 ).im x
190 |.im for length) and height of this glochid-
ium and noted that Lea's figure does not rep-
resent its shape or size accurately. The
glochidium of P. fasciolaris is much smaller
than that of Ligumia recta rather than, as pic-
tured by Lea (1858), larger than that species.
Ptychobranchus occidentalis (Conrad, 1836)
(Fig. 36A-D)
Material Examined
OSUM - 45361 .14, 45361 .17 Current River
"between Mo. Rt. 106 and Van Buren," about
19 mi. of stream, Shannon/Carter Co., Mis-
souri, 24 October 1977, R. D. Oesch.
Description
Glochidium subelliptical, higher than long,
length 197 to 203 цт (200 ± 4.24 ¡am, n = 2),
height 234 to 241 |im (238 ± 4.95 |.im, n = 2).
Dorsal margin slightly curved, 101 to 104 |im
(1 02 ± 1 .53 |jm, n = 3) in length. Anterior and
posterior margins equally and gently curving
to their maximum inflation at about 60% from
dorsal to ventral; ventral margin broadly
curved. Dorsal alae absent; valve surface
sparsely pitted. Loose-looped sculpture cov-
ering exterior valve surface (Fig. 36C). Cen-
tral ligament 44 to 48 цт (46 ± 2.83 цт, n =
2) in length, centered about 47% from the
posterior margin. Anterior ligament 30 to 34
|jm (32 ± 2.83 |iim, n = 2) in length; posterior
ligament 23 to 26 ¡am (25 ± 2.12 цт, n = 2)
long. Ventral flange poorly developed, very
narrow, covered with micropoints. Micropoints
lanceolate and unorganized.
Remarks
This glochidium is slightly larger than that of
P. fasciolaris and is further distinguished from
that species by its larger and more numerous
micropoints. No published figure of this glo-
chidium was found.
Ptycliobranctius greeni (Conrad, 1 834)
(Fig. 36E-H)
Material Examined
OSUM 19025.2 - Conasauga River above
lower Kings Bridge, 5 mi. WNW of Eton, Mur-
ray/Whitfield Co., Georgia, 25 September
1966, D. H. Stansbery & H. D. Athearn.
Description
Glochidium subelliptical, higher than long,
symmetrical, length 183 to 190 ¡.im (187 ±
3.61 (jm, n = 3), height 226 to 228 |im (227 ±
1.15 |.im, n = 3). Dorsal margin sligfitly curved,
90 to 96 |im (93 ± 3.06 цт, n = 3) in length.
Anterior and posterior margins equal and
slightly curved to their point of maximum infla-
tion at about 70% from dorsal to ventral. Ven-
tral margin broadly rounded. Surface of valve
sparsely pitted. Loose-looped sculpture cov-
ering exterior surface of valve (Fig. 36G). Dor-
sal alae present but very small, oriented al-
most perpendicular to hinge. Central ligament
about 49 )im long centered about 48% from
posterior to anterior. Anterior ligament about
26 |jm long; posterior ligament 21 |im in
length. Micropoints occur on ventral rim of
valve, small, lanceolate, unorganized. Ventral
flange not observed.
Remarks
This glochidium is about the same size as
that of P. fasciolaris; however, it can be distin-
guished from that species by its dorsal alae
and micropoints. This glochidium is figured by
Lea (1858: pi. 5, fig. 16, as Unio woodwar-
dianus, - P. greeni. fide Ortmann, 1923-
1924). His figure is essentially correct except
for size. Glochidia of Lampsilis ovata and L.
fasciola are larger than those of P. greeni
rather than smaller as figured by Lea.
Ptychobranchus subtentum (Say, 1 825)
(Fig. 37A-F)
Material Examined
OSUM 43156.5 - Clinch River at mouth of
Copper Creek, 1 .3 mi. S of Clinchport, 9.3 mi.
W of Gate City, Scott Co., Virginia, 21 October
1978, D. H. Stansbery et al.
Description
Glochidium subelliptical, higher than long,
symmetrical, length 181 to 195 цт (190 ±
5.54 |.im, n = 5), height 236 to 251 цт (241 ±
FIG. 36. Glochidium of Ptychobranchus occidentalis ( A-D) and Ptychobranchus greeni (E-H); A. exterior
valve OSUM 45361.17. bar length = 35|im: B. interior valve, OSUM 45361.14. bar length = 35 цт; С. exte-
rior valve sculpture, OSUM 45361 .17, bar length = 1 цт; D. hinge, OSUM 45361 .17. bar length = 1 5 |am; E.
exterior valve, OSUM 19025.2, bar length = 35 цт; F. interior valve, OSUM 19025.2, bar length = 35 |лт;
G. exterior valve sculpture, OSUM 19025.2, bar length = 1 |am; H. micropoints, OSUM 19025.2, bar length
= 2|im.
64
HOGGARTH
FIG. 37. Glochidium of Ptychobranchus subtentum OSUM 431 56.5: A. exterior valve, bar length = 35 ^m; B.
interior valve, bar length = 35 |.im; С exterior valve sculpture, bar length = 1 ¡.im; D. hinge, bar length =
20 |.im; E. micropoints, bar length = 5 (.im; F. interior valve pitting, bar length = 10 цт.
5.76 |.im, n = 5). Dorsal margin slightly curved,
82 to 90 |.im (85 ± 2.88 |im, n = 6) in length.
Anterior and posterior margins about straight
to their maximum inflation at about 60% from
dorsal to ventral, where they begin to curve to
form a broadly rounded ventral margin. Valve
surface sparsely pitted, loose-looped sculp-
ture covering exterior surface of valve (Fig.
37C). Adductor muscle scar present (Fig.
36B,D,F). Dorsal alae short, poorly devel-
oped. Central ligament 36 to 38 цт (37 ±
1.15 цт, n = 3) in length, centered about 48%
from posterior to anterior. Anterior ligament 25
to 26 |.im (25 ± 0.58 |.irn, n = 3) long; posterior
GLOCHIDIA OF UNIONIDAE
65
ligament 20 to 24 цт (22 ± 2.08 ¡im, n = 3) in
iengtli. Micropoints small, numerous, unorga-
nized, covering a large portion of ventral rim,
leaving very narrow unsculptured distal mar-
gin.
Remarks
This glochidium can be distinguished by its
relatively short hinge line, simple micropoints
and small dorsal alae. Ortmann (1912) noted
that the glochidium of P. subtentum is larger
than that of P. fasciolaris. He gave 180 цт x
220 |im for the length and height of P. subten-
tum glochidia. This glochidium is figured by
Ortmann (1912: pi. 29, fig. 5).
Obliquaria reflexa Rafinesque, 1820
(Fig. 38A-D)
Material Examined
OSUM 54361.1 - Mississippi River, R.Mi.
564.5-566.1 , about 3.5 mi. SW of Galena (IL),
Jo Daviess/Jackson Co., Illinois/Iowa, 7-8
August 1979, R. B. Lewis et al.
Description
Glochidium subrotund, symmetrical, length
213 to 219 |Lim (217 ± 3.46 |jm, n = 3), height
206 to 221 [im (215 ± 8.14 цт, n = 3). Dorsal
margin slightly curved outward, 1 1 9 to 1 27 ¡am
(122 ± 4.16 цт, n = 3) in length. Lateral and
ventral valve margins more or less round in
outline, with maximum inflation of both side
margins at about 50% from dorsal to ventral.
Exterior surface malleated and pitted, except
along valve margin, where shell fairly smooth.
Within this smooth marginal area, longitudinal
ridges present (Fig. 38C). Dorsal alae absent.
Central ligament 52 to 54 ¡.im (53 ± 1.41 ¡am,
n = 2) in length, centered about 46% from
posterior to anterior. Anterior ligament 37 to
44 jam (41 ± 4.95 \xm, n = 2) long; posterior
ligament 30 to 31 цт (31 ± 0.71 |im, n = 2) in
length. Large, bluntly pyramidal micropoints
occurring on ventral rim and on narrow ventral
flange, decreasing in size distally, arranged in
broken vertical rows that extend two thirds
length of flange, leaving a narrow unsculp-
tured distal flange edge. Micropoints extend-
ing laterally to about point of maximum lateral
inflation of valves.
Remarks
Lefevre & Curtis (1910, 1912), Surber
(1912), and Ortmann (1912) gave the follow-
ing measurements for this glochidium; 230
цт X 225 |im, 225 |im x 235 цт, and 220 цт
X 220 |im. Ortmann (1919) noted that the
shape of this glochidium, "may best be com-
pared with a circle a small section of which is
cut off." This glochidium can be distinguished
by its shape and micropoints. It was figured by
Lefevre & Curtis (1910; fig. M, 1912; fig. IN),
Surber (1912; pi. 2, fig. 39), and Ortmann
(1912; pi. 20, fig. 1).
Cyprogenia stegaria (Rafinesque, 1820)
(Fig. 39A-D)
Material Examined
OSUM 6298.21 - Green River at Munford-
ville. Hart Co., Kentucky, 22 October 1 961 , D.
H. Stansbery.
Description
Glochidium elongate-oval, subrotund,
length 204 to 208 |.im (206 ± 2.08 цт, n = 3),
height 1 64 to 1 70 |.im (1 67 ± 3.06 цт, n = 3).
Dorsal margin straight, long, with a length of
113to120|jm(116 ± 3.00 |.im,n = 5). Anterior
and posterior margins curving greatly to create
mirror images of each other, with maximum lat-
eral inflation at about 50% from dorsal to ven-
tral. Exterior surface smooth, lacking mal-
leations, with only a few pits. As pointed out
by Sterki (1 898), concentric ridges occur near
margins of valve. Valve disc smooth, except
for loose-looped exterior valve sculpture (Fig.
39C). Dorsal alae absent. Central ligament 35
to 40 |лт (38 ± 2.65 \im, n = 3) in length, cen-
tered about 45% from posterior to anterior.
Posterior ligament 30 to 33 \im (32 ± 1 .53 ¡am,
n = 3) in length; anterior ligament 43 to 47 jam
(45 ± 2.08 |im, n = 3) long. Micropoints lance-
olate, bluntly pointed, located on the ventral
rim of valve. Ventral flange not observed.
Remarks
This glochidium can be distinguished from
other species by its more broadly rounded
valve outline. Sterki (1898), Surber (1912),
and Ortmann (1912) gave the following mea-
66
HOGGARTH
FIG. 38. Glochidium of Obliquaria reflexa, OSUM 54361 .1 ; A. exterior valve, bar length = 35 |am; B. interior
valve, bar length = 35 ¡.im; С ventral view, bar length = 25 urn; D. micropoints, bar length = 7 ¡.im.
surements for length and height; 210 цт x
170).im,210|.imx 185|.im, and 180дтх 150
цт. This glochidium was figured by Ortmann
(1911: pi. 19, fig. 6, as С /rroratà) and Surber
(1912: pi. 1, fig. 11, as С ¡rrorata).
Cyprogenia aberti (Conrad, 1 850)
(Fig.39E-l)
Material Examined
OSUM - 48067 St. Francis River along Mo.
Rt. E, 2.5 mi. NNE of French Mills, 11.5 mi.
SW of Fredericktown, SE 1/4, Sec. 3, T32N,
R5E, Madison Co., Missouri, 5 October 1977,
W. L. Pflieger & T. Grace.
Description
Glochidium elongate oval, subrotund,
length 200 to 218 (.im (208 ± 7.03 |.im, n = 6),
height 143 to 161 цт (154 ± 7.79 цт, n = 6).
Dorsal margin straight, 125 to 136 цт (131 ±
4.18 |.im, n = 7) long. Anterior and posterior
margins equally curved to their maximum lat-
eral inflation at about 50% from dorsal to ven-
tral. Ventral margin flatly curved, dorsal alae
absent. Central ligament 28 to 36 цт (31 ±
3.11 i-im, n = 5) long, centered about 46%
from posterior to anterior. Anterior ligament 48
to 61 \.[m (54 ± 6.04 \.im. n = 5) in length; pos-
terior ligament 41 to 49 ¡.im (44 ± 3.13 цт. n
= 5) long. Micropoints lanceolate, bluntly
FIG. 39. Glochidium of Cyprogenia stegaria (A-E) OSUM 6298.21, and Cyprogenia aberíi (F-l) OSUM
48067: A. exterior valve, bar length = 30 цт: В. interior valve, bar length = 30 цт; С. exterior valve sculp-
ture, bar length = 2 цт: D. hinge, bar length = 20 ).im: E. exterior valve, bar length = 30 |агл; F. interior valve,
bar length = 30 цт; G. micropoints, bar length = 5 |.im; H. micropoints, bar length = 5 цт; I. hinge, bar length
= 20|im.
68
HOGGARTH
pointed, unorganized on ventral margin. Ven-
tral flange only poorly developed.
Remarks
This glochidium is essentially like that of С
stegaria. although it is slightly more de-
pressed than that species. The only other
glochidium that approaches this shape is that
of Dromus dramas. However, the glochidium
of D. dramas would not be confused with ei-
ther of these because of its extreme valve de-
pression. No published figure of the glochid-
ium of С aberti\Nas found.
Dromus dramas (Lea, 1 834)
(Fig. 40A-E)
Material Examined
OSUM 20407.1 - Powell River 2.5 mi. ENE
of Hoop, 11.5 mi. NE of Tazewell, Clairborne
Co., Tennessee. 20 October 1968, G. F.
Ahrens. OSUM 23200.9 - Powell River at
Hoop ("Brooks Bridge"), 9.5 mi. NE of
Tazewell, Clairborne Co.. Tennessee, 19 Oc-
tober 1969, D. H. Stansbery et al.
Description
Glochidium fabelliform or bean-shaped,
much longer than high, with a straight dorsal
margin, narrowly rounded anterior and poste-
rior margins, broadly curved ventral margin.
Glochidium 219 to 230 |.im (224 ± 5.09 цт,
n = 5) long, 114 to 120 |.im (118 ± 2.49 цт,
n = 5) in height. Dorsal margin 160 to 182
цт (174 ± 8.65 цт, n = 6) long. Anterior and
posterior margins equal, with their points of
maximum inflation occurring about 30%
from dorsal to ventral. Exterior valve surface
smooth, with pits only evident in internal view
(Figs. 40B-D). Pits absent at valve mar-
gins, but present in central portion of valve
(Fig. 40C). Dorsal alae absent. Central liga-
ment 39 to 40 |.im (39 ± 0.58 ,um, n = 3),
centered at about 45% from posterior to ante-
rior. Anterior ligament 76 to 82 um (79 ± 3.06
|im, n = 3) in length; posterior ligament 51 to
61 |im (56 ± 4.58 |.im, n = 3) in length. Nu-
merous bluntly pointed micropoints present
on valve margin. Ventral flange only poorly
developed.
Remarks
This glochidium is easily distinguished by
its shape. It represents the end of an evolu-
tionary line from the nearly round glochidium
of O. reflexa. through the progressively more
depressed valve of С stegaria and C. aberti,
to the extremely depressed valves of this
species. Surber (1912), Lefevre & Curtis
(1912), and Ortmann (1912) gave 190 [im x
1 00 i^m for length and height. This glochidium
was figured by Surber (1912: pi. 1, fig. 13),
Lefevre & Curtis (1912: fig. IM) and Ortmann
(1912: pi. 29, fig. 7).
Actinanaias pectarasa (Conrad, 1834)
(Fig. 41A-G)
Material Examined
A. pectarasa: OSUM 24337 - Middle Fork
Holston River, 3.7 mi. S of Glade Spring at Va.
Rt. 91 bridge (Craigs Bridge), Washington Co.,
Virginia, 16 September 1968, D. H. Stansbery
& W. J. Clench. OSUM 48748.3 - Clinch River
below footbridge at Slant. 7.1 mi. NNW of Gate
City, 8.1 mi. ENE of Clinchport, Dekalb Twp.,
Scott Co., Virginia, 22 December 1980, С С.
Coney. A. I. cahnata: МАИ 842.1 - Fish Creek
at bridge 2.0 mi. NW of Edgerton, 11.9 mi. W
of Bryan, Sec. 20, T6N, R1 E, St. Joseph Twp.,
Williams Co., Ohio, 29 October 1985, D. H.
Stansbery et al.
Description
Glochidium subelliptical, nearly symmetri-
cal, length 244 to 253 цт (248 ± 4.58 цт,
n = 3), height 260 to 270 f.im (267 ± 5.77 цт,
n = 3). Dorsal margin straight, 139 to 151 \xm
(144 ± 5.20 |.im, n = 3) in length. Anterior and
posterior margins gently and evenly curving:
ventral margin semicircular. Exterior surface
of valve malleated and pitted, except at its
margin, where longitudinal ridges occur (Fig.
41 С D). Dorsal alae well developed, about 38
|.im in length (Fig. 41 A, C). Beaded exterior
valve sculpture covering surface of valve (Fig.
41 G). Central ligament 58 to 62 ,um (60 ±
3.21 j.im, n = 3) in length, centered about 46%
from posterior to anterior. Anterior ligament 46
to 52 цт (48 ± 3.21 цт, n = 3) long: posterior
ligament 31 to 39 .um (35 ± 3.61 urn, n = 3) in
length. Micropoints, lanceolate, located on
ventral margin and on a narrow ventral flange.
GLOCHIDIA OF UNIONIDAE
69
FIG. 40. Glochidium of Dromus dramas: A. exterior valve, OSUM 20407.1, bar length = 30 цт; В. interior
valve, OSUM 20407.1 , bar length = 30 |агл: С. interior valve pitting. OSUM 23200.9. bar length = 5 \xm; D. in-
terior valve pitting, OSUM 20407.1 , bar length = 3 цт: E. micropoints, OSUM 23200.9, bar length = 2 ¡am.
Ventral flange slightly produced centrally to
give the abducted valve a beak-like appear-
ance (Fig. 41 C, D); this region of flange
densely covered with micropoints and proba-
bly facilitates attachment by digging deeply
into host tissue.
Remarks
The glochidium of A. I. carinata is nearly
identical to that of A. pectorosa. except it is
smaller. A single glochidium of A. I. carinata
gave the following measurements: length, 220
цт; height, 243 (.im; hinge length, 125|.im; an-
terior ligament length, 38 цт; central ligament
length, 57 цт; and posterior ligament length,
30 um. Ortmann (1912) gave 220 цт x 240 |.im
for length and height measurements for A. li-
gamentina and 250 цт x 290 \xrr\ for Nephro-
najas pendix (= A. pectorosa. fide Ortmann &
Walker, 1922). The glochidium of A. liga-
mentina was figured by Lea (1858: pi. 5, fig.
18), Ortmann (1911: p1. 89, fig. 16), and
Surber (1 91 2: pi . 2, fig. 1 8). The glochidium of
A. pectorosa was figured by Ortmann (1912:
pi . 1 9, fig. 1 2, as N. pendix).
Obovaria retusa (Lamarck, 1819)
(Fig. 42A-D)
Material Examined
UMMZ Uncataloged Tennessee River (Ken-
tucky Lake), U.S. Rt. 70 bridge. New John-
70
HOGGARTH
FIG. 41 . Glochidium of Actinonaias pectorosa;A. exterior valve, OSUM 48748.3, bar length = 35 цт; В. in-
terior valve, OSUM 48748.3, bar length = 40 мт; С. lateral view, OSUM 24337, bar length = 60 цт: D. mi-
cropoints, OSUM 24337, bar length = 5 pm; E. micropoints, OSUM 24337, bar length = 2 цт; F. hinge, OSUM
48748.3, bar length = 20 цт; G. exterior valve sculpture, OSUM 48748.3, bar length = 1 цт.
FIG. 42. Glochidium of Obovaria retusa UMMZ Uncataloged (A-D) and Obovaria olivaria OSUM 51282.2
(E-H): A. exterior valve, bar length = 40 ¡im; В. interior valve, bar length = 40 цт; С. exterior valve sculp-
ture, bar length = 1 um; D. hinge, bar length = 20 цт; E. exterior valve, bar length = 40 цт; F. interior valve,
bar length = 40 цт; G. exterior valve sculpture, bar length = 1 ¡лт; H. hinge, bar length = 1 5 цт.
72
HOGGARTH
sonville, Humphreys Co., Tennessee, October
1958. J. Bates.
Description
Glochidium subelliptical, slightly asymmet-
ric, length 21 8 to 223 цт (221 ± 3.54 |лт, n -
2), height 272 to 278 цт (275 ± 4.24 |ит, n =
2). Dorsal margin slightly curved, 115 to 119
|am (1 1 7 ± 2.08 ¡.im, n = 3) in length. Dorsal half
of posterior margin straight, oblique to the dor-
sal line. Ventral half of posterior margin gently
curved initially, then straight to run more or less
perpendicular to dorsal line. Anterior margin
gently curved throughout its length, with max-
imum lateral inflation occurring at about 70-
80% from dorsal to ventral. Ventral margin
semicircular in outline. Exterior valve surface
finely malleated and evenly pitted, except at
valve margin, where surface is smooth. Dorsal
alae about 39 \xvr\ in length. Beaded to loose-
looped sculpture covering exterior surface of
valve (Fig. 42C). Central ligament about 45 |im
in length, centered at about 45% from poste-
rior to anterior. Posterior ligament about 36 цт
long; anterior ligament about 38 lam long. Mi-
cropoints lanceolate, arranged in broken verti-
cal rows on ventral rim of valve and on narrow
ventral flange.
Remarks
Surber (1912) gave measurements for
length and height of 240 цт x 295 |im,
whereas Ortmann (1912) gave 220 цт x 270
jam. This rather wide discrepancy is not ad-
dressed by either author and demonstrates
the difficulty of trying to determine the species
of a glochidium based solely on size. In the
case of O. retusa, the moderately sized dorsal
alae and valve shape distinguish the glochid-
ium from species other than those of Obo-
varia. Its beaded to loose-looped sculpture
distinguish this species from the other mem-
bers of the genus. This glochidium was fig-
ured by Lea (1858: pi. 5, fig. 7), Surber (1912:
pi. 3, fig. 47), and Ortmann (1912: pi. 19, fig.
9). Lea's figure shows the anterior and poste-
rior margins evenly curved, whereas Surber's
figure shows the correct outline.
Obovaria olivaría (Rafinesque, 1820)
(Fig. 42E-H)
Material Examined
OSUM 51282.2 Mississippi River, R.Mi.
634.7, at U.S. Rt. 18 bridge, main channel.
1.5 mi. W of Prairie du Chien, Crawford Co.,
Wisconsin, 1 5 May 1 981 , M. E. Havlik et al.
Description
Glochidium subelliptical, slightly asymmet-
ric, length 1 98 to 206 цт (202 ± 4.00 |im, n =
3), height 254 to 261 цт (258 ± 3.51 цт, n =
3). Dorsal margin slightly curved, 101 to 118
|.im (109 ± 5.81 \лт, n = 5) long. As in O. re-
tusa, posterior margin is mostly straight sided,
with a bend about half the distance from dor-
sal to ventral. Anterior margin gently curved;
ventral margin semicircular. Dorsal alae well
developed, about 37 цт in length. Loose-
looped sculpture covering exterior surface of
valve (Fig. 42G). Central ligament 34 to 44 ¡am
(39 ± 5.03 |im, n = 3) in length, centered
about 45% from posterior to anterior. Poste-
rior ligament 26 to 29 цт (27 ± 1 .53 |im, n =
3) long; anterior ligament 32 to 43 цт (39 ±
5.86 |.im, n = 3) long. Micropoints lanceolate,
located on ventral rim and on narrow ventral
flange.
Remarks
This glochidium is very similar to that of O.
retusa. However, it is smaller and it has a dif-
ferent exterior valve sculpture. This glochid-
ium was figured by Surber (1912: pi. 2, fig. 25,
as O. ellipsis, = O. olivaría, fide Ortmann &
Walker, 1922) and Ortmann (1912: pi. 19, fig.
11). Surber gave length and height measure-
ments of 210 цт X 265 цт and Ortmann
(1912) recorded length and height measure-
ments of 1 90 |jm X 220 |лт. Ortmann's figures
are considerably smaller than Surber's or
mine.
Obovaria subrotunda (Rafinesque, 1820)
(Fig. 43A-E)
Material Examined
MAH 659.2 - Big Darby Creek at Scioto-
Darby (Mt. Sterling-Commercial Pt.) Rd.
bridge, 3.4 mi. S of Orient, 1 5.3 mi. NW of Cir-
cleville, Scioto/Darby Twp., Pickaway Co.,
Ohio, 18 May 1984, M. A. Hoggarth & G. T
Watters. MAH 805.1 - Fish Creek at Co. Rt.
49 bridge, 0.4 mi. above its mouth, 1.1 mi. N
of Edgerton, 10.4 mi. W of Bryan, St. Joseph
Twp., Williams Co., Ohio, 29 October 1985, D.
H. Stansbery et al.
GLOCHIDIAOF UNIONIDAE
73
FIG. 43. Glochidium of Obovaria subroturda; A. exterior valve, MAH 805.1, bar length = 30 цт; В. interior
valve, MAH 805.1 , bar length = 35 цт: С. micropoints, MAH 805.1 , bar length = 3 цт; D. hinge, MAH 659.2,
bar length = 1 5 цт; E. exterior valve sculpture, MAH 805.1 , bar length = 1 ¡jm.
Description
Glochidium subelliptical, length 174 to 180
|im (177 ± 2.70 j.im, n = 5), height 197 to 210
|im (204 ± 5.63 jim, n = 5). Dorsal margin
slightly curved, 85 to 95 |дт (91 ± 3.50 ¡am, n
= 6) in length. Posterior margin straight,
oblique dorsally; ventral portion of posterior
margin curved. Anterior margin gently curved
throughout its length, with maximum lateral in-
flation at about 60% from dorsal to ventral.
Ventral margin broadly rounded. Dorsal alae
about 29 ¡ivn in length, and loose-looped valve
sculpture covering exterior surface of the
valve (Fig. 43E). Central ligament 40 to 43 цт
(42 ± 2.12 |дт, n = 2) in length, centered
about 48% from posterior to anterior. Anterior
ligament 24 to 29 |дт (27 ± 3.54 цт, n = 2)
long; posterior ligament 21 to 25 цт (23 ±
2.83 |дт, n = 2) in length. Micropoints lanceo-
late, arranged in broken vertical rows on ven-
tral rim and on short ventral flange. Unsculp-
tured distal flange margin narrow.
Remarks
Surber (1915) described this glochidium as,
"semielliptical in shape; ventral margin
74
HOGGARTH
rounded; hinge line long and slightly de-
pressed near center; size medium." He sug-
gested that this glochidium is intermediate be-
tween those of O. retusa and O. olivaria but
differs from both by its smaller size. This study
supports Surber's observations. This glochid-
ium shares moderately long dorsal alae with
O. retusa. and moderately dense loose-
looped sculpture with O. olivaria. Ortmann
(1912) gave 200 |.im x 230 цт for length and
height, whereas Surber (1915) gave 1 70 |Lim x
215 |.im. This glochidium was figured by Ort-
mann (1911: pi. 89, fig. 15, as O. circulus, = O.
subrotunda. fide Ortmann & Walker, 1922),
and Surber (1915; pi. 1 , fig. 8, as O. circulus).
Obovaria jacksoniana (Frierson, 1912)
(Fig. 44A-D,G)
Material Examined
OSUM 50233.8 - Sipsey River below Rt. 21
bridge, 1.3 mi. W of Brownville, 16.8 mi. NW
of Tuscaloosa, Tuscaloosa Co., Alabama, 10
October 1981, L. M. Koch.
Description
Glochidium subelliptical, almost symmetri-
cal, length 1 75 to 1 87 |im (1 82 ± 5.74 цт, n =
4), height 230 to 243 цт (236 ± 6.56 ¡.im, n =
4). Dorsal margin slightly curved, 89 to 1 00 цт
(95 ± 4.57 jjm, n = 4) long. Posterior and an-
terior margins about equal, except posterior
margin slightly more produced as a result of a
larger angle of divergence. Posterior margin
curving ventrally at about 50% from dorsal to
ventral, becoming almost perpendicular to
dorsal margin, anterior margin gently rounded.
Exterior surface smooth with just a few pits
and very little malleation, except at the umbo.
Loose-looped sculpture covering exterior sur-
face (Fig. 44C). Dorsal alae with a length of
about 34 |.im. Central ligament 41 to 42 ¡.im (42
± 0.71 jim, n = 2) in length, centered 46% from
posterior to anterior. Anterior ligament 26 to 31
цт (29 ± 3.54 pm, n = 2) long; posterior liga-
ment 21 to 25 |.im (23 ± 2.83 |.im, n = 2) long.
Micropoints lanceolate, located on ventral rim
of valve and on short ventral flange.
Remarks
This glochidium is very similar to the other
members of the genus. It is intermediate in
size between O. olivarla and O. subrotunda.
and much smaller than O. retusa. No pub-
lished figure of this glochidium was found.
Obovaria unicolor (Lea, 1845)
(Fig. 44E, F, H)
Material Examined
OSUM 33158 - Sabine River at La. Rt. 8
and Tex. Rt. 3 bridge, 1 .3 mi. W of Burr Ferry,
Vernon Parish/Newton Co., Louisiana/Texas,
14 July 1968, D. H. Stansbery et al. OSUM
47696. 6-Pearl River, 0.8 mi. SSE of Drysdale,
6.7 mi. SW of Carthage, Sec. 25/36, T10N,
R6E, Lake Co., Mississippi, 5 October 1979,
R. G. Rummel & P Hartfield.
Description
Glochidium subelliptical, almost symmetri-
cal, length 1 68 to 1 80 цт (1 72 ± 4.36 (im, n =
5), height 218 to 234 (.im (225 ± 5.86 цт, n =
5). Dorsal line slightly curved, 85 to 96 цт (90
± 4.22 j.im, n = 6) long. Anterior and posterior
margins about equal, straight sided dorsally,
slightly curved ventrally, as in O. jacksoniana.
Ventral margin broadly rounded. Exterior sur-
face mostly smooth, with malleations at umbo
and only a few pits. Loose-looped sculpture
covering exterior surface of valve Dorsal alae
about 32 j.im in length. Central ligament 32 to
33 цт (32 ± 0.58 цт, n = 3) in length, centered
about 45% from posterior to anterior. Anterior
ligament 32 to 41 ,um (35 ± 4.62 (.im, n = 3)
long; posterior ligament 23 to 28 ¡am (25 ±
2.65 um, n = 3) long. Micropoints lanceolate,
arranged in incomplete vertical rows, located
on ventral rim of valve and on short ventral
flange.
Remarks
This glochidium is essentially identical to
that of O. jacksoniana. My measurements
would indicate that the glochidium of O. uni-
color is smaller than that of Q. jacksoniana.
but even here there is overlap. This glochid-
ium is figured by Ortmann (1912; pi. 19, fig.
10), who gave length and height measure-
ments of 160 цт X 210 |im.
Ellipsaria lineolata (Rafinesque, 1820)
(Fig. 45A-F)
Material Examined
OSUM 1984; 14 - Ohio River, R.Mi. 443.0-
445.0, from 0.4 mi. NW of Moscow, Ohio, to
Point Pleasant, Ohio, 5.0-6.8 mi. SE of New
Richland, Ohio, Pendleton/Campbell Co.,
GLOCHIDIAOF UNIONIDAE
75
FIG. 44. Glochidium of Obovaria jacksoniana (A-D, G) OSUM 50233.8 and Obovaria unicolor (E, F, H); A.
exterior valve, bar length = 30 \.im; B. interior valve, bar length = 30 |.im; С exterior valve sculpture, bar length
= 1 цт; D. hinge, bar length = 15 |am; E. exterior valve, OSUM 33158, bar length = 30 |лт; F. interior valve,
OSUM 331 58, bar length = 30|im;G. micropoints, OSUM 33158, bar length = 2 цт; H. micropoints, OSUM
47696.6, bar length = 3 |.im.
76
HOGGARTH
Kentucky, 16 November 1984, D. H. Stans-
bery et al.
Description
Glochidium subiiguiate, much higher than
long, length 229 to 245 (.im (237 ± 5.06 |im,
n = 7), height 310 to 325 цт (321 ± 5.15 цт,
n = 7). Dorsal margin slightly curved, 87 to 96
lam (91 ± 3.55 цт, n = 7) long. Anterior and
posterior margins more or less straight to
slightly incurved dorsally. At about 40% from
dorsal to ventral, margins again becoming
straight, slightly divergent. Point of maximum
lateral inflation at about 80% from dorsal to
ventral. Ventral margin broad, over twice
length of hinge margin, but more narrowly
curved than found in Obovaria, Actinonaias ox
Ptychobranchus. External valve surface
finely malleated and evenly pitted, except
along valve margins, where pits and mal-
leations are absent (Fig. 45C). Vermiculate
sculpture covering exterior surface of valve
(Fig. 45F); dorsal alae very small and ori-
ented at a 45° angle to hinge. Central liga-
ment 35 to 51 |im (43 ± 8.42 \im, n = 4) long,
centered about 48% from posterior to ante-
rior. Posterior ligament 20 to 24 |.im (22 ±
1.63 |im, n = 4) long; anterior ligament 23 to
29 |jm (26 ± 2.75 цт, n = 4) in length. Mi-
cropoints lanceolate, arranged in complete
vertical rows on ventral valve margin and on
wide ventral flange, decreasing in size dis-
tally, covering proximal two thirds of ventral
flange, leaving a narrow unsculptured distal
flange margin. The ventral flange becomes
folded laterally so that it forms small hook-like
points (Fig. 45C).
Leptodea fragilis (Rafinesque, 1820)
(Fig. 46A-E)
Material Examined
MAH 626.1 - Big Darby Creek at Scioto-
Darby (Mt. Sterling-Commercial Pt.) Rd.
bridge, 3.4 mi. S of Orient, 15.3 mi. SW of
Columbus, Scioto/Darby Twp., Pickaway Co.,
Ohio, 13 April 1984, M. A. Hoggarth & G. T.
Watte rs.
Description
Glochidium subelliptical, very small, length
72 to 73 |im (72 ± 0.58 |im, n = 3), height 80
to 83 |im (83 ± 1.53 цт, n = 3). Dorsal mar-
gin straight, 30 to 35 цт (33 ± 2.89 |im, n =
3) long. Anterior and posterior margins about
equally divergent dorsally, ventrally curved.
Ventral margin broadly rounded, joining lat-
eral margins at their point of maximum infla-
tion, near 70% from dorsal to ventral. Exterior
surface only malleated near umbo; however,
valve surface otherwise densely pitted, ex-
cept along valve margin (Fig. 46A, C) and at
adductor muscle insertion (Fig. 46A, B). Exte-
rior surface lightly covered with loose-looped
valve sculpture (Fig. 46E); dorsal alae very
small. When fully adducted, valves gape in
lateral view (Fig. 46C). Central ligament about
11 |im long, centered about 45% from poste-
rior to anterior. Posterior ligament 8 to 10 )ит
(9 ± 1.41 цт, n = 2) in length; anterior liga-
ment 11 to 14 (.im (13 ± 2.12 цт, n = 2) long.
Micropoints wide, lamellate plates arranged in
complete vertical rows on ventral valve mar-
gin and on narrow ventral flange. Unsculp-
tured distal flange margin narrow.
Remarks
Lefevre & Curtis (1910), Surber (1 912), and
Ortmann (1912) gave the following measure-
ments for length and height: 230 цт x 31 цт,
230 |jm X 330 |im, and 260 цт x 350 цт. This
glochidium can be distinguished from all oth-
ers by its fan-shaped outline, small dorsal
alae, exterior valve sculpture and its distinct
hook-like folding of the ventral flange. This
glochidium is figured by Lea (1 858: pi. 5, fig. 6,
as Uniosecuris, = E. lineolata. f/de Ortmann &
Walker, 1922), Lefevre & Curtis (1910: fig. H,
1912: fig. 1H, as Plagióla securis), Ortmann
(1 91 1 : pi. 89, fig. 1 7, as P. securis), and Surber
(1 912: pi. 2, fig. 1 4, as P securis).
Remarks
This glochidium will not be confused with
any other. Its size, round subelliptical shape,
small dorsal alae, and lamellate micropoints
distinguish it from all others. Lefevre & Curtis
(1910, 1912), Surber (1912), and Ortmann
(1912) gave the following measurements for
length and height: 75 цт x 85 yivn, 70 |.im x 95
|лт, and 80 цт x 90 |.im. This glochidium is fig-
ured by Lefevre & Curtis (1910: fig. K; 1912:
fig. IK, as Lampsilis gracilis. = L. fragilis. fide
Ortmann & Walker, 1922), Ortmann (1911: pi.
89, fig. 19, as Paraptera gracilis), Coker &
Surber (1911: pi. 1, fig. 2, 2a, as L. gracilis).
and Surber (1912: pi. 2, fig. 28, as L. gracilis).
GLOCHIDIAOF UNIONIDAE
77
FIG. 45. Glochidium of Ellipsaha lineolata, OSUM:1984:14; A. exterior valve, bar length = 45 |am; B. interior
valve, bar length = 45 |am; С lateral view, bar length = 35 jam; d. micropoints, bar length = 5 ¡am; E. hinge,
bar length = 15 jam; F. exterior valve sculpture, bar length = 1 цт.
Leptodea ochracea (Say, 1817)
(Fig.47A-E)
Material Examined
MAH 896 - Great Herring Pond at Carter's
Brook Rd., 4.0 mi. NNE of Buzzards Bay, 3.0
mi. NW of Sagamore, Plymouth Co., Massa-
chusetts, 31 July 1985, G.T. Watters et al.
Description
Glochidium subelliptical to subligulate,
length 241 to 246 |.im (243 ± 1 .95 |im, n = 5),
height 289 to 294 |.im (291 ± 1 .92 |.im, n = 5).
Dorsal margin straight, short, 102 to 110 ¡im
(1 07 ± 3.90 |am, n = 6) in length. Anterior and
posterior margins equal, more or less straight
to slightly incurved dorsally, bending ventrally
78
HOGGARTH
^m^g_¿^j\__^
а
1
^Яе^^^ЯШЁш*
^^^^ь,
1
' ^
^^ЖЯ
FIG. 46. Glochidium of Leptodea fragilis, MAH 626.1 ; A. exterior valve, bar length = 10 цт: B. interior valve,
bar length = 1 цт; С. lateral view, bar length = 1 цт: D. micropoints, bar length = 2 |.im: E. exterior valve
sculpture, bar length = 1 ¡.im.
about 50% from dorsal margin, slightly curved
to their maximum inflation at about 80% from
dorsal to ventral. Ventral margin gently and
evenly curved throughout its length. Exterior
surface malleated and pitted, except along
valve margins where longitudinal ridges occur
(Fig. 47A, C). Small dorsal alae occur at dor-
sal lateral margin of valve. Lateral valve gape
moderate. Central ligament 39 to 47 цт (43 ±
3.65 jim, n = 4) in length, centered about 48%
from posterior to anterior. Anterior ligament 32
to 37 |.im (35 ± 2.22 |.im, n = 4) long; postehor
ligament 28 to 34 }.im (31 ± 2.58 цт, n = 4) in
length. Micropoints narrow lamellate plates
arranged in complete vertical rows on ventral
flange and ventral valve rim, becoming
smaller, more lanceolate distally covering
about four fifths of the surface of flange. Un-
sculptured distal flange edge very narrow.
Remarks
The adult shell of L. ochracea is so similar
to that of Lampsilis cariosa that the adults are
often confused. The glochidia are very dis-
similar, however, and are easily distinguished
by valve outline, development of dorsal alae,
presence of lateral valve gape, and micro-
GLOCHIDIAOF UNIONIDAE
79
.)
V
1
Ы \
с
FIG. 47. Glochidium of Leptodea ochracea, MAH 896; A. exterior valve, bar length = 35 цгл; В. interior valve,
bar length = 35 цт; С. lateral view, bar length = 40 j.im; D. hinge, bar length = 15 цт; E. micropoints, bar
length = 5 |дт.
point Structure (see glochidium of L. cariosa
below). The glochidium of L. ochracea was
figured by Porter & Horn (1981 : fig. 7).
This species is often included in the genus
Lampsilis. primarily because the adult shell re-
sembles that of other members of that genus
and because of the pronounced sexual dimor-
phism of the shell (Simpson, 1900, 1914;
Johnson, 1947, 1970; Burch, 1975; Clarke,
1 981 b). However, Morrison (1 975) was unable
to find mantle flaps in female L. ochracea.
Nonetheless, Fuller & Bereza (1975) sug-
gested that this was not sufficient evidence to
re-classify L. ochracea. and Porter & Horn
(1 981 ) concluded that because the shape and
size of the glochidium of ochracea is similar to
that of cariosa, the species should remain in
Lampsilis. In fact, the glochidium of this
species does not ally ochracea with any
species of Lampsilis examined (see below).
80
HOGGARTH
Because the anatomy of this species is not
that of Lampsilis. and the glochidium, although
larger, is only closely allied to L. fragilis, I agree
with Morrison (1 975) that this species belongs
in Leptodea.
Potamilus ohiensis (Rafinesque, 1820)
(Fig. 48A-E)
Material Examined
OSUM 54520.1 - Big Blue River below
bridge, 3.0 mi. NNW of Barneston, 4.0 mi. SB
of Wymore, T2N, R7E, Gage Co., Nebraska,
7 September 1981, F. E. Hoke.
Description
Glochidium ax-head shaped, dorsally ligu-
late, becoming very broad ventrally (Fig.
47A), length 120 to 126 цт (123 ± 3.05 |im,
n = 3), height 175 to 187 |im (181 ± 6.03 |im,
n = 3). Dorsal margin straight, very short, 42
to 50 |im (45 ± 3.42 |im, n = 5) in length. An-
terior and posterior margins equal, straight
sided dorsally, crescent shaped ventrally, out-
wardly curved to their maximum inflation at
about 90% from dorsal to ventral. Exterior sur-
face smooth, without malleations or pits, and
loose-looped exterior valve sculpture cover-
ing the valve (Fig. 48D). Dorsal alae small, lat-
eral valve gape very large (Fig. 48B). Central
ligament 1 6 to 21 цт (1 9 ± 3.54 ^im, n = 2) in
length, centered about 50% from both lateral
margins. Anterior ligament, 1 3 to 1 5 |лт (1 4 ±
1 .41 |am, n = 2) long; posterior ligament 1 3 to
14 [im (14 ± 0.71 цт, n = 2) long. Lamellate
micropoints present on ventral rim and on nar-
row ventral flange. Lanceolate hooks absent.
Remarks
Lea (1 858, 1 863) described this glochidium
as wedge shaped, with a hook-like process at
each corner of the ventral margin. However,
Coker & Surber (1 91 1 ) reported that the spines
or hooks were absent, and Surber (1912) re-
ported, "Glochidium without spines (?)." This
glochidium can be distinguished from that of
most other members of the genus Potamilus
by the absence of spines. They are further dis-
tinguished by micropoint structure, exterior
valve sculpture and shape, especially the
roundness of the ventral margin. This glochid-
ium shares these characters with Leptodea
fragilis, but can be distinguished from that
species by its ax-head shape. The glochidium
of P. ohiensis \Nas figured by Lea (1858: pi. 5,
fig.24),Coker&Surber(1911:pl. 1,fig. 1, la),
Surber (1 91 2: pi. 1 , fig. 1 0), and Arey (1 921 : pi.
2, figs. 5, 6;1924: pi. 1, fig. 3). Lea incorrectly
figured the glochidium with hooks, whereas
Coker & Surber (1911), Surber (1912), and
Arey (1 921 ) did not. The shape of the glochid-
ium is correctly drawn in each figure.
Potamilus amphichaena (Frierson, 1898)
(Fig. 49A-E)
Material Examined
OSUM 33163.13 - Sabine River at La. Rt.
8 and Tx. Rt. 9 bridge, 1 .3 mi. W of Burr Ferry,
Vernon Parish / Newton Co., LouisianaЯexas,
14 July 1968, D. H. Stansbery et a!.
Description
Glochidium ax-head shaped, length 111 to
113 цт (112 ± 1.41 |im, n = 2), height 170 to
171 цт (1 71 ± 0.71 |am, n = 2). Dorsal margin
short, 40 to 41 цт (41 ± 0.71 цт, n = 2) in
length; anterior and posterior margins lie par-
allel dorsally, becoming strongly curved ven-
trally. Anterior and posterior margins meeting
broadly rounded ventral margin at point of
maximum lateral inflation at about 90% from
dorsal to ventral. Dorsal alae small. Exterior
surface of valve with loose-looped sculpture
(Fig. 49D). The hinge of a single specimen
gave the following measurements: anterior lig-
ament, 11 |.im; central ligament, 20 ¡am; poste-
rior ligament, 10 ¡am. The central ligament is
centered about 49% from posterior to anterior.
Micropoints lamellate, as in L. fragilis and P.
ohiensis. arranged in vertical rows on ventral
valve margin and on short ventral flange.
Lanceolate hooks absent.
Remarks
This glochidium is almost identical with that
of P. ohiensis but can be distinguished by its
slightly more broadly rounded ventral margin.
No previously published figure of this glochid-
ium was found.
Potamilus alatus (Say, 1817)
(Fig. 50A-F)
Material Examined
OSUM: 1983:58 - Muskingum River, R.Mi.
31 .8-33.4, 1 .4 mi. N of Luke Chute Lock and
GLOCHIDIAOF UNIONIDAE
81
FIG. 48. Glochidium of Lastena ohiensis, OSUM 54520.1; A. exterior valve, bar length = 25 |дгл; В. lateral
view, bar length = 30 |am; C. hinge, bar length = 1 цт; D. exterior valve sculpture, bar length = 2 jam; E. mi-
cropoints, bar length = 5 |лт.
Dam, 4.3 mi. W of Beverly, Windsor/Waterford
Twp., Morgan/Washington Co., Ohio, 25 Sep-
tember 1983, W. N. Kasson & K. Perkins.
OSUM 55465 - Clinch River, R.Mi. 213.0 at
Clinchport, just above swinging bridge, Scott
Co., Virginia, 5 October 1985, G. T. Walters.
Description
Glochidium ligulate, ax-head shaped, much
higher than long, length 206 to 227 цт (21 6 ±
9.15 |im, n = 5), height 371 to 386 (.im (378 ±
7.66 цт, П = 5). Dorsal margin straight, al-
though appearing slightly curved in exterior
view due to deep umbo cavity (Fig. 50C), 96 to
109 |im (102 ± 5.93 |im, n = 5) in length, or
less than half total length of valve. Anterior and
posterior margins equal, parallel sided for
about 70% their length, gently outwardly
curved to their point of maximum inflation at
about 95% from dorsal to ventral. Ventral mar-
gin slightly curved. Adductor muscle scar very
rough, with numerous ridges; however, pit
density not appearing to be reduced. Dorsal
alae small (Fig. 50A); lateral valve gape pro-
nounced (Fig. 50C); exterior surface covered
with vermiculate sculpture (Fig. 50E). Central
82
HOGGARTH
FIG. 49. Glochidium of Lastena amphichaena. OSUM 331 63.1 3; A, exterior valve, bar length = 25 цт; В. mi-
cropoints, bar length = 5 j.im; С hinge, bar length = 1 цт; D. exterior valve sculpture, bar length = 2 цт; E.
rлicropoints, bar length = 2 ).im.
ligament 42 [xm in the two specimens mea-
sured, centered about 44% from posterior to
anterior. Anterior ligament 35 to 38 цт (37 ±
2.1 2 |лт, n = 2) in length; posterior ligament 1 9
to 29 |im (24 ± 7.07 ¡am, n = 2) long. Lanceo-
late hooks present at lateral margins of ventral
flange, long, slightly inwardly curved, attenu-
ate. Micropoints arranged in complete vertical
rows on the ventral rim of valve and on rather
wide ventral flange, bluntly pointed, becoming
gradually smaller toward distal edge of flange.
covering most of flange, leaving narrow to ab-
sent unsculptured distal flange edge.
Remarks
This glochidium was figured by Lea (1858:
pi. 5, fig. 25), Lefevre & Curtis (1910: fig. D;
1912: fig. ID), Ortmann (1911 : pi. 89, fig. 18),
Coker & Surber (1911: pi. 1, fig. 3). Surber
(1912: pi. 1, fig. 8). and Utterback (1915-6:
figs. 9a, b). Length and height measurements
GLOCHIDIAOF UNIONIDAE
83
FIG. 50. Glochidium of Potamilus alatus; A. exterior valve, OSUM 55465, bar length = 50 цт; В. interior
valve, OSUM 55465, bar length = 50 цгп; С. lateral view, OSUM 55465, bar length = 50 цгл; D. lanceolate
hook, 0SUM:1 983:58, bar length = 5 |.im; E. exterior valve sculpture, OSUM 55465, bar length = 1 цт; F.
hinge, OSUM 55465, bar length = 15 цт.
84
HOGGARTH
were given by Lefevre & Curtis (1910, 1912),
Surber (1912), and Ortmann (1912), 230 цт
X 410 цт, 220 цт x 380 ¡.im, and 200 }.im x
380 |.im. This glocliidium can be distinguished
from other species by its size, shape, and mi-
cropoint structure.
The glochidium of Potamilus capax is also
similar to that of P. alatus (Cummings et al.,
1 990). The glochidium of P. capax is about the
same size as that of P. ohiensis (105 |лт in
length and 185 цт in height, Coker & Surber,
1911). This glochidium possess lanceolate
hooks, vermiculate exterior valve sculpture,
and a narrowly curved ventral margin. The
glochidium of P. capax is figured by Cum-
mings et al. (1990: fig. 6).
Potamilus purpuratus (Lamarck, 1819)
(Fig. 51A-F)
Material Examined
OSUM 15738.2 - Brazos River at foot of
Whitney Dam, about 20 mi. NW of Waco,
Bosque/Hill Co., Texas, 20 March 1966, С В.
Stein.
Description
Glochidium ligulate, ax-head shaped, much
higher than long, length 1 90 to 200 цт (1 95 ±
5.00 |im, n = 3), height 347 to 356 цт (350 ±
4.93 |im, n = 3). Dorsal margin short, straight,
100 to 109 |im (105 ± 3.87 |.im, n = 4) long.
Anterior and posterior margins parallel to
about 80% from dorsal to ventral, becoming
evenly curved to point of maximum lateral in-
flation at about 95% from the dorsal margin.
Ventral margin only slightly curved. Exterior
surface of valve malleated, pitted dorsally,
ventrally smooth, with sparse pitting in area of
adductor muscle scar compared to remainder
of valve. Dorsal alae small (Fig. 51 A, C), lat-
eral valve gape wide, and exterior surface
covered with vermiculate sculpture (Fig. 51 E).
Central ligament 37 to 41 цт (39 ± 2.83 цт,
n = 2) long, centered about 46% from poste-
rior to anterior. Anterior ligament 37 i^m in
length; posterior ligament 28 to 29 ¡am (29 ±
0.71 |im, n = 2) long. Lanceolate hooks lo-
cated at lateral margins of ventral flange. Mi-
cropoints arranged in complete vertical rows
on ventral rim of valve and on wide ventral
flange. Micropoints bluntly pointed (Fig. 51 D),
becoming gradually smaller toward distal
edge of flange, covering a large portion of
flange, leaving narrow unsculptured distal
edge.
Remarks
This glochidium can be distinguished from
that of P. alatus by its micropoint structure and
its size. This glochidium was figured by Lea
(1874: pi. 21, fig. 13) and Surber (1915: pi. 1,
fig. 5).
Ligumia recta (Lamarck, 1819)
(Fig. 52A-E)
Material Examined
0SUM:1 984:2 - Mississippi River, R.Mi.
635.6, E channel, 1.1 mi. NW of Prairie du
Chien, Crawford Co., Wisconsin, 29 April
1984, D. H. Stansbery et al.
Description
Glochidium subelliptical, length 205 to 219
j^m (211 ± 6.02 |лт, n - 5), height 257 to 265
fim (260 ± 3.08 }im, n = 5). Dorsal margin
slightly curved, 105 to 115 |im (109 ± 3.24
)im, n = 7) in length. Anterior and posterior
margins equally curved, valve outline sym-
metrical. Exterior surface of valve malleated
and pitted, except along valve margin, where
shell is smooth. Dorsal alae long, well devel-
oped, about 52 цт in length. Rough exterior
surface sculpture covering valve (Fig. 52E).
Central ligament 35 to 40 ¡am (38 ± 3.54 цт,
n = 2) long, centered about 45% from poste-
rior to anterior. Anterior ligament 41 to 45 |лт
(43 ± 2.83 |im, n = 2) in length; posterior liga-
ment 30 to 34 lam (32 ± 2.83 |am, n = 2) long.
Micropoints numerous, lanceolate, located on
ventral margin of valve and on rather wide
ventral flange (not figured). Center of ventral
margin almost beak-like, similar to that of A.
pectorosa (Fig. 41 C).
Remarks
This glochidium has the same valve outline
as that of A. pectorosa. However, the glochid-
ium of L. recta can be distinguished from that
glochidium by its larger size, broader ventral
flange, and longer dorsal alae. The glochid-
ium of L. rec/a was figured by Lea (1858: pi. 5,
fig. 11), Lefevre & Curtis (1910: fig. L, 1912:
fig. 1 L), Ortmann (1 91 1 : pi. 89, fig. 21 ), Surber
(1912: pi. 2, fig. 17), Isom & Hudson (1982:
fig. 1), and Isom (1983: figs, la, b).
GLOCHIDIAOF UNIONIDAE
85
FIG. 51 . Glochidium of Potamilus purpuratus, OSUM 1 5738.2; A. exterior valve, bar length = 50 lam; B. inte-
rior valve, bar length = 50 |jm; С lateral view, bar length = 50 lam; D. micropoints, bar length = 5 цт; E. ex-
terior valve sculpture, bar length = 1 цт; F. hinge, bar length = 15 цпп.
86
HOGGARTH
FIG. 52. Glochidium of Ligumia recta. 0SUM:1 984:2; A. exterior valve, bar length = 30 |.im: B. exterior valve,
bar length = 35 jam; С lateral view, bar length = 35 |.im: D. micropoints, bar length = 3 |um: E. micropoints,
bar length = 5 pm.
Venustaconcha ellipsiformis ellipsiformis
(Conrad, 1834)
(Fig. 53A-G)
Material Examined
MAH 947.2 - Kalamazoo River above St.
Rt. 60 bridge, 3.0 mi. WSW of Spring Arbor,
12.0 mi. WSW of Jackson, Jackson Co.,
Michigan, 13 May 1986, M. A. Hoggarth.
Description
Glochidium subeiliptical, symmetrical,
length 223 to 230 |.im (226 ± 2.94 цт, n = 4),
height 280 to 287 цт (285 ± 3.32 pm, n = 4).
GLOCHIDIAOF UNIONIDAE
87
FIG. 53. Glochidium of Venustaconcha e. ellipsiformis. MAH 947.2; A. exterior valve, bar length = 35 |am; B.
exterior valve, bar length = 35 |.im; С micropoints, bar length = 2 |.im; D. micropoints, bar length = 2 |am; E.
exterior valve sculpture, bar length = 1 ¡.im; F. interior valve pitting and adductor muscle scar, bar length =
10 |лт; G. hinge, bar length = 15 \xm.
Dorsal margin straight, 102 to 1 1 цт (1 04 ±
2.48 цт, П = 6) in length. Anterior and poste-
rior margins equally and gently curving to their
maximum inflation at about 70% from dorsal
to ventral. Exterior surface of valve finely mal-
leated near umbo, with fine loose-looped ex-
terior valve sculpture (Fig. 53E). Adductor
muscle scar indicated by numerous small
ridges and reduced pitting (Fig. 53F). Central
ligament 38 to 40 |am (40 ± 3.21 [im, n = 3)
88
HOGGARTH
long, centered 47% from posterior to anterior.
Posterior ligament 28 to 32 цт (30 ± 2.08
|дт, n = 3) long; anterior ligament 34 to 39 |.im
(36 ± 2.65 цт, n = 3) in length. Micropoints
lanceolate, located on ventral rim of valve and
on moderately wide ventral flange, arranged
in broken vertical rows that extend about half
distance of flange, leaving a wide unsculp-
tured distal flange margin.
Remarks
This glochidium is identical to that of L.
recta, except it is larger. It was figured by Lea
(1858: pi. 5, fig. 9, as Unio spatulatus, = V. el-
lipsiformis, fide Simpson, 1900) and van der
Schalle (1963: pi. 1 , center, left).
Villosa trabalis (Conrad, 1834)
(Fig. 54A, B, D, E)
Material Examined
V. trabalis: OSUM 9516.49 - Rockcastle
River, Rt. 80 W of London, Rockcastle/Laurel
Co., Kentucky, 26 October 1 963, С В. Stein &
D. H. Stansbery. V. perpurpurea (Fig. 53C):
OSUM 16262 - Clinch River at St. Rt. 460
bridge at Richlands, Tazewell Co., Virginia, 6
October 1965, D. H. Stansbery & J. J. Jenkin-
son.
Description
Glochidium subelliptical, higher than long,
with a short, straight dorsal margin, gently
curved, almost equal lateral margins and
broadly rounded ventral margin. Glochidium
21 1 to 21 6 цт (214 ± 2.36 |im, n = 4) in length,
277 to 280 цт (278 ± 1 .50 }.im, n = 4) in height.
Dorsal margin 94 to 99 цт (96 ± 2.1 7 ¡.im, n =
5) in length. Exterior valve surface mostly
smooth, with relief occurring at pits, which are
surrounded by smooth, circular discs (Fig.
54E). Dorsal alae moderate in length. Central
ligament 31 to 36 |.im (33 ± 2.89 |.im, n = 3)
long , centered about 45% from posterior to an-
terior. Posterior ligament 25 to 28 ).im (27 ±
1 .53 |.im, n = 3) long; anterior ligament 36 to 40
jim (37 ± 2.31 цт, n = 3) in length. Micropoints
lanceolate, numerous, unorganized, found on
ventral rim of valve and on distal half of wide
ventral flange, leaving wide unsculptured dis-
tal flange margin.
Remarks
Surber (1912) gave 193 цт x 255 |im for
the length and height of this glochidium, and
Ortmann (1912) gave 220 цт x 270 цт. This
glochidium is similar in shape, size and dorsal
alae structure to Obovaria, but it can be dis-
tinguished by its rather wide ventral flange.
The glochidium of Villosa perpurpurea (Lea,
1861 ) is virtually identical to that of V. trabalis,
although smaller: length, 165 |.im; height, 241
|.im; hinge length, 88 |лт. The glochidium of V.
iraba//s was figured by Surber (1912: pi. 3,fig.
40) and Ortmann (1912: pi. 20, fig. 4).
Villosa villosa (Wright, 1898)
(Fig. 55A-G)
Material Examined
OSUM 45940.3, 45940.7 - Santa Fe River
at U.S. Rt. 41/441 bridge, 2.0 mi. NNW of
High Spring, 27.3 mi. NW of Gainsville, Sec.
27/28, Alachua/Columbia Co., Florida, 4 Au-
gust 1975, J. M. Condit & E. P. Keferl.
Description
Glochidium subelliptical (Fig. 54C) to sub-
spatulate (Fig. 55A, B, D), length 240 to 250
|jm (245 ± 3.89 цт, n = 7), height 296 to 308
|im (303 ± 4.26 |.im, n = 7). Dorsal margin
straight, 105 to 116 цт (111 ± 3.65цт,п = 7)
in length. Glochidium usually characterized by
a gently curved ventral margin and lateral
margins that are oblique to the dorsal margin
dorsally and perpendicular ventrally. Dorsal
alae long, extending about half distance of
dorsal oblique section of lateral margins.
Valve finely malleated, densely pitted, except
along margins; exterior sculpturing rough but
not quite beaded, best described as very fine
pustules (Fig. 55G). Central ligament 37 to 42
|.im (40 ± 3.54 цт, n = 2) long, centered about
49% from posterior to anterior. Anterior liga-
ment 35 to 43 цт (38 ± 4.24 (.im, n = 2) long;
posterior ligament about 33 цт long (both
specimens had identical posterior ligament
lengths). Micropoints lanceolate, arranged in
more or less complete vertical rows on ventral
flange and on the rim of ventral margin of
valve. Micropoints becoming smaller from
proximal to distal portion of flange, covering
about 75% of flange surface, leaving narrow
unsculptured distal flange margin.
GLOCHIDIA OF UNIONIDAE
89
FIG. 54. Glochidium of Villosa trabalis (A, B, D, E) OSUM 9516.49, and Villosa perpurpurea (C) OSUM
1 6262; A. exterior valve, bar length = 35 |jin; B. interior valve, bar length = 35 pm; С exterior valve, bar length
= 40 |im; D. micropoints, bar length = 2 |im; E. exterior valve sculpture, bar length = 1 цт.
Remarks
The outline of the valve of this glochidium
distinguish it from the glochidia previously de-
scribed, but not from other members of this
genus (see Villosa vibex and Villosa iris iris
below) and all members of the genus Lamp-
silis. This appears to be the final change in
glochidial shape within this lineage. This
glochidium is figured here for the first time.
Villosa vibex {Conrad, 1834)
(Fig. 56A-F)
Material Examined
OSUM 54631 - Yellow River at U.S. Rt. 84
bridge, 3.0 mi. SE of Sanford, 5.0 mi. W of
Opp, Sec. 33, T4N, R17E, Covington Co., Al-
abama, 2 June 1969, H. Harima & B. Wall.
OSUM 241 24 - Santa Fe River at U.S. Rt. 27
90
HOGGARTH
FIG. 55. Glochidium of Villosa villosa:A. exterior valve, OSUM 45940.7, bar length = 40 |.im; B. interior valve,
OSUM 45940.3, bar length = 40 цт; С. exterior valve, OSUM 45940.7, bar length = 40 ¡.im: D. interior valve.
OSUM 45940.7, bar length = 40 цт: E. micropoints, OSUM 45940.3. bar length = 5 [im: F. micropoints.
OSUM 45940.3, bar length = 2 цт; G. exterior valve sculpture, OSUM 45940.7, bar length = 1 urn.
GLOCHIDIAOF UNIONIDAE
91
FIG. 56. Glochidium of Villosa vibex: A. exterior valve, OSUM 24124, bar length = 40 цт: В. interior valve,
OSUM 24124, bar length = 40 |.im; С adductor гли5с1е insertion, OSUM 54631, bar length = 5 |am; D. mi-
cropoints, OSUM 54631 , bar length = 5 |im; E. exterior valve sculpture, OSUM 54631 , bar length = 2 |лт; F.
exterior valve sculpture, OSUM 24124, bar length = 2 |.im.
bridge, 2.5 mi. NW of High Springs, 7.5 mi. SE
of Fort White, Columbia/Alachua Co., Florida,
24 February 1970, J. J. Jenkinson. OSUM
48628 - "Nichols Cr." Monroe Co., Missis-
sippi, in the Tombigbee River drainage (Mo-
bile River), 18 March 1972, collector un-
known, ex J. D. Williams.
Description
Glochidium subspatulate, length 224 to 239
|im (230 ± 5.42 |im, n - 8), height 295 to 304
цт (300 ± 4.17 |jm, n = 8). Dorsal margin
straight, 1 00 to 1 20 цт (11 ± 7.79 цт, n = 8)
long. Lateral margins straight, divergent dor-
92
HOGGARTH
sally, ventrally parallel; ventral margin gently
curved. Malleations and pits more or less uni-
form over surface of valve; fine sculpture of
exterior surface rough (Fig. 56E, F). Mem-
brane covering exterior surface torn in Figure
56F, intact in Figure 56E. No adductor muscle
scar evident, but just as in Alasmidonta mar-
ginata, adductor muscle inserting on interior
surface of valve, rather than on walls of pits
(Fig. 56C). Long dorsal alae like those in V.
villosa. Central ligament 44 to 45 |.im (45 ±
0.71 цт, n = 2) long, centered about 48%
from posterior to anterior. Anterior ligament 29
to 32 |Lim (31 ± 2.1 2 цт, n = 2) in length; pos-
terior ligament 28 to 29 ¡im (29 ± 0.71 цт, n
= 2) long. Micropoints pyramidal, arranged in
broken vertical rows on wide ventral flange
and on ventral rim of valve, covering proximal
half of ventral flange and leaving wide un-
sculptured distal flange margin.
Remarks
This glochidium was figured by Lea (1858:
pi. 5, fig. 4, as Unio rutilans, = V. vibex, fide
Johnson, 1970; 1874: pi. 21, fig. 7, as Unio
sudus, = V. vibex, fide Johnson, 1970). The
1874 figure is closer to my figure of this
glochidium than the 1858 figure, except that
the ventral half of the figure is much too round.
This glochidium is very close to that of V. vil-
losa, but it can be distinguished by its bluntly
pyramidal micropoints and its wide unsculp-
tured distal flange margin.
Villosa iris iris (Lea, 1 829)
(Fig.57A-F)
qua!, dorsally divergent, ventrally parallel. An-
terior margin may be slightly more rounded
ventrally than the posterior margin; ventral
margin gently curved. Dorsal alae long, but
only about half as long as the dorsal portion of
lateral margins. Valve malleated and pitted;
umbo deeply folded (Fig. 57A, C, D). Concen-
tric ridges extending from umbo ventrally for a
short distance; loose-looped sculpture cover-
ing exterior surface of valve (Fig. 57E). Central
ligament 41 to 42 [xm (42 ± 0.58 цт, n = 3)
long, centered 49% from posterior to anterior.
Anterior ligament 38 to 40 ^im (39 ± 1 .00 |am,
n = 3) long; posterior ligament 33 to 34 цт (33
± 0.58 цт, n = 3) long. Micropoints lanceolate,
arranged in broken rows on ventral rim of valve
and on wide ventral flange, covering proximal
half of flange, leaving wide unsculptured distal
flange margin.
Remarks
This glochidium can be distinguished from
all other subspatulate glochidia examined by
its loose-looped exterior valve sculpture, its
wide unsculptured distal flange border, and
smaller dorsal alae. Surber (1912) gave
length and height measurement of 240 ¡am x
300 цт for this glochidium, and Ortmann
(1912) gave 220 цт x 280 |.im. This glochid-
ium was figured by Lea (1 858: pi. 5, fig. 1 4, as
Unio novi-eboraci, = V. iris, fide Simpson,
1900), Ortmann (1911: pi. 89, fig. 20), and
Surber(1912:pl. 3, fig.46).
Lampsilis teres teres (Rafinesque, 1820)
(Fig. 58A-E)
Material Examined
Material Examined
MAH 641.1 - Little Darby Creek at Co. Rt.
131 bridge (Grewell Rd.), 1.8 mi. E of Plum-
wood, 7.7 mi. NW of West Jefferson, Monroe
Twp., Madison Co., Ohio, 11 May 1984, M. A.
Hoggarth & G. T Watters. OSUM 55828.3
55828.5 - South Fork Clinch River at Four-
way, at E edge of Tazewell, at U.S. Rt. 19
bridge, Tazewell Co., Virginia, 13 October
1985, D. H. Stansbery.
Description
Glochidium subspatulate, length 21 7 to 232
}im (225 ± 5.84 [im, n = 8), height 289 to 305
|дт (296 ± 6.64 fim, n = 8). Dorsal margin
straight, 107to 115|.im (113 ± 2.51 цт, n = 8)
in length. Anterior and posterior margins sube-
OSUM 36409 - Tombigbee River 0.5 mi.
below mouth of Luxaplilla Creek, 3.2 mi. S of
Columbus, Sec. 32/5, T18S, R18W, Lowndes
Co., Mississippi, 5 October 1974, R. Grace &
T Whitfields. OSUM 36531.1 - Whitewater
River at Bollinger Mill Dam, at Burfordville,
T31N, R11E, Cape Girardeau Co., Missouri,
17 October 1974, F. Schilling & H. Kemper.
OSUM 51669.2 - Mississippi River, R.Mi.
632.1, E. Shore main channel, 2.9 mi. SW of
Prairie du Chien, Craford Co., Wisconsin, 10
June 1 981 , D. J. Heath & M. С Weisensei.
Description
Glochidium subspatulate, length 1 89 to 1 94
|im (191 ± 1.87 цт, n = 9), height 255 to 265
GLOCHIDIAOF UNIONIDAE
93
FIG. 57. Glochidium of Villosa i. iris: A. exterior valve, OSUM 55828.4, bar length = 50 цт; В. interior valve,
MAH 641 .1 , bar length = 40 цт; С. exterior valve, MAH 641 .1 , bar length = 40 iim; D. umbo, MAH 641 .1 , bar
length = 15 |.im; E. exterior valve sculpture, MAH 641.1, bar length = 1 ¡.im; F. micropoints, MAH 641.1, bar
length = 5 цт.
94
HOGGARTH
FIG. 58. Glochidium of Lampsilis t. teres; A. exterior valve, OSUM 51669.2, bar length = 35 |im; B. interior
valve, OSUM 51669.2, bar length = 40 |am; С hinge, OSUM 36409, bar length = 15 |.im; D. exterior surface
sculpture, OSUM 51669.2, bar length = 1 цт; E. micropoints, OSUM 51669.2, bar length = 5 |im.
¡am (258 ± 3.71 |im, n = 9). Dorsal margin
straight, 100 to 116 |jm (110 ± 3.64 цт, n =
14) long. Anterior and posterior margins
equal, dorsally divergent, parallel to slightly
convergent ventrally. Ventral margin gently
curved. Exterior valve surface malleated and
pitted; exterior surface rough (Fig. 58D). Dor-
sal alae about half as long as divergent por-
tion of lateral margins. Central ligament 27 to
39 |дт (34 ± 4.09 цт, n = 6) long, centered
about 45% from posterior to anterior. Anterior
ligament 41 to 54 цт (46 ± 5.08 цт, n = 6)
long; posterior ligament 30 to 32 цт (31 ±
0.82 )im, n = 6) in length. Micropoints lanceo-
late (deformed in Fig. 58E), sharply pointed,
restricted to proximal half of ventral flange
and ventral rim of valve. This leaves a wide
unsculptured distal flange margin.
Remarks
Surber figured this glochidium (1912: pi. 2,
fig. 22, as L. fallaciosa. = L. teres, fide John-
son, 1972) and gave length and height mea-
surements of 200 цт X 240 ).im. This glochid-
ium is much smaller than any other recorded
for Lampsilis. except L. t. anodontoides (see
below).
GLOCHIDIAOF UNIONIDAE
95
FIG. 59. Glochidium of Lampsilis t. anodontoides: A. exterior valve, OSUM 35612, bar length = 40 ¡.im; B. in-
terior valve, OSUM 35612, bar length = 35 ¡.im; С hinge, OSUM 41762.2, bar length = 15 ¡am; D. exterior
valve sculpture, OSUM 41762.2, bar length = 1 |дт; E. micropoints, OSUM 41762.2, bar length = 5 |im.
Lampsilis teres anodontoides (Lea, 1 831 )
(Fig. 59A-E)
Material Examined
OSUIVI 35612 - Uphapee Creek at 1-85
bridge, 3.4 mi. N of Tuskegee, 6.3 mi. S of No-
tasulga. Sec. 7, T17N, R24E, Macon Co., Al-
abama, 4 April 1968, J. S. Ramsey et al.
OSUM 41762.2 - St. Francis River 2.2 mi. N
of Parkin, 33.0 mi. WNW of Memphis (TN),
Sec. 21, T8N, R5E, Cross Co., Arkansas, 14
March 1978, D. H. Stansbery et al.
Description
Glochidium subspatulate, length 1 87 to 207
цт (199 ± 7.19 (im, n = 7), height 249 to 255
\.im (251 ± 2.87 цт, n = 7). Dorsal margin
straight, 1 05 to 1 1 8 цт (1 1 1 ± 4.42 цт, n = 9)
long. Valve outline like that of L. t. teres; lat-
eral margins dorsally divergent, becoming
96
HOGGARTH
parallel, convergent ventrally; ventral margin
gently curved. Exterior surface of valve mal-
leated, especially near umbo; pits numerous.
As in L. t. teres, dorsal alae about half as long
as dorsal portion of lateral margins. Exterior
surface rough under high magnification (Fig.
59D). Central ligament 36 to 42 цт (38 ±
2.45 цт, n = 5) long, centered about 45%
from posterior margin. Anterior ligament 44 to
50 цт (46 ± 2.88 |.im, n = 5) long; posterioR
ligament 30 to 33 ).im (31 ± 1.14 цт, n = 5)
long. Micropoints lanceolate, located on ven-
tral rim of valve and on wide ventral flange,
arranged in incomplete vertical rows; un-
sculptured distal flange margin moderate in
length.
Remarks
This glochidium is identical to that of L. t.
teres. Surber (1912) gave length and height
measurements of 185 |.im x 210 цт for this
glochidium, and Ortmann (1912) gave 200 |.im
X 260 i-im. Surber's measurements seem too
small, even for this glochidium, and his figure
(1912: pi. 2, fig. 21) shows the ventral margin
as semicircular rather than gently curved.
These discrepancies suggest that his material
was not mature. This glochidium was also fig-
ured by Lea (1858: pi. 5, fig. 2) and Ortmann
(1912: pi. 20, fig. 9). Lea's figure is much
closer to Ortmann's and mine than to Surber's
figure.
Lampsilis radiata radiata (Gmelin, 1791)
(Fig. 60A-E)
Material Examined
L. r. radiata: MAH 897.1 - Great Herring
Pond at Carter's Brook Rd. 4.0 mi. NNE of
Buzzards Bay, 3.0 mi. NW of Sagamore, Ply-
mouth Co., Massachusetts, 31 July 1985, G.
T Watters et al. L. г. tuteóla: MAH 321 .6 - Big
Darby Creek at Scioto-Darby (Mt. Sterling-
Commercial Pt.) Rd. bridge, 3.4 mi. S of Ori-
ent, 15.3 mi. SW of Columbus, Scioto/Darby
Twp., Pickaway Co., Ohio, 7 January 1983,
M. A. Hoggarth & G. T Watters. MAH 587.
1 - Big Darby Creek at access 0.9 mi. N of
Harrisburg, 1.7 mi. NW of Orient, Pleasant
Twp., Franklin Co., Ohio, 27 September 1 983,
M. A. Hoggarth. MAH 628.1 - Big Darby
Creek at Scioto-Darby (Mt. Sterling-Commer-
cial Pt.) Rd. bridge, 3.4 mi. S of Orient, 15.3
mi. SW of Columbus, Scioto/Darby Twp.,
Pickaway Co., Ohio, 13 April 1984, M.A. Hog-
garth & G. T Watters. MAH 652 - Little Darby
Creek at Little Darby Rd access, 2.7 mi. SE of
Plumwood, 6.4 mi. NW of West Jefferson,
Monroe Twp., Madison Co., Ohio, 17 May
1984, M. A. Hoggarth & G. T Watters. MAH
721 .2 - Fish Creek at St. Rt. 49 bridge, 0.4 mi.
above its mouth, 1.1 mi. N of Edgerton, 10.4
mi. W of Bryan, St. Joseph Twp., Williams
Co., Ohio, 1 October 1985, M. A. Hoggarth &
D. Rice.
Description
Glochidium subspatulate, length 250 to 260
|im (255 ± 4.43 цт, n = 4), height 295 to 311
}.im (303 ± 6.58 |im, n = 4). Dorsal margin
straight, 121 to 129 ¡.im (125 ± 3.78 ¡jm, n =
4) in length. Lateral margins dorsally diver-
gent, ventrally parallel, ventral margin gently
curved. Dorsal alae long; fine structure of ex-
terior surface of valve rough (Fig. 60D). Exte-
rior surface finely malleated. Valve at umbo
deeply folded. Central ligament 45 to 47 цт
(46 ± 1.00 |Lim, n = 3) long, centered about
42% from posterior to anterior. Anterior liga-
ment 46 to 50 цт (48 ± 2.08 цт, n = 3) long;
posterior ligament 27 to 32 цт (29 ± 2.52 цт,
n = 3) long. Micropoints lanceolate, located on
rim of ventral margin and on wide ventral
flange, arranged in broken vertical rows and
covering proximal half of ventral flange. Un-
sculptured distal flange margin wide.
Morphometries from the subspecies L. r. lu-
teola are: length, 227 to 235 цт (231 ± 2.39
цт, n = 8); height, 280 to 295 цт (285 ± 6.63
|.im,n = 8); hinge length 1 07 to 123 (.im (115 ±
5.23 |im, n = 12); anterior ligament length 37
to 42 |дт (39 ± 1.94 цт, n = 6); central liga-
ment length 40 to 46 [im (42 ± 2.58 цт, n =
6); posterior ligament length 32 to 40 цт (36
± 3.39 |.im, n = 6).
Remarks
Surber (1912) and Ortmann (1912) gave
length and height measurements for the
glochidium of L. r. luteola: Ortmann's figures
are 230 цт x 280 ,um (almost identical to
mine), and Surber gave 250 ¡.im x 290 цт (al-
most the same as my measurements for L. r.
radiata). This species is as far ranging as
Pyganodon grandis and appears to have just
as wide a range in glochidial size. The
glochidium of L. r. radiata was figured by Lea
(1858: pi. 5, fig. 20), Wiles (1975: fig. 5), and
GLOCHIDIAOF UNIONIDAE
97
FIG. 60. Glochidium of Lampsilis r. radiata, MAH 897.1 ; A. exterior valve, bar length = 45 ¡am; B. interior valve,
bar length = 45 |am; С hinge, bar length = 20 цт; D. exterior valve sculpture, bar length = 1 цт; E. micro-
points, bar length = 10 цт.
Calloway & Turner (1 979: pi. 3, figs. 2, 4, 6, 9).
The glochidium of L. r. luteola was figured by
Lea (1958; pi. 5, fig. 10), Surber (1912: pi. 2,
fig. 1 5), and Arey (1 921 : pi. 2, fig. 3, 4; 1 924:
pl. 1,fig. 1).
Lampsilis abrupta (Say, 1 831 )
(Fig. 61A-F)
Material Examined
OSUM 1 3303 - Gasconade River 3 mi. N of
Mount Sterling, Gasconade Co., Missouri, 17
September 1964, D. H. Stansbery & J. J.
Jenkinson. OSUM 38841 - Kanawha River
immediately below Kanawha Falls, 0.3 mi.
SSE of Glen Ferris, 1.4 mi. SW of Gauley
Bridge, Falls Twp., Fayette Co., West Virginia,
27 November 1976, K. G. Borror et al.
Description
Glochidium subspatulate, length 207 to 214
jim (210 ± 2.66 |im, n = 6), height 251 to 259
¡am (254 ± 3.13 |.im, n = 6). Dorsal margin
straight, 96 to 112 цт (102 ± 5.62 цт, n - 7)
98
HOGGARTH
FIG. 61. Glochidium of Lampsilis abrupta: A. exterior valve, OSUM 13303. bar length = 40 цт: В. interior
valve, OSUM 38841, bar length = 40 цт: С. micropoints, OSUM 13303. bar length = 2 цт: D. micropoints,
озим 38841, bar length = 2 |.im; E. exterior valve sculpture. OSUM 13303. bar length = 1 цт: F. hinge,
OSUM 13303, bar length = 15 |.im.
in length. Exterior surface finely malleated,
wrinkled near umbo. Fine structure of exterior
surface of valve rough (Fig. 61 E). Pits few in
number, more or less evenly distributed
throughout valve. Dorsal alae long, extending
about one fifth height of valve. Hinge of a sin-
gle specimen with the following ligament
lengths: anterior ligament, 46 ¡.im: central lig-
ament, 38 j.im; posterior ligament, 27 ¡.im.
Central ligament centered about 44% from
posterior to anterior. Micropoints lanceolate,
arranged in broken vertical rows on wide ven-
GLOCHIDIAOF UNIONIDAE
99
tral flange and on rim of ventral margin, cov-
ering proximal half of ventral flange, leaving
wide unsculptured distal flange margin.
Remarks
Lea (1863) stated that this glochidium is,
"almost exactly the same with multiradlatus"
[= Lampsilis fasciola. fide Ortmann & Walker,
1922]; however, the glochidium of L. abrupta
is much smaller than that of L. fasciola. Ort-
mann (1912) gave length and height mea-
surements of 1 90 (.im X 21 i-im and 200 цт x
250 i-im for the glochidium of L. orbiculata {=
L. abrupta, fide Stansbery et al., 1985). My
figures are nearer his second set of measure-
ments, and the glochidia were all about the
same size. There is often differential matura-
tion of glochidia and Ortmann's smaller indi-
viduals may have been immature. This glo-
chidium was figured by Ortmann (1 91 1 : pi. 89,
fig. 22, as L. orbiculata).
Lampsilis higginsi {Lea, 1857)
(Fig. 62A-F)
Material Examined
OSUM 49024.1 Mississippi River, R.Mi.
633.3, E channel, N end of Indian Isle, 2.1 mi.
SW of Prairie du Chien, Sec. 1/2, T6N, R7W,
Craford Co., Wisconsin, 11 November 1980,
A. Reed, donor: M. E. Havlik.
Description
Glochidium subspatulate, length 21 4 to 21 7
lam (216 ± 2.12 |im, n = 2), height 254 to 257
|дт (256 ± 2.12 цт, n = 2). Dorsal margin
straight, 1 08 to 1 1 8 ¡jm (1 1 1 ± 4.72 цт, n = 4)
in length. Anterior and posterior margins dor-
sally divergent, subparallel ventrally. Ventral
margin gently curved. Surface of valve finely
malleated, with concentric ridges near umbo.
Dorsal alae strongly curved, extending about
half distance of dorsally divergent portions of
lateral margins. Fine structure of exterior sur-
face rough (Fig. 62E). Central ligament 39 to
47 |am (43 ± 5.66 |im, n = 2) long, centered
about 47% from posterior to anterior. Anterior
ligament 38 to 40 цт (39 ± 1 .41 цт, n = 2) in
length; posterior ligament about 31 \.im long.
Micropoints lanceolate, arranged in incom-
plete rows on ventral flange and on rim of ven-
tral margin. Micropoints decreasing in size dis-
tally, covering about three quarters of flange
surface. Unsculptured distal flange margin
moderate in width.
Remarks
Waller et al. (1988) gave almost identical
morphometric data for this glochidium. They
also found that the glochidium of L. recta had
similar morphometries. They suggest, how-
ever, that the glochidia are also the same
shape and can be distinguished only by the ex-
tent of the development of dorsal alae and
placement of the central hinge ligament. Their
conclusion regarding shape was based on
their inability to distinguish any differences
in outline upon overlaying light microscopy
transparencies. I found that the glochidium of
L. recta can best be described as subelliptical,
with a rounded ventral border and equally
curved lateral margins (similar to that of A.
pectorosa), whereas the glochidium of L. hig-
ginsi is subspatulate, with a gently curved ven-
tral margin and lateral margins that are dor-
sally divergent and ventrally parallel. The
anterior margin of L. higginsi is slightly more
rounded than the posterior margin, and the
margins are clearly not equal. These addi-
tional differences help distinguish these glo-
chidia. This glochidium was figured by Waller
et al. (1 988: figs. 2, 4), who gave length, height
and hinge length measurements of 215 jam x
259|.imx 110 (.im.
Lampsilis ovata (Say, 1817)
(Fig. 63A-F)
Material Examined
L. ovata: OSUM 43164.1 - Clinch River
above mouth of Copper Creek, 1.3 mi. S of
Clinchport, 9.3 mi. W of Gate City, Scoot Co.,
Virginia, 21 October 1978, D. H. Stansbery et
al. L. ornata: OSUM 54661 -Amite River 1 mi.
NNW of Denham Springs, about 13 mi. WNW
of Livingston, T6S, R3E, Ward 2, Livingston
Parish, Louisiana, 1 October 1967, E. N. Lam-
bremont et al. L. satura: OSUM 38973.51,
38973.52 - West Fork San Jacinto River
below 1-45 bridge, 4.5 mi. S of Conroe, 36 mi.
N of Houston, Montgomery Co., Texas, 20
July 1968, D. H. Stansbery et al.
Description
Glochidium broadly subspatulate, length
about 232 (.im, height 271 to 276 цт (274 ±
100
HOGGARTH
FIG. 62. Glochidium of Lampsilis higginsi, OSUM 49024.1 ; A. exterior valve, bar length = 35 ит; B^interior
valve, bar length = 35 цт; С. lateral valve view, bar length = 35 um; D. micropoints, bar length = 2 цт, t. ex-
terior valve sculpture, bar length = 1 цт; F. hinge, bar length = 1 5 (.im.
GLOCHIDIAOF UNIONIDAE
101
FIG. 63. Glochidium of Lampsilis ovata. OSUM 43164,1; A. exterior valve, bar length = 40 цт; В. interior
valve, bar length = 40 цт; С. micropoints, bar length = 5 |im; D. exterior valve sculpture, bar length = 1 ¡ivn;
E. hinge, bar length = 20 |.im; F. hinge, bar length = 20 (.im.
3.54 |.im, n = 2). Dorsal margin straight, 11 3 to
119 lam (116 ± 2.58 )im, n = 4) in length. Dor-
sal portions of lateral margins strongly diver-
gent; ventral portions of these margins sub-
parallel; anterior margin slightly curved.
Dorsal alae extending about three quarters
length of dorsal divergent portion of lateral
margins, ventral margin gently curved. Fine
structure of exterior surface of valve rough
(Fig. 63E). Central ligament 43 to 44 цт (44
± 0.71 |.im, n = 2) long, centered about 49%
from posterior to anterior. Anterior ligament 39
to 40 |дт (40 ± 0.71 цт, n = 2) long; posterior
ligament 34 to 35 |im (35 ± 0.71 |im, n = 2) in
102
HOGGARTH
length. Micropoints bluntly pyramidal to lance-
olate, arranged in broken vertical rows on
ventral flange and on rim of ventral margin,
covering tfiree quarters of ventral flange, leav-
ing moderately wide unsculptured distal
flange margin.
Remarks
The glochidium of L. ovata is distinguished
from those previously described by its broadly
subspatulate shape. However, this shape is
also found in the glochidia of the other mem-
bers of the L. ovata complex (i.e., L. ventri-
cosa, L. satura and L. ornata). Relative size
may be the only way to distinguish these
species. Morphometric data for Lampsilis
satura is: length, 220 to 223 (.im (222 ±2.12
|im, n = 2); height, 268 to 269 цт (269 ± 0.71
|im, n = 2); hinge length, 113 to 117 цт (115
± 1 .82 |im, n = 4); anterior ligament length, 42
to 43 |im (43 ± 0.71 цт, n = 2); central liga-
ment length, 40 ¡am; and posterior ligament
length, 33 to 35 цт (34 ± 1 .41 |im, n = 2). The
position of the central ligament is about 48%
from posterior to anterior. Morphometric data
for Lampsilis ornata: length, 198 to 204 |.im
(202 ± 3.21 lam, n - 3); height, 257 to 260 |.im
(258 ± 1.53 )im, n = 3); hinge length, 96 to
100 |im (98 ± 2.08 jam, n = 3). A single spec-
imen gave the following ligament lengths: an-
terior ligament, 38 цт; central ligament, 33
|im; posterior ligament, 29 ¡am. The midpoint
of the central ligament is about 46% from pos-
terior to anterior. The glochidium of L. ovata
was figured by Lea (1858: p1.5, fig. 15) and
Isom (1983: fig. 1c).
Lampsilis ventricosa (Barnes, 1823)
(Fig. 64A-G)
Material Examined
OSUM 44619 - Green River below Lock 5
dam at Glenmore, 1 2 mi. N of Bowling Green,
Warren Co., Kentucky, 21 October 1979, D.
H. Stansbery et al. МАИ 846.1 - Fish Creek at
bridge 2.0 mi. NW of Edgerton, 1 1 .9 mi. W of
Bryan, Sec. 20, T6N, RÍE, St. Joseph Twp.,
Williams Co., Ohio, 29 October 1985, D. H.
Stansbery et. al. MAH 954.1, 954.2, 954.3,
954.4, 954.5 & 954.6-Sugar River 300 m
downstream of Lake Belle View Dam, 150 m
upstream of St. Rt. 69 bridge, at Belleville,
Sec. 34, T5N, R8E, Dane Co., Wisconsin, 15
May 1986, M. A. Hoggarth & D. J. Heath.
Description
Glochidium broadly subspatulate, length of
240 to 255 цт (249 ± 4.46 цт, n = 1 8), height
274 to 293 [xm (283 ± 6.30 |.im, n - 18). Dor-
sal margin straight, 1 04 to 1 1 8 ¡am (1 1 1 ± 4.06
).im, n = 21) in length. Shape very much like
that of L. ovata. Straight dorsal portion of lat-
eral margin strongly divergent; ventral portion
of lateral margin almost parallel; ventral mar-
gin gently curved. Dorsal alae long, strongly
arched (Figs. 64A, C). Fine structure of exte-
rior valve sculpture rough (Fig. 64F). Central
ligament 37 to 42 цт (39 ± 1.98 цт, n = 8)
in length, centered about 47% from posterior
to anterior. One female produced glochidia
with a more posterior central ligament (43%)
whereas another produced glochidia with a
more anterior central ligament (50%). Anterior
ligament 38 to 50 цт (44 ± 4.82 цт, n = 8)
long; posterior ligament 29 to 33 \лт (32 ±
1.41 \im, n = 8) in length. Micropoints lanceo-
late to bluntly pyramidal, arranged in broken
vertical rows on wide ventral flange and on
rim of ventral margin, covering about 75% of
flange and leaving moderately wide unsculp-
tured distal flange margin.
Remarks
The glochidium of L. ventricosa is longer
and higher than that of L. ovata, but both
species have about the same hinge length.
This glochidium was figured by Lea (1858:
pi .5, fig. 1 3, as Unio occidens, = L. ventricosa,
f/de Johnson, 1970), Ortmann (1911: p1.89,
fig. 23), Surber (1912: p1. 2, fig. 24), and
Waller et al. (1988: figs. 5, 6). Surber gave
length and height measurements of 200 ¡am x
250 цт for this glochidium; Ortmann (1912)
gave 250 |лт x 290 цт; and Waller et al. gave
21 6 í-im X 257 [im. My measurements are 249
|jm X 283 цт. Waller et al. and Surber col-
lected their material from the Mississippi River,
whereas Ortmann's and my material came
from smaller streams.
Lampsilis reeviana brevicula (Call, 1 887)
(Fig. 65A-E)
Material Examined
OSUM 45363.20, 45363.22, 45363.35 &
45363.50 - Current River "between Mo. Rt.
106 and Van Buren," about 9 mi. of stream.
Shannon/Carter Co., Missouri, 24 October
1981, R. D. Oesch.
GLOCHIDIAOF UNIONIDAE
103
FIG. 64. Glochidium of Lampsilis ventricosa; A. exterior valve, MAH 954.6, bar length = 40 цт: В. interior
valve, MAH 954.3, bar length = 40 цт; С. lateral view, OSUM 44619. bar length = 40 цт; D. rnicropoints,
MAH 954.3, bar length = 2 ¡.im; E. micropoints, MAH 954.4, bar length = 2 |im; F. exterior valve sculpture,
MAH 954.1, bar length = 1 цт; G. hinge, MAH 954.6, bar length = 15 цт.
104
HOGGARTH
FIG. 65. Glochidium of Lampsilis breviculata: i\. exterior valve, OSUM 45363.35, bar length = 40 цт; В. inte-
rior valve, OSUM 45363.20, bar length = 40 |jm; С hinge, OSUM 45363.50, bar length = 15 цт; D. exterior
valve sculpture. OSUM 45363.50, bar length = 1 цш; E. micropoints, OSUM 45363.35, bar length = 5 цт.
Description
Glochidium broadly subspatulate length
230 to 245 цт (235 ± 6.05 )im, n = 8), height
286 to 297 |jm (290 ± 4.37 |im, n - 8). Dorsal
margin straight, 113 to 127 цт (119 ± 4.50
|am, n = 9) long. Lateral margins straight,
strongly divergent dorsally, subparallel ven-
trally. Anterior margin slightly more curved
ventrally than the posterior margin; ventral
margin gently curved throughout its length.
Dorsal alae about as long as the divergent
dorsal portion of lateral margins. Fine struc-
ture of the exterior surface rough (Fig. 65D).
Central ligament 37 to 48 |im (43 ± 4.39 |im,
n = 5) in length, centered about 47% from
posterior to anterior. Anterior ligament 38 to
50 цт (43 ± 5.45 |.im, n = 5) long; posterior
ligament 32 to 36 цт (34 ± 2.1 9 цт, n = 5) in
length. Micropoints lanceolate, arranged in
broken vertical rows on ventral rim of valve
and on a wide ventral flange, covering proxi-
mal half of ventral flange, leaving a wide un-
sculptured distal flange margin.
Remarks
This glochidium was figured by Surber
(1915: pi. 1, fig. 14), who gave length and
height measurements of 230 |im x 290 цт.
GLOCHIDIAOF UNIONIDAE
105
Lampsilis crocata (Lea, 1841 )
(Fig. 66A-F)
Material Examined
OSUM 54485.1, 54485.2 - Tar River at
U.S. Rt. 64 bridge, 2.4 mi. SW of Spring Hope,
10.8 mi. mi. WSW of Nashville, Nash Co.,
Tennessee, 1973, M. Imlay. OSUM 42060.1 -
Waccamaw River just below Lake Waccamaw
Dam, 6.2 mi. SE of Hallsboro, 11 .7 mi. ESE of
Whiteville, Columbus Co., North Carolina, 21
July 1978, D. H. Stansbery et al.
Description
Glochidium narrowly subspatulate, length
235 to 249 |.im (242 ± 6.23 (im, n - 8), height
287 to 303 [im (293 ± 5.03 цт, n = 8). Dorsal
margin straight, 110 to 125 цт (118 ± 4.81
|im, n = 8) in length. Lateral margins weakly di-
vergent dorsally, ventrally subparallel. Ventral
margin evenly and broadly curved; dorsal alae
long, strongly arched (Fig. 66A, C) Fine struc-
ture of exterior surface of valve rough (Fig.
66E). Central ligament 52 to 56 |jm (54 ± 2.83
)im, n = 2) long, centered about 46% from pos-
terior to anterior. Anterior ligament 36 to 39 цт
(38 ± 2.12 |im, n = 2) long; posterior ligament
about 32 |im long. Micropoints bluntly lanceo-
late, covering about 75% of ventral flange,
arranged in broken vertical rows on ventral
flange and on rim of ventral margin, remaining
about equal in length from proximal to distal.
Unsculptured distal flange margin narrow.
Remarks
The shape of this glochidium is like that of
L. teres, but it is much larger. No published
figure of this glochidium was found.
Lampsilis cariosa (Say, 1817)
(Fig. 67A-D)
Material Examined
OSUM 54500 - Tar River at U.S. Rt. 15
bridge, 2.7 mi. SW of Clay, 6.7 mi. SSW of Ox-
ford, Granville Co., North Carolina, 1973, M.
Imlay.
Description
Glochidium narrowly subspatulate, length
about 241 |nm, height about 314 ,um, hinge
length about 109 |am. Dorsal margin straight;
ventral margin evenly and broadly curved; lat-
eral margins about equally divergent dorsally
and ventrally parallel. Dorsal alae long. Fine
structure of exterior surface of valve rough
(Fig. 67C). Valve pitted, except along mar-
gins; pitting reduced in adductor muscle scar
(Fig. 67B). Umbo deeply folded and the cen-
ter of numerous concentric ridges. Hinge liga-
ments not seen. Micropoints lanceolate, re-
stricted to proximal half of ventral flange and
on rim of ventral valve margin, arranged in
broken vertical rows. Unsculptured distal
edge of ventral flange wide.
Remarks
It is clear that this glochidium is very differ-
ent from that of Leptodea ochracea. The long
dorsal alae, lanceolate micropoints, and rough
exterior valve sculpture place the glochidium
of L. cariosa firmly in Lampsilis, whereas the
small dorsal alae, lamellate micropoints,
loose-looped exterior sculpture, and lateral
valve gape separate the glochidium of Lep-
todea ochracea from Lampsilis. No published
figure of this glochidium was found.
Lampsilis fasciola Rafinesque, 1820
(Fig. 68A-F)
Material Examined
OSUM 25467 - Kanawha River immedi-
ately below Kanawha Falls, 1.2 mi. below
Gauley Bridge, Falls Twp., Fayette Co., West
Virginia, 1 October 1970, D. H. Stansbery &
W. J. Clench. MAH 562.3 - Big Darby Creek
at and above Scioto-Darby (Mt. Sterling-Com-
mercial Pt.) Rd. bridge, 3.4 mi. S of Orient,
1 5.3 mi. SW of Columbus, Scioto/Darby Twp.,
Pickaway Co., Ohio, 19 August 1983, M. A.
Hoggarth. OSUM 55977 - Olentangy River
just above Powell Rd. bridge, 1.6 mi. E of
Powell, 9.8 mi. S of Delaware, T3N, R19W,
Delaware Co., Ohio, 29 September 1983, С
В. Stein et al. OSUM 55033.2 - Clinch River,
R.Mi. 269.7, 1 .0 mi. above Carbo, 1 .8 mi. W of
Cleveland, 6.2 mi. NW of Lebanon, Castle-
wood Twp., Russell Co., Virginia, 26 July
1985, D. H. Stansbery & С В. Stein.
Description
Glochidium narrowly subspatulate, length
240 to 250 цт (247 ± 4.47 ¡am, n = 5) height
287 to 295 \хг(\ (290 ± 3.08 цт, n = 5). Dorsal
106
HOGGARTH
FIG. 66. Glochidium of Lampsilis crocata;fK. exterior valve. OSUM 42060.1 , bar length = 40 цт; В. interior
valve, OSUM 42060.1, bar length = 50 цт; С. lateral view. OSUM 54485.1, bar length = 50 |.im; D. micro-
points, OSUM 54485.1 , bar length = 2 pm; E. exterior valve sculpture, OSUM 54485.1 , bar length = 1 цгл; F.
micropoints, OSUM 42060.1, bar length = 5 pm.
\
GLOCHIDIA OF UNIONIDAE
107
FIG. 67. Glochidium of Lampsilis cariosa, OSUM 54500; A. exterior valve, bar length = 45 цт; В. interior
valve pitting and adductor muscle scar, bar length = 15 |дт; С. exterior valve sculpture, bar length = 1 |im;
D. micropoints, bar length = 5 |.im.
margin straight, 107 to 115 цт (110 ± 2.95
цт, П = 5) in length. Posterior margin strongly
divergent from dorsal margin for about half its
length. Near its midpoint, posterior margin
bending ventrally, straightening out to run
more or less perpendicular to hinge. Anterior
margin diverging from dorsal margin at a
lesser angle, straight for about three quarters
of its length, curving continuously to ventral
margin. Ventral margin gently and evenly
curved. Dorsal alae long, with a strong arch
(Fig. 68A, C). Fine structure of exterior sur-
face of the valve rough (Fig. 68E). Central
ligament about 44 цт long, centered about
48% from posterior margin. Anterior ligament
about 38 цт long; posterior ligament about 33
цт long. Micropoints lanceolate, arranged in
broken vertical rows on ventral rim of valve
and on a wide ventral flange. Micropoints de-
creasing in size distally on flange covering
about three quarters of its proximal surface,
leaving narrow unsculptured distal flange
margin.
Remarks
The glochidium of L. fasciola was figured by
Lea (1858; pi. 5, fig. 17, as Unio multiradia-
108
HOGGARTH
^_ ' ^^.^1
/
\
'
% ^
A
FIG. 68. Glochidium of Lampsilis fasciola; A. exterior valve, OSUM 55033.2, bar length = 45 цт: В. interior
valve, OSUM 55033.2, bar length = 45 цт; С. lateral view, OSUM 25467, bar length = 40 цт; D. micropoints,
OSUM 55033.2, bar length = 5 |.im; E. extehor valve sculpture, OSUM 55033.2, bar length = 1 цт; F. hinge,
OSUM 55033.2, bar length = 15 цт.
tus, = L. fasciola, fide Ortmann & Walker,
1922) and Surber (1915: p1. 1, fig. 2, as L.
multiradiatus). This glochidium is like other
narrowly subspatulate glochidia, except that
the anterior margin is slightly more rounded.
In this regard, this glochidium resembles that
of Obovaria, but it will not be confused with
Obovaria because of it posterior margin, wide
ventral flange, and rough exterior valve sculp-
ture. The glochidium of Obovaria has a curved
posterior margin, narrow ventral flange, and
loose-looped exterior valve sculpture.
GLOCHIDIAOF UNIONIDAE
109
FIG. 69. Glochidium of Epioblasma triquetra, MAH 588.1; A. exterior valve, bar length = 35 |jm; B. interior
valve, bar length = 40 цт; С. interior valve, bar length = 40 цт; D. supernumerary hook and micropoints, bar
length = 2 |дт; E. micropoints, bar length = 5 цт.
Epioblasma triquetra (Rafinesque, 1820)
(Fig. 69A-E)
Material Examined
IVIAH 588.1 - Big Darby Creek at access
0.9 mi. N of Harrisburg, 1 .7 mi. NW of Orient,
Pleasant Twp., Franklin Co., Ohiio, 27 Sep-
tember 1983, M. A. Hoggarth. MAH 631.1 -
Big Darby Creek at Scioto-Darby (Mt. Ster-
ling-Commercial Pt.) Rd. bridge, 3.4 mi. S of
Orient, 15.3 mi. SW of Columbus, Scioto/
Darby Twp., Pickaway Co., Ofiio, 13 April
1984, M. A. Hoggarth & G. T. Watters.
Description
Glochidium depressed subelliptical, length
208 to 217 |im (214 ± 3.54 }лт, n = 5), height
205 to 21 4 цт (21 1 ± 4.09 цт, n = 5). Dorsal
margin straight, 149 to 156 \im (152 ± 3.05
цт, n = 5) in length. Lateral margins equally
and gently curved; ventral margin broadly
curved. Exterior surface sparsely malleated,
110
HOGGARTH
pits few. Exterior valve sculpture loose-
looped. Dorsal alae absent (Fig. 69A). Central
ligament 48 to 53 |.im (50 ± 2.89 |.im, n = 3) in
length, centered about 44% from posterior to
anterior. Anterior ligament 56 to 65 ¡.im (60 ±
4.51 |im, n = 3) long; posterior ligament 39 to
48 |.im (43 ± 4.73 |.im, n =; 3) in length. Micro-
points blunt, irregular in shape. Ventral flange
narrow; distal unsculptured flange edge wide.
Supernumerary hooks (Fig. 69D) about 2 ¡.im
in length, on unsculptured portion of flange.
Remarks
This glochidium was figured by Lea (1858:
pi. 5, fig. 19, as Unio triangularis, = E. trique-
tra. fide Ortmann & Walker, 1922) and Ort-
mann (1911: pi. 89, fig. 24). Ortmann (1919)
note that Lea's figure was incorrect and gave
length and height measurements of 210 |.im x
210 \xn\ (Ortmann, 1912). Lea's figure does
not show the morphological depression of the
valve.
Epioblasma brevidens (Lea, 1 831 )
(Fig. 70A-F)
Material Examined
OSUM 16173 - Cedar Creek about 1 mi.
SE of Mingo at bridge, Tishomingo Co., Mis-
sissippi, 5 November 1965, R Yokley & B. G.
Isom.
Description
Glochidium depressed subelliptical, length
213to220|.im (216 ± 3.11 |im, n = 4), height
205 to 21 4 (.im (21 ± 4.24 цт, n = 4). Dorsal
margin straight, 141 to 153 |im (147 ± 5.06
\im, n = 4) long. Lateral margins equally and
evenly curved; ventral margin broadly curved.
Exterior surface sparsely malleated, pits few.
Valve pitting reduced in adductor muscle scar;
however, irregular ridges, probably to in-
crease surface area for attachment of large
adductor muscle, found within muscle scar
(Fig. 70B, D). Fine structure of exterior sur-
face of valve rough (Fig. 70E), Dorsal alae ab-
sent. Central ligament 48 to 49 yim (49 ± 0.71
|im, n = 2) in length, centered about 46% from
posterior margin. Anterior ligament 49 to 52
цт (51 ± 2.12 )Lim, n = 2) in length; posterior
ligament 41 to 47 цт (44 ± 4.24 |.im, n = 2)
long. Micropoints small, triangular, arranged
in horizontal rows on ventral flange and on
ventral rim of valve. Supernumerary hooks tri-
angular (Fig. 70B) to lanceolate (Fig. 70C).
Unsculptured distal margin of ventral flange
narrow.
Remarks
This glochidium is similar to that of E. tri-
quetra. They are of about the same shape and
size and have similar micropoints and central
ligament positions. However, the glochidium
of E. brevidens has rough exterior valve
sculpture, whereas that of E. triquetra has
loose-looped exterior valve sculpture.
Epioblasma capsaeformis (Lea, 1834)
(Fig. 71A-G)
Material Examined
OSUM 42007 - Clinch River at Clinchport,
above swinging bridge, 2.3 mi. N of Speer's
Ferry, Scoot Co., Virginia, 2 July 1978, С R.
Ciola & K. L. Ciola.
Description
Glochidium depressed subelliptical, length
240 to 252 цт (246 ± 3.88 цт, n = 8), height
226 to 238 цт (234 ± 4.34 цт, n = 8). Dorsal
margin straight, 154 to 173 цт (162 ± 6.33
jim, n = 8) long. Lateral margins about equal
and gently curved (Fig. 71 A, B) to more
broadly curved (Fig. 71 E). Ventral margin
broadly curved. Dorsal alae absent. Loose-
looped sculpture covering valve exterior (Fig.
71 F). Adductor muscle scar large, with only a
few pits and with numerous small ridges (Fig.
71 B, E). Central ligament 51 to 60 цт (57 ±
3.43 jim, n = 6) in length, centered about 46%
from posterior to anterior. Anterior ligament 52
to 64 j.im (59 ± 4.45 |дт, n = 6) long; posterior
ligament 40 to 57 цт (47 ± 5.89 |im, n = 6) in
length. Micropoints blunt, irregular (Fig. 71 C),
arranged in broken horizontal rows on a nar-
row ventral flange. Near proximal border, mi-
cropoints coalescing to form broken ridges.
Distal unsculptured margin narrow. Attenuate
supernumerary hooks present on otherwise
unsculptured distal flange margin (Fig. 71 D).
Remarks
This glochidium is slightly larger than that of
E triquetra or E brevidens. but it is otherwise
very similar. No published figure of this glo-
chidium was found.
1
GLOCHIDIAOF UNIONIDAE
111
FIG. 70. Glochidium of Epioblasma brevidens. OSUM 16173; A. exterior valve, bar length = 40 цт; В. inte-
rior valve, bar length = 40 цт; С. supernumerary hook and micropoints, bar length = 5 |im: D. adductor mus-
cle scar, bar length = 20 (.im: E. exterior valve sculpture, bar length = 1 ¡.im: F. hinge, bar length = 25 |.im.
Epioblasma rangiana (Lea, 1839)
(Fig. 72A-F)
Material Examined
MAH 632.1 - Big Darby Creek at Scioto-
Darby (Mt. Sterling-Commercial Pt.) Rd.
bridge, 3.4 mi. S of Orient, 15.3 mi. SW of
Columbus, Scioto/Darby Twp., Pickaway Co.,
Ohio, 13 April 1983, M. A. Hoggarth & G. T.
Waiters. MAH 701 - Big Darby Creek at
Scioto-Darby (Mt. Sterling-Commercial Pt.)
112
HOGGARTH
FIG. 71 . Glochidium of Epioblasma capsaeformls. OSUM 42007; A. exterior valve, bar length = 35 цт; В. in-
terior valve, bar length = 35 цт: С. micropoints, bar length = 5 цт: D. зирегпиглегагу hook, bar length =
5 цт; E. interior valve, bar length = 35 цт; F. exterior valve sculpture, bar length = 1 ).im; G. hinge, bar length
= 25цт.
Rd. bridge, 3.4 mi. S of Orient, 1 5.3 mi. SW of
Columbus, Scioto/Darby Twp., Pickaway Co.,
Ohio, 6 September 1985, M. A. Hoggartli.
Description
Glochidium depressed subelliptical, length
238 to 258 |im (249 ± 7.34 \xm, n = 8), height
21 to 238 um (224 ± 9.34 ¡jm, n = 8). Dorsal
margin straight, 160 to 188 |дт (170 ± 6.30
цт, n = 8) in length. Lateral margins and ven-
tral margin are about equally rounded. Shape
of this glochidium approaching that of Obli-
quaria reflexa (subrotund). Dorsal alae ab-
sent. Surface sculpturing loose-looped (Fig.
72F). Central ligament 53 to 59 ¡.im (57 ± 3.21
|.im, n = 3) long, centered about 46% from
posterior to anterior. Anterior ligament 60 to
77 |лт (68 ± 8.62 ¡.tm, n = 3) long; posterior
ligament 50 to 55 |.im (52 ± 5.21 (.im, n = 3) in
length. Micropoints blunt, located on a narrow
ventral flange and on ventral rim of valve (Fig.
72E), covering about 50% of flange, leaving a
wide unsculptured distal flange margin. Su-
pernumerary hooks present as triangular ex-
tensions of this unsculptured area of flange.
GLOCHIDIAOF UNIONIDAE
113
FIG. 72. Glochidium of Epioblasma torulosa rangiana;A. exterior valve, MAH 632.1 , bar length = 40 цт; В.
interior valve, MAH 632.1 , bar length = 35 |.im; С interior valve, MAH 632.1 , bar length = 35 цт; D. hinge,
MAH 701 , bar length = 25 цт; E. micropoints and supernumerary hooks, MAH 701 , bar length = 5 ¡im; F. ex-
terior valve sculpture, MAH 632.1 , bar length = 1 |im.
Remarks
This glochidium can be distinguished by its
extreme morphological depression. It was fig-
ured by Lea (1858: pi. 5, fig. 21, as Unio per-
plexus, = E. rangiana, fide Ortmann & Walker,
1922), who provided length and height mea-
surements of 224 |im X 240 |im. Lea's figure
is correctly drawn, so it must be assumed that
his valve length equals my valve height and
that his valve height equals my valve length.
His measurements would then agree with
mine and Ortmann's (1912) 260 цт x 230 |nm.
DISCUSSION
Anyone interested in the literature dealing
with the glochidium of a particular species is
referred to the remarks section above for that
species, and this information will not be re-
peated here. Instead, this discussion will re-
114
HOGGARTH
late glochidial structure to an understanding
of relationships within the Unionidae and pro-
vide some suggestions for future investiga-
tions. Larval stages have been instrumental in
our understanding of relationships among
higher taxa of invertebrates (phyla, classes
and orders) and have played increasing roles
in lower level taxonomic decisions, even
within the Unionidae (Sterki, 1903; Ortmann,
1912: Morrison, 1955).
Hoeh (1990) proposed a radically new
arrangement of the genus Anodonta, which
has resulted in the splitting of the that genus
into Anodonta s.S.. Utterbackia, and Pygan-
odon in North America. The genus Anodonta
was restricted to A. cygena and related
species found in Europe and along the east
coast of North America. Glochidial characters,
support this reclassification of the genus An-
odonta. with one exception. Hoeh (1990) in-
cludes the species implicata in the genus An-
odonta. but glochidial valve shape, fine
structure of the styliform hook, and exterior
valve sculpture appear to separate this
species from from all other species of An-
odonta, Utterbackia and Pyganodon and sug-
gests that additional work to refine the rela-
tionship between this species and the other
members of this group is warranted. Also, it
might be noted that glochidial morphology
confirms the close relationship between the
members of these genera and A. ferrusa-
cianus.
Glochidial characters assist in confirming
the relationship between the Ambleminae and
the Lampsilinae. Davis & Fuller (1981) con-
clude that the Ambleminae and Lampsilinae
are monophyletic, and the lack of a clean mor-
phological break in glochidial structure ap-
pears to agree with that conclusion. Glochidial
morphology of the more primitive lampsiline
species, such as P. fasciolaris. agree almost
completely with that of some members of the
amblemine species, such as E. dilatata ana P.
dombeyana. It is suggested that an examina-
tion of more of the glochidia of the amblemine
species would be helpful to confirm this con-
clusion.
Perhaps, however, the most powerful dem-
onstration of the use of glochidial characters
in the determination of relationship is the
confirmation of ochracea in the genus Lep-
todea. The species has been assigned to the
genus Lampsilis because of adult shell mor-
phology, including pronounced sexual dimor-
phism (Simpson, 1900, 1914; Johnson. 1947,
1970; Burch, 1975; Fuller & Bereza, 1975;
Clarke, 1 981 b; Porter & Horn 1 981 ). Morrison
(1975), on the other hand, was unable to lo-
cate mantle flaps on gravid females and
placed the species in Leptodea. The size and
shape of the glochidium of this species con-
firm Morrison's conclusion, which is also in
agreement with most modern arrangements
of unionid taxa (e.g., Turgeon et al. 1988).
The glochidia described above suggest
other taxonomic relationships within the Un-
ionidae which have not been proposed pre-
viously. Within the tribe Alasmidontini, two
types of glochidia occur. One has a de-
pressed, pyriform shape, with two row of mi-
crstylets on the styliform hook and looped ex-
terior valve sculpture. Furthermore, these
glochidia posses exceptionally large adductor
muscles in cross-sectional area and dorsal
placement of the adductor muscle (Hoggarth
& Gaunt, 1988). Species with this type of
glochidium included all members of the genus
Strophitus examined, A. viridis. A. heterodon,
L. compressa. L. subviridis. L. iiolstonia and P.
fabula. The other type of glochidium is high,
pyriform, with at least four rows of micro-
styletes on the hook and beaded to rosette ex-
terior valve sculpture. These species, which
include A. undulata. A. marginata. A. confrigo-
sus. L. costata and L. complanata. have a
smaller adductor muscle in cross-sectional
area and the adductor muscle is placed further
from the dorsal shell margin (Hoggarth &
Gaunt, 1988). Additional study using anatomi-
cal or molecular methods might confirm closer
relationships among the members of these
two groups and thereby support a different tax-
onomy of the tribe.
Glochidial structures suggest a much differ-
ent arraignment of the species of Potamilus
and Leptodea than currently accepted (e.g.,
Turgeon at al., 1988). Furthermore, E. lineo-
lata is shown to belong to this group of taxa
as it posses lateral valve gape and small,
obliquely oriented dorsal alae. Ortmann
(1912) was the first to suggest this associa-
tion based on the presence of lateral valve
gape in the glochidia of Potamilus and Ellip-
saria.
The genus Potamilus is currently defined
on the basis of a single character; the pres-
ence of a ligulate glochidium. Ortmann (1912)
stated, ". . .this genus [Potamilus] stands in
all characters except the glochidia, by the side
of Paraptera [= Leptodea]." The current study
suggests that there are many other, more
subtle, glochidial characters that can be used
to define the genus. If the genus is restricted
GLOCHIDIA OF UNIONIDAE
115
to alatus. purpuratus. and capax (the glochid-
¡um of capax was described by Cummings et
al., 1990), then it can be defined as having
ligulate glochidia with lanceolate hooks at the
lateral margins of the ventral flange. The
glochidia of these species are generally large
(210 ,um X 380 ,um in length and height) (ex-
cept capax). and they posses vermiculate ex-
terior valve sculpture. Two species currently
in the genus, P. ohiensis and P. amphichaena.
lack these characters. The glochidia of these
two species, although possessing lateral
valve gape, lack lanceolate hooks at the ven-
tral margin of the valve and posses looped
rather than vermiculate exterior valve sculp-
ture. They are much more rounded at the ven-
tral margin than any other member of the
genus: they are smaller (about the same size
as capax): and they have lamellate rather
than lanceolate micropoints on the ventral
flange. These two species share these glo-
chidial characters with L. fragilis and L. ochra-
cea. I believe a study of adult anatomical
characters could provide useful data that
would support a different relationship among
these taxa. Roe & Lydeard (1998) found mol-
ecular genetic characters support the separa-
tion of ohiensis and amphichaena from alatus
and purpuratus.
Finally, it is suggested that an examination
of the glochidia of additional species would
suggest other taxonomic questions and/or
provide support for currently accepted views
of unionid taxonomy. An examination of the
glochidia of the Ambleminae would be partic-
ularly helpful. Fewer amblemine glochidia
have been examined with SEM than any other
group primarily, because it is more difficult to
tell when amblemine females are gravid and
because they hold mature glochidia for a
shorter period of time. Still, such structures as
the coronal micropoints of Q. infucata and the
exaggerated larval threads of /W. nervosa and
M. boyiiiniana suggest a fruitful area of inves-
tigation.
ACKNOWLEDGEMENTS
This work formed part of a Ph.D. disserta-
tion submitted to The Ohio State University. I
wish to thank my advisor, Dr. David H. Stans-
bery for his support, for use of the collection at
OSU and for the many discussions he has
had with me concerning glochidia. I wish to
thank the National Science Foundation (grant
BSR-8401209) for financial support and Dr.
George Davis and two anonymous reviewers
for helpful criticism of a previous draft. I thank
Dr. J. B. Burch. David Heath, and Kevin Cum-
mings for providing access to specimens, and
I must thank all those who have contributed
specimens to this study. Their names, too nu-
merous to list here, are found in the material
examined sections of each species account.
LITERATURE CITED
AREY, L. В., 1921. An experimental study on
glochidia and the factors underlying encystment.
Journal of Experimental Zoology. 33: 463-499.
AREY, L. В., 1924, Glochidial cuticuiae. teeth, and
the mechanics of attachment. Journal of Mor-
phology 39: 323-335.
AREY, L. В., 1932a, The formation and structure of
the glochidial cyst. Biological Bulletin. 62:
212-221.
AREY, L. В., 1932b, The nutrition of glochidia dur-
ing metamorphosis. Journal of Morptiology 53:
201-221.
BRAUN, M.. 1878, Ueber die postembryonale Ent-
wicklung unserer Susswassermuschein {An-
odonta). Zoologischer Anzeiger 1878: 7-10.
BURCH, J. В., 1975, Freshwater unionacean clams
(Mollusca: Pelecypoda) of North America. Mala-
cological Publications. Hamburg, Michigan.
204 pp.
CALLOWAY С В. & R. D. TURNER, 1979, New
techniques for preparing shells of bivalve larvae
for examination with the scanning electron micro-
scope. Bulletin of the American Malacological
Union. \o\ 1978: 17-24.
CARUS, G. G., 1832, Neue Untersuchungen über
die Entwickeiungsgeschichte unserer Flussmu-
schel. Nova Acta Physico-media Academiae
Caesareae Leopoldino-Carolinae Nature Cu-
hosorum. 16: 1-87.
CLARKE, A. H., 1981a, The thbe Alasmidontini
(Unionidae: Anodontinae), Part I: Regias. Alasmi-
donta. and Arcidens. Smithsonian Contributions
to Zoology 326: 1-101.
CLARKE, A. H., 1981b, The freshwater molluses of
Canada. National Museum of Natural Sciences,
National Museums of Canada, Ottawa, Canada,
446 pp.
CLARKE, A. H., 1985, The thbe Alasmidontini
(Unionidae: Anodontinae), Part II: Lasmigona
and Simpsonaias. Smithsonian Contributions to
Zoology 399: 1 -75.
COKER. R. E. & T SURBER. 1911, Anote on the
metamorphosis of the mussel Lampsilis laevis-
simus. Biological Bulletin. 20: 179-182.
COPE, O. В., 1959, New parasite records from
stickleback and salmon in an Alaska stream.
Transactions of the American Microscopical Soci-
ety 78■.^57-^62.
116
HOGGARTH
CUMMINGS, К. S., M. E. RETZER, С A. MAYER &
L. M. PAGE, 1990, Life history aspects and status
of thie federally endangered fat pocketbook,
Potamilus capax (Green. 1832) (Mollusca; Union-
idae) in the lower Wabash River, Illinois and Indi-
ana. Illinois Natural History Survey Center for Bio-
diversity Techinical Report. 1990(1 ): 1 -37.
DAVIS, G. M. & S. L. H. FULLER, 1 981 , Genetic re-
lationships among Recent Unionacea (Bivalvia)
of North Аглег1са. Malacologia. 20: 217-253.
FLEMMING, W., 1875, Studien in der Entwick-
lungsgesch der Najaden. Sitzungsberichte der
Mathematisch-Naturwissenschaftlichen Classe
der Kaiserlichen Akademie der Wissenschaften.
71:81-212.
FULLER, S. L. H. & D. J. BEREZA, 1 975, The value
of anatomical characters in naiad taxonomy (Bi-
valvia: Unionacea). Bulletin of the American Mal-
ocological Union, for 1974: 21 -22.
GIUSTI, F., 1973, The minute shell structure of the
glochidium of some species of the genera Unio.
Potomida and Anodonta (Bivalvia, Unionacea).
Malacologia. 1 4: 291 -301 .
GIUSTI, F., L. CASTAGNOLO, L. MORETTI FA-
RINA & A. RENZONI, 1975, The reproductive
cycle and the glochidium of Anodonta cygnea L.
from Lago Trasimeno (Central Italy). Monitore
Zoológico Italiano, (п. s.) 9: 99-118.
HAAS, F., 1910, On Unió. Margaritana. Pseudoan-
odonta and their occurrence in the Thames Val-
ley. Proceedings of the Malacological Society of
London. 9: 106-112.
HARMS, W., 1907a, Ueber die postembryonale
Entwicklung von Anodonta piscinalis. Zoologi-
scher Anzeiger 31 : 801 -814.
HARMS, W., 1907b, Zur Biologie und Entwick-
lungsgeschichte der Flussperlmuschel. Zoologi-
scher Anzeiger 31 : 814-824.
HARMS, W., 1907c, Die Entwicklungsgeschichte
der Najaden und ihr Parasitismus. Sitzungs-
berichte der Gesellschaft zur Beförderung der
Gesammten Naturwissenschaften zu Mahrburg.
for 8 May 1907:70-94.
HARMS, W., 1908, Die postembryonale Entwick-
lung von Unio pictorum und U. túmidas. Zoolo-
gischer Anzeigen 32: 693.
HARMS, W., 1909, Postembryonale Entwicklungs-
geschichte der Unionidae. Zooligische Jahr-
bucher. Abteilung für Anatomie und Ontogenis,
28: 325-386.
HEARD, W. H. & R. H. GUCKERT 1970, A re-eval-
uation of the Recent Unionacea (Pelecypoda) of
North America. Malacologia. 10: 333-355.
HISCOCK, I. D., 1951, A note on the life history of
the Australian freshwater mussel, Hyridella aus-
tralis Lam. Transactions of the Royal Society of
South Australia. 74: 146-148.
HOEH, W. R., 1990, Phylogenetic relationships
among eastern North American Anodonta (Bi-
valvia: Unionidae). Malacological Review. 23:
63-82.
HOGGARTH, M. A., 1987, Determination of ante-
rior-posterior orientation of glochidia by the ex-
amination of glochidial valves present within the
umbos of juvenile unionid clams (Mollusca: Bi-
valvia). Ohio Journal of Science. 87: 93-95.
HOGGARTH, M. A. & A. S. GAUNT 1 988, Mechan-
ics of glochidial attachment (Mollusca: Bivalvia:
Unionidae). Journal of Morphology, 198: 71-
81.
HOWARD, A. D., 1912, The catfish as a host for
fresh-water mussels. Transactions of the Ameri-
can Fisheries Society. 42: 65-70.
HOWARD, A. D., 1914a, Some cases of narrowly
restricted parasitism among commercial species
of fresh water mussels. Transactions of the Amer-
ican Fisheries Society. 44: 41 -44.
HOWARD, A. D., 1914b, A new record in rearing
fresh-water pearl mussels. Transactions of the
American Fisheries Society. 44: АЪ-Л1.
HOWARD, A. D., 1914c, Experiments in propaga-
tion of fresh-water mussels of the Quadrula
group. Report of the U.S. Commission of Fish-
eries, for 1913, Appendix IV: 1 -52.
HOWARD, A. D., 1 91 5, Some exceptional cases of
breeding among the Unionidae. Nautilus. 29:
4-11.
HOWARD, A. D., 1951 , A river mussel parasitic on
a salamander. Natural History Miscellanea, 77:
1-6.
HOWARD, A. D. & B. J. ANSON, 1922, Phases in
the parasitism of the Unionidae. Journal of Para-
sitology 9: 68-82.
INABA, S., 1941, A preliminary note on the
glochidia of Japanese freshwater mussels. Anno-
tationes Zoologicae Japonenses. 20: 14-23.
ISOM, B. G., 1983, Potential uses of in i//iro culture
of freshwater mussel glochidia for conservation.
Pp. 42-45, in: A. С MILLER, Compiler, Report of
freshwater mussels workshop. 26-27 October,
1982. U.S. Army Engineer Watenways Experi-
ment Station Environmental Laboratory, Vicks-
burg, Mississippi.
ISOM, B. G. & R. G. HUDSON, 1 982, In vitro culture
of parasitic freshwater mussel glochidia. Nau-
tilus. 96: 147-151.
JOHNSON, R. I., 1946, Anodonta implicata Say.
Museum of Comparative Zoology. Occasional
Papers on Mollusks. 1 : 1 09- 1 1 6.
JOHNSON, R. I., 1947, Lampsilis cariosa Say and
Lampsilis ochracea Say. Museum of Compara-
tive Zoology. Occasional Papers on Mollusks. 1 :
145-156.
JOHNSON, R. I., 1967, Additions to the unionid
fauna of the Gulf drainage of Alabama, Georgia
and Florida (Mollusca:Bivalvia). Breviora. 270:
1-21,
JOHNSON, R. I., 1970, The systematic and zoo-
geography of the Unionidae (Mollusca:Bivalvia)
of the southern Atlantic slope region. Bulletin of
the Museum of Comparative Zoology. 140(6):
263-450.
JOHNSON, R. I., 1972, The Unionidae (Mollusca:
Bivalvia) of peninsular Florida. Bulletin of the
Florida State Museum. Biological Sciences, 16:
181-249.
KAT, P. W., 1984, Parasitism and the Unionacea
(Bivalvia). Biological Review. 59: 189-207.
GLOCHIDIAOF UNIONIDAE
117
LEA, I., 1858, Descriptions of the embryonic forms
of thirty-eight species of Unionidae. Journal of the
Academy of Natural Sciences of Philadelpfiia,
(n.s.)4:43-50.
LEA, I., 1863, Descriptions of the soft parts of one
hundred and forty-three species and some em-
bryonic forms of Unionidae of the United States.
Journal of the Academy of Natural Sciences of
Philadelphia, (n.s.) 5: 401 -456.
LEA, I., 1874, Supplement to Isaac Lea's paper on
Unionidae. Journal of the Academy of Natural
Sciences of Philadelphia, (n.s.) 7: 55-69.
LEEUWENHOEK, A., 1695, Arcana naturae de-
tecta, henricum a kroonevelt. 568 pp.
LEFEVRE, G. & W. С CURTIS, 1910, Experiments
in the artificial propagation of fresh-water mus-
sels. Bulletin of the U.S. Bureau of Fisheries, 28:
615-626.
LEFEVRE, G. & W. С CURTIS, 1912, Studies on
the reproduction and artificial propagation of
freshwater mussels. Bulletin of the U.S. Bureau
of Fisheries, 30: 105-201.
LEYDIG, F, 1866, Mittheilung über den Para-
sitismus junger Unioniden an Fischen in Noll.
Tubingen, Inaugural-Dissertation. Frankfort-am-
Main.
LILLE, F R., 1895, The embryology of the Union-
idae. Journal of Morphology. 10: 1-100.
MCMICHAEL, D. F & I. D. HISCOCK, 1958, A
monograph of the freshwater mussels (Mollusca:
Pelecypoda) of the Australian Region. Australian
Journal of Mahne and Freshwater Research, 9:
372-503.
MORRISON, J. R E., 1955, Family relationships in
the North American freshwater mussels. Bulletin
of the American Malacological Union, for 1955:
16-17.
MORRISON, J. R E., 1973, The families of the
pearly freshwater mussels. Bulletin of the Ameh-
can Malacological Union, for 1972: 45-46.
MORRISON, J. R E., 1975, Maryland and Virginia
mussels of Lister. Bulletin of the American Mala-
cological Union, for 1974: 36-39.
ORTMANN, A. E., 1911, A monograph of the na-
jades of Pennsylvania. I. Anatomical introduc-
tion. The system of the North American najades.
Memoirs of the Carnegie Museum, 4: 279-
347.
ORTMANN, A. E., 1912, Notes upon the families
and genera of the najades. Annals of the
Carnegie Museum. 8: 222-365.
ORTMANN, A. E., 1918, The anatomy of two
African náyades. Unió caff er and Spatha wahl-
bergi. Nautilus, 31 :75-78.
ORTMANN, A. E., 1919, A monograph of the na-
iades of Pennsylvania. Part III. Systematic ac-
count of the genera and species. Memoirs of the
Carnegie Museum, 8: 1 -384.
ORTMANN, A. E., 1923-1924, Notes on the
anatomy and taxonomy of certain Lampsilinae
from the Gulf drainage. Nautilus, 37: 56-50,
99-104, 137-144.
ORTMANN, A. E. & B. WALKER, 1922, On the
nomenclature of certain North American naiades.
Occasional Papers of the Museum of Zoology.
University of Michigan. 112: 1-75.
PORTER, H. J. & K. J. HORN, 1981, Freshwater
mussel glochidia from Lake Waccamaw Colum-
bus County, North Carolina. Bulletin of the Amer-
ican Malacological Union, for 1980: 13-17.
RAND, T. G. & M. WILES, 1982, Species differ-
ences of the glochidia of Anodonta Cataracta Say,
1817 and Anodonta implicata Say, 1829 (Mol-
lusca: Unionidae) by scanning electron mi-
croscopy. Canadian Journal of Zoology, 60:
1722-1727.
RATHKE, J., 1797, Om Dammuslinger. Naturhisto-
rie Selskabets Skhfter (Kjobenhaun), 4: 139-
179.
ROE, K. J. & С LYDEARD, 1998. Molecular sys-
tematics of the freshwater mussel genus
Potamilus (Bivalvia: Unionidae). Malacologia,
39(1-2): 195-205.
SCHIERHOLZ, С, 1878, Zur Entwicklungsge-
schichte der Teich- und Flussmuschel. Zeitschrift
für wissenschaftliche Zoologie, 31 : 482-484.
SCHIERHOLZ, C, 1888, Über Entwicklung der
Unioniden. Denkschnften der Kaiserlichen
Akademie der Wissenschaften (Wein), Mathema-
tisch-naturwissenschaftliche, 55: 183-214.
SCHIMDT, F, 1885, Beitrag zur kenntniss der
postembryonalen Entwicklung der Najaden.
Archiv für Naturgeschichte. 51 : 201 -234.
SIMPSON, С T., 1900, Synopsis of the naiades, or
pearly fresh-water mussels. Proceedings of the
U.S. National Museum, 22: 501 -1044.
SIMPSON, С T., 1914, Л descriptive catalogue of
the naiades or pearly freshwater mussels. Bryant
Walker, Detroit, Michigan, 1,539 pp.
STANSBERY D. H., K. E. NEWMAN, K. G. BOR-
ROR & С В. STEIN, 1985, Literature records of
bivalve mollusks of the Muskingum River system,
Ohio. Ohio Biological Survey Technical Report, 8:
1-472.
STERKI, v., 1898, Some Observations on the gen-
ital organs of Unionidae, with reference to classi-
fication. Nautilus, 12: 18-32.
STERKI, v., 1903, Notes on the unionidae and their
classification. American Nationalist, 37: 103-
113.
SURBER, T, 1912, Identification of the glochidia of
freshwater mussels. Report of the U.S. Commis-
sioner of Fisheries for 1912 and Special Papers,
pp. 1-10.
SURBER, T., 1913, Notes on the natural hosts of
fresh-water mussels. Bulletin of the U.S. Bureau
of Fisheries, 32: 110-116.
SURBER, T., 1 91 5, Identification of the glochidia of
fresh-water mussels. Report of the U.S. Commis-
sioner of Fisheries, for 1914, Appendix V, pp.
1-9.
TOMPA, A. S., 1979, Life-cycle completion of the
freshwater clam Lasmigona compressa (Bivalvia:
Unionidae) in an experimental host, Lebistes
reticu latus. The Veliger, 22: 188-190.
TRAUTMAN, M. В., 1981, The fishes of Ohio. The
Ohio State Univers Press. 782 pp.
TUCKER. M. E., 1927, Morphology of the glochid-
118
HOGGARTH
ium and juvenile of the mussel Anodonta imbe-
cilis. Transactions of the American Microscopical
Society. 46: 286-293.
TUCKER, M. E., 1928, Studies on the life cycles of
two species of fresh-water mussels belonging to
the genus Anodonta. Biological Bulletin. 54:
117-127.
TURGEON, D. D., A. E. BOGAN, E. V. COAN, W. K.
EMERSON, W. G. LYONS, W. L. PRATT, С F. E.
ROPER. A. SCHELTEMA, F. G. THOMPSON &
J. D. WILLIAMS, 1988, Common and scientific
names of aquatic invertebrates from the united
States and Canada: mollusks. American Fish-
eries Society. Special Publication. 16: 1 -277.
UTTERBACK, W. !.. 1915-1916, The naiades of
Missouri. American Midland Naturalist. 4: 41 -53,
97-152, 182-204, 244-273, 311-327, 339-
354, 387-400, 432-464.
VAN DER SOMALIE, H., 1963, The distribution,
ecology and life history of the mussel, Actinona-
ias elllpsiformis (Conrad), in Michigan. Occa-
sional Papers. Museum of Zoology. University of
Michigan. 633: 1-17.
WALLER, D. L., L. E. HOLLAND-BARTELS & L. G.
MITCHELL, 1988, Morphology of glochidia of
Lampsllis higginsi (Bivalvia: Unionidae) com-
pared with three related species. American Mala-
colgical Bulletin. 6: 39-43.
WILES, M., 1975, The glochidia of certain Union-
idae (Mollusca) in Nova Scotia and their fish
hosts. Canadian Journal of Zoology. 53: 33-41 .
WOOD, E. M., 1974, Development and morphology
of the glochidium larvae of Anodonta cygnea
(Mollusca: Bivalvia). Journal of Zoology. London.
173: 1-13.
ZALE, A. V. & R. J. NEVES, 1982, Identification of a
fish host for Alasmidonta т/лол (Mollusca: Union-
idae). American Midland Naturalist. 107: 386-
388.
ZS.-NAGY, I. & E. LABOS, 1969, Light and electron
microscopical investigations on the adductor
muscle and nervous elements in the larva of An-
odonta cygnea L. Annals of Biology. Tihany 36:
123-133.
Revised ms. accepted 10 January 1999
MALACOLOGIA, 1999, 41(1): 119-137
GROWTH PATTERN AND DYNAMICS OF A SOUTHERN PERIPHERAL
POPULATION OF PISIDIUM AMNICUM {MÖLLER, 1774)
(BIVALVIA: SPHAERIIDAE) IN SPAIN
R. Araujo\ M. A. Ramos^ & R. Molinet^
ABSTRACT
The dynamics of a Spanish population of P. amnicum. one of the southernmost populations of
the species in the world, is presented. P. amnicum is the only living European species of the sub-
genus Pisidium s. s. The study is based on monthly samples, from June 1 990 to May 1 991 , from
the Miño River, Galicia, northwestern Spain. Specimens from all size classes were dissected and
all embryo/larval stages were counted and measured. Studies were conducted to find out the
growth pattern of this population using the von Bertalanffy model. Pisidium amnicum in Spain is
semelparous and univoltine. Only two cohorts coexisted from late spring to late summer. The life
span of the species is about 15 months. The parental generation disappear in August. Juvenile
recruitment occurs in April-May when water temperature ranges between 15-20°C. The mini-
mum observed size (shell length) of a gravid specimen was 3-4 mm. Fertilized eggs were
brooded for approximately nine months in the inner demibranches until they reached up to 2 mm.
The bigger clams had, in the same month, more and bigger embryos (or larvae), than the smaller
ones. Not all the initial embryos completed their development. Nevertheless, this phenomenon
of intramarsupial suppression does not seem to be very important in the Spanish population. A
main feature of the Spanish population of P. amnicum is the high number of larvae incubated dur-
ing the months immediately before birth.
Key words: Pisidium amnicum. population dynamics, growth, Spain, peripheral population.
INTRODUCTION
The natural species range is the area in
which it is well adapted both morphologically
as ecologically. However, it is known that en-
vironmental factors may be extreme at the
edge of a species distribution and may pre-
vent it from extending its range. Under such
conditions peripheral populations may show
restriction to particular biotopes (Ford, 1964),
tendency to isolation (Mayr, 1 963), changes in
the genetic structure of the populations
(Ramos, 1985), as well as in other physiolog-
ical aspects, such as development, fecundity
or life span (Möller & Swaddle, 1997).
All known species of the family Spaheriidae
are hermaphroditic, and incubate fertilized
eggs in brood sacs developed in the inner gill.
Therefore, they are an excellent material for
studying reproductive biology. Taking into ac-
count that some reproductive aspects have
been mentioned as very important not only in
the subgeneric characterization (Heard, 1965)
but also at the specific level, it seems impor-
tant to know the specific variation in these
characters along a geographical range.
Pisidium amnicum is the only European
species of the subgenus Pisidium s. s. Only a
few papers include data on the biology and re-
production of this species (Odhner, 1929;
Meier-Brook, 1970; Holopainen, 1979; Holo-
painen & Hanski, 1986) or deal specifically
with its population dinamics and growth (Da-
neel & Hinz, 1976; Bass, 1979, Vincent et al.,
1981), but they all are referred to northern or
central paleartic populations and do not in-
clude data on larval stages prior to shell for-
mation.
This paper describes the dynamics of a
Spanish isolated population, the southern-
most known of the species in Europe and in
the world, with the exception of the North
African population cited by Kuiper (1972).
Therefore, its reproductive characteristics can
also be used as a sensitive measure to as-
certain the extent to which this peripheral pop-
ulation has adapted to local conditions. Data
such as life span, birth period, size at sexual
^Museo Nacional de Ciencias Naturales С. S. I. С. José Gutiérrez Abascal 2. 28006 Ivladrid. Spain; rafael@mncn.csic.es
^Programa de Cooperación Técnica para la Pesca CEE-VECEP ALA 92/43. Centro Profesional Santa Paula, Torre A.
Avenida Circunvalación, Santa Paula, Caracas, Venezuela.
119
120
ARAUJOETAL.
FIG. 1 . Live Pisidium amnicum specimen from the Miño River, Galicia, Spain.
maturity, and percentage and size of incu-
bated larvae, when compared with the avail-
able data from central and northern popula-
tions, should allow us to determine if the
geographical position of this population at the
edge of the species range represents margin-
ality in the ecological and/or physiological
sense, or whether the response remains es-
sentially similar in all the populations. Studies
were conducted for the first time to discover
and describe the growth pattern of a popula-
tion of P. amnicum. dissecting specimens
from all size classes, and counting and mea-
suring all embryo/larval stages.
MATERIAL AND METHODS
Specimens of Pisidium amnicum (Fig. 1)
were collected monthly between June 1990
and May 1991 in the Miño River in north-
western Iberian Peninsula (Fig. 2). From
June to August the sample locality was the
Miño River near Goian, Pontevedra, close to
the Spain-Portugal ferryboat docks. As no
more live specimens of the species were
found after September (Araujo, et al., 1993).
the site was changed to 15 km upstream on
the same river (Fig. 2). The two sites are
on the same shore of the river, which is 400 m
wide, and are exposed to tidal influences as
they are 10 and 25 km, respectively, from the
Atlantic Ocean.
The sampling method consisted of drag-
ging the bottom with a dredge in which sand
and mud were retained so that specimens of
all size classes were collected. Animals were
sorted from the sediment using a 1-mm mesh
sieve. They were carried alive on ice to the
laboratory in plastic jars containing water from
the sample site. Artificial aeration was pro-
vided for five sec every eight h. In the labora-
tory, specimens of each monthly sample were
measured (maximum shell length) and sorted
into 1-mm size classes.
To discover the ratio of gravid animals in
each size class and month, ten animals (when
possible) from each class and month were
dissected. To obtain data on number and size
of incubated embryos (or larvae), ten speci-
mens from each size class per month, from
June to December, were prepared in the field
using the method developed by Eggleton
(Heard, 1965), which consists of isolating
each animal in a glass tube filled with water.
When the tisúes decayed, the larval shells
were counted and measured. Once it was re-
alized that this technique does not permit the
recovery of embryonic phases prior to shell
formation, data for all months were obtained
directly from the animals. After dissection,
embryos and larvae present in each gill were
POPULATION DYNAMICS OF PISIDIUM AMNICUM
121
-^ PORTUGAL
5 Km
FIG. 2. Map of sample sites.
counted and measured (maximum length).
The sizes of the embryos and larvae were ob-
tained by recording the largest and smallest in
each gill. Observations, dissections, counting
and measuring were carried out under a
Bausch and Lomb stereomicroscope with 1 0x
oculars, 1x to 7x zoom and micrometric ocu-
lar. Due to the extremely small size of the in-
cubated phases after fertilization, values for
June to September are less reliable than
those for the rest of the year.
The frequency of gravid animals was calcu-
lated from the data of dissected animals, and
the number and mean size of the incubated
embryos and larvae refer to the number of
gravid animals. In order to check if there was
lateralization, that is differences in the number
and sizes of the incubated phases between
gills, and as these variables are not normally
distributed, a repeated measures ANOVA test
was performed using SuperAnova. Descrip-
tive statistics were carried out using StatView
4.1 . Both software packages are from Abacus
Concepts by Macintosh. The possible influ-
ence on the specimens' gravidity, coded as
gravid (1) and non-gravid (0), of both month
and the specimen size, as well as the interac-
tion between these two factors, was explored
by means of a logistic regression analysis
using the SPSS 6.0 statistical package. The
effect of these factors (specimen size and
month) on the number of incubated embryos
for each of the gravid specimens was tested
by an ANCOVA performed with the Statistica
4.1 package from Statsoft. The variances of
the dependent variable (number of embryos)
were not homogeneous, so the squared-root
transformation was used in the analysis. In
the analysis, data from June to August were
excluded due to very low sample size (n < 1 0).
122
ARAUJOETAL
The same statistical procedure was used to
study the relation of these factors to the size
of the incubated embryos.
The von Bertalanffy model (Bertalanffy,
1934) was used for analysis of growth. Al-
though this model was designed specifically
for fisheries, it has been applied to freshwater
bivalves (Morton, 1969. 1977). This mathe-
matical model expresses length (L) as a func-
tion of age (t) as follows: Lt = L,^, (1-е '^""'o*),
where Ц^, = the mean length of an "infinitely"
old animal, К = the curvature parameter
showing the speed at which the animal be-
comes L|^, and tQ = the parameter of initial
condition, is the time at which the animal has
zero length. It does not have biological signif-
icance.
As no data about age were available, for-
mula parameter values were estimated from
size class composition of the monthly sam-
ples. With these data, the Bhattacharya
(1967) method was used to identify both the
different cohorts alive during the year and the
mean of the normal curve adjusted to each
generational group. These values were the
basis to calculate the parameters of the von
Bertalanffy formula. They were estimated
using ELEFAN (Gayanilo et al., 1988) soft-
ware. As the time between samples was con-
stant (one month) the value of Ц^, was esti-
mated using the graphic method of Ford
(1933) and Walford (1946), as discussed in
Sparreetal., (1989).
While collecting specimens, physical and
chemical characteristics of the water were
recorded at the sample site. Temperature (±
0.5°C), dissolved oxygen (± 1 .0 ppm) and pH
(± 0.1 pH) were monitored using an Horiba U-
7 water checker. Conductivity (|.is ± 0.3%)
was measured using a Crison 523 conduc-
tivimeter. Alkalinity, calcium, total water hard-
ness and carbonate hardness values were
obtained in situ with Merck Aquamerck kits.
The influence of the measured water physical
parameters on gravidity was explored using a
logistic regression analysis.
RESULTS
Population Dynamics and Growth
The raw data on collected, dissected and
gravid specimens, and the number and size of
the incubated phases for each gravid animal
from each size class and month are in Araujo
(1995). Appendix 1 summarizes this informa-
tion.
Monthly histograms of the different size
classes (Fig. 3) showed that there was only
one reproductive period per year. Juveniles
1 -2 mm long only appeared in May, when the
next size class (2-3 mm) reaches its highest
annual frequency. This suggests that birth oc-
curs between April and May. Between June
and July, there was notable mortality in the old-
est classes (Fig. 3). The largest individuals
(10-11 mm) represent a minimum ratio in June
and August, and were not present the rest of
the year. There is hardly any growth in winter.
In fact, the Bhattacharya (1967) method iden-
tified only two cohorts that coexisted from late
spring to late summer (Table 1). Recruitment
occurred in May, and no members of the
parental generation appeared in August, indi-
cating that the life span of each generation is
about 15 months. Values in Table 1 show that
animals were born proportionally very big, and
that their growth was fast, reaching a mean
size between 3.3 and 4.5 mm in summer when
the reproductive period began. This growth
continued until December delaying until Feb-
ruary, when it started again but more slowly.
Animals were largest at the time of juvenile re-
lease, and stabilized at around 8 mm until the
disappearance of the adults.
The growth curve of this population was
tested using a Ford-Walford plot (Table 2).
The corresponding regression analysis (Fig.
4) gave the formula: L(t -(- 1) = 1.499 +
0.81 8L(t), which approaches the 45° line as
expected if growth fits the von Bertalanffy for-
mula. Therefore, the value of the asymptotic
length (Ц^, = 8.245 mm) was employed to cal-
culate the dependent variable of regression in
the von Bertalanffy method, the data from
which appear in Table 2. Values for the inter-
cept and slope of this regression are a = 0. 1 83
and b = 0.201 , respectively. Parameters of the
growth formula of von Bertalanffy are: L,^, =
8.254: К = 0.201 : to = 0.913 years.
Figure 5 shows the monthly frequencies of
gravid animals in each size class. No gravid
animals under 3-4 mm have ever been found.
Therefore, the minimum size to be gravid ap-
pears to be 3-4 mm, although this size was
only recorded in October (50% of collected
specimens). From June (just after juvenile re-
cruitment) to September, some gravid ani-
mals in classes up to 7 mm were found, but
from October to May practically the whole
adult population was gravid. The logistic re-
gression to test the influence of the size and
the period of the year on gravidity revealed
a significant (X^,2 = 182.66: p < 10"^) model
POPULATION DYNAMICS OF PISIDIUM AMNICUM
January n = 164 •'"'У
50
40
30
20
10
120
100
80
60
40
20
n=260
1
I
123
February
60
50
40
30
20
10
n = 163
80
60
40
20
Jh
n = 170
ll
L
August
25
20
15 I
10
5
n = 93
JdiL
^ in <© h.
A 4 Л ¿
(¿ к ¿ Т
о» о ^
September
150
100
50
О
n=350
Jl
April
100
60
60
40
20
n=258
October
10
n=215
.lliü
May
n=285
November
80
60
40
20
n = 202
^M
I
It
5 Î s
n = 404
JJL
70
60
50
40
30
20
10
n = 173
j|
1
FIG. 3. Size histograms of the population structure of P. amnicum in the Miño River between June 1990 and
May 1991.
124
ARAUJOETAL.
TABLE 1. Mean length (mm) of the cohorts identi-
fied by the Bhattacharya (1967) method on the size
distribution of P. amnicum.
Cohort 1 (mm)
Cohort 2 (mm)
Months
X
X
January
7.00
February
7.27
March
7.47
April
7.66
May
2.42
7.83
June
3.27
7.90
July
4.21
7.72
August
4.57
7.92
September
5.77
October
6.12
November
7.11
December
7.22
TABLE 2. Variables used for the Ford-Walford and
von Bertalanffy regression analyses. *t expressed
in months.
Ford-Walford Plot
von Bertalanffy Plot
Time (t)*
Lt
L(t + 1)
Ln(1-(Lt/LJ
1
2.42
3.27
0.3470
2
3.27
4.21
0.5045
3
4.21
4.57
0.7135
4
4.57
5.77
0.8067
5
5.77
6.12
1.2008
6
6.12
7.11
1.3527
7
7.11
7.22
1.9761
8
7.22
7.00
2.0772
9
7.00
7.27
1 .8843
10
7.27
7.47
2.1267
11
7.47
7.66
2.3539
12
7.66
7.83
2.6314
13
7.83
7.90
2.9685
14
7.90
7.73
3.1489
15
7.73
7.92
2.7568
16
7.92
that correctly classified 87.2% of individuals.
The model accuracy was higher for "gravids"
(90.6%) than "non-gravids" (72.3%). The sea-
sonal variation observed in Figure 5 is signifi-
cant (Wald = 52.33; df = 11 ; p < 10^^), time
being the main factor influencing gravidity in
this species as can be deduced from the par-
tial correlation (Грд^ = 0.302). The effect of in-
dividual size on gravidity, although weaker
than that of seasonal variation (Wald = 10.42;
df = 1 ; p = 0.001 ; r^^^ - 0.1 59), is positive, and
the probability of a clam being gravid is higher
among the biggest specimens throughout the
year. The effect of size on gravidity does not
change during the year (interaction between
month and size; Wald = 1.914; df = 11; p >
0.95).
The repeated measures ANOVA test gave
no significant differences either in number
(F^ 263 = 0-009, p = 0.922) or size (F^ 2бз =
0.182, p = 0.67) of incubated phases between
the two gills. Thus, the corresponding values
for the two gills of each individual were added
for analysis.
The ANCOVA test showed that both, the
month (F(8 254^ = 10.27; p < 10"^) and the
mother size (F(i 254) = 324.50; p < 1 0"^) highly
influence the number of incubated embryos
(Table 3, Fig. 6). However, in this case the ef-
fect of time is not homogeneous within all age
classes, because the interaction between
these two factors is highly significant (F(8 246)
= 10.686; p < 10"^). From December to May,
we found a progressive decrease in the total
number of embryos (larvae), especially in the
smaller classes. It is minimum in July and Au-
gust, and thereafter the number of embryos
increased following the oocyte fertilization pe-
riod, which occurs from June to mid-autumn
(Araujo & Ramos, 1999). Figure 7 illustrates
how the seasonal variation is lower among
the smaller gravid specimens. The influence
of mother size is positive (r = 0.749). In other
words, in the same month, as a whole, lar-
ger clams contained more embryos (or lar-
vae) than smaller clams. To show this temp-
oral variation, the standardized regression
coefficients were calculated for all the
months. (The period June-August was ex-
cluded because of the lack of enough gravid
specimens.) Their values were positive and
similar for all the months except May when
negative values (Table 4, Fig. 8) were re-
corded.
Regarding the size of the incubated phases
(Table 5, Fig. 9), the ANCOVA analysis re-
sults, although weaker than the ones regard-
ing number of embryos, suggest that both
month (F,
= 1 73.45; p< 10"
v5 ,
and mother
0.51) have
(8,255)
(1,255)
an effect. The effect of the latter is not homo-
geneous, as it changes monthly (F^g247) =
1.99; p = 0.05); in a given month, larger
classes incubated larger embryos (or larvae)
than smaller ones. The monthly variation of
this regression can be observed in Figure 10.
The progressive increase after August was
proportional to each size class, with a maxi-
mum in April-May just before birth. In the last
month of incubation, larval growth was high-
est.
POPULATION DYNAMICS OF PISIDIUM AMNICUM
125
FIG. 4. Ford-Walford plot of P. amnicum from the
Miño River in which the mean length of each age
grouping (Lt) has been plotted against the suc-
ceeding age grouping (Lt + 1 ).
Influence of Abiotic Factors
Monthly physico-chemical water values are
shown in Table 6. There were significant sea-
sonal variations in conductivity that may be
due to tidal influences, total hardness, alkalin-
ity, calcium values, and water temperature.
The latter increased progressively from April-
May (15-20"C) to July (27X). Conductivity
values are low, as occurs in rivers on granitic
soils poor in soluble salts. This parameter re-
mained constant between 60 and 70 mScm"^
at 25°C during the year, increasing in June to
reach 225 mScm"' in August, and decreasing
to 112 mScm"^ in September. A summer in-
crease also occurs in total hardness and al-
kalinity. Calcium values change in summer,
reaching a maximum in September.
The influence of the physical water param-
eters on the reproductive cycle of P. amnicum
was analyzed using logistic regression (dis-
solved oxygen was excluded, because of the
lack of measurements for two months). The
stepwise procedure (backward) obtained a
highly significant model (X^^ = 168.41; p =
10'"), accounting for 88.30% of the observed
variability in gravidity. The accuracy was
higher when explaining "gravids" (91.37%)
than "non-gravids" (75%). The model identi-
fied four water variables as significant (Table
7), water temperature being the most impor-
tant, as shown by the partial correlation coef-
ficients, although calcium, pH and alkalinity
were also significant.
To test to what extent the reproductive cycle
of P. amnicum (measured by the frequency of
gravid specimens during the year), is influ-
enced by abiotic factors, a multiple regression
analysis was performed using the residuals
obtained from the logistic model (gravidity = f
(mother size, month)) as the dependent vari-
able. The result of the regression analysis
was not significant (r = 0.028; F^^ 33^, = 0.039;
p = 0.99), which means that none' of tne tested
water parameters were directly influencing
the reproduction of the species after removing
the effect of mother size and month of the
year, the previous significances being a sec-
ondary effect of the seasonal variation of
these parameters and their influence on other
intrinsic biotic factors.
DISCUSSION
Reproductive Strategies
In the Spanish study population, P. amnicum
had only one annual reproductive cycle with
births and juvenile recruitment in April-May,
when water temperature ranged between
1 5°C and 20°C. Danneel & Hinz's (1 976) find-
ings for a German population were similar, al-
though the new generation appeared in May-
June, a little later than the Spanish one. The
apparent absence of gravid animals in August
in the German population, and in June and
July in the English one (Bass, 1 979) was prob-
ably a consequence of the method used, as
larval stages prior to shell formation were ig-
nored. With very few P. amnicum specimens
from two different localities in Germany, Meier-
Brook (1970) cited a total of eight gravid indi-
viduals containing embryos between 0.25 and
0.4 mm long in September, and concluded that
the reproductive period was synchronous and
began in autumn; also data for six individuals
collected at the beginning of spring in England
agree with this, "they had big embryos be-
tween 0.5 and 1." Similar data were reported
by Thiel (1928) in Germany and Odhner
(1929) in Sweden, whereas in a Finnish lake
juveniles were born later on in summer (July)
(Holopainen, 1979). Table 8 provides a com-
parative summary of the data on the five stud-
ied populations of P. amnicum.
The species in Europe seems to be very
conservative in reproductive cycle with syn-
chrony among populations: it begins in sum-
mer, the newborn appearing in spring, with
some differences depending on the country.
Such variation seems to follow a latitudinal
dine, which might be related to water temper-
ature. One cycle follows as soon as the for-
mer has finished. Bass (1 979) reported that in
126
ARAUJOETAL.
% gravid animals
JANUARY
| 111111111П1 1 111111Т111п пгтгг1гтптттт1п пгтт тттт1пттттпт ^
FEBRUARY
i Ill llTfTnTTriiiniîTfTïmf 11ПТ1Т1ТП1Т1 ТТтт1ТГГТтттт т т1ттт1тттггт1Т
MARCH
iiiiiimmiiiiiiiiiiiiiiiiiiiiiiiiiiiiilimiiÉi
1Т1Тт ПТПТТт1ттгттттт
APRIL
sraa«JWA;i:^ta«a-j«^i:.'a>--''
iiimmiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiililiiiiiiiiiiiiii
''777?77777777777777777?77?7?7777777?77/Z7777777777777777777}77777777?7777777777Z.
MAY
^^¡^^»»>vz>í>í.r>ovj<>í;<>v>y>í.^^
JUNE
miiiiiiiiiiii
&v>vz^/>vx>'>v>v;
JULY
AUGUST
SEPTEMBER
pcriTi
mrnxiiiiiiminiiiiiiiinniiiiTTiiiiim iiiiiiiiiiiiiiiiii
OCTŒER
¡1ГГ|ТТТТТ1ТП1ПП1ППи1Т1ТТТ11111111ГТТ1П 1111111111111ПТ1ППТТГ1Т1ТТГтГГГ
NOVEMBER
TTimiinnnTimminiTiiiriiiinm iim miiimTtTTTTniTimTTinnnTTT
DECEMBER
mirmillimilllllllllimmillllllinnTIinTTnnmnniiiiiiiiiiiiimn
Ш
m,
=
111
""]
H
■
■
D
■
_l
Ю
00
-vl
en
СЛ
^
CO
ГО
_A
cp
-J.
CO
00
-Ч
a>
СЛ
■^
00
M
FIG. 5. Monthly variation, by size class, in the frequencies of gravid animals in relation to the number of an-
imals dissected.
POPULATION DYNAMICS OF PISIDIUM AMNICUM
127
TABLE 3. Mean number and standard deviation of the embryos (larvae) occurring in the gravid animals.
CLASS
(mm)
JANUARY
FEBRUARY
MARCH
APRIL
MAY
JUNE
1-2
2-3
—
—
—
—
—
3-4
—
—
—
—
—
—
4-5
7.00
—
—
—
—
—
5-6
7.9 ± 2.28
10.5 ± 3.94
3.67 ± 3.21
11.5 ± 0.71
11.00
—
6-7
16 ± 4.92
16.2 ± 5.29
15.5 ±4.95
15.5 ± 5.27
19 ±7.07
—
7-8
34.33 ± 11.18
36.1 ± 14.43
30.3 ± 12.41
29.3 ± 9.43
16.17 ± 16.13
10.00
8-9
44.1 ± 10.94
48.3 ± 14.69
44.6 ± 15.85
43.1 ± 9.37
11.67 ± 5.03
6 ± 4.24
9-10
73.00
56 ± 1.41
63 ± 5.66
42.2 ± 11.61
5.5 ± 6.36
11.00
10-11
—
—
—
—
—
TOTAL
26.07 ± 18.09
31.08 ± 19
29.74 ± 18.11
30.08 ± 14.56
13.71 ± 11.4
8.25 ± 3.59
CLASS
(mm)
JULY AUGUST SEPTEMBER
OCTOBER
NOVEMBER
DECEMBER
TOTAL
1-2
2-3
- -
I
:
I
—
:
3-4
— —
—
6.5 ± 7.78
—
—
6.5 ± 7.78
4-5
— —
13.00
13.33 ± 2,07
15 ±2.83
—
13 ± 2.87
5-6
— —
12 ± 4.63
18.5 ± 5.95
15.33 ± 2.52
14.5 ± 3.69
11.89 ± 5.6
6-7
— —
18.33 ± 6.11
28.83 ± 6.97
28.33 ± 4.84
23.11 ± 9.95
19.33 ± 7.73
7-8
— —
25.6 ± 2.89
31.83 ± 6.49
37.5 ± 6.56
42.17 ± 6.62
31.45 ± 12.33
8-9
3 ± 2 2.33 ±
1.53 25.67 ± 4.93
35.67 ± 7.5
41.17 ± 11.65
43.33 ± 9.65
36.97 ± 17.61
9-10
— —
42.00
—
40.00
44.33 ± 16.62
42.17 ± 20.36
10-11
— —
—
—
—
—
—
TOTAL
3 ± 2 2.33 ±
1.53 21.94 ±8.64
24.44 ± 11.06
31.58 ±11.92
32.57 ± 14.71
—
England a small proportion of the largest
adults initiated a second reproductive cycle,
but died before the brood was fully developed.
None of the European populations show a de-
cline in fertility in relation to age, as occurs in
the related genus Sphaerium (Meier-Brook,
1970).
Avery close similarity can also be found be-
tween the reproductive strategies of P. am-
nicum and the vicahant North American
species P. dubium. Both species are of similar
size, share the same incubation and birth
months, and, in both cases, the mean number
and size of larvae increased with the size of
the parents, although the number of incu-
bated larvae was considerably higher in the
Spanish population. The strategy of the Cana-
dian population of P. amnicum (Vincent et al.,
1981) also fits perfectly with the one of P. du-
bium, with only two exceptions: P. dubium at-
tains its sexual maturity in its first year of life,
and its number of larvae is considerably
higher, suggesting that these two populations
might belong to the same taxon.
Meier-Brook (1970) suggested that, with
the exception of P. amnicum for which no data
were available, sexual maturity in populations
of Pisidium occurs when individuals reach 50
or 60% of maximum size. According to this
rule, as the maximum size of the Spanish
population ranges from 10-11 mm, it follows
that sexual maturation would occur in a size
class near 5-6 mm, which is higher than the
4-5 mm that we found to be the minimum
needed to be gravid (with the exception of two
gravid specimens of 3-4 mm collected in Oc-
tober). However, it applies perfectly to the
maximum theoretical length (8.25 mm) esti-
mated by the von Bertalanffy method. More
surprising are the proportions reported by Da-
neel & Hinz (1976) (Table 8). Thus, although
most of the available data (Table 8) seem to
follow Meier-Brook's rule, in some popula-
tions maturity is reached before specimens
are 40-50% of maximum size. Then, if any
rule were applicable to P. amnicum, we would
expect a positive correlation between mini-
mum gravid length and maximum adult size
throughout the species range (Table 8), which
was not found.
The main feature of the Spanish population
of P. amnicum seems to be the high number
of larvae incubated during the months imme-
diately before birth. The difference with all the
other previous data may be explained by dif-
ferences related to temperature and/or lati-
tude. In warmer climates, the same species
becomes bigger, incubates a greater number
of larvae (maximum of 73 in Spain and 12 in
Finland), and the larvae are also bigger. Such
128 ARAUJOETAL
mean number of embryos
ooooooooo
\ — I
mnrnTmnnmriTinnnnnmimmlimmimmiiiiiiiimiii
FEBRUARY
MARCH
IrrmTniiiiiimiiiiiiiiiTimiiiiiTirmmTnmiiii
^i
^^^
lllli
iimiimmiiimiimiiiimmmmmmmm
III
iiniiiiiiiimmT
MAY
mm
JUNE
imniiiii
JULY
13
AUGUST
SEPTEMBER
OCTOBER
NOVEMBER
DECEMBER
jnmi iimimiiTmnnmiiimiiiiii
г ■ :': — ■ ]
И II I ПИП
himnniimnnnnnnnmínmmni
1
II
Г1
Щ
■
■
■
CD
00
•Ч
O)
S"
■t^
W 1
i CD
1
CD
00
■>J
Ô)
Ul
к
FIG. 6. Monthly variation, by size class, in the mean number of embryos (larvae) occurring in gravid animals.
POPULATION DYNAMICS OF PISIDIUM AMNICUM
129
80
•
/U
•
bU
•
50
-
•
40
30
•
•
•
•
•
1
•
1
•
•
i
10
•
•
•
•
1
•
t
•
•
•
t
•
•
Size classes (mm)
FIG. 7. Variation in the mean number of embryos
(larvae) in each size class with parental size.
TABLE 4. Monthly standardized regression coeffi-
cients of the variation in the number of incubated
embryos in relation with the mother size.
Month
BETA
St. Err. of BETA
January
0.898
0.070
February
0.837
0.091
March
0.842
0.094
April
0.800
0.101
May
-0.325
0.272
September
0.824
0.157
October
0.846
0.097
November
0.794
0.129
December
0.733
0.133
north-south clinal reproductive behaviour was
apparently not found in preliminary studies
with the related North American species P.
idahoense Roper, in which the northern popu-
lations were smaller but had more progeny
(Heard, 1965). On the other hand, as the life
span of P. amnlcum in Spain appears to be
about 15 months, our results do not confirm
Bass's idea (1 979) of an extended life span or
a successful second brood in southern areas.
The scarce data on the Finnish population,
one of the northermost in the species range,
indicate a reduction in fertility in relation to the
central European populations. This reduction
may be accompanied by a long life span, al-
lowing iteropahty, as reported by Holopainen &
Hanski (1 986) and Vincent et al., (1 981 ) in Fin-
land and Canada, respectively. Given that the
Spanish population broods the maximum
number of larvae in the species range, and
that the maximum embryo length is similar to
all other reported populations, we may state
that in P. amnicum there is no trade-off be-
tween litter size and embryo size as was pro-
posed by Holopainen & Hanski (1986) for the
genus Pisidlum. However, the characteristics
of the Spanish population as a whole reflect its
breeding success. The species seems to be so
well adapted to local conditions in this range
margin that it is possible to speculate that such
attributes might allow the species to expand its
range if the opportunity presented itself. In
fact, the Goian (Fig. 2) colony, which experi-
enced a drastic reduction after colonization by
Corbicula fluminea (Müller, 1774) (Araujo et
al., 1993), has re-established itself, sharing its
habitat with the latter species as we have ob-
served recently (unpublished data).
Study of the phenomenon that Meier-Brook
(1977) called intramarsupial suppression of
fetal development has deserved special atten-
tion. This author cited 50% of the brood dying
before birth in P obtusale and P. lilljeborgii,
whereas it is very variable in P amnicum
(0-54%) according to Danneel & Hinz (1976).
Our results show a progressive increase in
embryos from June to December, which
means that settlement of eggs or zygotes may
occur over several months. On the other hand,
we observed that from December, and in sev-
eral size classes, the embryo number may
decrease, suggesting that not all the initial em-
bryos complete their development. Neverthe-
less, this phenomenon does not seem to be
very important in the Spanish population. The
idea suggested by Araujo & Ramos ( 1 997) that
ova fertilization occurs in the gills and not in the
spermoviduct as was proposed (Okada, 1 935;
Odhner, 1929; Meier-Brook, 1970), may pro-
vide a new perspective, suggesting that the so
called "embryos" present in the gill during the
first months of the reproductive period may be
eggs not yet impregnated. This may explain
the differences found by other authors when
comparing the initial and final number of larval
stages in the gills.
Although there is the same high mortality
following the birth of juveniles, the corre-
sponding ratio of the size classes over 6 mm
in the Spanish population is higher than in the
others during the year.
Influence of Environmental Factors
None of the water measured parameters in
itself seems to directly influence the reproduc-
tive cycle of the Spanish population of P am-
nlcum as measured by specimen gravidity,
although seasonal variation in water tempera-
ture, calcium, pH, and alkalinity might be im-
130
ARAUJOETAL.
December
FIG. 8. Monthly variation in the relation between the mean number of embryos and parental size.
TABLE 5. Mean lenght (mm) and standard deviation of the embryos (larvae) occurring in the gravid animals
CLASS
(mm)
JANUARY
FEBRUARY
MARCH
APRIL
MAY
JUNE
1-2
2-3
3-4
4-5
-
—
-
—
—
—
0.5
—
—
—
—
_
5-6
0.52 ± 0.12
0.55
±0.1
0.45
± 0.39
0.76 ± 0.04
1.3
—
6-7
0.66 ± 0.19
0.73
± 0.88
0.84
± 0.09
0.88 ± 0.16
1.37 ± 0.19
—
7-8
0.73 ± 0.14
0.87
± 0.11
0.92
± 0.09
1.23 ± 0.25
1.61 ± 0.4
0.08
8-9
0.79 ± 0.17
0.9
± 0.1
0.94
± 0.13
1.2 ± 0.22
2.02 ± 0.08
0.09 ± 0.04
9-10
0.85
1
± 0.14
1.09
± 0.13
1.08 ± 0.15
2.26 ± 0.09
0.1
10-11
—
—
—
—
—
—
TOTAL
0.67 ± 0.17
0.8
± 0.16
0.87
± 0.2
1.08 ± 0.25
1.77 ± 0.4
0.09 ± 0.02
CLASS
(mm)
JULY AUGUST
SEPTEMBER
OCTOBER NOVEMBER
DECEMBER
TOTAL
1-2
2-3
—
-
-
-
-
-
-
-
3-4
-
-
-
0.08 ±
-
0.08 ±
4-5
—
—
0.08
0.12 ±
0.04 0.13 ±
0.07
—
0.15 ± 0.13
5-6
—
—
0.09 ± 0.01
0.18 ±
0.10 0.25 ±
0.09
0.42 ± 0.07
0.46 ± 0.24
6-7
—
—
0.12 Í
t 0.02
0.22 ±
0.02 0.37 ±
0.07
0.42 ± 0.14
0.62 ± 0.29
7-8
—
—
0.09 =
: 0.01
0.22 ±
0.04 0.34 ±
0.03
0.55 ± 0.15
0.75 ± 0.47
8-9
0.08
± 0.01 0.08
±0.02
0.14 d
: 0.12
0.25 ±
0.03 0.34 ±
0.05
0.6 ± 0.08
0.75 ± 0.47
9-10
—
—
0.18
-
0.35
0.45 ± 0.15
0.84 ± 0.54
10-11
—
—
-
-
—
—
—
TOTAL
0.08
± 0.01 0.08
± 0.02
0.11 ± 0.05
0.19 ±
0.07 0.32 ±
0.09
0.49 ± 0.14
—
portant for their influence on other species
intrinsic factors (i.e., mother size, previous epi-
sodes of reproduction). Moreover, since P. am-
nicum incubates its larvae within a ctenidial
marsupium, larvae release and subsequent
growth could be induced, both directly and in-
directly, by these factors.
Temperature increases progressively over
April-May, when births occur, to July, when fer-
tilization begins. The 27'C maximum is more
or less maintained in August and September,
and it is followed by a sudden decrease of
about 9 С in October, precisely when the pe-
riod of maximum population growth ends. It
could be argued that temperature has so im-
portant relation with birth that larvae big
enough to be born in April did not because
temperature was below 15'C. A relationship
between temperature and growth in freshwa-
ter bivalves has also been described in the Asi-
atic clam Corbicula fluminea (Morton, 1977;
Eng, 1979; Joy, 1985; Ituarte, 1985).
POPULATION DYNAMICS OF PISIDIUM AMNICUM 131
mean size of the incubated embryos
JAN'UAR^
FEBRUAR'»
ÎTTiiiirinmirTiiiiiirTrnii
ппитпитпигтгтпттлт
MARCH
В
rniiiiiTimiriiiiiiimTTrmnmii
APRIL iggiiiigggggggi^^
[тппптпттттп i тттттптгтп i ii i
MAY
AUGUST
SEPTEMBER
OCTOBER
NOVEMBER
DECEMBER
h
Ii
+
lllllllllini
IIIIIIIIIIIIIITT
JUNE L_
JULY
im
M п г m m
ЧО 00 -J о '-П -t>- OJ
>— чО oc ~J C> '-Л -t^
FIG. 9. Monthly variation, by size class, in mean embryos (larvae) size in gravid animals.
132
ARAUJOETAL
Size classes
FIG. 10. Monthly variation in the relation between mean embryo size and parental size.
TABLE 6. Montly and mean values of the chemical and physical properties of the water at the sample site
Dissolved
Conductivity
Total
Carbonate
Temperature
oxygen
(las cm-1)
Calcium
Alkalinity
hardness
hardness
Month
(^C)
ph
(ppm)
(mg/l)
(mmol/l)
(°dh)
(°dh)
January
8.13
6.33
12.23
53.00
8.00
0.50
1.20
1.00
February
9.17
7.13
—
60.00
9.00
0.50
1.75
1.40
March
10.80
6.00
14.77
60.00
10.00
0.35
1.30
1.40
April
13.97
6.07
13.97
68.00
11.00
0.50
1.30
1.40
May
18.83
5.97
9.77
78.00
7.00
0.60
1.50
1.70
June
22.73
5.87
11.23
84.80
8.00
0.40
1.20
1.50
July
26.97
8.17
8.43
149.00
7.00
0.45
1.60
1.50
August
26.17
7.33
—
225.00
13.00
0.80
2.00
1.50
September
26.27
7.37
9.93
112.00
20.00
0.80
1.90
1.70
October
17.67
8.27
8.43
92.50
10.00
0.75
1.55
1.50
November
14.80
7.23
9.77
82.00
10.00
0.65
1.85
1.60
December
11.63
7.20
9.93
76.00
13.00
0.50
1.70
1.40
Mean
17.26
6.91
10.85
95.03
10.50
0.57
1.57
1.47
sd
6.62
0.81
2.07
46.53
3.45
0.15
0.27
0.17
TABLE 7. Water parameters selected by the logistic
regression analysis as the most important influenc-
ing the gravidity of P. amnicum
Variable
Wald
df
sig.
Alkalinity
3.97
1
0.05
-0.077
Calcium
16.44
1
0.001
0.209
pH
10.27
1
0.001
0.158
Temperature
53.25
1
10-^
-0.40
Calcium, alkalinity and pH increased in
summer suggesting that a relation with the
higher growth speed in these months may
also exist. Our results show a certain corre-
spondence between the increase in the val-
ues of the abovementioned parameters and
the disappearance of the adults.
Finally, Hornbach & Cox (1987) suggested
that in Pisidium casertanum (Poli) there exists
a possitive relation between high calcium and
alkalinity values and adult shell length, and
probably also embryo number and size. If we
compare the values of these two parameters
for the Spanish P. amnicum (alkalinity: mean
= 0.57 mmol/l; calcium: mean := 10.50 mg/l)
with those of the Finnish population (alkalinity:
0.17-0.19 mmol/l; calcium = 6.1 mg/l) (Holo-
painen, 1979), the only available ones, we
could explain the biggest population parame-
ters found in Spain (Table 6) as a function of
the relatively "high" calcium and alkalinity val-
ues.
AKNOWLEDGEMENTS
We are very indebted to Dr. Emilio Rolan
and his wife for the facilities provided during
the monthly sampling field trips, and to Diego
Moreno for his help in the field. We are also
POPULATION DYNAMICS OF PISIDIUM AMNICUM
133
TABLE 8. Reproductive and growth features of several populations of P. amnicum.{^) Several specimens live
longer but do not breed. (2) Recorded two months before birth in adults bigger than 4 mm. (3) Not all months
recorded. (4) Recorded in the month immediately after birth. (5) Recorded in the first year of life.(*) Values
dependant on parents' size and date.
Germany
England
Finland
Canada
Danneell &
Bass
Holopainen
Vincent et al.
Spain This
Hinz (1976)
(1979)
(1979)
(1981)
study
Maximum adult length (mm)
8.9
8.5
8.8
9
10-11
Life span (months)
12
12(1)
—
36
15
Birth period
May-June
May
July
June
April- May
Minimum gravid length (mm)
2-3
4
4.3
4
3-4
% of gravid animals (2)
80-100%
1 00%
—
42-62
100%
Maximum number of
34
37
12
18
73
embryos/larvae*
Maximum length of larvae (mm)*
2
1.4
0.6(3)
—
2.4
Lateralization
no
—
—
—
no
% of new born in relation to
< 50%
< 50%
—
62%
< 50%
whole population (4)
Premature births
yes
probably
—
—
yes
Annual mean growth of adults
3
4
—
2-3 (5)
5-6
(mm)
Winter interruption of larval
yes
no
—
-
no
growth
very grateful to Dr. Luis M. Carrascal for his in-
valuable help and advice as regards the sta-
tistical analysis of the data. Two anonymous
reviewers made interesting comments which
improved the manuscript. Thanks also to Dr.
Pablo Penchaszadeh for his comments and
help in the preparation of the manuscript and
to Lesley Ashcroft for the revision of the Eng-
lish version. This work received financial sup-
port from the Project "Fauna Ibérica И" (SEUI,
DGICYTPB89 0081).
LITERATURE CITED
ARAUJO, R., D. MORENO & M. A. RAMOS, 1993,
The Asiatic clam Corbicula fluminea (Müller,
1774) (Bivalvia: Corbiculidae) in Europe. Ameri-
can Malacological Bulletin. 1 0: 39-49.
ARAUJO, R., 1995, Contribución a la taxonomía у
biogeografía de la familia Sphaehidae en la
Península Ibérica e Islas Baleares con especial
referencia a la biología de Pisidium amnicum. Ph.
D. thesis, Universidad Complutense de Madrid,
393 pp.
ARAUJO, R. & M. A. RAMOS, 1997, Evidence of in-
trafollicular fertilization in Pisidium amnicum
(Müller, 1774) (Mollusca: Bivalvia). Invertebrate
Reproduction and Development, 32(3): 267-272.
ARAUJO, R. & M. A. RAMOS, 1999, Histological
description of the gonad, reproductive cycle and
fertilization of Pisidium amnicum (Müller, 1774)
(Bivalvia: Sphaeriidae). The Veliger, 42(2): 124-
131
BASS, J. A. В., 1979, Growth and fecundity of Pi-
sidium amnicum (Müller) (Bivalvia: Sphaeriidae)
inTheTadnoll Brook, Dorset, England. Journal of
Conchology. 30: 129-134.
BERTALANFFY, L. VSON, 1934, Untersuchungen
über die Gesetzlichkeiten des Wachstums. 1 . All-
gemeine Grundlagen der Theorie. Roux Arctiiv
für Entwicklungsmechanik der Organismen. 1 31 :
613-653.
BHATTACHARYA, С G., 1967, Asimple method of
resolution of a distribution into Gaussian compo-
nents. Biometrics. 23: 115-135.
DANNEEL, I. & W. HINZ, 1 976, Zur Biologie von Pi-
sidium amnicum О. F. Müller (Bivalvia). Archiv für
Hydrobiologie. 77(2): 213-225.
ENG, L. L., 1 979, Population dynamics of the Asiatic
clam, Corbicula fluminea (Müller), in the concrete-
lined Delta-Mendota Canal of Central California,
pp. 250-287, in: J. С britton, ed.. Proceedings
of the International Corbicula Symposium, 1st,
1977, Texas Christian University. Fort Worth,
Texas USA.
FORD, E., 1933, An account of the herring investi-
gations conducted at Plymouth during the years
from 1 924 to 1 933. Journal of the Mahne Biolog-
ical Association of the United Kingdom, 19: 305-
384.
FORD, E. B. 1964, Ecological genetics. London:
Methuen.
GAYANILO. F. С JR., M. SORIANO & D. PAULY
1988, A draft guide to the COMPLEAT ELEFAN.
ICLARM Software Project. 2: 65 pp. and 10
diskettes.
HEARD, W. H., 1965, Comparative life histories of
North American pill clams (Sphaeriidae: Pisid-
ium). Malacologia, 2(3): 381 -411.
134
ARAUJOETAL
HOLOPAINEN, I. J., 1979, Population dynamics
and production of Pisidium species (Bivalvia:
Sphaeriidae) in the oligotrophic and mesohumic
lake Pääjärvi, southern Finland. Archiv für Hydro-
biologie. Supplement. 54(4): 446-508.
HOLOPAINEN, I. J. & I. HANSKI, 1986, Life history
variation in Pisidium (Bivalvia: Pisidiidae). Hol-
arctic Ecology. 9; 85-98.
HORNBACH, D. J. & С COX, 1987, Environmental
influences on life history traits in Pisidium caser-
tanum (Bivalvia: Pisidiidae): field and laboratory
experimentation. American Malacological Bul-
letin, bil): 49-64.
ITUARTE, С F., 1985, Growth dynamics in a nat-
ural population of Corbicula fluminea (Bivalvia:
Sphaeriacea) at Punta Atalaya, Rio de la Plata,
Argentina. Studies of the Neotropical Fauna and
Environment. 20: 217-225.
JOY, J. E., 1985, A 40 week study on growth of the
Asian clam, Corbicula fluminea (Müller), in the
Kanawha River, West Virginia. The Nautilus. 99:
110-116.
KUIPER, J. G. J., 1972, Une récolte de Pisidium
dans le Moyen Atlas. Résultats de la mission bi-
ologique au Maroc de l'Université de Grand Bel-
gique. Publication no. 9. Basteria. 36: 189-198.
MAYR, E. 1963, Animal species and evolution.
Cambridge, Mass.: Harvard University Press.
MEIER-BROOK, C, 1970, Untersuchungen zur Bi-
ologie einiger Pisidium-Arten (Mollusca: Eulamel-
libranchiata: Sphaeriidae). Archiv für Hydrobiolo-
gie. Supplement. 38(1/2): 73-150.
MEIER-BROOK, С, 1977, Intramarsupial suppres-
sion of fetal development in sphaerüd clams.
Malacological Review. 10: 53-58.
MÖLLER, A. P. & J. P. SWADDLE, 1997, Asymme-
try developmental stability and evolution. Oxford
University Press. Oxford.
MORTON, В., 1969, Studies in the biology of Dreis-
sena polymorpha Pall. III. population dynamics.
Proceedings of the Malacological Society of Lon-
don. 38: 47^ -482.
MORTON, В., 1977, The population dynamics of
Corbicula fluminea (Bivalvia: Corbiculacea) in
Plover Cover, Hong Kong. Journal of Zoology,
London. 181:21-42.
ODHNER, N. H., 1929, Die Molluskenfauna des
Täkern. Sjön Täkerns Fauna och Flora. Utgiven
av K. Svenska Vetenskapsakademien. 8: 129
pp.,7pl.
OKADA, K., 1935, Some notes on Musculium het-
erodon (Pilsbry), a freshwater bivalve III. Fertil-
ization and segmentation. Science Reports of the
Tohoku Imperial University. Biology, 10: 467-
483.
RAMOS, M. A. 1 985, Shell polymorphism in a south-
ern peripheral population of Cepaea nemoralis
(L.) (Pulmonata, Helicidae) in Spain. Biological
Journal of the Linnean Society. 25: 1 97-208.
SPARRE, P., E. URSIN & S. С. VENEMA, 1989, In-
troduction to tropical fish stock assessment. Part
1- Manual. FAO Fisheries Technical Paper
306(1): 337 pp.
THIEL, M. е., 1928, Zur Biologie unserer Süb-
wassermuscheln. Zeitschhft für Morphologie und
Ökologie der Tiere. 13: 65-116.
VINCENT, В., G. VAILLANCOURT & N. LAFON-
TAINE, 1981, Cycle de développement, crois-
sance et production de Pisidium amnicum (Mol-
lusca: Bivalvia) dans le Saint-Laurent (Ouébec).
Canadian Journal of Zoology 59: 2350-2359.
WALFORD, L. A., 1946, A new graphie method of
describing the growth of animals. Biological Bul-
letin. Mahne Biological Laboratory Woods Hole,
90: 141-147.
Revised ms. accepted 10 August 1998
POPULATION DYNAMICS OF PISIDIUM AMNICUM
135
APPENDIX 1. Summarized information on the specimens collected, dissected and gravid, and of the total
mean number and mean size of the incubated embryos (or larvae). Frequency of gravid animals at 1%
freq.
total
mean
mean
collected
dis-
freq.
of
number of
length of
month
class
collected
%
sected
gravid
gravid
embryos
embryos
embryos
January
1-2
2-3
3-4
1
0.61
4-5
4
2.44
1
1
1
7
7
0.5
5-6
34
20.73
10
10
1
79
7.9
0.52
6-7
44
26.83
10
10
1
160
16
0.66
7-8
41
25.00
10
9
0.9
309
34.33
0.73
8-9
37
22.56
10
10
1
441
44.1
0.79
9-10
3
1.83
1
1
1
73
73
0.85
10-11
Total January
164
100
42
41
0.98
1069
26,07
0.67
February
1-2
2-3
3-4
4-5
1
0.61
5-6
18
11.04
7
6
0.86
63
10.5
0.55
6-7
44
26.99
10
10
1
162
16.2
0.73
7-8
60
36.81
10
10
1
361
36.1
0.87
8-9
35
21.47
10
10
1
483
48.3
0.9
9-10
5
3.07
2
2
1
112
56
1
10-11
Total February
163
100
39
38
0.97
1181
31.08
0.8
March
1-2
2-3
3-4
4-5
1
0.59
5-6
11
6.47
3
2
0.66
11
3.67
0.45
6-7
35
20.59
10
10
1
155
15.5
0.84
7-8
75
44.12
10
10
1
303
30.3
0.92
8-9
41
24.12
10
10
1
446
44.6
0.94
9-10
7
4.12
2
2
1
126
63
1.09
10-11
Total March
170
100
35
34
0.97
1041
29.74
0.87
April
1-2
2-3
3-4
4-5
1
0.39
5-6
5
1.94
2
2
23
11.5
0.76
6-7
62
24.03
10
10
155
15.5
0.88
7-8
91
35.27
10
10
293
29.3
1.23
8-9
82
31.78
10
10
431
43.1
1.2
9-10
17
6.59
5
5
211
42.2
1.08
10-11
Total April
258
100
37
37
1
1113
30.08
1.08
May
1-2
52
18.25
10
2-3
106
37.19
11
3-4
36
12.63
10
4-5
5-6
3
1.05
1
1
1
11
11
1.3
6-7
12
4.21
3
1
0.33
38
38
2.73
7-8
35
12.28
10
6
0.6
97
16.17
1.61
8-9
34
11.93
10
3
0.3
35
11.67
2.02
9-10
7
2.46
2
2
1
11
5.5
1.72
10-11
Total May
285
100
57
13
0.23
192
14.77
1.78
{continued)
136
ARAUJOETAL
APPENDIX 1, (Continued)
freq.
total
mean
mean
collected
dis-
freq.
of
number of
length of
month
class
collected
%
sected
gravid
gravid
embryos
embryos
embryos
June
1-2
2-3
32
7.92
3-4
87
21.53
4-5
4
0.99
5-6
8
1.98
2
6-7
35
8.66
5
7-8
106
26.24
5
1
0.2
10
10
0.06
8-9
101
25
5
2
0.4
12
6
0.07
9-10
30
7.43
5
1
0.2
11
11
0.08
10-11
1
0.25
Total June
404
100
22
4
0.18
33
8.25
0.07
July
1-2
2-3
3-4
66
25.38
4
4-5
120
46.15
4
5-6
10
3.85
6-7
18
6.92
3
7-8
24
9.23
5
8-9
12
4.62
4
3
0.75
9
3
0.06
9-10
10
3.85
10-11m
Total July
260
100
20
3
0.15
9
3
0.06
August
1-2
2-3
3-4
4
4.3
4-5
17
18.28
5
5-6
6
6.45
1
6-7
11
11.83
5
7-8
25
26.88
4
8-9
22
23.66
4
3
0.75
7
2.33
0.08
9-10
7
7.53
2
10-11
1
1.08
Total August
93
100
21
3
0.14
7
2.33
0.08
September
1-2
2-3
3-4
13
3.71
2
4-5
37
10.57
8
1
0.125
13
13
0.08
5-6
67
19.14
10
3
0.3
26
4.33
0.03
6-7
143
40.86
6
3
0.5
55
18.33
0.12
7-8
65
18.57
6
5
0.83
128
25.6
0.09
8-9
19
5.43
5
3
0.6
77
25.67
0.14
9-10
6
1.71
2
1
0.5
42
42
0.18
10-11
Total Septembe
r
350
100
39
16
0.41
341
17.95
0.09
October
1-2
2-3
3-4
12
5.58
4
2
0.5
13
6.5
0.06
4-5
50
23.26
6
6
80
13.33
0.12
5-6
52
24.19
6
6
111
18.5
0.18
6-7
28
13.02
6
6
173
28.83
0.22
7-8
40
18.6
6
6
191
31.83
0.22
8-9
32
14.88
6
6
214
35.67
0.25
9-10
1
0.47
10-11
Total October
215
100
34
32
0.94
782
24.44
0.19
POPULATION DYNAMICS OF PISIDIUM AMNICUM
137
APPENDIX 1, (Continued)
freq.
total
mean num-
mean
collected
dis-
freq.
of
ber of
length of
month
class
collected
%
sected
gravid
gravid
embryos
embryos
embryos
November
1-2
2-3
3-4
2
0.99
4-5
9
4.46
2
2
30
15
0.13
5-6
19
9.41
3
3
46
15.33
0.25
6-7
49
24.26
6
6
170
28.33
0.37
7-8
85
42.08
6
6
225
37.5
0.34
8-9
33
16.34
6
6
247
41.17
0.34
9-10
5
2.48
1
1
40
40
0.35
10-11
Total November
202
100
24
24
1
758
31.58
0.32
December
1-2
2-3
3-4
1
0.58
4-5
5
2.89
5-6
15
8.67
4
4
58
14.5
0.42
6-7
43
24.86
9
9
208
23.11
0.42
7-8
69
39.88
6
6
253
42.17
0.55
8-9
36
20.81
6
6
260
43.33
0.6
9-10
4
2.31
3
3
133
44.33
0.45
10-11
Total December
173
100
28
28
1
912
32.57
0.49
TOTAL
2737
398
276
0.69
7438
26.95
0.66
MALACOLOGIA, 1999, 41(1): 139-145
RELATIONSHIPS BETWEEN LENGTH OF PREY/PREDATOR FOR THE MOST
IMPORTANT PREY OF THE CUTTLEFISH SEPIA OFFICINALIS L.
(MOLLUSCA: CEPHALOPODA)
Alexia Blanc, Guy Pinczon du Sel & Jacques Daguzan
Laboratoire de Zoologie et d'Ecophysiologie, Bat 13, Campus de Beaulieu, Université de
Rennes I, 35042 Rennes Cedex, France; alexia.blanc@univ-rennes1.fr
ABSTRACT
As is typical of cepinalopods, the cuttlefish Sepia officinalis L. is an opportunist predator. This
study shows that Sepia selects the length of its prey taking into account its own size. In the nat-
ural environment, the most important prey are fishes and crabs. For each group, we have linked
the prey remains (found regularly in the stomach contents of the cuttlefish) to total prey length.
Prey remains consist of the merus of pereiopods 2 to 5 for crabs and otoliths for fishes. These
pieces are not attacked by enzymes during their transit through the digestive tract.
Fishes become prey when they measure between 25-80% of the DML (Dorsal Mantle Length)
of the cephalopod. The cuttlefish attack crabs when they measure between 20-40% of their own
DML. Cuttlefish attack crabs by jumping or pouncing without using their tentacles.
INTRODUCTION
Throughout its geographical distribution,
the diet of the cuttlefish Sepia officinalis L.
consists of fishes and crustaceans (Brachy-
ura and Macrura): in the English Channel
(Richard, 1 971 ), in the Bay of Biscay (Pinczon
du Sel, 1 996), on the Spanish coast (Castro &
Guerra, 1989), on the Italian coast (Scalera
Liaci & Piscitelli, 1982), and on the Tunisian
coast (Najai & Ktari, 1 979). In the Ria de Vigo,
Guerra (1 985) showed that the composition of
the diet of Sepia officinalis changes with
growth (Amphipoda and Caridea for individu-
als with a DML less than 45 mm, and Porcel-
lanidae, Brachyura, and fishes when the DML
is above 55 mm). Castro & Guerra (1990) re-
marked that in the diet of Sepia officinalis the
proportion of crustaceans decreases whereas
that of fishes increases with cuttlefish growth.
However, the decrease of crustaceans in gen-
eral did not modify the importance of Brachy-
ura for larger cuttlefish.
Only a few studies have been carried out to
determine the relationships between the
length of the prey and the length of the cuttle-
fish. Moreover, these studies have been car-
ried out only in the laboratory (Boulet, 1964;
Duval etal., 1984).
Fishes and Brachyura were identified re-
spectively by the otoliths and principally by
the periopods or walking legs. We next linked
these to the size of the prey. Thus, we can
begin to elucidate the relationship between
length or size of the predator to its prey.
MATERIALS AND METHODS
All cuttlefish examined for this study were
collected by 2 trawls (10 x 3 x 0.6 m with 20
mm mesh size and 3.5 x 1.2 x 0.4 m with 5
mm mesh size to capture young) in the Bay of
Biscay from June 1 990 to July 1 992, thus tak-
ing into account all the stages between hatch-
ing and death. Only two Sepia were under 10
cm, 19 between 10-14 cm, 12 between 15-
19 cm, 15 between 20-24 cm, and 12 Sepia
greater than 25 cm. The prey items was iden-
tified under binocular microscope. The stom-
ach contents were examined in all 60 cuttle-
fish.
Crabs parts were present in 28 and fishes
in 32 of the cuttlefish studied. For fishes and
crabs, when the weight of the stomach con-
tents is significant there is only one type of
prey present in the stomach (Pinczon du Sel,
1996). In five cuttlefish, there were two prey
items in the stomach (two fishes or fishes with
Brachyura not Identified).
Carcinus maenas L. (Crustacea: Brachyura)
In the Bay of Biscay, the most important
Brachyura taken by Sepia officinalis is Carci-
nus maenas L. (Pinczon du Sel, 1996). To ap-
139
140
BLANC, PINCZON DU SEL & DAGUZAN
preciate the size range of the captured crabs,
a total of 100 crabs were measured and
weighed. References to size are based on the
maximum length and width of the carapace
and maximum width of the merus of perei-
opods 2 to 5 (Fig. 1 ). This portion of the walk-
ing legs was often found in the stomach con-
tents. In a laboratory study, cuttlefish eat only
52.5% of the crabs and do not consume the
legs (Pinczon du Sel, 1996). All the measure-
ments were made to the nearest 0.1 mm, and
weights to the nearest 0.1 g.
The merus of the walking legs identified in
the stomach contents were measured. Then,
the width of the crab was estimated from the
allometric relation as well as the surface area
of the céphalothorax. When several merus
were present in the same gut content, the
pairs were reconstituted by minimizing the
variations. The size used to calculate crab
length was the mean width of each pair. Next,
the width of the céphalothorax was linked to
the size of the merus (Fig. 2) and to the sur-
face area of the céphalothorax (surface area
was calculated by length times width) (Fig. 3).
Duval et al. (1 984) studied the method used
by cuttlefish to capture crabs in relation to the
surface area of the céphalothorax. The com-
parison index is R = weight of the cuttlefish
(g) / surface of the céphalothorax of the crab
(mm^). Duval et al. (1984) noted that when R
was greater than 1 (i.e., the crab was "small"),
the crab was captured using the tentacules.
However, when R was less than 1 (i.e., the
crab was large), capture was done by jumping
or pouncing without using the tentacules (data
not submitted).
Fishes
A diversity of fishes constitute the second
most important prey for cuttlefish. The re-
mains of fish prey are identified in stomach
contents by the presence of otoliths, bones of
the inner ear, shape. Moreover, as is typical of
bones, there is an allomethc relationship be-
tween the length of the otolith and the total
length (Lt: from head to caudal fork) and
weight of the fish.
This study was carried out on seven
species of fishes, which represented the most
important elements found in the stomach con-
tents (Table 1, Fig. 4). We compare otoliths
found in stomach contents with a collection of
otoliths made for several fish species of the
Bay of Biscay.
Before using the allometric relationship be-
FIG. 1. Measurements made on the crab Carcinus
maenas L. (scale bar = 2 cm). LC, length of the
céphalothorax; WC, width of the céphalothorax;
WM, width of the merus of P3 (pereiopods 3) (draw-
ing from Christiansen, 1969)
0.2 0.3 0.4 0.5 0.6
width of crab merus (cm)
FIG. 2. Carcinus maenas L. Allometric relationship
between width of the merus of pereiopods 2 to 5
and width of the céphalothorax. Y = 13.082 x +
1.040; r = 0.965
tween length of otolith and length of fish, we
had to be sure that the otoliths were not
eroded by digestive enzymes during their
passage through the cuttlefish digestive tract.
As a control, fishes were weighed, measured
and the length of the otolith estimated and fed
to cuttlefish. When otoliths were recovered in
the faeces, the two lengths could be com-
pared.
All otoliths found in the cuttlefish stomach
content were identified and measured to the
nearest 0.01 mm. When several otoliths were
present in the same stomach, they were all
measured and the pairs were reconstituted by
minimizing the variations. Then, the allometric
CUTTLEFISH PREY/PREDATOR-LENGTH RELATIONSHIPS
141
width (cm)
FIG. 3. Carcinus maenas L, Allometric relationship
between width of the céphalothorax and its surface
area. Y = 0.739 x 2.003; r = 0.995
relations allowed us to estimate the length
and weight of the fishes caught by Sepia.
RESULTS
Of the 60 cuttlefish examined, 28 contained
crabs parts in the stomach. The cuttlefish that
contained crab pereiopods ranged in size
1 1 -29 cm DML and 200-2200 g in total body
net weight.
The ratios of prey to predator sizes (width of
the céphalothorax of prey to DML of predator
= WC/DML) ranged from 20% about 35% of
the predator DML (Fig. 5). Thus, the lengths of
captured crabs increase as the length of the
predator increases. Therefore, there was a
slight trend for crab size to decline as cuttle-
fish size increases. The size range of wild
crabs was between 30-80 mm in carapace
width. As the cuttlefish grow, if they do not
have enough large crabs, they eat smaller
ones. The ratios of weights range are be-
tween 0.1 -4.0% of the weight of the cuttlefish,
with a greater inter-individual variation. Some
adult cuttlefish (w = 1 .875 kg) eat small crabs
(w = 38.04 g) and vice versa (Fig. 6).
The ratio R (weight of cuttlefish to surface
area of crab céphalothorax) is given in the
Table 2. The majority of the ratios are less
than 1 and the crabs are considered to be
"large".
The size range of cuttlefish with otoliths in
their gut was between 4 cm to 29 cm, with a
weight between 1 8 g to 2025 g. The results of
our study on the digestion of otoliths indicates
that they are not attacked by the cuttlefishes'
digestive enzymes (Table 3). Thus, we can
use the allometric relationship between length
of otolith and length of fish to estimate the size
of the captured fish.
The majority of the size ratios (length of fish
to DML of cuttlefish) are between 25-80%
(Fig. 7) in Morbihan Bay (breeding area) as
well as in the Morbraz (growth area). These
are both areas of the Bay of Biscay. Only
three cuttlefish had otoliths in their stomach
contents, indicating fishes with length ratios
greater than 80% of the DML. Moreover, we
can see that among these three cases, two
show a ratio greater than 100% (116 and
120%). These three fish belong to the same
species with a greatly elongate shape, the eel
Anguilla anguilla L.
The majority of cuttlefish (73%) eat fish with
an estimated weight of less than 15 g (Fig. 8).
Sepia seems to be more selective of fishes
than crabs. Anguillidae, Gadidae and Labri-
dae eaten were greater than 15 g, Gobiidae
and Atherinidae less than 1 5 g.
DISCUSSION
Although the cuttlefish. Sepia officinalis is
opportunist in its choice of prey-species (Najai
& Ktari, 1979), it seems to be more selective
when it comes to prey size. Size ratios for
fishes ranges from 25-80% of the DML of the
predator and between 20-40% for the crabs.
Sepia officinalis L. can also seize its prey with
its arms by jumping on it (Wilson, 1946; Mes-
senger, 1968).
The prey's capacity to defend or to escape
capture increases with prey size; larger prey
are also more difficult for the cuttlefish to at-
tack and capture successfully. In Sepia ele-
gans, there was a relationship between tenta-
cle club length and body size. Individual
cuttlefish equipped with comparatively longer
clubs can capture larger prey (Bello, 1991).
Boulet (1964) observed that the ratio be-
tween the size of the crab and of the cuttlefish
may inhibit predatory behaviour of the cepha-
lopod; if the crab is too big the cuttlefish will
not attempt to capture it.
A study on the energy expended in prey cap-
ture and the energy gained from ingestion and
assimilation could perhaps explain the larger
size limit. In the case of Octopus vulgaris, the
stimulus for prédation is controlled by the
repletion ratio of the crop (number of full stom-
achs of the octopus to number of total number
of stomach contents examined (Young, 1964;
142
BLANC, PINCZON DU SEL& DAGUZAN
TABLE 1 . Allometric relationship between otolith length and the length and weight of several species of fish
prey.
Family
Species
L(mm) = f(Lo(mm))
r
n
References
0.94
25
Monaix, 1992
0.92
58
Pinczon du Sel, 1996
0.99
30
La Мао, 1985
0.88
28
Pinczon du Sel, 1996
0.99
32
Pinczon du Sel, 1996
Anguillidae
Gadidae
Atherinidae
Callonymidae
Gobiidae
Anguilla anguilla L.
Trisopterus lu se и s L.
Atherina presbyter Cuvier
Callyonymus lyra L.
G obi и s s p./
Pomatochistus sp.
L= 186.73LO- 85.35
L = 31 .877LO- 89.924
L = 31.91Lo-0.35
L = 87.323Lo- 72.783
L = 29.461 Lo- 9.238
Lo = length of the otolith (see also Fig. 4)
CaLLIONWIIDAE'S OTOLITH
Atherimdaf's otolith
о I о
FIG. 4. Measurements made on fish otoliths. Lo, length of otolith.
40%-
■
■
30%-
■
■
■
■
■ ■
Ш
■
20%-
Ш
1 ■
■
■
Ш
»--
..Л.
Ш
10%-
1
1
1
1
10
15
20
25
3(
200-
cuttlefish DML (cm)
FIG. 5. Ratio of the "width of the céphalothorax of
the captured crab ^ DML of the cuttlefish" plotted
against cuttlefish (predator) DML.
g, 150-1
0)
g 100-
"О
Е 50-
IxfS-?
.#
cuttlefish weigth (g)
.^
.#
FIG. 6. Relationship between cuttlefish weight and
estimated crab weight.
CUTTLEFISH PREY/PREDATOR-LENGTH RELATIONSHIPS
143
TABLE 2. Estimates of the surface area of the céphalothorax of crabs captured by cuttlefish in their natural
environment and the calculation of the ratio R.
Sepia
officinalis
Carcinus maenas
DML
Weight
Width of the
Carapace surface
Crab
Ratio
(cm)
g)
carapace (mm)
area (mm^)
weight (g)
R
11
200
29.8
632.9
1.74
0.32
13
250
33.3
796.3
3.26
0.31
14
372
40.6
1178.2
8.86
0.32
15
325
51.6
1888.3
27.11
0.17
17
400
62.1
2716.6
61.33
0.15
18
525
43.8
1364.3
12.66
0.38
20
1078
73.2
3758.5
123.84
0.29
21
840
56.2
2232.5
39.65
0.38
22
1100
69.3
3371.8
98.13
0.33
23
1246
36.6
957.7
5.27
1.30
24
1100
55.0
2181.6
37.64
0.50
24
1413
58.8
2441.9
48.44
0.58
25
1220
52.9
1983.7
30.35
0.62
25
1265
78.6
4320.4
166.43
0.29
26
1560
65.3
3005.8
76.52
0.52
10
145
30.4
666.3
2.01
0.22
13
225
43.8
1364.3
12.66
0.16
13
254
34.9
871.9
4.16
0.29
14
325
30.9
689.1
2.20
0.47
15
325
40.4
1163.3
8.59
0.28
16
450
53.3
2012.8
31.37
0.22
18
600
50.8
1832.1
25.30
0.33
23
1157
67.3
3186.2
86.84
0.36
24
1375
64.2
2900.1
70.77
0.47
25
1344
46.4
1529.6
16.62
0.88
25
1550
62.3
2739.3
62.45
0.57
28
2200
81.4
3409.6
193.20
0.64
29
1875
55.7
2191.7
38.04
0.86
R = weight of the cuttlefish (g). surface of the céphalothorax (mm ) (Duval et al., 1984).
Nixon, 1965, 1966). Mather (1980) obtained
the same results with a second species of oc-
topus, Octopus joubinii.
There is another hypothesis that must also
be taken into account in the attack and cap-
ture of larger prey. This relates to the effi-
ciency of the toxic secretions from the poste-
rior salivary glands of Sepia on larger fishes
or crabs. The injection of a toxic saliva pro-
vokes the rapid paralysis of the crab (Chich-
ery & Chichery, 1992).
For each type of prey (crabs and fishes), the
length ratios remain constant in relation to the
DML of the predator. The anatomical charac-
teristics of the digestive tract constrains the
ability of the cuttlefish to break down prey tis-
sue into particles that can pass through the oe-
sophagus where it penetrates the central ner-
vous system. Guerra et al. (1 988) showed that
the antero-postehor length of the buccal mass
increases in size as the oesophagus grows,
until the cuttlefish reaches a DML of 150 mm.
This implies that the size of ingested particles
is limited by these two parameters. In other
words, the larger the prey is, the more the buc-
cal mass -the beaks and radula-must work
to reduce the prey before ingestion take place.
Fishes appear to be preferentially captured
by the tentacle strategy, whereas crabs are
captured by both strategies (Messenger,
1968). The tentacle strategy was often used
for smaller crabs (Chichery & Chichery, 1 991 ).
Messenger (1968) demonstrated in a labo-
ratory study that crabs can be attacked by two
methods -a jump or tentacular strike. Duval
et al. (1984) indicated that the attack behav-
iour is dependent on the ratio R (weight of the
cuttlefish/surface area of the crabs' céphalo-
thorax). Estimates based on stomach con-
tents show that the majority of crabs are cap-
tured by the jump method. Control of the two
first parts of the attack (attention and position-
ing) is a visually controlled loop system (Mes-
senger, 1968). The prey and its movements
144
BLANC, PINCZON DU SEL & DAGUZAN
TABLE 3. Comparison of estimated otolith lengtlns after digestion of thie fisfi prey by tfie cuttlefisfi (n = 14)
Total length
Otolith length
Otolith length
Teleostan
of the
estimated
following
family
Species
fish (mm)
(mm)
digestion (mm)
Gadidae
Trisopterus luscus
200
9.29
8.70
Gadidae
Trisopterus luscus
180
8.47
8.60*
Gadidae
Trisopterus luscus
215
9.57
9.80*
Gadidae
Trisopterus luscus
230
10.04
10.60*
Gadidae
Trisopterus luscus
94
5.77
5.30
Callionymidae
Callionymus lyra
205
3.18
3.55*
Callionymidae
Callionymus lyra
230
3.47
3.41
Callionymidae
Callionymus lyra
185
2.95
3.02*
Callionymidae
Callionymus lyra
255
3.75
3.55
Callionymidae
Callionymus lyra
275
3.98
3.58
Gobiidae
Gobius sp. or Pomatoschistus
sp.
91
3.40
3.55*
Gobiidae
Gobius sp. or Pomatoschistus
sp.
73
2.79
2.78
Gobiidae
Gobius sp. or Pomatoschistus
sp.
103
3.81
3.71
Gobiidae
Gobius sp. or Pomatoschistus
sp.
54
2.15
2.14
Gobiidae
Gobius sp. or Pomatoschistus
sp.
89
3.33
3.30
Gobiidae
Gobius sp. or Pomatoschistus
sp.
102
3.78
3.30
Mean ± sd
4.98 ± 2.72
4.96 ± 2.77
*Some data on otolith length following digestion were greater than these of estimated otolith length. The latter came from the
equations given in Table 1 . There were also variations.
10 15 20
cuttlefish DML (cm)
FIG. 7. Ratio of the "length of the captured fish ^
DML of the cuttlefish" plotted against DML cuttlefish
(predator) DML.
are identified by the cuttlefish before begin-
ning an attack. The possibility of identification
of the prey in terms of type and size, possibil
with an accurate spatial location, increases
the rate of attack success. The tentacle strat-
egy is the best adapted for prey that possess
a rapid escape response (Duval et al., 1984;
Chichery & Chichery, 1992).
Hurley (1976) noted that the young squid
Loligo opalescens showed an absence of se-
lectivity for prey length. But this author also
noticed that, although success of the hunt de-
60 П
B,
Ш
3
Où
1
50-
40-
В
■
Ш
E
30-
20-
BW
■
■ E
шЕ
10-
и
Л
OH
■^ ■
— ^—
V-
— 1
s^ Ф* г^
cuttlefish weight (g)
FIG. 8. Relationship between cuttlefish weight and
estimated fish weight.
E: eels G: gobies
B; bibs S: sand smelts
BW: bailan wrasses
pended on prey size. Other factors have to be
taken into account to explain the variations in
behaviour ^the prey species hunted, and the
age, experience, and motivation of the preda-
tor.
Otoliths are routenely used to identify fishes
and to estimate their sizes. Digestive en-
zymes do not attack these hard structures.
CUTTLEFISH PREY/PREDATOR-LENGTH RELATIONSHIPS
145
Jobling & Breiby (1986) noticed the same re-
sult with the squid Todarodes sagittatus. They
estimated that the pH (5.2 to 6.3) of the di-
gestive tract of oceanic squid was not acidic
enough to dissolve these structures.
The only fish that is attacked even if the
length ratio is unfavourable to the predator is
the eel. Our observations in the laboratory re-
vealed that the tentacular method undergoes
some modifications. In these case, the head
of the eel is not targeted first. The first bite is
made on the spinal column. Subsequently,
manipulation to bring the head of the fish to
the mouth can be done because risks of flight
have been minimized.
ACKNOWLEDGEMENTS
We are grateful to Dr. A. Leroux for his ad-
vice and L. Allano for his technical assistance.
We also thank the two referees for their com-
ments and advice.
LITERATURE CITED
BELLO G., 1991, Relationship between tentacle
club length and body size in Sepia elegans. pp.
92-97, in: E. boucaud-camou, ed., Acta of the
first international symposium on the cuttlefish
Sepia, Centre de Publications de l'Université de
Caen.
BOULET, P. С, 1964, La prédation chez la seiche.
Actualités l\/lannes. 8(2): 26-32.
CASTRO, B. G. & A. GUERRA, 1989, Feeding pat-
tern of Sepia officinalis (Cephalopoda: Sepi-
oidea) in the Ria de Vigo (NW Spain). Journal of
Marine Biological Association of the United King-
dom. 69: 545-553.
CASTRO, B. G. & A. GUERRA, 1990, The diet of
Sepia officinalis L. and Sepia elegans Blainville
1827 (Cephalopoda, Sepioidea) from the Ria de
Vigo (NW Spain). Sciencia Marina. 54(4): 375-
388.
CHICHERY, R. & M. P CHICHERY, 1991, The
predatory behaviour of Sepia officinalis, pp. 233-
239, in: E. BOUCAUD-CAMOU, ed.. Acta of the first
international symposium on the cuttlefish Sepia.
Centre de Publications de l'Université de Caen.
CHICHERY R. & M. P CHICHERY, 1992, Learning
performances and aging in cuttlefish {Sepia offic-
inalis). Experimental Gerontology. 27: 233-239.
CHRISTIANSEN, M. E., 1969. Decapoda Brachy-
ura. Marine Invertebrates of Scandinavia n'2,
Scandinavia University Books, Universtets for-
laget, Oslo, 143 pp.
DUVAL, P, M. P CHICHERY & R. CHICHERY,
1984, Prey capture by the cuttlefish (Sepia offici-
nalis): an experimental study of two strategies.
Behavioural Processes, 9: 13-21.
GUERRA, A. 1985. Food of the cuttlefish Sepia of-
ficinalis and Sepia elegans in the Ria de Vigo
(NW Spain) (Mollusca: Cephalopoda). Journal of
Zoo/ogfy, 207:511-519.
GUERRA, A., M. NIXON & B. G. CASTRO, 1988,
Initial stages of food ingestion by Sepia officinalis
(Mollusca: Cephalopoda). Journal of Zoology
214: 189-197.
HURLEY, A. C, 1976, Feeding behavior, food con-
sumption, growth and respiration of the squid
Loligo opalescens raised in the laboratory. Fish-
ery Bulletin, 74{^)■. 176-182.
JOBLING, M. & A. BREIBY 1986, The use and
abuse of fish otoliths in studies of feeding habits
of marine piscivores. Scarcia, 71 : 265-274.
LA MAO, P., 1985, Peuplements piscicole et
teuthologique du bassin maritime de la Ranee:
Impact de l'aménagement marémoteur Thèse de
Doctorat de 3ème cycle. ENSA Rennes I. 1 25 pp.
MATHER, J. A., 1 980, Some aspects of food intake
in Octopus joubini. The Veliger 22(3): 286-290.
MESSENGER, J. В., 1968, The visual attack of the
cuttlefish Sepia officinalis. Animal Behaviour, 16:
342-357.
MOUNAIX, В., 1992, Intercalibration et validation
des méthodes d'estimations de l'âge de l'anguille
européenne (Anguilla anguilla L.). Application au
bassin versant de la Vilaine, Bretagne. Publica-
tions du Département d'Halieutique, ENSA
Rennes, 14: 1-150.
NAJAI, S. & M. H. KTARI, 1 979, Etude du régime al-
imentaire de la seiche commune Sepia officinalis
L. du Golfe de Tunis. Bulletin Institut National Sci-
ences Techniques Océanographique Pêche,
Salammbô, Tunisie, 6(1 -4): 53-61 .
NIXON, M., 1965, Some observation on food intake
and learning in Octopus vulgaris. Journal of Zool-
ogy 34: 329-339.
NIXON, M., 1966, Changes in body weight and in-
take of food by Octopus vulgaris. Journal of Zool-
ogy 150: 1-9.
PINCZON DU SEL, G., 1996. Régime alimentaire
de la seiche Sepia officinalis L. (Mollusque,
Céphalopode, Sepiidae) dans le secteur Bre-
tagne Sud. Thèse d'Université Rennes I, 270p.
RICHARD, A., 1971, Contribution à l'étude expéri-
mentale de la croissance et de la maturation sex-
uelle chez le céphalopode Sepia officinalis L.
(Mollusque, Céphalopode), Thèse Doctorat d'E-
tat, Univ Lille, 264 p.
SCALERA LIACI, L. & G. PISCITELLI, 1982, Ali-
mentazione di Sepia officinalis L. nella laguna di
Lesina. Bollettino del Musei Istiti Biologici dell'
Universita di Genova, 50 (Suppl.): 398.
YOUNG, J. Z., 1964, A model of the brain. Oxford:
Clarendon Press.
WILSON, D. P., 1 946, A note on the capture of prey
by Sepia officinalis. Journal of the Marine Biolog-
ical Association of the United Kingdom, 26: 421 -
425.
Revised ms. accepted 14 October 1998
MALACOLOGIA, 1999, 41(1): 147-150
THE ACCUMULATION OF TAXONOMIC KNOWLEDGE: THE HISTORY OF
SPECIES DESCRIPTIONS OF SOME PREDATORY GASTROPODS
Geerat J. Vermeij
Department of Geology and Center for Population Biology. University of California at Davis
One Sfiields Avenue Davis, California, 95616 USA; vermeij(g)geology ucdavis.edu
ABSTRACT
In order to assess how perceived temporal and spatial patterns of distribution of species de-
pend on taxonomic knowledge, I compiled dates of description of all 558 known gastropod
species with shells having a labral tooth. Peaks in the number of described species were reached
in the intervals of 1 820- 1 859 and 1 980- 1 999 for Recent species, and 1 920- 1 939 for fossil taxa.
Distributional patterns were already evident by 1859, when about half of the living and 13% of
the fossil species had been described; but geographical and temporal ranges are very sensitive
to accumulated taxonomic knowledge, and have changed substantially as more species were
described. An asymptote in the number of species has not yet been approached either in these
gastropods or in molluscs and other organisms generally. This fact underscores the importance
of continuing support for taxonomic research.
Key words: taxonomy, gastropods, labral tooth, history.
INTRODUCTION
For some years I have been studying the
functional morphology, taxonomy, phylogeny,
and distribution in time and space of predatory
gastropods with a labral tooth, a sharp, ven-
trally and sometimes anteriorly directed pro-
trusion of the abapical or medial sector of the
shell's outer lip. In all studies of this kind, ques-
tions arise about the completeness of taxo-
nomic data, and about the robustness of the
patterns that those data reveal. One approach
to tackling these questions is to document the
history of description of species since the be-
ginnings of formal zoological taxonomy in
1758. If the number of species described per
time interval falls off, we may infer that an as-
ymptotic number is being approached, which
would mean either that we are close to dis-
covering all existing or knowable species, or
that taxonomic interest is waning. This latter
possibility is potentially a matter of concern,
because it affects directly the amount and
quality of fundamental data on which most
studies of ecology, evolution, biogeography,
stratigraphy, and many other fields depend.
Here I report the results of my compilation
of the times of description of all fossil and
living gastropod species with a labral tooth
known to me. Although these gastropods rep-
resent a tiny and potentially unrepresentative
fraction of gastropods as a whole, to say noth-
ing of living things generally, they reveal pat-
terns of description that are both surprising
and informative.
METHODS
I have assembled a list of all fossil and liv-
ing gastropod species with a labral tooth. A
species was included if I observed a labral
tooth on specimens, or if authors describing
or discussing the species in question clearly
indicated the presence of a labral tooth. In a
large number of cases, the original describers
were unaware of, or failed to record, the pres-
ence or characteristics of the labral tooth. My
own examination of material in museum col-
lections therefore revealed many cases that
would not have come to light from a search of
the literature alone.
For each species, I used the earliest pub-
lished name, even if that name is not the cur-
rently accepted one because of homonymy.
Names that on the basis of my studies or tax-
onomic revisions by other authors appear to
be synonyms were not included.
I grouped dates of publication into 20-year
time intervals. For each interval, I compiled
separately the number of living and fossil
species.
147
148
VERMEIJ
RESULTS AND DISCUSSION
My current tally shows that 558 species (21 9
Recent and 339 fossil) have a labral tooth.
These species range in age from Late Creta-
ceous (Campanian) to Recent, and belong to
ten families. These include the tonnoidean
Ranellidae, and the neogastropod families
Muhcidae, Buccinidae, Melongenidae, Nas-
sariidae, Fasciolariidae, Pseudolividae, Tur-
binellidae, Lividae, and Marginellidae.
The temporal pattern of description of Re-
cent species differs from that of fossil ones
(Fig. 1). For Recent species, there was a
broad peak from 1820 to 1859, and a very
conspicuous maximum in the present 20-year
interval, 1980-1999. The first half of the 20th
century (1900-1959) represents a broad
trough in the number of described Recent
species. The fossil data reveal a general rise
in the number of described species per inter-
val to a peak in the years 1920-1939, fol-
lowed by more than half a century of stability
at a high level (about 30 species per 20-year
Interval) of taxonomic activity.
A striking result of this study is that the most
recent time interval (1980-1999) has con-
tributed a sizable fraction of our accumulated
taxonomic knowledge. Of the 219 Recent
species, 48 (21.9%) have been described
during this interval. Among the 339 fossil spe-
cies, the corresponding number is 34 (9.4%).
For Recent species, taxonomic knowledge
has doubled since 1 860; for fossil species, the
number has doubled since 1920.
As Sepkoski et al. (1981) and Sepkoski
(1993) have pointed out in their studies of
the number of families and genera of marine
fossil invertebrates, increases in taxonomic
knowledge do not necessarily require revision
in our perception of temporal and spatial pat-
terns of distribution of species. A general in-
crease in the number of taxa over geological
time, and the existence of a latitudinal in-
crease in diversity from the poles to the equa-
tor, are robust patterns that were discernible
by 1859, the date of publication of Charles
Darwin's Origin of Species. Similarly, the
broad outlines of distribution of gastropods
with a labral tooth were already apparent by
this date when, according to my data, about
half the living species with a labral tooth and
13% of the presently known fossil taxa had
been named. By the middle of the 19th cen-
tury, observers able to examine specimens
could already have concluded that the inci-
dence of labral teeth is higher among living
predatory gastropods on the Pacific than on
the Atlantic side of the Americas, and that the
incidence in Europe declined from the Mio-
cene to the Recent.
Other details of distribution, especially first
appearances and unusual geographical
records, are very sensitive to accumulated
knowledge. For example, only one of 30 Cre-
taceous gastropods with a labral tooth -Sl/c-
cinopsis parryi Conrad, 1 857 -was described
before 1859, and this species was based on
material so poorly preserved that no contem-
porary scholar could have inferred the pres-
ence of a labral tooth. Muricids with a labral
tooth from the Oligocène, the earliest epoch
from which this structure is known in this fam-
ily, were not described until 1 91 8. In that year,
Clark described Thais pacl<i, but its labral
tooth was not recognized until 1993 (Amano
et al., 1 993). A second Oligocène muricid with
a labral tooth was described by Vokes (1963),
but its age was originally thought to be early
Miocene. Until about 1970, therefore, taxono-
mists might legitimately have concluded that
muricids with a labral tooth did not originate
until early Miocene time.
The continuing high rate of discovery of
species with a labral tooth mirrors a general
pattern for molluscs (Bouchet, 1997) and for
organisms generally (Winston & Metzger,
1 998). This is surprising in view of the fact that
most labral-tooth-bearing gastropods are rel-
atively large (greater than 2 cm in height) and
thus more easily collected than are micromol-
luscs.
The fossil data in Figure 1 are likely to be
less representative for the description of fossil
taxa in general than are the data for Recent
species. Large parts of the world remained
essentially unexplored paleontologically until
the present century. This is especially true for
older rocks and for small fossils. Despite this,
the vigor of descriptive paleontological activity
in the latest 20-year interval as indicated in
Figure 1 is noteworthy.
The early peak (1820-1859) in taxonomic
description of Recent gastropods with a labral
tooth probably reflects the collective efforts of
major scientific expeditions throughout the
world. Many of the common shallow-water
taxa became known during this time. The
1980-1999 maximum, preceded by a rise in
taxonomic activity that began in the 1960s,
likely represents the efforts of divers and of
expanded trawling and dredging operations. It
TAXONOMIC DESCRIPTION THROUGH TIME
149
1740-1759
17604779
17804799
18004819
19804999
10 20 30 40 50 60 70
Number of species described
FIG. 1 . Number of gastropod species with a labra! tooth described per 20-year interval, 1 740-present.
150
VERMEIJ
also demonstrates unequivocally that interest
in molluscan taxonomy is as strong as ever.
The pattern of description of fossil species
probably reflects exploratory activities by ge-
ologists and paleontologists working in the oil
industry or for other commercial interests. It is
noteworthy that, whereas the number of de-
scribed Recent species reached a broad low
in the decades during and between the great
world wars, the fossil data show no such ap-
parent effect of worldwide conflict. There is no
dramatic upsurge in the number of described
fossil species in the most recent time interval
as there is for Recent species, but neither is
there the decline that might be expected if tax-
onomic expertise and the number of working
taxonomists were declining.
In the future, the rate of species description
is likely to dwindle for either or both of two rea-
sons. The first, and most worrying, is the long
feared reduction in funding for, and interest in,
taxonomy. This problem may be mitigated
somewhat, at least for shell-bearing molluscs,
by the increased participation of highly knowl-
edgeable amateurs. According to my data,
amateurs account for about one-quarter of the
authors of labral-tooth-beahng gastropod
species since 1980. The second reason is
thiat fewer species remain to be discovered.
Potentially counteracting this trend is the find-
ing that taxa previously interpreted as repre-
senting a single species in fact comprise two
or more genetically distinct species. In any
case, the limited data in Figure 1 give no indi-
cation that either taxonomic activity or the rate
of discovery of species id declining or reach-
ing asymptotic values.
The fact that taxonomic data are funda-
mental to all branches of comparative and his-
torical biology and that large numbers of taxa
remain to be described provides a strong ar-
gument for continued support of what many
observers disparage as mundane, descriptive
science. Phylogenetic and ecological studies
are only as good as their underlying data and
assumptions. Continued research and publi-
cation on taxonomy should remain an impor-
tant component of the study of the history and
distribution of living things.
ACKNOWLEDGMENTS
This research was funded by Grant 97-
06749 from the National Science Foundation.
I thank Janice Cooper, Janice Fong, and Mary
Graziöse for technical assistance.
LITERATURE CITED
AMANO, К., G. J. VERMEIJ & К. MARITA, 1993,
Early evolution and distribution of the gastropod
genus Nucella, with special reference to Miocene
species from Japan. Transactions and Proceed-
ings of the Palaeontological Society of Japan,
(n.s.)171:237-248.
BOUCHET P., 1 997, Inventorying the molluscan di-
versity of the world: what is our rate of progress?
Ttie Veligen 40Л -^^ .
CLARK, B. L., 1918, The San Lorenzo Series of
middle California: a stratigraphie and palaeonto-
logic study of the San Lorenzo Oligocène series
of the general region of Mount Diablo, California.
University of California Publications. Bulletin of
ttie Department of Geology. 1 1 :45-234.
CONRAD, T. A., 1857, Descriptions of Cretaceous
and Tertiary fossils. Pp. 141-174, in w. н. emory,
ed., Reports on the United States and Mexican
boundary survey, made under the direction of the
Secretary of the Interior. Volume 1 , Part 2. Wash-
ington, D.C.
SEPKOSKI, J. J., JR., 1993, Ten years in the li-
brary: new data confirm paleontological patterns.
Paleobiology 19:43-51.
SEPKOSKI, J. J. JR., R. K. BAMBACH, D. M.
RAUP & J. W. VALENTINE. 1981, Phanerozoic
marine diversity and the fossil record. Nature.
293:435-437.
VOKES, E. H., 1963, Notes on Cenozoic Muricidae
from the western Atlantic region, with descrip-
tions of new taxa. Tulane Studies in Geology, 1 :
151-163.
WINSTON, J. E. & K. L. METZGER, 1998, Trends
in taxonomy revealed by the published literature.
BioScience. 28: 125-128.
Revised ms. accepted 16 December 1998
MALACOLOGIA, 1999, 41(1): 151-162
THE ROLE OF SUBSTRATUM STABILITY IN DETERMINING ZEBRA MUSSEL
LOADONUNIONIDS
Susan A. Toczylowski, R. Douglas Hunter & Lisa M. Armes
Department of Biological Sciences, Oakland University. Rochester Michigan 48309 - 4476
USA; hunter@oakland.edu
ABSTRACT
Data reported herein do not support the existence of preference for or attraction to unionids by
settling or migrating zebra mussels compared to alternative hard substrata. Despite claims and
inferences in the Dreissena literature suggesting that unionids are preferred substrata, higher
Dreissena loads on unionids compared to alternative hard substrata can be explained by mech-
anisms other than preference. Our data indicate that substratum conditions are often critical in
determining the relative zebra mussel loads that accrue on unionids. On stable and relatively
hard lake/river bottoms, zebra mussel loads on unionids tend to be similar to those on other hard
substrata. However, on bottoms mostly composed of very soft or unstable substrata, discreet
hard objects become silted-over and/or buried, hence sub-optimal for zebra mussels. Under
such conditions, unionids develop a higher load of zebra mussels due to their ability to maintain
position in relation to the sediment/water interface. We conclude that high Dreissena loads on
unionids relative to other substrata are not a matter of preference for or attraction to the union-
ids, but are the outcome of differential survival/emigration of the Dreissena due to unstable or
changing bottom conditions.
Key words: Dreissena. zebra mussel, substratum, preference, unionids, stability.
INTRODUCTION
Among the many ecological changes attrib-
uted to the zebra mussel invasion of North
American lakes, none is more readily appar-
ent than the virtual elimination of entire
unionid communities (Schloesser & Kovalak,
1991: Schloesser & Nalepa, 1994: Nalepa,
1994). This consequence is especially unfor-
tunate due to the already precarious status of
the majority of unionid taxa (Williams et al.,
1993: Steins Flack, 1997).
The common impression from examination
of unionids in lakes recently invaded by zebra
mussels is that they are highly suitable as
substrata for attachment. This often gives the
appearance that unionids are being singled
out by zebra mussels by some mechanism
of preference (Lewandowski, 1976; Mackie,
1 993: Ricciardi et al., 1 995, 1 996). One of the
difficulties in using the word preference in the
context of settling zebra mussels is that it
specifies active choice, a choice based on the
ability to discriminate among alternatives. The
underlying mechanism would of necessity in-
volve an ability to detect positive or negative
stimuli from amongst alternatives (e.g., differ-
ent kinds of substrata) and to choose based
on the perception of those qualities. With the
example of settling larvae in marine ecosys-
tems, preference is typically based on detec-
tion of chemical or textural cues originating
from the biofilm or from components of the
biofouling community (Morse, 1991; Ro-
driguez et al., 1993). However, it is possible
for a different mechanism, not involving pref-
erence, to result in clumped or aggregated
distribution of a species. In such cases, larvae
settling on both suitable and non-suitable sub-
strata would experience post-settlement dif-
ferential mortality with higher mortality among
the latter group. The resulting greater aggre-
gation on suitable substrata might give the im-
pression of preference, but in the absence of
active choice the resulting distribution would
simply be the outcome of differential mortality
on dissimilar substrata.
Although it is often taken for granted that
zebra mussels "prefer" hard substrata, they
have been reported living on soft substrata
and can achieve surprisingly high densities
(Hunter & Bailey, 1992; Dermott & Munawar,
1993; Coakley et al., 1997). Despite these re-
ports, the highest densities are on hard sub-
strata, and it is likely that they can colonize
soft substrata only from an initial seed object
151
152
TOCZYLOWSKI ETAL.
and if the bottom is undisturbed by current.
Zebra mussels are clearly epifaunal and the
morphological adaptations for this mode of
existence are obvious: presence of a byssus
and heteromyarian shell morphology (Morton
& Yonge, 1964; Yonge & Campbell, 1968).
One form of preference that has been doc-
umented for zebra mussels is settlement on
or near conspecifics. This was first suggested
by Lewandowski (1976) and supported by the
field studies of Wainman et al. (1996), Chase
& Bailey (1996), and Toczylowski & Hunter
(1 997). Although none of these studies exper-
imentally addressed the release of chemical
cues by established adults, the results of
Wainman et al. (1996) suggest that it is more
likely that conspecific shell surface chemistry
was the basis for veliger preference.
Some of the recent literature largely rein-
forces the perception of unionid preference by
D. polymorpha, without specifically identifying
the mechanism by which preferred substrata
are recognized (Ricciardi et al., 1 995, 1 996). A
previous study (Toczylowski & Hunter, 1997)
found no evidence to support zebra mussel at-
traction to unionids based on field studies of
Dreissena seWWng on unionid and non-unionid
test surfaces. The present work extends and
refines those findings and examines the ef-
fects of substratum stability on differential mor-
tality of zebra mussels attached to stones as
compared to unionids.
We hypothesize that zebra mussels do not
actively choose unionids over other hard sub-
strata. Higher Dreissena density on unionids
in relation to other hard substrata, where it oc-
curs, is the outcome of differential mortality re-
sulting from bottom instability.
EXPERIMENTAL DESIGN AND METHODS
Site Descriptions
All of the field sites were in the upper Clin-
ton River or in Loon Lake, a lake connected to
the Clinton River located in Oakland County in
southeastern Michigan (Fig. 1). The upper
mainstem of the Clinton River between Loon
Lake and Dawson's Millpond Outlet (DM0) is
mostly less than 10 m in width, with discharge
rates averaging 1.42 and 1.75 m^/s for the
1995 and 1996 calendar years respectively
(Blumer et al., 1997). The range in 1996 was
0.28 - 15.01 m^/s. The dominant unionid
species in this region of the river, Ptycho-
branchus fasciolaris (23%), Elliptio dilatata
(21%), and Strophitus undulatus (15%), are
indicative of a hydrological tendency toward a
stable river as opposed to an event river
(Hunter et al., 1 997; DiMaio & Corkum, 1 995).
The water is relatively clear with the silt load
sufficiently low that one can see the bottom in
most places. Mean seston load for 1995 from
12 monthly measurements was 8.7 mg/L dry
mass (range = 0.7 - 37.4 mg/L) and values for
1 996 were similar (Hunter et al., 1 997). In this
stretch of the river, the watershed is mostly
residential with a few small towns and no
heavy industry. The river bottom is mostly
hard consisting of cobble and/or sand with
soft silt elsewhere.
Loon Lake is a mesotrophic lake with an
area of 0.95 km^ and maximum depth of 21 m.
The bottom is mostly soft mud except near
shore where it is mostly cobble and sand ex-
cept for a few marshy areas. Macrophyte
growth along the shore is moderate.
Natural Density of Zebra Mussels on
Unionids and Stones
To determine the field density of zebra mus-
sels on unionids and stones, samples of nat-
ural substrata were collected at DM0 on 16
October 1996 (Fig. 1). Unionids and stones
were from an area where the water was about
0.5 m deep and the bottom was partially cob-
ble and partially sand. Bottom sediments in
this area were subject to considerable move-
ment during periods of high water flow; conse-
quently hard objects, such as stones, were al-
ternately buried in sand or exposed as cobble.
At this site, the most abundant three unionids
were ElUptio dilatata. Lampsilis siliquoidea.
and Ptyctiobranchus fasciolaris.
The density measured in this instance,
refers to number or biomass of zebra mussels
per unit of unionid or stone surface area, not
per unit total bottom area. Although the latter
meaning is the standard one, our modified
density (= suitable substrate density) allows a
more direct comparison between different
types of discreet hard surfaces. Ten unionids
along with ten stones of approximately the
same exposed surface area as the unionids
were selected. Both the unionids and the
stones were carefully transported to the lab
where all attached zebra mussels were re-
moved with a scalpel under a dissecting mi-
croscope to ensure that small individuals
were included. After removal, the zebra mus-
sels were counted and their length was mea-
sured using digital calipers. For each of the
two surface types, area above the sedi-
ZEBRA MUSSEL LOAD ON UNIONIDS
153
Dawson's Mlllpond
Outlet
(DM0)
Sylvan
Lake
^Clinton
River
FIG. 1 . Map of the upper mainstem of the Clinton River in southeastern Michigan with the sample and study
sites indicated.
154
TOCZYLOWSKI ETAL.
ment/water interface was obtained using a foil
method. This method is essentially the same
as that used in Mackie (1993). Because
buried areas on the test surfaces became
darkened, this made the sediment/water in-
terface clearly visible, so the area exposed to
zebra mussel settling was easy to see. The
method involved removing attached zebra
mussels, covering the exposed area with alu-
minum foil, removing the foil, and trimming
and flattening it. After placing the foil on met-
ric graph paper, the outline was traced and
the surface area counted. Data from this
study were compared using an unpaired t test
to determine if there were significant differ-
ences between zebra mussel density on
stones and unionids. Mean values given in
the text are accompanied by standard error.
Settling and Migration Preference Study
An experimental test of preference utilized
test quadrats placed in Loon Lake and in the
Clinton River at Drayton Plains Nature Center
(DPNC; asterisks mark these locations in Fig.
1). Quadrats at Loon Lake were placed in
water of 1.2 - 1.4 m deep and ~10 m from
shore, where the bottom sediment was soft
and silty, lacking in cobble or stones. The
DPNC quadrats were placed at a depth of
about 1 .0 m at the center of a slow-flowing
stretch of the river where the bottom was also
mostly silt. Quadrats were circular and en-
closed an area of about 0.75 m^. Each
quadrat was an open-topped enclosure using
plastic garden edging as a boundary, with
about 3 cm above and about 9 cm below the
sediment, to facilitate recovery of unionids
later in the summer. Each quadrat contained
four replicates of each of the following test
surfaces; unionids with no attached zebra
mussels (= control), unionids with marked
zebra mussels, control stones, and stones
with marked zebra mussels. Although they dif-
fered slightly in size, the surface area of each
object was measured at the end of the exper-
iment and the density of attached zebra mus-
sels expressed per unit area. Average total
exposed unionid surface area was 31 cm^
and that of stones was 43 cm^.
The species composition of test unionids at
Loon Lake was Pyganodon grandis. Lampsilis
siliquoidea, and Elliptio dilatata in a ratio of
4:3:1 and at a density of 8/quadrat. Unionids
placed in Clinton River quadrats were col-
lected from the river nearby and included
Strophitus undulatus, Ptychobranchus fascio-
laris, and Pyganodon grandis in a ratio of ap-
proximately 4:2:1, also at a density of
8/quadrat.
The unionids and stones with zebra mus-
sels each had about 5 - 30 zebra mussels at
the outset of the experiment. Each of these
zebra mussels was marked with a dot of
enamel model paint; those on unionids were
marked with red paint and those on stones
were marked with blue. All unionids and
stones had an identification number painted
on them at the start of the experiment. By
marking the zebra mussels attached to the
test surfaces at the beginning (= residents), it
was possible to identify non-residents that im-
migrated onto the test surfaces during the ex-
periment.
Test surfaces were put into Loon Lake on 6
June 1996 and remained in situior 103 days.
At the Clinton River site they were put in on 1
June 1996 and were in situ for 99 days. This
period coincided with the major period of the
veliger presence and spat settling for Loon
Lake. After completion of the experiment, all
attached zebra mussels were removed and
counted. Area of the test surfaces above the
sediment/water interface was measured
using the foil method. For unionids with at-
tached zebra mussels, the surface area of the
original resident zebra mussels was included
in the total surface area. However, for the
density and biomass data, the ohginal at-
tached zebra mussels were not included.
The mean of four replicates of each of the
four test surfaces was statistically analyzed
using two-way ANOVA, with treatment and
site as the factors. To evaluate the relation-
ship between number of attached zebra mus-
sels at the start and number of immigrating
zebra mussels that attached during the study,
a regression ANOVA was used that calculated
an F-statistic using sum of squares and mean
square. This indicated how important the in-
dependent variable was in explaining the be-
havior of the dependent variable. Emigration
data was expressed as % decrease of resi-
dent mussels for plotting, but two-factor
ANOVA was done on arcsin transformed data
(p' = arcsin [sqrt p]), the results of which are
reported in Table IB.
Zebra Mussel Survival and Migration on
Unstable Substrata
To determine survival and migration of
zebra mussels on unionids and stones lo-
cated on shifting river bottom, we performed a
field experiment in the Clinton River at the
CLR site (Cooley Lake Road Bridge. Fig. 1).
ZEBRA MUSSEL LOAD ON UNIONIDS
155
Flow rate at this site is moderately fast for the
upper Clinton River, and the bottom is rela-
tively unstable. At a given point in time, most
of the bottom is sand with a few areas of cob-
ble, the boundaries of which change over the
year. Depressions in the sand are common
here, and their size and location also shift.
The test surfaces were located on the up-
stream slopes of such depressions. Test sur-
faces consisting of unionids and stones of ap-
proximately the same size were collected and
given an identification mark. The test unionids
were collected on site and included Villosa
iris, Elliptio dilátala, Pyganodon grandis,
Strophitus undulatus, Lampsilis siliquoidea,
and Ptychobranchus fasclolaris in ratios of
20:3:3:2:2:1 . Unionids were allocated to repli-
cates in approximate proportion to their nat-
ural abundance. Zebra mussels on both test
surfaces were counted and marked at the
start of the experiment, so that any unmarked
zebra mussels found at the end of the experi-
ment could be recognized as immigrants. To
ensure recovery at the conclusion of the ex-
periment, unionids were tethered to a stake in
the river bottom using nylon string attached to
one valve with cyanoacrylate glue. Stones
were not tethered, and were simply placed on
the sand bottom. Ten unionids and ten stones
were placed around each of the three repli-
cate stakes on 23 June 1 997 and remained in
situ for 78 days, after which they were re-
trieved and brought to the laboratory. In the
lab, the number of resident and non-resident
zebra mussels was recorded for each test
surface. Surface area was not measured due
to the inability to determine the exact amount
of the test surface that was above the sedi-
ment-water interface (i.e., the sediment was
not chemically reducing at this site). The data
were analyzed with an unpaired t test to de-
tect significant differences using mean num-
ber of zebra mussels per test surface at each
stake, hence N = 6 (i.e., 3 replicates x 2 treat-
ments/replicate). When the total number of
zebra mussels (residents + immigrants) was
expressed as a % change on unionids and
stones, the decimal equivalents were arcsin
transformed before using the t test; p' = arcsin
(sqrt p).
RESULTS
Natural Density of Zebra Mussels on Union-
ids and Stones
Unionids at DM0 had a zebra mussel den-
sity of 1.40/cm^, which was significantly
3.5-,
3.0
^g 2.5
"^ 2.0
0)
J3
E 1.5,
^ 1.0-
0.5-
Loon Lake
a
T
Unionid
Stone
3.5
3.0-
^c 2.5-
О
"i: 2.0-
0)
E 1.5^
^ 1.0.
0.5-
В
Clinton River at DMO
У///////////////////////////Л
Unionid
Stone
FIG. 2. Natural density of zebra rnussels (mean ±
SE) on unionids and on stones from (A) Loon Lake
and (B) Clinton River at DMO (Dawson's Millpond
Outlet). Histograms with the same letter are not sig-
nificantly different from each other.
greater than the zebra mussel density ob-
served on stones, 0.14/cm^ (Fig. 2B; t^^g^ =
5.332; P = 0.0001). There was also a signifi-
cant difference in wet biomass; unionids had
600.3 mg zebra mussels/cm^, stones had
78.0 mg/cm^ (t^^g, = 4.568; P - 0.0002).
Hence, the river results did not agree with
those from the Lake. DMO is a river site that
is frequently exposed to high current due to
the proximity of a dam used to maintain
nearby lake levels. As a result, it is a high-en-
ergy location with an unstable bottom. The re-
sults we obtained at this site contrasted with
those obtained from Loon Lake a year earlier
(Toczylowski & Hunter, 1997).
Settling Preference Experiment
All test surfaces in Loon Lake, whether
stones, unionids, with or without resident
zebra mussels, attracted a significantly
greater density of unmarked (= immigrant)
156
TOCZYLOWSKI ETAL.
(Л
3 2
E
¿ 1
0)
N
Unionid with zm
Stone with zm
-- ^
250
^ 200
E
OÎ 150
to
^ 100
E
о
CÛ 50
Lake
В
River
t^ Unionid control
Unionid with zm
11,2^
Lake
River
FIG. 3. Interaction plots for the settling preference
study. (A) Numerical density of zebra mussels in
Loon Lake and Clinton River treatments. (B) Bio-
mass density of zebra mussels in Loon Lake and
Clinton River treatments. The four treatments used
at each site are identified on the plots.
zebra mussels/cm^ than those in the Clinton
River (Fig. ЗА). The lake test surfaces aver-
aged five times as many newly recruited
zebra mussels as their river counterparts. As
used here, immigrants denotes both spat
zebra mussels that have recently settled out
of the plankton directly onto the test surface,
as well as later stage individuals that settled
elsewhere and subsequently relocated onto
the test surfaces. This latter group consists of
young-of-the-year juveniles as well as adults
that settled in previous years. No attempt was
made to separate these groups in this data
set; spat and later stages were lumped as "im-
migrants". This higher rate of accumulation of
zebra mussels on lake compared to river sub-
strata is simply the result of higher settling
rates due to higher veliger density in the lake
(Hunter et al., 1997; unpublished data for
1996).
There were also significant differences in
immigrant zebra mussel density among the
test surface types (Table 1 A). Unionids in the
lake had about a 4x higher density of immi-
grant zebra mussels than did the stones.
However, there was no such difference at the
river site; that is, both unionids and stones ap-
peared to attract similar numbers of immi-
grants. It also appeared to be of little impor-
tance whether the test surface had resident
zebra mussels or not. That is, immigration of
zebra mussels was not enhanced by the pres-
ence of previously settled (resident) con-
specifics. There was also a significant interac-
tion between the main effects (Table 1A).
Hence, zebra mussel density was responsive
to surface type at the lake site, but not at the
river site.
The biomass data showed a similar pattern
of significance as was seen in the numerical
density data above (Fig. 3B). There were
strong site effects; unionids had higher zebra
mussel loads than did stones, and presence
of resident zebra mussels had little effect
(two-factor ANOVA F
(1, 112)
= 29.6; P <
0.0001). Unionid test surfaces had a signifi-
cantly greater biomass of immigrants (ap-
proximately five times greater) than the river
unionid surfaces (F^g ^^^, = 7.3; P = 0.0002).
There was also a significant interaction be-
tween main effects (F^g ,^2) = 6.5; P = 0.0005).
Density differences between lake and river
are great for unionids but small for stones.
In order to evaluate the effects of emigra-
tion of resident zebra mussels and whether it
differed by test surface type, the percent de-
crease of marked individuals was calculated.
There was a significant difference in percent
decrease by site and surface type (Table 1 B).
Loon Lake had a larger percent decrease of
resident zebra mussels on both unionid and
stone test surfaces compared to those in the
river (Fig. 4). Stone surfaces had a signifi-
cantly greater overall percent decrease in res-
ident mussels regardless of site (Table 1 B).
However, in contrast to number and biomass
data, the interaction of the two factors was not
significant.
The number of resident zebra mussels at
the start of the study was not a good predictor
of the number of mussels that subsequently
immigrated to that surface at the lake site;
however, the river site showed the opposite
result. At Loon Lake, neither unionids (F 3^^ =
0.391 ; P = 0.537) nor stones {F^^z) = 2-"'Û5; P
= 0.162) attracted more immigrants if the
number of resident mussels at the start was
ZEBRA MUSSEL LOAD ON UNIONIDS
157
TABLE 1 . Analysis of variance tables based on settling preference data. (A) Results of a three-
factor ANOVA on number of immigrant zebra mussels per cm^. (B) Results of a two-factor
ANOVA on percent decrease (emigration) of resident zebra mussels from test settling sur-
faces.
A. Immigration: no./cm^
Source of variation
Deg
rees of freedom
Mean square
F
Main effects
site
1
122.6
29.2"
surface
1
90.4
21.6**
res/non-res
1
0.3
0.8 ns
Two-way interaction
site X surface
1
54.3
12.9*
surf X res/non-res
1
1.5
0.3 ns
res/non-res x site
1
14.1
3.4 ns
Three-way interaction
site x surface x res/non
-res
1
8.1
1.9 ns
Error
104
4.2
Total
111
6.8
*P = 0.0005; **P< 0.0001
B.
Emigration:
% decrease of residents
Source of variation
Deg
rees of freedom
Mean square
F
Main effects
site
1
1.758
14.99*
surface type
1
1.160
9.89**
Two-Way interaction
site X surface type
1
0.243
2.07 ns
Error
56
0.117
Total
59
3.278
*P = 0.0003; **P = 0.003; ns =
p>
0.05
80-
Í 60-
O 40-
(U
CO
CO
03
T3
Stone
Unionid
Lake
River
FIG. 4. Emigration of zebra mussels as % decrease
(mean ± SE) of residents from stones and unionids
by the end of the settling preference study. See
Table 1 В for statistical analysis.
greater (Fig. 5A). When unionids and stones
at the river site were analyzed, there was a
significant correlation between the two vari-
ables (unionids, F,2g) - 528.6; P = 0.0001;
stones, F(28, = 210.1 ; P = 0.0001 ; Fig. 5B). In
other words, at the river site, having more res-
ident mussels at the start significantly in-
creased the number of zebra mussels that re-
cruited to the test surfaces. However, at the
lake site, where recruitment was about an
order of magnitude higher, there was no such
relationship. The mean length of recruited
zebra mussels did not differ between surface
type (t(g) - -2.205; P = 0.696).
Zebra Mussel Survival and Migration on
Unionids and Stones on Unstable Substrata
Test unionids and stones at the start of this
experiment had virtually identical mean num-
bers of attached zebra mussels by object:
unionids, 6.48 ± 0.435 zebra mussels, and
stones, 6.45 ± 0.449 zebra mussels (t^gg) =
0.041 ; P = 0.967). After 78 days on the unsta-
ble river bottom at CLR, unionids had a signif-
icantly higher number of zebra mussels (t^^j ==
4.855; P = 0.008), averaging 8.4 ± 0.91 zebra
mussels, whereas stones averaged 1.9 ±
0.50 zebra mussels (Fig. 6A). This gain of
zebra mussels by unionids occurred for two
158
TOCZYLOWSKI ET AL.
О)
E
о
ó
О)
(О
со
Е
со
к_
JD
Ф
N
■«— •
С
СО
1_
Е
«4—
О
ó
600
500-
400-
_ço
О)
ел
СО
13
Е
со
1_
JD
0)
i^ 300 H
С
СО
Я^ 200-
Loon Lake
Unionid: Y = 127.46- 1.323Х; г = 0.113
Stone: Y = 29.38 + 2.092X; г = 0.302
5 10 15 20 25
No. of zebra mussels at start
30
В
зо-п Clinton River
Unionid: у = -0.111 +0.764X; r = 0.975 о о
XX
су^ Stone:
^ Y = -0.438 + 0.71 5X; r = 0.941
10 15 20 25 30 35
No. of zebra mussels at start
FIG. 5. Scatter plot of the number of zebra mussels immigrating to unionids and to stones at Loon Lake (A)
and the Clinton River (B) in relation to the number of resident zebra mussels on these test objects at the start
of the settling preference study. Lines are fitted by least squares regression.
ZEBRA MUSSEL LOAD ON UNIONIDS
159
10 -1
t =4.855; P = 0.008
Unionid
В
3.0
i2 2.5
с
^ 2.0-
E
Z 1.5-
1.0-
0.5-
Stone
t =3.326; P = 0.0292
Unionid
Stone
FIG. 6. (A) Total number of zebra mussels (mean ±
SE) on unionid and stone test objects at the end of
the experiment at CLR (Cooley Lake Road; unsta-
ble substratum site). (B) Number of immigrant
zebra mussels (mean ± SE) that were present on
test objects at the end of the experiment.
reasons. First, resident zebra mussels on
unionids (those marked at the start) showed
little tendency to emigrate, so that the mean
number of residents on unionids at the end of
the experiment was 6.1 compared to 6.5 at
the start. In contrast, the number of resident
zebra mussels on stones at the end of the ex-
periment averaged 1.8, a mean decrease of
4.7 zebra mussels on each stone. Second,
there was greater immigration of zebra mus-
sels to unionids than there was to stones, so
that at the end of the experiment, the unionids
averaged 2.3 ± 0.48 immigrants per unionid
compared to 0.04 ± 0.037 immigrants for
stones (Fig. 6B). These values were signifi-
cantly different: t^^, = 3.326, P = 0.029.
It is not known if the decrease in resident
zebra mussels was due to mortality or emi-
gration. Evidence for the former came from
field observations that many of the stones
were entirely buried, and these often had
empty shells of resident zebra mussels still at-
tached. These differences between unionids
and stones in terms of survival and behavior
of attached zebra mussels led to large per-
centage changes by the end of the experi-
ment (Fig. 7). Average percent change in
zebra mussels attached to unionids was both
positive and significantly higher (24.5 ±
8.66%) than that for stones, the latter de-
creasing by 73.1 ± 7.36% (t,^. = 5.14; P =
0.007).
DISCUSSION
Our data show that unionids in both lake
and river conditions are not preferentially set-
tled on or colonized by zebra mussels com-
pared to other hard substrata. We found no
evidence to suggest that unionids are more
attractive to either migrating zebra mussels or
to settling veligers than are alternative hard
substrata nearby. This finding agrees with
Toczylowski & Hunter (1997), who examined
a variety of substrata, including natural (live
unionids, unionid shells, wood) and artificial
surfaces (tiles, plastic mesh) and reached the
same conclusion. The density of zebra mus-
sels on unionids is approximately the same as
it is on other natural hard substrata located in
the same area, the outcome that would be
predicted if there were no preference. These
data were obtained for lake and river bottoms
that are relatively firm and in which inanimate
hard substrata do not become buried.
However, under different bottom conditions,
we obtained different results. When the bot-
tom is soft enough for hard substrata to be-
come buried, or if the bottom sediment is un-
stable due to a strong current, as in a river,
then any hard surfaces that are buried or
silted-over become sub-optimal for zebra
mussels. In these conditions, unionids main-
tain their position with respect to the sedi-
ment-water interface, and the zebra mussels
attached to the posterior end are carried
above the sediment. Zebra mussels on
stones or other kinds of inanimate substrata
become buried and either die or are forced to
emigrate. Therefore, under conditions of un-
stable substrata, greater loads of zebra mus-
sels will develop on unionids than on inani-
mate hard substrata due to differential
survival of the immigrant zebra mussels.
This study provides one observation and
160
TOCZYLOWSKI ETAL.
чи-
"Г
t^^=3.60; Р = 0.023
20-
n
^^^^^^;
и -
^^^^^^^
-20-
^^^^^
-40^
^^^^
-60-
^^^Щ
-80-
J-
1 пп
i ии-
Unionid
1
Stone
FIG. 7. Percent change (mean ± SE) in the total
number of zebra mussels (residents -i- immigrants)
on unionids and on stones at CLR (unstable sub-
stratum site) over the course of the experiment.
two experiments that support our hypothesis.
The observation is from measurements of
natural zebra mussel density on unionids and
on stones from a river site with an unstable
bottom (DM0). Test stones and unionids from
this site were located on patchy sand and
cobble where sand movement alternately
buried and exposed the cobble, rendering the
stones less suitable for zebra mussels. The
result was that unionids carried significantly
higher zebra mussel loads per unit area of ex-
posed surface than did stones from the same
site. This result differs from that reported in an
identical study in Loon Lake in 1995 (Toczy-
lowski & Hunter, 1997).
We believe the different outcomes of these
two studies can be explained by different sub-
stratum stability. The lake work was done on a
predominantly hard substratum (cobble), on
which test stones had little tendency to sink
and maintained their position relative to the
sediment/water interface. Unionids, which
were similarly positioned, had loads that were
basically the same as those on the stones. At
the river site, with its unstable bottom, zebra
mussels located on stones were buried and
died, whereas those on unionids survived by
remaining above the sediment.
Our hypothesis was further tested by ex-
periments conducted at both lake and river
sites. Pre-marked test substrata placed on an
unstable bottom in the Clinton River at CLR
confirmed that zebra mussels previously at-
tached to unionids were not only more likely to
remain attached over the summer than their
counterparts attached to stones, but were
also more likely to be joined by immigrants.
On stones, not only were there negligible
numbers of immigrants, but previously at-
tached zebra mussels died or emigrated due
to burial under bottom sediments.
A further test was provided by experiments
on a soft bottom site in Loon Lake and a sim-
ilar site in the Clinton River (DPNC). Under
such conditions, we would predict that union-
ids should accumulate a higher zebra mussel
load than stones, due to siltation of the stone
surfaces that occurred during the course of
the experiment. Our data supported the hy-
pothesis in both in lake and river conditions;
however, the results were more pronounced
under lake conditions despite the similarity of
substratum type. We believe this was due to
the presence of low current at the river site.
Current may have reduced silt accumulation
on upper stone surfaces and prevented re-
duced O2 levels from occurring at the sedi-
ment/water interface. These improved bottom
conditions would reduce the negative effects
to which zebra mussels on the river bottom
were exposed, reducing mortality. These
same subtle lake/river bottom differences
would also explain our finding that emigra-
tion/mortality of previously marked resident
zebra mussels from lake test surfaces was
higher than that from river test surfaces.
More recently, studies by Ricciardi et al.
(1995, 1996) have purported to demonstrate
that unionids are preferred substrata, and
consequently zebra mussels reach greater
densities on unionids than on the bottom in
general. A model presented in Ricciardi et al.
(1995) in support of preference was tested
using data from sites that were mostly soft
substrata. Under these conditions, the prefer-
ence by Dreissena for unionids is simply a
choice between hard and soft substrata. This
model has tested the obvious: that zebra
mussels reach higher density on hard, com-
pared to soft substrata. The control that was
not done would have examined zebra mussel
load at the same sites on stones, bricks, or
other hard substrata of comparable size and
surface texture. Although their model is useful
in extrapolating from general zebra mussel
density to zebra mussel loads on unionids, it
does not provide any useful insight into the
real issue of preference: do zebra mussels
select unionids over other natural hard sur-
faces?
In a further examination of the impact of
Dreissena on unionids, Ricciardi et al. (1996)
offered more evidence in support of prefer-
ence by showing that there was higher zebra
ZEBRA MUSSEL LOAD ON UNIONIDS
161
mussel density on unionids (in the mud bot-
tom of a canal) than on the concrete walls of
the canal. The problem with these data is that
Dreissena density on the wall is based on the
total wall area, whereas the density of Dreis-
sena on unionids is based only on the area of
unionid shell, not on total area of the bottom
(unionids plus mud). Using the terminology of
Bailey et al. (1995), what Ricciardi et al.
(1996) have done is to express zebra mussel
density on unionids in terms of number per
unit unionid surface (BMI/RSA) and zebra
mussel density on the concrete wall in terms
of number per unit wall area (BMI/BA). It is not
surprising that the former density is greater.
Because unionid shells are discreet optimal
surfaces for zebra mussel settling and sur-
vival and are dispersed in a habitat that is
generally sub-optimal, the result is intense
crowding. In contrast, the entire wall is suit-
able substratum.
We have shown the absence of preference
where the substratum is relatively firm and an
appearance of preference where the substra-
tum is soft and/or unstable. In this latter case,
it is likely that higher zebra mussel densities
on unionids are simply the outcome of the
ability of unionids to actively maintain their
posterior shell surface in an exposed position
on the bottom. Unionids are a relatively stable
surface due to their responsiveness to chang-
ing bottom conditions. Therefore, the devel-
opment of higher Dreissena densities on
unionids is not an outcome of preference in-
volving substratum choice using sensory per-
ception and subsequent taxes or directed
movement. Instead, our data indicate that
these density differences have arisen from
lower mortality and emigration rates on a re-
sponsive live surface compared to that on un-
responsive, inanimate hard surfaces.
Although it is well known that unionids move
both horizontally and vertically (extent of pro-
trusion from the sediment), the conditions
stimulating these responses and the adaptive
value of specific movements are largely un-
clear and have received scant attention in
most species (Balfour & Smock, 1995; Amyot
& Downing, 1997). Vertical movement involv-
ing shifts from epibenthic to endobenthic posi-
tion are known to occur in some species. The
best studied species is Elliptio complánala,
which showed seasonal movement resulting
in most of the population becoming endoben-
thic in winter and epibenthic in summer (Amyot
& Downing, 1997). In terms of the present
study, should any of the species that were test
unionids become endobenthic in winter it
could be a means of eliminating or reducing
their load of zebra mussels. The authors have
observed unionids in the field that are almost
entirely buried, with only the siphons visible.
Yet adjacent to the siphons are a few zebra
mussels attached to the extreme posterior
margins of the unionid shell, where they have
avoided burial but remain at the sediment/
water interface. Because they are epifaunal,
zebra mussels are likely to be less tolerant of
burial than unionids, hence might be partially
or entirely eliminated over winter when much
of the unionid community is endobenthic (un-
published observations). Although as yet we
have no overwinter survival data for zebra
mussels attached to unionids, we do have ev-
idence from zebra mussels loads in fall and the
following spring that indicate there is no signif-
icant reduction of loads over winter (Hunter et
al., 1997). It is entirely possible that once a sig-
nificant load of zebra mussels has accumu-
lated, it may act as a kind of stop, limiting the
ability of the unionid to bury itself in the sub-
stratum. If this occurs, it would prevent fresh-
water mussels that vertically migrate from be-
coming endobenthic.
ACKNOWLEDGMENTS
We acknowledge the Drayton Plains Nature
Center for allowing access to the Clinton
River from their property. Nicole Rudolph pro-
vided help with recovery of test surfaces from
lake and river. Jon Guilliat and Mike Armes
provided field assistance; Deborah Bishop
and Michael Attan provided laboratory assis-
tance. Thanks to the Michigan Department of
Natural Resources Wildlife Division for sup-
porting this work through Natural Heritage
Small Grants in 1996 and 1997, and to the
MDNR Fisheries Division for loaning us a boat
and trailer in 1996 and 1997. Thanks are also
due to the MDNR Parks and Recreation Divi-
sion for providing free access to the Loon
Lake and Cass Lake boat ramps.
LITERATURE CITED
AMYOT J. & J. DOWNING, 1997, Seasonal varia-
tion in vertical and horizontal movement of the
freshwater bivalve Elliptio complanata (Mollusca:
Unionidae). Freshwater Biology, 37: 345-354.
BAILEY R. C, M. CHASE & J-P. BECHTOLD,
1995. An improved technique for estimating the
162
TOCZYLOWSKI ETAL.
density of benthic macroinvertebrates on cobble.
Journal of Freshwater Ecology. 10: 189-192.
BALFOUR, D. L. & L. A. SMOCK, 1995, Distribu-
tion, age structure, and movements of the fresh-
water mussel Elliptio complanata (Mollusca:
Unionidae) in a headwater stream. Journal of
Freshwater Ecology. 10: 255-268.
BLUMER, S. P., T. E. BEHRENDT, J. M. ELLIS, R.
J. MINNERICK, R. L. LEUVOY & С R. WHITED,
1997, Water resources data — Michigan water
year 1997, U. S. Geological Survey, Water Re-
sources Division, Lansing, Michigan, USGS-
WDR-MI97-1.
CHASE, M. E. & R. С BAILEY 1996, Recruitment
of Dreissena polymorpha: áoes the presence and
density of conspecifics determine the recruitment
density and pattern in a population? Malacologia.
38: 19-31.
COAKLEY J. P., G. R. BROWN, S. E. lOANNOU &
M. N. CHARLTON, 1997. Colonization patterns
and densities of zebra mussel Dreissena in
muddy offshore sediments of western Lake Erie,
Canada. Water, Air, and Soil Pollution, 99: 623-
632.
DERMOTT R. & M. MUNAWAR. 1993, Invasion of
Lake Erie offshore sediments by Dreissena. and
its ecological implications. Canadian Journal of
Fisheries and Aquatic Science. 50: 2298-2304.
Dl МАЮ, J. & L. D CORKUM, 1995, Relationship
between the spatial distribution of freshwater
mussels (Bivalvia: Unionidae) and hydrological
variability of rivers. Canadian Journal of Zoology,
73:663-671.
HUNTER, R. D. & J. F. BAILEY 1992, Dreissena
polymorpha (zebra mussel): colonization of soft
substrata and some effects on unionid bivalves.
The Nautilus. 106:60-67.
HUNTER, R. D., S. A. TOCZYLOWSKI & M. G.
JANECH, 1997, Zebra mussels in a small river:
impact on unionids. Pp. 161-186, in: F. d'itri, ed.,
Zebra mussels and aquatic nuisance species.
Ann Arbor Press Inc., Chelsea, Michigan.
LEWANDOWSKI. K., 1976, Unionidae as a sub-
stratum for Dreissena polymorpha Pall. Polskie
Archiwum Hydrobiologii. 23: 409-420.
MACKIE, G. L., 1993, Biology of the zebra mussel
{Dreissena polymorpha) and observations of
mussel colonization on unionid bivalves in Lake
St. Clair of the Great Lakes. Pp. 153-165, in: т. F.
NALEPA & D. w. SCHLOESSER, ed.. Zebra mussels:
biology, impacts, and control. Lewis Publishers,
Boca Raton, Florida.
MORSE, A. N. C, 1991, How do planktonic larvae
know where to settle? American Scientist. 79:
154-167.
MORTON, J. E. & С M. YONGE, 1964, Classifica-
tion and structure of the mollusca. Pp. 1-57, in
K. M. WILBUR & с M. YONGE, eds., Physiology of
Mollusca, Academic Press, New York, New York.
NALEPA, T F., 1994, Decline of native unionid bi-
valves in Lake St. Clair after infestation by the
zebra mussel, Dreissena polymorpha. Canadian
Journal of Fisheries and Aquatic Sciences, 51:
2227-2233.
RICCIARDI, A., F. G. WHORISKEY & J. B. RAS-
MUSSEN, 1995, Predicting the intensity and im-
pact of Dreissena infestation on native unionid bi-
valves from Dreissena field density. Canadian
Journal of Fisheries and Aquatic Sciences, 52:
1449-1461.
RICCIARDI, A., F. G. WHORISKEY & J. B. RAS-
MUSSEN, 1996, Impact of the Dreissena inva-
sion on native unionids in the upper St. Lawrence
River. Canadian Journal of Fishehes and Aquatic
Sciences. 53: 1434-1444.
RODRIGUEZ. S. R., F. P. OJEDA & N. I. INE-
STROSA, 1993. Settlement of benthic marine in-
vertebrates. Marine Ecology Progress Series, 97:
193-207.
SCHLOESSER, D. W. & W. P. KOVALAK, 1 991 , In-
festation of unionids by Dreissena polymorpha in
a power plant canal in Lake Erie. Journal of Shell-
fish Research. 10: 355-359.
SCHLOESSER, D. W. & T F. NALEPA, 1994, Dra-
matic decline of unionid bivalves in offshore wa-
ters of western Lake Erie after infestation by the
zebra mussel, Dreissena polymorpha. Canadian
Journal of Fisheries and Aquatic Sciences, 51 :
2234-2242.
STEIN, B. A. & S. R. FLACK, 1997, Species report
card: the state of U. S. plants and animals. The
Nature Conservancy, Arlington, Virginia.
TOCZYLOWSKI, S. A. & R. D. HUNTER, 1997, Do
zebra mussels preferentially settle on unionids
and/or adult conspecifics? Pp. 125-140, in: F.
D'ITRI, ed.. Zebra mussels and aquatic nuisance
species. Ann Arbor Press Inc., Chelsea, Michi-
gan.
WAINMAN, B. C, S. S. HINCKS, N. K. KAUSHICK
& G. L. MACKIE, 1996, Biofilm and substrate
preference in the dreissenid larvae of Lake Erie.
Canadian Journal of Fisheries and Aquatic Sci-
ences. 53: 134-140.
WILLIAMS, J. D., M. L. WARREN, K. S. CUM-
MINGS. J. L. HARRIS & R. J. NEVES, 1992,
Conservation status of freshwater mussels of the
United States and Canada. Canadian Journal of
Fisheries. 18: 6-22.
YONGE, С M. & J. I. CAMPBELL, 1968, On the
heteromyarian condition in the bivalvia with spe-
cial reference to Dreissena polymorpha and cer-
tain Mytilaceae. Transactions Royal Society of
Edinburgh. 68: 4-42.
Revised ms. accepted 16 December 1998
MALACOLOGIA, 1999, 41(1): 163-174
GANULA GADIRANA N. SR, A NEW HYGROMIIDAE FROM SOUTHERN SPAIN
(PULMONATA: HELICOIDEA)
Benito Muñoz, ^ Arturo Almodovar,^ & José R. Arrebola^
ABSTRACT
Ganula gadirana Muñoz, Almodovar & Arrebola, is proposed as a new species from the south-
ernmost corner in the Iberian Peninsula. This species, sometimes erroneously recorded as Helix
(= Ganula) lanuginosa Boissy, 1835, is characterized by a globose conical-depressed shell, with
ovate aperture, small umbilicus, and periostracum with long, persistent hairs; it has a right om-
matophore retractor between the penis and the vagina, a penial nerve arising from the right cere-
bral ganglion, ring-shaped glandular area on the distal penis wall, a fenestrate and elongate pe-
nial papilla, and a circumvaginal tuft of long, annulated digitiform glands. A dart-sac complex is
on one side of the vagina and is formed by an outer dart-bearing stylophore and inner apically
bilobed dartless stylophore; each open separately into a canaliculate deep groove that divides
two large pleats, distally detached from the vaginal wall. Comparisons with Ichnusotricha and
Ganula suggest that the new species may be closely related to G. lanuginosa (Boissy, 1835).
The bilobate distal portion of the inner stylophore, the larger number of tufts in the digitiform
glands, the fenestrate penial papilla, the penial glandular area, and the radular formula to distin-
guish G. gadirana from G. lanuginosa.
Key words: Ganula gadirana. Gastropoda, Stylommatophora, Hygromiidae, Spain.
INTRODUCTION
Many new generic taxa have been recently
described by the Giusti-Manganelli's team for
the western Mediterranean area, most of
them being monotypical: Helicotricha carusoi
from the Aeolian Islands (Giusti et al., 1992),
Cilliellopsis oglasae from Montecristo Island
(Giusti & Manganelli, 1990), Schileykiella ior
Helix parlatoris Bivona, 1839, and Helix
reinae Pfeiffer, 1856, from Sicily (Manganelli
et al., 1989), Ichnusotricha berninii anä Nien-
huisiella antonellae from Sardinia (Giusti &
Manganelli, 1987), and Tyrrheniella josephi
from Sardinia/Capraia Islands (Giusti & Man-
ganelli, 1989). The number of new morpho-
logical patterns is striking, and more new
generic taxa can be expected.
Ganula lanuginosa, according to biblio-
graphical records, seems to be distributed in
four areas: eastern Balearic Islands (type lo-
cality of Helix lanuginosa Boissy, 1835) Sar-
dinia (Giusti & Manganelli, 1987), southern
Spain (Servain, 1880), and northwestern
Africa (Terver, 1839; Bourguignat, 1864; Le-
tourneux & Bourguignat, 1887). This species
could be of great zoogeographical value if its
presence in these areas is anatomically con-
firmed (Giusti & Manganelli, 1984), although
Balearic records could be the result of an in-
troduction from northwestern Africa during the
Middle Ages (Gasull, 1963).
Nevertheless, the anatomical confirmation
has been carried out only for the Balearic Is-
lands (Hesse, 1931; Gittenberger, 1968) and
Sardinia (Giusti & Manganelli, 1987). Search-
ing to confirm its presence in southern Spain
where it was repeatedly cited (Servain, 1880
Sacchi, 1956, 1957; Sacchi & Nos, 1958
Gasull, 1963), was unsuccessfull. In nearly
identical shells, we have found snails with
genitalia similar to those in Ganula lanugi-
nosa, but differing in the configuration of the
dart-sac complex and in a larger number of
tufts in the digitiform glands. These charac-
ters support introduction of a new species
from southern Spain: Ganula gadirana.
MATERIALSAND METHODS
Living specimens were drowned in tap-
water and preserved in 70% ethanol. Re-
moved bodies were dissected and studied by
optical stereomicroscopes (Zeiss and Nikon).
Genital system and other structures were
drawn using a camera lucida (Zeiss and
Nikon). Nerve rings were removed from the
'Dpto. Biología Animal I (Zoología), Universidad Complutense de Madrid, Avda. Complutense s/n, 28040-Madrid, Spain;
titomu@eucmax.sim.ucm.es
^Dpto. Fisiología Animal y Biología, Universidad de Sevilla, Avda. Reina f^ercedes, 6,4101 2-Sevilla, Spain
163
164
MUÑOZ ETAL
connective tissue with a pointed watchmaker
forceps prior to drawings.
Serial microscopical sections were coloured
with Mallory's stain and photographed with a
Zeiss Stemi SV6 photostereomicroscope.
Radulae were removed from buccal bulbs by
hot digestion in KOH and then washed in pure
ethanol. Some radulae were mounted on
metallic blocks with electron-conductive glue,
coated with gold, and photographed in a JEOL
820 SEM, and others were stained by a modi-
fied hemalumpicroindigocarmine method and
observed using a Zeiss stereomicroscope.
Shells were photographed with the same SEM
or with a Pentax P30N camera.
Shell parameters were measured in adult
specimens using a Zeiss stereomicroscope
with millimethc lens. Features of the central
nervous system were described according to
the character states proposed by Tillier (1 989:
40-42).
The nomenclature for the genital system
(that used by other authors in brackets) is the
following: outer stylophore (= dart sac or lower
stylophore), inner dartless stylophore (= ac-
cessory sac or upper stylophore), penial
papilla (= verge), digitiform glands (= mucous
glands).
SYSTEMATIC DESCRIPTION
Ganula gadirana Muñoz, Almodovar &
Arrebola, п. sp.
Helix lanuginosa. Servain, 1880: 51 {non
Boissy) ["Environs d'Algesiras"].
Fruticicola lanuginosa. Sacchi, 1956: 17
{non Boissy) [only, "nelle regioni tra Malaga e
Algesiras"].
Fruticicola lanuginosa. Sacchi, 1957: 19
{non Boissy) [only: "presse Malaga e presso
Algesiras"].
Fruticola lanuginosa, Sacchi & Nos, 1958:
93 {non Boissy) [only: "region d'Algesiras"].
Shell (Figs. 1, 2). Shell medium sized, glo-
bose-depressed, conical-convex above, in-
flated below, uniform light brown translucent
(Fig. 1A-B). Spire low conical, consisting of
5-5 1/2 convex, regularly increasing whorls
separated by deep sutures. Apex small, pro-
truding apex; protoconch with 1 1/4-1 1/2
whorls, striated, with hair scars (Fig. 2A). Last
whorl large, one and one half times broader
than the penultimate whorl, rounded at pe-
riphery, and variably descending near aper-
ture. Umbilicus very small (0.6-0.8 mm),
deep, partly covered by the reflected margin
of peristome. Aperture oblique, oval or nearly
circular, without internal lip, but with a whitish
band (reddish in external view). Peristome in-
terrupted straight, smooth, not thickened, with
very separated, non-convergent marginal
edges; columellar edge gently curved, wid-
ened at origin, very reflected over umbilicus.
External teleoconch surface with long (0.4
mm), erect, persistent hairs in transverse
rows 0.4 mm far apart, with a density of 1 4- 1 7
hairs/mm^ (Fig. 2B); microsculpture of teleo-
conch formed by fine, small crests among
hairs; hair scars curved.
Dimensions (n = 16): Shell diameter, 11.0-
14.0 mm (holotype, 13.0 mm); shell height,
8.0-10.9 mm (holotype, 9.8 mm); aperture
maximum diameter, 6.8-8.2 mm (holotype,
7.9 mm).
Foot (Fig. 2C). Sole with a tripartite appear-
ance (also visible in preserved specimens):
central zone light, lateral zones darker.
Lung roof (Fig. ЗА) with an irregular pattern
of small, blackish, irregular spots, which are
larger over kidney borders.
Kidney (Fig. 3B) with numerous, long,
raised internal pleats, more numerous on
ureteric side; primary ureter broad, mostly sit-
uated on the kidney; secondary ureter, closed
at first, but open 1 mm from the ureteric angle
(in front of heart ventricle) to form a broad
ureteric groove.
Heart oval, somewhat longer than half
length of kidney. Slender primary pulmonary
vein with inconspicuous secondary veins.
Mantle collar (Fig. 3C). Left lateral lobe,
thin, small, with lobed upper border and
straight lower border. Right lateral lobe long,
thick, with an upper corner forming an anal
lobe. Both left and right dorsal lobes with free
marginal borders. Subpneumostomal lobe
separating anal and pneumostomal orifices.
Central nervous system (Fig. 3D-F). Cere-
bral commissure shorter (CC3) and right cere-
bropedal connective somewhat longer
(CPD2) than right cerebral ganglion width;
both right and left cerebropedal connectives
of similar length (CPR2), somewhat longer
than cerebropleural connectives; both right
and left pleural ganglions closer to pedal gan-
glions than cerebral ganglions (PLD1 , PLG1 );
visceral ganglion in median plane of pedal
ganglions (VG2); right parietal ganglion in
contact with both visceral and right pleural
ganglions (PAD2), apparently fused with the
latter; left parietal ganglion only in contact with
visceral ganglion (PAG3), both ganglia appar-
ently fused (FG3).
Genital system (Figs. 4A-B, 5A-E, 6A-E).
GANULA GADIRANA N. SP
165
FIG. 1. Shells of Ganula gadirana. n. sp., from the type locality. A. Holotype; B. Two paratypes (upper col-
lected on 30 October 1992, and lower on 14 May 1992).
Multilobate gonad and hermaphrodite duct
without special features (not figured). Albu-
men gland relatively short. FPSC (Fig. 5A),
with both seminal receptacle and fertilization
pouch simple, sac-like, the former shorter,
slender. Ovispermiduct (Fig. 4A), wide, cir-
cumvoluted, with prostatic and uterine parts
well differentiated. Vas deferens slender.
Reniai complex (Figs. 4A-B, 5D) consisting
of flagellum, epiphallus (i.e., from end of vas
deferens to attachment point of penial retrac-
tor muscle), and penis (i.e., from point of
attachment of penial retractor muscle to geni-
tal atrium); penial retractor muscle attached to
penial complex proximal to point in which
base of penial papilla is contained; a strong
muscular band (Fig. 4B) extends from outer
penial wall of proximal penis to wall of genital
atrium, bending the penial complex; flagellum
short, 1/3 of epiphallus length, conical in
shape, with thick walls; epiphallus cylindrical,
somewhat longer than penis, its inner wall
pleated; penis with distal yellowish, ring-
shaped, glandular area (Fig. 5C-D); penial
papilla long, cylindrical (Fig. 5B), with three
fenestrations through which wide distal lacu-
nae inside penial papilla walls communi-
cate with penial cavity; a nearly isolated
central canal traverses the penial papilla, its
walls fixed by radial septa to walls of penial
papilla.
Female part (Fig. 4A-B) consisting of a
short free oviduct (half length of bursa copu-
latrix duct); bursa copulathx large, shoe-
shaped, with short pedunculus; vagina short,
with digitiform glands and dart-sac complex,
the latter located on one side. Digitiform
glands (Fig. 5E) composed of many (14-20),
long, unbranched or basally branched tubes
disposed all around vagina; tubes with an an-
nulate appearance, the inner secretory ep-
ithelium forming small ridges (Fig. 7A, G).
Dart-sac complex (Figs. 6A-B, D-G, 7A-F)
short, broad, composed of two stylophores;
outer stylophore basally surrounded by
groove of vaginal wall (Fig. 6A-B); inner sty-
lophore broader, dartless, with thick walls and
narrow inner cavity, extended into two apical
166
MUÑOZ ETAL
; _.:^r\-
1
^шл
^
^ * \
1
^
^Pçy^l^
SIbé
Ш
f
^_1Щ^
п^НШвн
j^^^
ià
ik N v%r%lîf'*^
FIG. 2. Shell (SEM micrographs) and pedal sole of Ganula gadirana, n. sp. A. Protoconch of a paratype from
Cortijo de Ahojiz; В. Microsculpture and one hair of teleoconch (same locality); С Ventral view of pedal sole
showing the tripartite sole (from type locality).
lobes (Fig. 6D, 7A). Inner stylophore opening
into vagina through a wide orifice, far from
outer stylophore aperture (Fig. 6B, 7C). Dart
(Fig. 6C) smooth, straight, circular in section,
but with flattened, keeled tip, elongate, ex-
tending out of outer stylophore. Inner vagina
(Fig. 6B, F-G, 7C-F) with two thick, large
pleats, which distally fuse to form a tongue-
like structure, U-shaped in traverse section,
its tip detached from vaginal walls; outer sty-
lophore and inner stylophore open far apart
inside groove of tongue-like structure.
Juvenile specimens (Fig. 6E-G) with dart-
sac complex placed at half of vagina length,
with outer stylophore shorter than bilobed
inner stylophore and tongue-like structure
present inside vagina.
Jaw (Fig. 3H) odontognathous, with 24 ribs,
central larger.
Radula (Fig. 8) consisting of many rows of
teeth each with a formula of 35-37 -i- С -i-
35-37, with the lateral/marginal transition to-
wards the 15-1 6th tooth. Central tooth with
wide basal plate, pointed vertices and body
with large mesocone and two small ectocones
nearly fused to base of mesocone. Lateral
teeth with wide basal plates and body with en-
docone, large, pointed mesocone and short
ectocone; both mesocone and ectocone pro-
gressively slender towards radular margin,
mesocones with little lateral protuberance and
pointed ectocones split into two points.
Other anatomical characters. Right om-
matophore retractor running between penis
and vagina. Penial nerve apparently arising
from nght cerebral ganglion.
Type Locality
Arroyo de la Cabañuela, Puerto de Bolonia,
Tarifa (Cádiz, Spain, UTM: 30STE552990).
Type Material
Holotype (shell and dissected specimen) -
30 October 1992, A. Almodovar, В. Muñoz
and P. Refoyo leg. Museo Nacional de Cien-
cias Naturales, Madrid, Spain.
Paratypes:
-Type locality, 15 paratypes: 11 specimens
(10 dissected), 30 October 1992, A. Almo-
dovar, В. Muñoz and P. Refoyo leg., 1
shell, 9 May 1994, A. Almodovar leg.; 3
GANULA GADIRANA N. SP
167
FIG. 3. Some body parts of Ganula gadirana. n. sp., from Los Barrios, 16 May 1993 (A-G) and the type lo-
cality (H). A. Pigmentary patches on the pulmonary cover (external view); B. Kidney, ureters and pericardium;
С Mantle collar; D-E. Left and right views of central nervous system; F. Posterodorsal view of central ner-
vous system from another specimen; G. Position of salivary glands in respect to buccal bulb; H. Jaw. Abré-
viations: an, anal lobe; bb, buccal bulb; ce, cerebral ganglion; dsg, ducts of salivary glands; in, intestine; ki,
kidney; Id, left dorsal lobe; II, left lateral; oe, oesophagus; pa, parietal ganglion; pc, pericardium; pe, pedal
ganglion; pi, pleural ganglion; pu, primary ureter; pv, pulmonary vein; rd, right dorsal lobe; re, rectum; ri, right
lateral lobe; sg, salivary glands; sp, supneumostomal lobe; su, secondary ureter; vi, visceral ganglion. Scale
bar, 1mm.
168
MUÑOZ ETAL
FIG. 4. Genital system of Ganula gadirana. n. sp. A. Specimen from type locality (gonad excluded); B. Spec-
imen from Los Barrios (distal ducts), 3 November 1991. Abréviations: ag, albumen gland; apb. atrio-penial
muscular band; is, inner stylophore; be, bursa copulatrix; dbc, duct of bursa copulatrix; dg, digitiform glands;
OS, outer stylophore; dsc, dart sac complex; e, epiphallus; f, flagellum; to, free oviduct; ga, genital atrium; hd,
hermaphrodite duct; p, penis; po, prostatic part of ovispermiduct; pr, penial retractor muscle; uo, uterine part
of ovispermiduct; v, vagina; vd, vas deferens. Scale bar, 1 mm.
GANULA GADIRANA N. SP
169
FIG. 5. Some genital parts of one specimen of Ganula gadirana. n. sp., from Los Barrios, 16 May 1993. A.
Fertilization pouch and seminal receptacle complex (external view) and four transversal sections: B. Reniai
papilla within penis and transversal sections at different levels (1 .33 times enlarged): С Glandular area on
distal part of penis (internal view): D. Penial complex (contracted) and transversal sections (1.5 times en-
larged) of epiphallus and flagellum: E. Shape and disposition of digitiform glands on vagina. Abréviations:
apb, atho-penial muscular band: dl, distal lacuna: e, epiphallus: f, flagellum: fp, fertilization pouch: gr, glan-
dular ring: hd. hermaphrodite duct: os, ovispermiduct: p, penis: pf, papillär fenestration: sd, seminal duct: sr,
seminal receptacle. Scale bar, 1 mm.
170
MUÑOZ ETAL
FIG. 6. Some genital parts of Ganula gadirana. n. sp. A. External views of dart-sac complex in a contracted
specimen from Los Barrios, 16 May 1993; B. Same dart-sac complex in longitudinal section: С Dart of a
specimen from type locality; D. Inner vaginal view of dart-sac complex and cavity of bilobed inner stylophore
from a specimen from Los Barrios, 5 February 1990; E. Genital system (gonad excluded) of a juvenile spec-
imen from Los Barrios, 1 6 May 1 993. Note position of dart-sac complex on vagina and bilobation of inner sty-
lopfiore; F Inner vaginal view; G. Longitudinal section of dart-sac complex of same juvenile specimen. Ab-
breviations as in Fig. 4. Scale bar, 1mm.
GANULA GADÍRANA N. SP
171
os ]s dg
A e
/'
OS
'^•^^Щ^£^
в
D
\
/
■/ ;
dbc
Л'
I
ga
/
G
FIG. 7. Outline of dart-sac complex (digitiform glands not figured) and transversal sections at different lev-
els. A. Bilobed cavity of inner stylophore and inner ridges of digitiform glands; B. Monoluminar cavity of inner
stylophore and opening of a digitiform gland inside vagina; C. Opening of inner stylophore inside groove
formed by vaginal pleats; D. Groove between inner and outer stylophiores forming a functional channel; E.
Opening of outer stylophore inside groove and fused pleats partially detached from vaginal wall; G. Enlarged
transversal section of digitiform glands showing inner ridges and secreted mucus.
specimens (dissected), 14 May 1994, A.
Almodovar leg.
-Cortijo de Ahojiz, between km 90-91 of
C-440 road to Los Barrios (Cádiz, UTM:
30STF701 0), 1 6 paratypes: 3 specimens (one
dissected), 5 February 1990; 6 specimens (2
dissected), 18 May 1991; 1 specimen (dis-
sected), 24 March 1991 ; 6 specimens (3 dis-
sected), 3 November 1991. All, J. Arrebola
leg.
-'Campo de Gibraltar' sawmill, Los Barrios
(Cádiz, UTM: 30STF7507), 30 paratypes: 21
specimens (mostly juveniles, 5 dissected) and
9 shells, 16 May 1993, E. Unamuno and J. С
Ruiz leg.
-Algeciras, near a shelter (Cádiz, UTM:
30STE7394), 1 paratype (damaged shell), 16
May 1993, E. Unamuno and J. С Ruiz leg.
-2 km towards Punta Paloma from N-340
road. Tarifa (Cádiz, UTM: 30STE5694), 10
paratypes (3 dissected), 6 December 1993, J.
Arrebola leg.
172
MUÑOZ ETAL
FIG. 8. SEM micrographs of the radula of Ganula gadirana. n. sp. A. Central tooth and lateral teeth; B. Last
lateral and marginal teeth.
Dehvatio Nominis
From the name of the Fenician colony
Gadir, which gave origin to Cádiz, in the
southernmost province of Spain, where the
described snails were collected.
Ecology
Ganula gadirana has been found under
stones, half buried in the ground or on herba-
ceous vegetation associated to mediter-
ranean bushes {Chamaerops humilis and Ne-
hum oleander) and close to periodically river
flows. The associated malacological fauna is
composed by Rumina decollata (Linnaeus,
1758), Ferussacia follicula (Gmelin, 1791),
Helix aspersa Müller, 1774, Cochlicella acuta
(Müller, 1774), Cochlicella barbara {Linnaeus,
1758), Gasulliella simplicula (Morelet, 1845),
Caracollina lenticula (Michaud, 1831), Xe-
rotricha apicina (Lamarck, 1822), Cernuella
virgata (Da Costa, 1 778), Theba pisana (Mül-
ler, 1774), Ótala láctea (Müller, 1774), Can-
didula д1дахИ{Р^еШг, 1848), Oestophora bar-
bula (Rossmässler, 1838), and Oestophora
íam /ел/ (Morelet, 1854). Specimens from Los
Barrios, collected on 1 6 July 1 993, were inten-
sively parasitized by small nematodes, which
were found in all growth stages inside the pul-
monary cavity.
all records from southern Spain referred to
this species have been assigned to Ganula
gadirana.
There are older conchological records for
Helix lanuginosa from northern Africa - from
Morocco (Hidalgo, 1909: "Muluya") to north-
western Algeria (Terver, 1839; Bourguignat,
1864), but with one record from Tunisia (Le-
tourneux & Bourguignat, 1887). Other two
nominal species are considered to be related
to G. lanuginosa - Helix flava Terver, 1839,
and Helix roseotlncta Forbes, 1838 (Clessin,
1881; Richardson, 1980). Helix flava from
"Gourayah," near Bougie (Terver, 1839), and
H. roseotinctawas cited from northeastern Al-
geria (Bourguignat, 1864) to northern Tunisia
(Letourneux & Bourguignat, 1887) (Bourguig-
nat, 1 864, considered the two species as syn-
onymous). Taking into account that Ganula
gadirana has been an overlooked species
identified as H. lanuginosa, there is a possi-
bility that some of the African records belong
to Ganula gadirana. Further studies must be
carried out to test this. Nevertheless, Giusti &
Manganelli (1 987) stated that H. flava is an Al-
gerian species clearly distinct from G. lanugi-
nosa and also from G. gadirana (Giusti, pers.
com.).
DISCUSSION
Geographical Distribution (Fig. 9)
All known localities are in the province of
Cádiz in the southernmost corner of Spain,
concentrated on a small region around the
northern side of the Gibraltar Strait. Because
the previously recorded presence of Ganula
lanuginosa has not been confirmed for that re-
gion after many searches (see Introduction),
The new species has similar features to
Ichnusotricha berninii Giusti & Manganelli,
1987, and Ganula lanuginosa (Boissy, 1835).
Ichnusotricha berninii has a similar shell
shape, but the sutures are more superficial,
the aperture is descendent, the umbilicus is
nearly closed, and the periostracal hairs are
very short. Anatomically, it differs from the
new species, because /. berninii has a long,
GANULA GADIRANA N. SP
173
— 5
5
— 7
5
2
1
I
^
■
О
IP
\
9
■
^
■^
5
tE
8
X
3¿5
''\r--
FIG. 9. Geographical distribution of Ganuia gadirana, n. sp. (square and upper left inset) and other endemic
Hygromiinae from the western Mediterranean area (larger letters): C, Cernuellopsis ghisottii Manganelli &
Giusti, 1987; G, Ganuia lanuginosa (Boissy, 1835); I, Ichnusotricha bern I ni i GlusW & Manganelli, 1987; N,
Nienhuisiella antonellae Giusti & Manganelli, 1987; Y, Cyrnotheba Corsica (Shuttleworth, 1843); X, Xero-
plana lacosteana (Bourguignat, 1864). Smaller letters from northern Africa show old conchologica! records
for Helix flava Terver, 1 839 (F), Helix lanuginosa Boissy, 1 835 (L), and Helix roseotincta Forbes, 1 838 (R).
non-fenestrated penial papilla; small, narrow
dart-sac complex placed far apart; shorter
digitiform glands: and long cylindrical vagina,
with two long, slender pleats fused distally to
form an apical tap or dart gun (Giusti & Man-
ganelli, 1987).
Ganuia lanuginosa has a nearly identical
shell, although it is convex-depressed above,
with more superficial sutures and less convex
whorls, shorter (length, 0.3 mm; in shells from
Mallorca), less spaced hairs (25-30/mm^),
wider umbilicus (1/10 of shell diameter), and
descending aperture. Anatomically, G. lanugi-
nosa and G. gadirana are very different, be-
cause G. lanuginosa has four bifurcated digi-
tiform glands, standard inner stylophore, and
non-fenestrated penial papilla (Gittenberger,
1968; Giusti & Manganelli, 1987). Ganuia
gadirana has 1 4-20 digitiform glands, and the
tubes have an annulate appearance, apically
bilobed inner stylophore, fenestrate penial
papilla, and penial glandular area.
On the other hand, in the pedal sole G.
gadirana appears tripartite, and the left pari-
etal ganglion is in contact with the visceral
ganglion; these features are unknown in G.
lanuginosa, but are present in a few species
of the family Hygromiidae. The tripartite ap-
pearance of the pedal sole appears frequently
among helicoids (Schileyko, 1978), and is il-
lustrated in a drawing of Leucozonella rubens,
although we do not know other similar cases.
The genera of Hygromiidae studied by Tillier
(1 989) have a more anterior position of the left
parietal ganglion, being in contact with both
visceral and left pleural ganglions.
Ganuia lanuginosa has radular formula of
32 + С + 32 (Giusti & Manganelli, 1987),
whereas G. gadirana has a formula of 35-
37 -b С -b 35-37; both species have a similar
central tooth, but G. lanuginosa has lateral
teeth with an apex formed by a wide, robust,
pointed mesocone and a short, sharp, robust
ectocone, whereas the new species has lat-
eral teeth with a large endocone, pointed
mesocone, and short ectocone. Two points on
the mesocone apex of the extreme marginal
teeth appear occasionally in G. lanuginosa,
whereas there is a little lateral protuberance in
G. gadirana.
ACKNOWLEDGEMENTS
Special thanks for comments on a previous
version and precise observations on the de-
scription of the genital system to Dr. Foico
Giusti (Siena, Italy). The authors wishes to
174
MUÑOZ ETAL
thank to Dr. С E. Prieto (Bilbao, Spain) for his
helpful comments on the description genital
system and for his drawing (Figs. 3, 4B, 5 and
6).
LITERATURE CITED
BOURGUIGNAT J. R.. 1864, Malacologie de l'Al-
gérie. 2 vols. Paris. 686 pp. + 58 pis.
CLESSIN, S., 1881, Nomenclátor Heliceorum
viventium quo continetur nomina omnium hujus
familiae generum et specierum hodie cognitarum
disposita ex affinitate naturali. Opus postum-
mum. Ed. Fischer, Kassel. 617 pp.
EMBERTON, K. С & S. TILLER, 1995, Letter: to
the editor. Clarification and evaluation of Tillier's
(1989) stylommatophoran monograph. Malacolo-
g/a, 36(1/2): 203-208.
GASULL, L., 1963, Algunos moluscos terrestres y
de agua dulce de Baleares. Boletín de la So-
ciedad de Historia Natural de Baleares. 9 (1 -4):
3-80.
GITTENBERGER, E., 1968, Zur Systematischen
Stellung von Helix lanuginosa Boissy, mit Neu-
beschreibung eines Subgenus. Boletín de la So-
ciedad de Historia Natural de Baleares. 14: 63-
68.
GIUSTI, F & G. MANGANELLI, 1984, Relation-
ships between geological land evolution and pre-
sent disthbution of terrestrial gastropods in the
western Mediterranean area. Pp: 70-92, In: A.
soLEM & A. с VAN BRUGGEN, eds., World-wide
snails. Biogeographical studies on non-marine
Mollusca. Leiden.
GIUSTI, F & G. MANGANELLI, 1987, Notulae
malacologicae, XXXVI. On some Hygromiidae
(Gastropoda: Helicoidea) living in Sardinia and in
Corsica. (Studies on the Sardinian and Corsican
Malacofauna VI). Bolletino Malacologico.
23(5-8): 123-206.
GIUSTI, F & G. MANGANELLI, 1989, Notulae
malacologicae, XLIV. A new Hygromiidae from
the tyrrhenian islands of Capraia and Sardinia
with notes on the genera Xeromicra and Xe-
rotricha (Pulmonata: Helicoidea) (Studies on the
Sardinian and corsican malacofauna, VIII). Bol-
letino Malacologico. 25 (1 -4): 23-62.
GIUSTI, F & G. MANGANELLI, 1990, Ciliellopsis
oglasae, a new hygromiid from Montecristo Is-
land (Tuscan Archipiélago, Italy) (Pulmonata: He-
licoidea). Journal of Conchology. 33: 269-277.
GIUSTI, R, G. MANGANELLI & J. V. CRISCI, 1992,
A new problematical Hygromiidae from the Aeo-
lian Islands (Italy) (Pulmonata: Helicoidea). Mala-
cologia. 34 {^ -2): 107-128.
HESSE, P, 1931, Zur Anatomie und Systematik
palearktischer Stylommatophoren. Zoológica. 31
(81): 1-118 + 16pls.
HIDALGO, J. G., 1909, Enumeración de los molus-
cos recogidos por la Comisión exploradora de
Marruecos. Boletín de la Sociedad Española de
Historia Natural. 9 (4): 211-213.
LETOURNEUX, A. & J. R. BOURGUIGNAT 1887,
Prodrome de la malacologie terrestre et fluviatile
de la Tunisie. Impr. Nationale, Paris. 166 pp.
MANGANELLI, G., I. SPARACIO & F GIUSTI,
1 989, New data on the systematics of two Sicilian
land snails. Helix parlatoris Bivona, 1839 and
Helix reinae L. Pfeiffer, 1856 and description of
Schileykiella n. gen. (Pulmonata: Hygromiidae).
Journal of Conchology. 33: 1 41 - 1 56.
RICHARDSON, L., 1980, Helicidae: Catalog of
Species. Tryonia. 3: iii -i- 1 -697.
SACCHI, С F, 1956, I Molluschi terrestri nelle re-
lazioni biogeografiche tra Italia ed Africa. Archiivi
di Botánica e Biogeografia Italiana, 1 (4): 1 -31 .
SACCHI, С F., 1957, Lineamenti biogeografici
della Spagna mediterránea su basi malacofau-
nistiche. Publicaciones del Instituto de Biología
Aplicada. 25: 5-48.
SACCHI, С. F & R. NOS, 1958, Ouelques distribu-
tions intéressantes des mollusques terrestres
ibériques. Publicaciones del Instituto de Biología
Aplicada. 27: 89-95.
SCHILEYKO, A. A., 1978, Nazemnye molliuski
nadsemejstva Helicoidea. Fauna SSSR Mol-
liuski. 3(6). Zoologicfieskij Institut, Akademija
Nauk SSSR. (n. s.) 117:1 -384 (in Russian).
SERVAIN, G., 1880. Etude sur les mollusques re-
cueillis en Espagne et en Portugal. Paris. 1 72 pp.
TERVER, M., 1839, Catalogue des mollusques ter-
restres et fluviátiles, observes dans les posses-
sions françaises au Nord de I Afrique. J. B. Bail-
liêre & Crochard, Paris. 40 pp. -t- 4 pis.
TILLIER, S., 1989, Comparative morphology, phy-
logeny and classification of land snails and slugs
(Gastropoda: Pulmonata: Stylommatophora).
Malacologia. 30(1-2): 1-303 (see also Ember-
ton &Tillier, 1995).
Revised ms. accepted 17 December 1998
MALACOLOGIA, 1999, 41(1): 175-186
COMPARATIVE KARYOLOGICAL STUDY OF CUPPED OYSTER SPECIES
Alexandra Leitao,^ Pierre Boudry,^ Jean-Philippe Labat^ &
Catherine Thiriot-Quiévreux^*
ABSTRACT
Chromosomes of six cupped oyster species were studied using karyometric analysis after
conventional Giemsa staining, and silver staining. Karyotypes of Crassotrea angulata (nine
metacentric and one submetacentric chromosome pairs), C. sikamea (nine metacentric and one
submetacentric chromosome pairs), С virginica (eight metacentric and two submetacentric
chromosome pairs), С ariakensis (eight metacentric and two submetacentric chromosome
pairs), C. grasar (six metacentric and four submetacentric chromosome pairs), and Saccostrea
commercialls (eight metacentric and two submetacentric chromosome pairs) are distinguishable
by the number and position of the submetacentric chromosome pair and by the location of nu-
cleolus organizer regions. Comparative karyological analysis of these six cupped oysters and of
С gigas was made using a Principal Component Analysis and a Hierarchical Clustering Analy-
sis. Crassostrea gasar appears isolated from the other oyster species. Then, two clusters are
separated. The first one groups С gigas. С angulata and С sikamea, in which С gigas is ple-
siomorphic. The second one consists of C. ariakensis. С virginica and S. commercialis. Results
are discussed with regards to oyster species relationships based on other genetic characters and
to hybridization possibilities.
Key words; cupped oyster, chromosome, karyotype, NORs, cytotaxonomy, Bivalvia.
INTRODUCTION
Chromosomes of Ostreidae have been
studied in 22 species (Nakamura, 1985; Vit-
turi et al., 1985; leyama, 1990). Cupped oys-
ter species of the genera Crassostrea and
Saccostrea show a common diploid chromo-
some number of 2n = 20, and their karyotypes
include only metacentric and submetacentric
chromosomes (Table 1). Interspecific differ-
ences consist of the occurrence and differing
proportions of these morphological types,
identified either by observation or after chro-
mosome measurements. Karyotype differ-
ences may be seen within a species (e.g., С
rhizophorae; Table 1 ) which could be due ei-
ther to intraspecific polymorphism or to the
different techniques used.
Oyster species might have become differ-
entiated through pericentric inversions of cen-
tric shifts. However, cytotaxonomic compari-
son needs to be based on karyological
analysis carried out by the same technique
and the same worker. For example, the con-
centration and time of incubation in the
colchicine and in the hypotonic treatment, re-
sulting in differential condensation or elonga-
tion of chromosomes (Sharma & Sharma,
1980), vary from one author to another. Kary-
ometric analysis brings a more quantitative
method to assess chromosome morphology,
but still depends on the condensation or elon-
gation of chromosomes.
Banding techniques have been found to be
very useful for the identification of individual
chromosomes and also of particular regions
of chromosomes. Few studies have looked at
banding patterns in the chromosomes of oys-
ters (Rodriguez-Romero et al., 1 979c; Insua &
Thiriot-Quiévreux, 1991; Li & Havenhand,
1997). Fluorescence in situ hybridization has
been tested in Crassostrea gigas (Clabby et
al., 1 996; Quo & Allen, 1 997). Selective stain-
ing of the nucleolus organiser regions (NORs)
has been shown to have potential as a cyto-
taxonomic tool (e.g., Amemiya & Gold, 1990).
Patterns of specific NORs have been de-
scribed in five species of oysters (Thiriot-
Quiévreux & Insua, 1992; Insua & Thiriot-
Quiévreux, 1993; Ladron de Guevara et al.,
1994). Identification of structural chromo-
some features is useful in hybrid breeding
programs and in oyster stock conservation.
In the present study, karyotypes and NORs
'corresponding author
ЮЬ5егуа1о1ге Océanologique, LIRA 2077, BP 28, 06230 Villefranche-sur-Mer, France
^Station IFREMER, Laboratoire de Génétique, Aquaculture et Pathologie, BP 1 33, 1 7390 La Tremblade, France
175
176
LEITAOETAL
TABLE 1 . Chromosome data
in cupped oysters.
Species
2n
karyotype
Origin
Authors
Crassostrea
С amasa (Iredale)
20
Australia
Menzel, 1968
С. angulata (Lamarck)
20
Portugal
Menzel, 1968
20
10m*
France (Barfleur)
Thiriot-Quiévreux, 1984
С. belcheria (Sowerby)
20
10 m-sm
Japan
leyama & Inaba, 1974
С. corîeziensis (Hertiein)
20
7m-3sm*
Mexico
Rodriguez-Romero et a!, 1979a
С. gigas (Th un berg)
20
8m-2sm
USA
Ahmed & Sparks. 1967
20
10m*
France (Barfleur)
Thiriot-Quiévreux, 1984
С. glomerata (Gould)
20
West Pakistan
Ahmed, 1973
С. gryphoides (Scholteim)
20
West Pakistan
Ahmed, 1973
С. /reda/e/ (Faustino)
20
Philippines
Menzel, 1968
С. rhizophorae (Guilding)
20
5m-5sm*
Mexico
Rodriguez-Romero et al., 1979b
20
8m-2sm*
Venezuela
Marquez, 1992
С. 'Tivularis (Gould)"
20
West Pakistan
Ahmed, 1973
Syn. C. ariakensis (Fujita)
20
lOm-sm
Japan
leyama, 1975
С sikamea (Amemiya)
20
West Pakistan
Ahmed, 1973
(Kumamoto variety of
С gigas)
С virginica (Gmeiin)
20
6m-4sm*
East coast USA
Longwell et al., 1967
20
6m-4sm*
Mexico
Rodriguez-Romero et al., 1978
20
6m-4sm*
Venezuela
Marquez, 1992
Saccostrea
S. commercialis (Iredale &
20
Australia
Menzel, 1968
Roughley)
S. cucullata (Born)
20
10m*
India
Goswami, 1992
S. echinata (Quoy &
20
lOm-sm
Japan
leyama & Inaba, 1974
Gaimard)
S. mordax (Gould)
20
lOm-sm
Japan
leyama & Inaba, 1974
2n; diploid chromosome number
*: after chromosome measurements
m: metacentric; sm: submetacentric
were studied in six species of cupped oysters:
Crassostrea angulata, С sikamea. C. vir-
ginica, С ariakensis, С gasar, and Sac-
costrea commercialis. These species originat-
ing from different areas were imported and
reared in common quarantine facilities. Com-
parative karyologicai analysis was made with
reference to C. gigas (Thiriot-Quiévreux,
1984, and unpublished data, 1997).
MATERIALS AND METHODS
Species Studied
Five cupped oyster species of the genus
Crassostrea and one of the genus Saccostrea
were studied, none native to Europe. Cras-
sostrea gigas and С angulata have been in-
troduced into the natural environment for
decades (Grizel & Héral, 1991) or centuries
(Boudry et al., 1998) respectively. The other
species were recently imported into France as
part of a genetic resources research program.
They have been strictly confined to the quar-
antine facilities of the IFREMER hatchery in
La Tremblade, Charente-Maritime, France,
according to international recommendations.
All the oysters studied were reared in the
same environmental conditions for at least
three months before sampling. The С angu-
lata oysters studied originate from the Rio
Sado estuary, Setubal, Portugal. The taxo-
nomic status of these oysters was assessed
using mitochondrial DNA markers as de-
scribed in Boudry et al. (1998). Crassostrea
sikamea were imported from Bodega Marine
Laboratory, University of California, USA.
Their taxonomic status was confirmed using
mitochondrial DNA markers as described in
Banks et al. (1993). Crassostrea virginica
were imported from a wild stock located in
Shippagan, New Brunswick, Canada. Cras-
sostrea ariakensis ("C. rivularis." auctt.) were
imported from the Shellfish Research Labora-
tory, Rutgers State University, New Jersey.
USA. This species was introduced from
Japan into the Northwest waters of the USA.
and its aquaculture potential has been re-
cently reviewed by Langdon & Robinson
KARYOLOGY OF CUPPED OYSTERS
177
(1996). Mangrove oysters, С gasar, were im-
ported from a wild stock located in Kafountine,
Casamance, Senegal. Saccostrea commer-
cialis were collected from the wild at Port
Stephens, New South Wales, Australia.
Because of the low number of animals
available, only two animals from each species
(except three of C. sikamea) were used for
this study.
Chromosome Preparations
Oysters were incubated for 7 h with 0.005%
colchicine in sea water. The gills were then dis-
sected out and treated for 30 min in 0.9%
sodium citrate in distilled water. The material
was then fixed in a freshly prepared solution of
absolute ethanol and acetic acid (3:1), with
three changes of 20 min duration each. Slide
preparation was made using an air-drying
technique (Thiriot-Quiévreux & Ayraud, 1 982).
For conventional karyotypes, slides were
stained directly with Giemsa (4%, pH 6.8) for
10 min. Photographs of suitable mitotic meta-
phases were taken with a Zeiss III photomi-
croscope, and after karyotyping, chromosome
measurements of ten metaphases in each
species were made with a digitizer table
(Summa Sketch II) interfaced with a Macin-
tosh. Data analysis was performed with an
Excel macro program. Terminology relating to
centromere position follows that of Levan et al.
(1964). NORs were silver-stained directly on
unstained slides using the technique of How-
ell & Black (1980), modified by Gold & Ellison
(1982).
Statistical Analysis
In order to evaluate the relationships be-
tween the six species studied here and С
gigas, a principal component analysis (PCA)
was carried out. The data set is a matrix of 70
objects, that is, ten metaphases in seven
species described by the centromeric index
values of ten chromosome pairs. Means of
centromeric index values for each species
were considered as supplementary objects
and were projected in the PCA space. The po-
sition (i.e., component score) of the ten
metaphases around this mean point gives in-
formation of the scattering of each species.
Their correlations give a criterion of their ex-
planation by the PCA axes considered. As a
second step, a hierarchical clustering analysis
(HCA) was performed between the species
described by their component scores on the
first four axes of the PCA, using the Ward ag-
glomeration method (Ward, 1963). This clus-
tering offers the possibility of representing the
distances between species by a dendogram.
PCA and HCA were computed with the SPAD
software (CESIA) (Lebart et al., 1995).
RESULTS
The results obtained for each species are
summarised in Table 2.
Crassostrea angulata
The karyotype (Fig. 1 A, Table 3) consists of
nine metacentric and one submetacentric (no.
8) chromosome pairs. Ag-NORs were found
terminally on the metacentric pair 10. The two
homologous chromosomes showed hetero-
morphism involving apparent NOR activity.
The most frequent case (69%) was one silver-
stained NOR chromosome (Fig. 2A).
Crassostrea sikamea
The karyotype (Fig. 1 B, Table 3) shows nine
metacentric and one submetacentric (no. 6)
chromosome pairs. Ag-NORs were found ter-
minally on the metacentric pairs nos. 9 and 1
(Fig. 2B). A variable number of one to three
Ag-NORs was observed. 54% of the silver
stained metaphases only showed NORs on
pair 10. The most frequent case (61%) was
one silver-stained NOR chromosome in pair
10.
Crassostrea virginica
The karyotype (Fig. 1С, Table 3) has eight
metacentric and two submetacentric (nos. 4
and 8) chromosome pairs. Ag-NORs were
found terminally on the short arms of meta-
centric pairs nos. 1 and 5 (Fig. 2C). A variable
number of one to three Ag-NORs was ob-
served. The most frequent case (52%) was
one silver-stained NOR chromosome in pair 1
and in pair 5.
Crassostrea ariakensis
The karyotype (Fig. 1 D, Table 3) consists of
eight metacentric and two submetacentric
(nos. 4 and 8) chromosome pairs. Ag-NORs
were found terminally on the metacentric
pairs 9 and 10. A variable number of one to
three Ag-NORs was observed (Fig. 2D). 68%
of the silver stained metaphases showed Ag-
NORs only on pair 10.
Crassostrea gasar
The karyotype (Fig. IE, Table 3) includes
six metacentric and four submetacentric (nos.
178
LEITAOETAL.
TABLE 2. Summary of karyological data of the six cupped oysters studied.
Species
No. metaphases
studied
No. karyotypes
studied
Giemsa
NOR
Giemsa
NOR
2n
Chromosome
type (no.
chromo. pairs)
m sm
No. of (haploid)
NOR-chromo-
somes
С angulata
C. sikamea
C. virginica
С ahakensis
С gasar
S. commercialis
42
32
29
30
33
34
31
62
57
46
55
35
13
15
10
17
13
13
12
8
7
20
20
20
20
20
20
1 (pair 10)
2 (pairs 9 and 10)
2 (pairs 1 and 5)
2 (pairs 9 and 10)
1 (pair 2)
2 (pairs 9 and 10)
*after chromosome measurements of 10 metaphases
m: metacentric; sm: submetacentric
2, 8, 9, and 10) chromosome pairs. Ag-NORs
were found terminally on the short arms of two
homologous chromosomes of the metacentric
pair 2 (Fig. 2E). Heteromorphism involving
NOR-size occurred in 49% of the metaphases
examined.
Saccostrea commercialis
The karyotype (Fig. IF, Table 3) shows
eight metacentric and two submetacentric
(nos. 4 and 7) chromosome pairs. Ag-NORs
were found terminally on the metacentric
pairs 9 and 10. A variable number of one to
three NORs were observed. 77% of the silver
stained metaphases showed Ag-NORs only
on pair 10 (Fig. 2F).
Comparative Karyological Analysis
Figure 3 shows ideograms constructed from
relative length and centromeric index values
(Table 3) of the six oyster species studied here
and of Crassostrea gigas. Chromosome mea-
surements of this later species were taken
from ten metaphases of animals collected at
La Tremblade in 1 997. Mean values of relative
length and centromeric index are similar to
those found in C. gigas from Barfleur (Thiriot-
Quiévreux, 1984). Crassosíreagfasar is distin-
guishable from the other species first, due to
the occurrence of four submetacentric chro-
mosome pairs. Crassostrea angulata and С
sikamea showed only one submetacentric
chromosome pair, whereas С virginica, С ari-
akensis and S. commercialis have two sub-
metacentric chromosome pairs. Crassostrea
angulata and С sikamea may be differentiated
by the different positions of the submetacentric
chromosome pair and by the Ag-NORs which
appear on pair 10 and on pairs 9 and 10 re-
spectively. Crassostrea virginica and С ariak-
ensis share a similar karyotype, but Ag-NORs
are observed in different locations (pairs 1 and
5, and pairs 9 and 10, respectively). Sac-
costrea commercialis is close to С ariakensis.
Their karyotypes differ by the position of the
second submetacentric pair and by the fre-
quencies of Ag-NORs observed on pair 10.
Crassostrea gigas has the most symmetrical
karyotype, with only metacentric chromosome
pairs.
Principal component analysis of the data
set of 70 objects (ten metaphases for seven
species described by centromeric index val-
ues of ten chromosome pairs) gives percent-
ages of variance for the first five axes of
31.74, 20.06, 12.61, 11.17 and 7.77 respec-
tively. The variance decreases progressively
from 5th axis. We have thus only considered
the information provided by the first four axes
as relevant. The 1/2 plan (Fig. 4) explains
51.80% of the variance. It shows the sepa-
rated position of С pasar (correlation with 1/2
plan of 0.98) without continuity with the other
species. The six other species overlap along
a continuum. Crassostrea gigas (correlation
of 0.87) is the most distant from this contin-
uum. Then, С ariakensis and С virginica are
very close and overlap a part of S. commer-
cialis (correlations of 0.57, 0.43 and 0.52, re-
spectively). Crassostrea sikamea (correlation
of 0.38) shows a larger scattering. Cras-
sostrea angulata is unexplained by this plan,
as shown by its correlation of 0.01. The 3/4
plan explained less of the total variance:
23.78%. There is a trend of separation be-
tween С virginica. С ariakensis and S. com-
mercialis (correlations of 0.40, 0.17 and 0.55
respectively). Crassostrea angulata and С
sikamea remain together (correlations of
0.48 and 0.55 respectively). Figure 5 shows
the dendogram of a Hierarchical Clustering
Analysis made using the information from the
в
KARYOLOGY OF CUPPED OYSTERS 1 79
1 2 3 4 5
6 7 Я 8 9 10
;^K M )!>t ÍS U
>X XX KU At KI
9(
С
II a:i fiA *,, ¿ai
^ix st ¿л ï»
»
JIK >1х p%jr^ /^êy. .-^
'>;к .'^'■(^ лх г./-
FIG. 1 . Giemsa-stained karyotypes of six cupped oysters. A: Crassostrea angulata. B: Crassostrea sikamea.
C: Crassostrea virginica. D: Crassostrea ariakensis. E: Crassostrea gasar F: Saccostrea commercialis. Ar-
rows show submetacentric chromosome pairs. Scale bar = 5 цт.
180
LEITAOETAL.
TABLE 3. Chromosome measurements and classification in ten cells of six cupped oyster species.
Chromosome
pair No.
Relative
length
Arm ral
tio
Centromei
ric index
Chromosome
Type
Mean
SD
Mean
SD
Mean
SD
С angulata
1
12.81
0.91
0.79
0.08
43.79
2.50
m
2
11.34
0.41
0.84
0.09
45.22
2.45
m
3
10.75
0.41
0.83
0.09
45.02
2.65
m
4
10.33
0.63
0.64
0.09
38.46
2.89
m
5
10.12
0.74
0.82
0.06
44.79
1.86
m
6
9.82
0.60
0.62
0.07
38.01
2.56
m
7
9.53
0.82
0.88
0.07
46.47
1.94
m
8
9.25
0.68
0.59
0.07
36.84
3.01
sm
9
8.98
0.66
0.68
0.12
40.07
3.83
m
10
7.08
0.67
0.75
0.10
42.36
3.44
m
C. sikamea
1
12.40
1.17
0.85
0.17
45.24
4.35
m
2
11.28
0.59
0.77
0.10
43.08
2.99
m
3
11.20
0.64
0.87
0.12
46.03
3.38
m
4
10.39
0.93
0.86
0.12
45.92
3.14
m
5
10.38
0.78
0.85
0.13
45.41
3.89
m
6
9.87
0.89
0.53
0.13
34.20
5.27
sm
7
9.50
0.76
0.79
0.13
43.72
4.36
m
8
9.27
1.19
0.79
0.08
43.52
2.48
m
9
8.61
0.89
0.82
0.20
44.28
6.39
m
10
7.10
1.05
0.95
0.18
47.75
4.98
m
С virginica
1
12.71
0.67
0.88
0.08
46.46
2.51
m
2
11.50
0.56
0.87
0.11
46.27
3.03
m
3
10.89
0.90
0.79
0.11
43.61
3.62
m
4
10.57
0.71
0.46
0.08
30.97
3.49
sm
5
10.33
0.47
0.82
0.08
44.63
2.34
m
6
9.62
0.61
0.74
0.09
42.18
3.19
m
7
9.38
0.92
0.79
0.16
43.52
5.43
m
8
9.16
0.69
0.43
0.10
29.34
4.57
sm
9
8.64
0.51
0.73
0.11
41.81
3.78
m
10
7.19
0.54
0.86
0.12
45.63
3.49
m
С ariakensis
1
12.10
0.63
0.82
0.07
44.65
2,09
m
2
11.34
0.43
0.77
0.10
42.98
3.47
m
3
10.67
0.46
0.78
0.07
43.71
2.10
m
4
10.54
0.59
0.51
0.05
33.77
2,32
sm
5
10.21
0.79
0.81
0.09
44.44
2,55
m
6
9.91
0.39
0.81
0.12
44.30
3,78
m
7
9.64
0.57
0.74
0.07
42.12
2,03
m
8
9.44
0.78
0.53
0.08
34.25
3,08
sm
9
9.09
0.90
0.79
0.17
43.41
5,00
m
10
7.07
0.43
0.76
0.12
42.64
3.86
m
С gfasar
1
11.36
0.63
0.79
0.10
43.64
3.11
m
2
11.19
0.74
0.38
0.05
27.52
2.34
sm
3
11.04
0.41
0.85
0.08
45.62
2.47
m
4
10.62
0.76
0.62
0.07
37.80
2.31
m-sm
5
10.54
0.51
0.90
0.09
46.95
2.45
m
6
9.97
0.43
0.86
0.08
45.77
2.33
m
7
9.64
0.54
0.84
0.12
45.21
3.54
m
8
9.54
0.64
0.45
0.07
30.78
3.43
sm
9
8.82
0.60
0.41
0.06
28.65
3.01
sm
10
7.28
0.56
0.39
0.08
27.89
3.78
sm
KARYOLOGY OF CUPPED OYSTERS
181
TABLES. (Continued)
Relative
length
Arm ratio
Centromeri
С index
ChromosomG
Ciiromosome
pair No.
Mean
SD
Mean
SD
Mean
SD
Type
S. commercialis
1
13.41
1.07
0.81
0.08
44.41
2.45
m
2
12.39
1.04
0.78
0.06
43.58
1.77
m
3
10.84
0.56
0.81
0.09
44.50
2.68
m
4
10.36
0.70
0.44
0.05
30.48
2.29
sm
5
9.89
0.74
0.80
0.08
44.07
2.58
m
6
9.82
0.66
0.81
0.11
44.32
3.48
m
7
9.13
0.98
0.48
0.07
32.24
2.98
sm
8
9.11
0.63
0.78
0.11
43.42
3.56
m
9
8.44
0.59
0.81
0.10
44.34
3.01
m
10
6.60
0.63
0.79
0.12
43.19
3.76
m
m: rnetacentric; sm: submetacentric
first four axes of the PCA. Crassostrea gasar
appears clearly separated from the other
species. Then, two clusters are differentiated,
one with the grouping of С virginica, С ariak-
ensls and S. commercialis. and the other with
two close species С angulata, C. sikamea
and C. gigas at a higher distance.
DISCUSSION
Our chromosome study of these six cupped
oyster species confirms the diploid chromo-
some number of 2n = 20 found up to now in all
cupped oysters examined (Table 1).
The karyotype of C. angulata differs from
that described in the animals reared at
Barfleur (Thiriot-Quiévreux, 1984). This differ-
ence could be due to the origin of animals.
Samples in this study came from the Bay of
Setubal, Portugal, and are considered as pure
C. angulata (Boudry et al., 1 998). The origin of
samples in the Barfleur study in unknown.
The karyotype of С virginica. showing two
submetacentric chromosome pairs, differs
from those with four submetacentric chromo-
some pairs described by Longwell et al.
(1967), Rodriguez-Romero et al. (1978), and
Marquez (1992). However, the position of
these submetacentric chromosome pairs is
different between these authors. Genetic dis-
continuity has been observed in this American
oyster along the Atlantic coast and the Gulf of
Mexico (Buroker, 1983; Reeb & Avise, 1990;
Hare & Avise, 1 996). The origin of our animals
is close to those studied by Longwell et al.
(1967). Therefore, the karyological variation
observed could be due either to the effect of
acclimation or to differences in the technique
(e.g., different concentrations of colchicine:
0.02% in the Longwell et a!. 1967 study and
0.005% in this study). Karyotypes made by
the same scientist, with the same techniques
carried out within a short period of time give a
more valid comparison than karyotypes made
by different authors.
Karyotypes of С sikamea, С ariakensis. C.
gasar 3.^0, S. commercialis are first described
here.
Our observations on Ag-NORs are original
in the six species studied. In С virginica,
Longwell & Stiles (1996) suggested that NOR
sites could be located on the secondary con-
striction observed on the longest metacentric
chromosome pair. Our results confirm the lo-
cation of Ag-NORs on this pair 1 , but another
Ag-NOR was observed on pair 5. Heteromor-
phism involving apparent NOR activity and
NOR-size is a common phenomenon in bi-
valves (Thiriot-Quiévreux & Insua, 1992;
Insua et al., 1994; Martinez-Exposito et al.,
1994). However, the number of Ag-NORs,
their chromosomal location and their position
within karyotypes are considered as species-
specific characters (Sumner, 1990). In this
study, the majority of species showed Ag-
NORs on pair 9 or pairs 9 and 10, in a fre-
quency that varies according to the species
considered. The position of NORs was differ-
ent in С virginica and С gasar. Ag-NORs al-
lowed the separation of C. angulata and C.
sikamea. and of С virginica and C. ariakensis
which have similar karyotypes.
Comparative karyological analysis (Figs.
3-5) highlights the isolation of C. gasar. Then
two clusters are separated. The first cluster
consists of C. gigas. С angulata and C.
sikamea. in which C. gigas, with the most sym-
metrical karyotype, could be considered as
plesiomorphic. Crassostrea gigas and C. an-
182
LEITÄO ET AL.
A
Il ъ
■■•'i
Щ
.^^!^
,^%^
''"'' ^■
-i^'^
.:■> Ö
tt
в
i* i^
« ' V
«%
f f
^' «-
1^
1^
Ha
С *| .
i'^.'
>iït
|l^^^
/% .»-%
aÍÍk.»
^Ik
Л^
A^
¿1^
» »
ПЛ
^
m-
i'
«
nt
*« cf
•y
^í
/t
Л Л
'1?
f ♦
■1
u
1С
II
if.
31
91
-9^ ^
FIG 2 Silver-stained karyotypes of six cupped oysters. A: Crassostrea angulata. B; Crassostrea sikamea.
C: Crassostrea virginica. D: Crassostrea ariakensis. E: Crassostrea gasar. F: Saccostrea commercialis. Ar-
rows show Ag-NORs. Scale bar = 5 цт.
KARYOLOGY OF CUPPED OYSTERS
183
C. gigas
С virginica
Xy
Шт
X
с. gasar
s. commercialis
X
X
X
Шт
gulata are often considered as a same species
(Menzel, 1974), as are С. gigas and С.
sikamea, of which the latter has sometimes
been considered as the "Kumamoto variety"
(Ahmed, 1973). Recent molecular genetic
studies have displayed differences between
C. gigas and С sikamea (Banks et al., 1994),
and between С gigas and C. angulata
(Boudry et al., 1998; Ó Foighil et al., 1998).
Our karyological study confirms these genetic
differences. The second cluster put together
C. ariakensis. С virginica and S. commer-
cialis. Molecular phylogenetics of cupped oys-
ters (Littlewood, 1994) distinguished two lin-
eages: (1) С gigas. С belcheri. "C. rivularis"
(= С ariakensis) and (2) С virginica, C. rhi-
zophorae and S. commercialis. Ó Foighil et
al. (1995) using mitochondrial 16S ribosomal
gene sequences confirm a genetic divergence
between С wrg /'л/са and two Asian congeners
С gigas and С ariakensis. Ladrón de Gue-
vara et al. (1996) suggested that С. virginica
showed the most primitive karyological fea-
tures when compared with C. rhizophorae and
С corteziensis. Our study is in agreement with
the relationship between С virginica and S.
commercialis, but does not agree on the posi-
tion of C. ariakensis. Multidisciplinary ap-
proaches would help in understanding evolu-
tionary relationships of oyster taxa.
Interspecific hybridizations have been pro-
duced in cupped oyster species (Gaffney &
Allen, 1993, provide a review). Crassostrea
gigas is known to hybridize with С angulata,
С sikamea and, rather less successfully, with
С ariakensis. These observations are in
agreement with our study. Looking at kary-
ological features, С ariakensis and S. com-
mercialis would be good candidates for hy-
bhdization, although this cross has apparently
never been thed. Ag-NORs observed in С vir-
ginica isolate this species from the others.
This could explain inviability of hybrids of C.
gigas and С ariakensis with С virginica
(Allen & Gaffney, 1 993). In the future, it will be
of great interest to study the mitotic and mei-
otic chromosomes of interspecific hybrids
such as С gigas x С sikamea or С gigas x
С ariakensis and their backcross offspring.
ACKNOWLEDGMENTS
Chromosome pair
FIG. 3. Ideograms of seven cupped oysters con-
structed from relative length and centromeric index
values. White chromosome: metacentric, grey chro-
mosome: submetacentric. Circles indicate Ag-
NORs, dark circles the most frequent case.
This work was supported by the CNRS
(URA 2077), IFREMER, and by the Région
Poitou-Charentes (Convention RPC-R-57),
the French-Portuguese cooperation (no. 158
CI ), and a research training project (FAIR GT
97-3599). We are very grateful to S. K. Allen,
184
LEITAOETAL
3 . о
2.0 -
1.0 -
О
о
о 0.0
-1.0-
2 .0
3.0 -2.0 -1.0 0.0 1.0
Axis 1 (31.74%)
2 . о
3 . о
4. о
FIG. 4. 1/2 plan determined by Principal Component Analysis of chromosome data. Small characters repre-
sent active objects, large characters indicate the mean for each species. A line is drawn around each species
to show the dispersion within species. AN: Crassotrea angulata. CO: Saccostrea commercialis. GA: Cras-
sostrea gasar. Gl: Crassostrea gigas. Afí: Crassostrea ariakensis. SI: Crassostrea sikamea. VI: Crassostrea
virginica.
VI
AR
CO
GI
AN
SI
GA
FIG. 5. Hierarchical Clustering Analysis showing the distances between the seven species from the first four
axes of the PCA. AN: Crassostrea angulata. CO: Saccostrea commercialis. GA: Crassostrea gasar. Gl: Cras-
sostrea gigas, AR: Crassostrea ariakensis. SI: Crassostrea sikamea. VI: Crassostrea virginica.
KARYOLOGY OF CUPPED OYSTERS
185
A. Mallet, W. Borgeson, F. Noble, J. Mazurié
for supplying live oysters. We thank S. Heur-
tebise and S. Sabini for excellent technical as-
sistance, V. Thiriot for collaboration in Figure
3 and H. McCombie for advice on the English.
LITERATURE CITED
AHMED. M.. 1973. Cytogenetics of oysters. Cytolo-
gia. 38: 337-346.
AHMED, M. & A. K. SPARKS. 1967, A preliminary
study of chromosomes of two species of oysters
{Ostrea lurida and Crassostrea gigas). Journal of
Fisheries Research Board Canada. 24: 2155-
2159.
ALLEN. S. K. & P. M. GAFFNEY, 1993, Genetic
confirmation of hybridization between Cras-
sostrea gigas (Thunberg) and Crassostrea rivu-
laris (Gould). Aquaculture. 113: 291 -300.
ALLEN. S. K., P. M. GAFFNEY. J. SCARPA & D.
BUSHEK. 1993, Inviabie hybrids of Crassostrea
virginica (Gmelin) and С rivularis (Gould) and С
gigas (Thunberg). Aquaculture. 113: 269-289.
AMEMIYA. С T & J. R. GOLD. 1990. Cytogenetic
studies in North American minnows (Cyprinidae).
XVII. Chromosomal NOR phenotypes of 12 spe-
cies, with comments on cytosystematic relation-
ship among 50 species. Hereditas. 112: 231-
247.
BANKS. M. A., D. HEDGECOCK & С WATERS.
1993. Discrimination between closely related Pa-
cific oyster species {Crassostrea) via mitochon-
drial DNA sequences coding for large subunit
rRNA. Molecular Mahne Biology and Biotechnol-
ogy. 2{3): 129-136.
BANKS. M. A.. D. J. McGOLDRICK, W. BORGE-
SON & D. HEDGECOCK. 1994. Gametic incom-
patibility and genetic divergence of Pacific and
Kumamoto oysters. Crassostrea gigas and С
sikamea. Marine Biology. 121 : 127-135.
BOUDRY P.. S. HEURTEBISE. B. COLLET F
CORNETTE & A. GERARD, 1998, Differentiation
between populations of the Portuguese oyster.
Crassostrea angulata (Lamarck) and the Pacific
oyster, Crassostrea gigas (Thunberg) revealed
by mtDNA RFLP analysis. Journal of Expehmen-
tal Marine Biology and Ecology 226: 279-291 .
BUROKER. N. E.. 1983. Population genetics of the
American oyster Crassostrea virginica along the
Atlantic coast and the Gulf of Mexico. Mahne Bi-
ology 75: 99-^^2.
CLABBY C, U. GOSWAMI, F FLAVIN, N. P.
WILKINS. J. A. HOUGHTON & R. POWELL.
1996. Cloning, characterisation and chromoso-
mal location of a satellite DNA from the Pacific
oyster, Crassostrea gigas. Gene. 168: 205-209.
GAFFNEY P M. & S. K. ALLEN, 1993, Hybridiza-
tion among Crassostrea gigas: a critical review.
Aquaculture. 116: 1-13.
GOLD. J. R. & J. R. ELLISON, 1982. Silver staining
for nucleolar organizing regions of vertebrate
chromosomes. Stain Technology. 58: 51 -55.
GOSWAMI, U.. 1992. Cytogenetical studies in Sac-
costrea cucullata from Goa waters with special
reference to chromosomal polymorphism. La
Kromosomo II, 66: 2229-2239.
GRIZEL, H. & M. HÉRAL, 1991. Introduction into
France of the Japanese oyster {Crassostrea
gigas). Journal du Conseil International pour l'Ex-
ploration de la Mer 47: 399-403.
GUO, X. & S. K. ALLEN. 1997. Fluorescence in situ
hybridization of vertebrate telomere sequence to
chromosome ends of the Pacific oyster. Cras-
sostrea gigas Thunberg. Journal of Shellfish ¡Re-
search. 16: 87-89.
HARE M. R & J. С AVISE, 1996, Molecular genetic
analysis of a stepped multilocus dine in the
American oyster {Crassostrea virginica). Evolu-
tion. 50: 23Q5-23^ 5.
HOWELL, W. M. & D. A. BLACK. 1982. Controlled
silver-staining of nucleolus organizer regions with
a protective colloidal developer: a 1 -step method.
Experientia. 36: 1014-1015.
lEYAMA. H., 1975. Chromosome numbers of three
species in three families of Pteriomorpha. Venus.
Japanese Journal of Malacology. 34: 26-32.
lEYAMA, H., 1990. Chromosomes of the oysters,
Hyotissa Imbhcata and Dendostrea folium (Bi-
valvia: Pteriomorpha). Venus. Japanese Journal
of Malacology 49: 63-68.
lEYAMA, H. & A. INABA. 1974. Chromosome num-
bers of ten species in 4 families of Pteriomorpha
(Bivalvia). Venus. Japanese Journal of Malacol-
ogy 33: 129-137.
INSUA, A. & С THIRIOT-OUIEVREUX, 1991. The
characterization of Ostrea denselamellosa (Mol-
lusca Bivalvia) chromosomes: karyotype, consti-
tutive heterochromatin and nucleolus organizer
regions. Aquaculture. 97: 317-325.
INSUA. A. & С THIRIOT-OUIEVREUX, 1 993, Kary-
otype and nucleolus organizer regions in Ostrea
puelchana (Bivalvia: Ostreidae). The Veliger 36:
215-219.
INSUA. A.. J. PLABAT& С THIRIOT-OUIEVREUX,
1994, Comparative analysis of karyotypes and
nucleolar organizer regions in different popula-
tions of Mytilus trossulus. Mytilus edulis and My-
tilus galloprovincialis. Journal of Molluscan Stud-
ies. 60: 359-370.
LADRÓN DE GUEVARA, В., F WINKLER & С.
PALMA, 1994, Karyotype description and the po-
sition of the nucleolar organizer regions (NOR) in
the Chilean oyster Tiostrea chilensis (Philippi)
Chanley and Dinamani. Pp. 399-405, In: A. R.
BEAUMONT, ed., Genetics and evolution of aquatic
organisms. Chapman & Hall.
LADRÓN DE GUEVARA. В., F WINKLER. F RO-
DRIGUEZ-ROMERO & С. PALMA-ROJAS.
1996. Comparative karyology of four American
oyster species. The Veliger 39: 260-266.
LANGDON. С J. & A. M. ROBINSON. 1 996, Aqua-
culture potential of the Suminoe oyster {Cras-
sostrea ariakensis Fugita 1913). Aquaculture,
144:321-338.
LEBART L.. A. MORINEAU & M. PICON, 1995,
Statistique exploratoire multidimensionnelle. Ed.
Dunod. 439 pp.
186
LEITAOETAL
LEVAN. A., К. FREDGA & A. A. SANDBERG, 1 964,
Nomenclature for centromere position in chromo-
somes. Hereditas, 52; 201 -220.
LI. X. X. & J. N. HAVENHAND. 1997. Karyotype,
nucleolus organizer regions and constitutive het-
erochromatin in O. angasi (Mollusca: Bivalvia):
evidence of taxonomic relationships with Ostrei-
dae. Marine Biology. 27: 443-449.
LITTLEWOOD, D. T. J., 1994. Molecular phyloge-
netics of cupped oysters based on Partial 28S
rRNA Gene Sequences. Molecular Phylogenet-
ics and Evolution. 3(3): 221 -229.
LONGWELL, A. C, S. S. STILES & D. G. SMITH,
1 967. Chromosome complement of the American
oyster Crassostrea virginica. as seen in meiotic
and cleaving eggs. Canadian Journal of Genetic
and Cytology. 9: 845-856.
LONGWELL, A. С & S. S. STILES, 1996, Chromo-
somes, biology and breeding. Pp. 443-465, In: v.
S. KENNEDY, R. I. E. NEWELL & A. EBLE, eds.. The
Eastern Oyster: Crassostrea virginica. College
Park Maryland. Maryland Sea Grant College XVI.
MARQUEZ, E.G., 1992, Comparación de cariotipos
entre las ostras Crassostrea virginica (Gmelin), С
rhizophorae (Guilding) y Pinctada imbrícala {Roo-
ing). Carribean Journal of Science. 28: 51 -55.
MARTÍNEZ-EXPÓSITO, M. J., J. J. PASANTES &
J. MÉNDEZ, 1994, NOR activity in larval and ju-
venile mussels (Mytilus galloprovincialis Lmk).
Journal of Experimental Marine Biology and
Ecology 175: 155-165.
MENZEL, R. W., 1968, Chromosome number in
nine families of marine pelecypod mollusks. The
Nautilus, 82: 45-50.
MENZEL, R. W., 1974, Portuguese and Japanese
oysters are the same species. Journal of the
Fisheries Research Board Canada. 31 : 453-456.
NAKAMURA, H., 1985, A review of molluscan
cytogenetic information based on CISMOCH-
Computerized index system for molluscan chro-
mosomes. Bivalvia, Polyplacophora and Cepha-
lopoda. Venus. Japanese Journal of Malacology.
44: 193-225.
Ó FOIGHIL, D., P. M. GAFFNEY & T. J. HILBISH,
1 995, Differences in mitochondrial 1 6 S ribosomal
gene sequences allow discrimination among
American [Crassostrea virginica (Gmelin)] and
Asian [C. gigas (Thunberg), С ariakensis\Nak\\/a]
oyster species. Journal of Experimental Marine
Biology and Ecology 1 92: 21 1 -220.
Ó FOIGHIL. D., R M. GAFFNEY. A. E. WILBUR &
T. J. HILBISH, 1998, Mitochondrial cytochrome
oxidase I gene sequences support an Asian ori-
gin for the Portuguese oyster, Crassostrea angu-
lata. Manne Biology 131(3): 497-503.
REEB, С A. & J. С AVISE, 1990, A genetic discon-
tinuity in a continuously distributed species: mito-
chondrial DNA in the American oyster, Cras-
sostrea virginica. Genetics. 124:397-406.
RODRIGUEZ-ROMERO, F.. M. URIBE-ALCOCER
& A. LAGUARDA-FIGUERAS, 1978, Cytogenetic
study of an oyster population of the species Cras-
sostrea virginica Gmelin from the coast of
Tabasco, Mexico. Venus, Japanese Journal of
Malacology 37: 83-86.
RODRIGUEZ-ROMERO, F., M. URIBE-ALCOCER
& A. LAGUARDA-FIGUERAS, 1979a, The kary-
otype of Crassostrea corteziensis Hertlein (Mol-
lusca: Ostreidae). Anales del Centro de Ciencias
del Mar y Limnología. 6: 15-18.
RODRIGUEZ-ROMERO. F.. M. URIBE-ALCOCER,
A. LAGUARDA-FIGUERAS & M. E. DIOPOTEX-
CHONG, 1979b, The karyotype of Crassostrea
rhizophorae (GuMlng). Venus, Japanese Journal
of Malacology 38: 135-140.
RODRIGUEZ-ROMERO, F., A. LAGUARDA-
FIGUERAS, M. URIBE-ALCOCER & M. L
ROJAS-LARA. 1979c, Distribution of "G" bands
in the karyotype of Crassostrea virginica. Venus,
Japanese Journal of Malacology 38: 1 80-1 84.
SHARMA, A. K. & A. SHARMA. 1 980, Chromosome
techniques, third edition. Buttenworths. 711 pp.
SUMNER, A. T., 1990, Chromosome banding. Un-
wing Hyman Ltd. 424 pp.
THIRIOT-QUIEVREUX, C, 1984, Analyse com-
parée des caryotypes d'Ostreidae (Bivalvia).
Cahiers de Biologie Mahne. 25: 407-418.
THIRIOT-QUIEVREUX, С. & N. AYRAUD. 1982,
Les caryotypes de quelques espèces de Bivalves
et de Gastéropodes marins. Manne Biology. 70:
165-172.
THIRIOT-QUIEVREUX, C. & A. INSUA, 1992. Nu-
cleolar organizer region variation in the chromo-
somes of three oyster species. Journal of Exper-
imental Marine Biology and Ecology 157: 33-
40.
VITTURI, R., R CARBONE & E. CATALANO, 1985,
The chromosomes of Pycnodonta cochlear (PoW)
(Mollusca, Pelecypoda). Biologisches Zentral-
blatt. 104: 177-182.
WARD, J. H., 1963, Hierarchical grouping to opti-
mize an objective function. Journal of the Ameh-
can Statistics Association. 58: 236-244.
Revised ms. accepted 8 June 1998
MALACOLOGIA, 1999, 41(1): 187-195
ULTRASTRUCTURAL AND CYTOCHEMICAL STUDY OF THE KIDNEY AND
NEPHRIDIAL GLAND CELLS OF THE MARINE PROSOBRANCH MOLLUSC
NUCELLA LAPILLUS{L.) IN RELATION TO FUNCTION
Vasilis K. Dimitriadis^ & Elizabeth B. Andrews^
ABSTRACT
There are three epithelial cell types over the dorsal wall folds of kidney of Nucella lapillus (L.).
The most numerous, excretory cells, are unciliated columnar cells, characterised by the pres-
ence of a variety of large membrane-bound cytoplasmic vacuoles, some containing granular ma-
terial that gives a positive reaction for periodate-reactive carbohydrates. The second type, re-
sorptive cells, are usually cone-shaped ciliated cells, with a well-developed endocytotic canal
system and numerous vesicles in their apical cytoplasm. The apical cytoplasm contains large
clusters of periodate-reactive granules, possibly glycogen and occasional large vacuoles con-
taining some granular material. The third type, "small mucous cells," are carbohydrate-contain-
ing cells with many small mucous granules, which react positively for sulphated and carboxylated
carbohydrates showing a reticulate positive reaction.
In the "secondary" folds, which are interspersed with the primary ones, the epithelium is
cuboidal and in some places almost squamous, and most of the cells display some cilia, well-
recognizable microvilli, many pinocytotic vesicles, phagosomes and elements of the canal sys-
tem on their apices. Many mitochondria and deposits of glycogen are also observed.
The nephridial gland epithelium is composed of ciliated resorptive cells and "small mucous
cells," very similar to those in the dorsal wall folds. The significance of the intracellular presence
of carbohydrates in relation to the ability of the tissues to absorb and store sugars is discussed.
INTRODUCTION
The questions addressed by Fretter & Gra-
ham (1962) regarding the complex kidney of
the carnivorous mesogastropods and neogas-
tropods are still not adequately answered. A
series of papers by Delhaye (1974-1976)
gives a comparative systematic, primarily his-
tological, survey of a range of archaeogas-
tropods and mesogastropods, while other in-
vestigations deal with the excretory system of
neogastropods, archaeogastropods, proso-
branchs or molluscs in general (Little & An-
drews, 1977; Andrews, 1981; 1985; Taylor &
Andrews, 1987; Andrews, 1988). However,
there is little information on kidney of neogas-
tropods at the ultrastructural level.
Andrews (1981) describes the cell types
composing the kidney epithelium of the car-
nivorous neogastropod Nucella lapillus (L)
and other monotocardians, such as Littorina,
Viviparus. Assiminea. Turritella and Lunatia.
According to this study, three main types of
epithelial cell have been identified in the two
regions of the kidney folds and nephridial
gland of Nucella lapillus. only one, the ciliated
resorptive cells, being common to both and
possibly implicated in reabsorption of organic
solutes on the basis of well-developed apical
microvilli with coated vesicles at their bases.
The vacuolated excretory cells appear to
have prominent microvilli, although coated
vesicles are rarely observed.
Data for Littorina indicate that the nephridial
gland is the major resorptive site for glucose
(Andrews & Taylor, 1990), which is consistent
with ultrastructural observations showing that
the dorsal wall is also involved in uptake of
solutes. The latter is believed to be responsi-
ble for the resorption of residual glucose and
for nitrogenous excretion (Taylor & Andrews,
1987; Andrews & Taylor, 1990). Thus, one of
the goals of the present study is the investi-
gation of the intracellular presence of carbo-
hydrates in the kidney and nephridial gland
epithelium of Nucella lapillus and the compar-
ison of the results with those obtained from
other gastropods.
^Department of Genetics. Development and Molecular Biology, School of Biology, Aristotle Univ of Thessaloniki, Thessa-
loniki, 54006, Greece: vdimitr@bio.auth.gr
^Biology Division, School of Biological Sciences, Royal Holloway and Bedford New College, University of London, Egham,
Surrey, TW20 OEX
187
188
DIMITRIADIS& ANDREWS
In the present study, the ultrastructure of
the kidney and nephridial gland cells of Nu-
cella laplllus is considered and compared with
analogous ones of other carnivorous or her-
bivorous molluscs. In addition, cytochemical
tests for carbohydrate presence and acid
phosphatase activity were applied.
MATERIALSAND METHODS
Specimens of Nucella lapillus from the Uni-
versity Marine Station, Millport, Scotland,
were kept in an aquarium with circulating arti-
ficial sea water at 10°C. Some of the speci-
mens were fixed for electron microscopy and
histochemistry on arrival and others were
used within a few days.
Electron Microscopy
Samples were fixed in 3% glutaraldehyde in
Sorensen's phosphate buffer, ph 7.2 adjusted
to 1100 mOsM with 14% sucrose. Material
was embedded in TAAB resin, and thin sec-
tions were stained in 1% aqueous uranyl ac-
etate and Reynolds' lead citrate.
Cytochemistry
For carbohydrate cytochemistry, finely
minced pieces of tissues were incubated
overnight after fixation in high iron diamine
(HID) (Spicer et al., 1 978; Sannes et al., 1 979)
using a medium containing 36 mg N,N-di-
methyl-m-phenylenediamine, 6 mg of the para
isomer and 0.45 ml ferric chloride 40%, in 15
ml d water. Other pieces were incubated in low
iron diamine (LID) (Takagi et al., 1982), using
a medium containing 27 mg N,N-dimethyl-m-
phenylenediamine, 4.5 mg of the para isomer,
and 0.45 ml ferric chloride 40% in 45 ml d
water. After incubation, the tissues were post-
fixed with 2% aqueous osmium tetroxide and
embedded in Spurr's resin. Thin sections of
these specimens were stained with the thio-
carbohydrazide-silver proteinate (TCH-SP)
sequence, using a medium containing 2%
thiocarbohydrazine in 20% acetic acid, to re-
veal periodate-reactive substances. Control
tissues were incubated in 1 M MgClj in place of
LID or HID. Specimens without osmium tetrox-
ide treatment were used for the postembed-
ding pehodate-thiocarbohydrazide-silver pro-
teinate (PA-TCH-SP) method (Thiéry, 1967).
Control sections were stained without the pe-
riodate treatment.
For acid phosphatase demonstration, a
modification (Lewis & Knight, 1992) of the
method proposed by Barka & Anderson
(1 962) was applied. Glutaraldehyde was used
as a fixative, and the tissues were incubated
in a medium containing 0.2M tris/maleate as a
buffer and 0.1 M ß-glycerophosphate as sub-
strate at 37° for 15-30 min. Control sections
were taken through an identical sequence ex-
cept that the substrate is omitted from the in-
cubation medium or had 0.01 M sodium fluo-
ride added.
RESULTS
Kidney
The dorsal wall of kidney of Nucella lapillus
(L.) (Fig. 1 ) consists of two sets of folds, "pri-
mary" folds, which receive blood from a
branch of the afferent renal vein running over
the ventral wall of the kidney sac, and smaller
"secondary" folds, which receive blood from a
branch running over its dorsal wall (Fretter &
Graham, 1962).
Three different cell types comprise the ep-
ithelium of the "primary" folds. The most com-
mon, excretory cells (Fig. 2), are columnar
cells, which present a variety of excretory vac-
uoles in their cytoplasm (Figs. 4, 8).
The vacuolated excretory cells have promi-
nent microvilli (Figs. 4, 8), which are lost in
later stages when apical blebs are formed.
The most conspicuous organelles of these
cells are a series of large excretory vacuoles
containing in certain cases a concentration of
finely fibrous excretory material (Fig. 4). Due
to the presence of the large vacuoles in the
apical and mid cytoplasmic region, the ovoid
nuclei are usually displaced in their base (Fig.
4). The fibrous material of the excretory vac-
uoles in most cases gives a positive reaction
for periodate-reactive carbohydrates (Fig. 8).
In certain cases, this material reacts positively
for acid phosphatase activity (Fig. 11). Other
clusters of fibrous and/or granular periodate-
reactive material is also present in the cyto-
plasm of these cells (Fig. 8). Small vesicles
are seen in the process of fusing with excre-
tory vacuoles of varying size and may appear
as evaginations of the vacuolar membrane.
The rough endoplasmic reticulum of the ex-
cretory cells is usually restricted to a narrow
zone of apical cytoplasm, often associated
with clusters of glycogen. The Golgi com-
plexes lie close to the nuclei and are small,
ULTRASTRUCTURAL AND CYTOCHEMICAL STUDY
189
pfk skf
varv
FIG. 1. Drawing of the kidney sac of Nucella lapil-
lus, opened to show double series of folds on the
wall, based on the original in Fretter & Graham
(1962). darv, dorsal branch of afferent renal vein;
hg, hypobranchial gland; ko, kidney opening; ng,
nephridial gland; pa, posterior aorta; pfk, primary
kidney folds; skf, secondary kidney folds; te, testis;
varv, ventral branch of afferent renal nein; ve, ven-
tricle. Bar = 1 mm.
conspicuous and rarely observed, whereas
the mitochondria are positioned parallel with
the infoldings of the basal plasma membrane.
In the basal cytoplasm, the excretory cells
display long slender basal processes, which
invaginate the basal lamina and extend into
underlying blood spaces (Fig. 9). Near the
basal membrane, in the narrow cytoplasmic
spaces between the adjacent infoldings and
very close to the blood spaces clusters of per-
iodate-reactive granules, mostly glycogen are
present (Fig. 9).
The second cell type in the "primary" dorsal
folds, the ciliated resorptive cells (Fig. 5), bear
apical microvilli and cilia. Usually, they appear
to be cone-shaped, with their distended
apices bulging into the lumen in umbrella-like
fashion over the neighbouring cells (Fig. 5).
The dense cytoplasm of the resorptive cells
possesses an endocytotic canal system and
many small vesicles (Fig. 10), some near the
apical membrane being coated. Pinocytosis
occurs at the base of the microvilli and the
pinocytic vesicles are coated (Fig. 10). Pino-
cytotic vesicles are coated on the cytoplasmic
face by closely spaced, bead-like particles.
The apical membrane itself bears a fine gly-
cocalyx and microvilli 0.7-1 .2 ¡am long are in-
terspersed amongst groups of long cilia 4.5-
5 }.im long.
In the apical cytoplasm there are some
lysosomes, usually in the form of residual
bodies and occasional large vacuoles, some
of them containing a little granular material.
Also in the apical cytoplasm are prominent
large clusters of periodate-reactive particles,
mostly glycogen (Fig. 10). Similar particles
are also present throughout the cytoplasm.
The third cell type, the "small mucous
cell" (Fig. 6), is very rare and consists of small
cells with small periodate-reactive secretory
granules, or cells having discharged their se-
cretion. The granules of these cells reacted
positively for sulphated and carboxylated car-
bohydrates showing a reticulate positive reac-
tion (not shown).
The "secondary" folds of dorsal wall of kid-
ney of N. lapillus are present in troughs be-
tween "primary" folds. In the "secondary" folds
the epithelium is cuboidal and in some places
almost squamous (Fig. 7), while the presence
of fewer cilia comparing to those of the pri-
mary kidney folds makes the recognition of
the cell limits under the scanning microscope
obscure, comparing to the "primary folds"
(Figs. 2, 3). In the apical portion, the cells
show many pinocytotic vesicles, phagosomes
and elements of the canal system. Many mi-
tochondria, lysosomes and deposits of perio-
date-reactive particles are also observed. In
the "secondary" folds, the presence of the
"small mucous cells" is less obvious than in
the "primary" ones.
Nephridial Gland
The nephridial gland of N. lapillus receives
post branchial blood from the auricle. In the
epithelium of this gland, the same type of cili-
ated resorptive cell covers the surface and the
more superficial parts of the tubules (Figs. 12,
13). Unlike the resorptive cells of the dorsal
wall folds, cone-shaped forms in the nephridial
gland are very rare, the cells consisting mostly
of cuboidal form. The resorptive cells show
apical microvilli and cilia and a well-developed
endocytotic canal system with small coated
vesicles. In the apical portion, the presence of
small, empty vacuoles is usually apparent
(Fig. 13). Periodate-reactive particles, mostly
glycogen, are also located in clusters usually
in the apical cytoplasm (Fig. 14).
Dense bodies frequently lie in the apical
portion of the cells and usually exhibit an elec-
190
DIMITRIADIS& ANDREWS
FIGS. 2 & 3. Scanning electron micrographs of the apical surface of "рг1глагу" (Fig. 2) and "secondary" folds
(Fig. 3) of the kidney dorsal wall. In the "secondary" folds, the decreased presence of cilia (arrow), compared
to the "primary" folds, makes the recognition of the cell limits not well recognizable. EC, excretory cell; RC,
resorptive cell. Fig. 2 Bar = 2.5 |.im. Fig. 3 Bar = 8 цт.
tron-dense matrix. The material of most
dense bodies gives a positive reaction for per-
iodate-reactive carbohydrates, as well as for
suiphated (Fig. 15) and carboxylated carbo-
hydrates. In certain cases, membranous rem-
nants of the dense bodies display a positive
acid phosphatase reaction (not shown).
Like the dorsal wall epithelium of the kidney,
in the nephridial gland there is a second cell
type, the "small mucous cells" consisting of
small cells with small periodate-reactive se-
cretory granules giving also a reticulate posi-
tive reaction for suiphated and carboxylated
carbohydrates (not shown).
Control sections of all cytochemical tech-
niques constantly lacked reaction product.
DISCUSSION
The kidney of N. lapillus is histologically
similar to that of other monotocardians proso-
branchs (Andrews, 1981) in that the epithe-
lium is composed of distinct vacuolated ex-
cretory and ciliated resorptive cells. The
morphological and functional parameters of
the excretory and ciliated cells of Nucella
lapillus documented by the results of the pres-
ent study, such as the structure of their excre-
tory vacuoles, canal system compartments,
and basal processes, are very similar to those
of the herbivorous mesogastropod Littorina lit-
torea (Andrews, 1981; Taylor & Andrews,
1987; Andrews, 1988), albeit that Nucella is
carnivorous. A differentiation in the density of
the excretory vacuoles contents was reported
by Andrews (1981) between Nucella and her-
bivores species, being denser in Nucella than
in the other species, as well as a differentia-
tion in the colour of the lysosomes of the cili-
ated cells, being greenish in the herbivores
species studied.
The vacuolated excretory cells of the dorsal
wall folds of the kidney of Nucella, like the cor-
responding cells in Littorina (Andrews, 1988),
contain excretory vacuoles with finely fibrous
excretory material, unlike those of archaeo-
gastropods and amphibious and terrestrial
species, which contain layered concentra-
tions. It is generally accepted that the excre-
tory vacuoles are formed and increase in size
after fusion of smaller vesicles and finally are
ultimately shed in blebs of apical cytoplasm
that are nipped off from the cell membrane
(Andrews, 1981, 1985). Delhaye (1976) dem-
onstrated pinocytosis at the bases of the ex-
cretory cells of Monodonta by which the cells
abstracted ferritin injected into the blood. A
second indication for this function is the clearly
well-recognisable long slender processes that
invaginate the basal lamina of the excretory
cells of Nucella, and as in Littorina (Andrews,
1988) and Monodonta (Andrews, 1985), per-
meate the underlying blood spaces, which
suggest transport of molecules across the
lamina.
The ciliated resorptive cells comprising the
epithelium of N. lapillus display many similari-
ties to those of the ciliated resorptive cells of
the dorsal wall folds. The ciliated cells in both
epithelia were regarded as cells of the same
cell type (Andrews, 1981; Andrews & Taylor,
1990). The results of the present study show
that the ciliated cells in the kidney folds differ
from those in the nephridial gland in certain
features, for example their cone shape, where
they are interspersed with excretory cells.
However, other differences, such as the num-
ber of their apical vesicles and lysosomes,
should be attributed to the different phases of
ULTRASTRUCTURAL AND CYTOCHEMICAL STUDY 1 91
Lu
i^
''^'- .f^"'''''^k'
Lu
EV
•A
N
Lf
'jtopH
^ -
Lu
EV *
Lu
V
â^-'à^h:'- ■ 'j.l# -1 k^<
N
FIG. 4. Apical portion of excretory cells in a "primary" fold of dorsal wall of kidney showing large excretory
vacuoles (EV). Lu, lumen; N, nucleus. Bar = 4 ¡am.
FIG. 5. The apex of a ciliated resorptive cell displaying microvilli and cilia, as well as apical vesicles (aster-
isk), is protruded into the lumen. Bar = 3.5 дт.
FIG. 6. A "small mucous cell" (asterisk) is positioned in the apical portion of the epithelium, possibly just be-
fore the secretion of the cell product. EV, excretory vacuole; Lu, lumen; N, nucleus. Bar = 3 [xm.
FIG. 7. In a "secondary" fold of the kidney dorsal wall, there is a transition in the form of the cells from colum-
nar-cuboidal towards cuboidal-squamous. Db, dense body; Lu, lumen; N, nucleus; V, apical vesicle. Bar =
5 um.
192
DIMITRIADIS & ANDREWS
FIG. 8. Cytochemistry for periodate-reactive carbohydrates. Excretory cells show periodate-reactive fibrous
material in large excretory vacuoles (EV). Other "primary" fold cells show large clusters of other periodate-
reactive material in their cytoplasm (arrow). Lu. lumen. Bar = 3 цт.
FIG. 9. Cytochemistry for periodate-reactive carbohydrates. Basal cytoplasm of excretory cell. The basal
membrane infoldings form long slender cytoplasmic processes, which permeate the blood spaces (BS) and
contain periodate-reactive carbohydrates (arrow). Bar = 0.6 |.im.
FIG. 10. Cytochemistry for periodate-reactive carbohydrate. Apical cytoplasm of a ciliated resorptive cell in
the kidney of Nucella lapillus. Clusters of periodate-reactive particles, mostly glycogen, are located in the cy-
toplasm. Note the coated vesicle (arrow) and the well-stained vesicles and tubules of the canal system (small
arrows). Bar = 0.8 цт.
FIG. 1 1 . Cytochemistry for acid phosphatase. Materials in an excretory vacuole of an excretory cell react pos-
itively for enzyme presence (arrow). Mv, microvilli. Bar = 1 ¡.im.
ULTRASTRUCTURAL AND CYTOCHEMICAL STUDY
193
*
•чЧ
h2 У
■л .,_^ ^^
vr-
Db
Lu
Db
•>
,.^ N
. ^ Ы
Ы
^ 13
tv
'MV
\
Lu
Lu
V
14
'.Й^Р"^
'4'
,.■/*■
■■ . ... '. i. .t '
49P*M'
rf'Db
15
FIG. 12. Ciliated resorptive cells and a "small mucous cell" (asterisk) in the nephridial gland of Nucella lapil-
lus. Db, dense body: Lu, lumen. Bar = 3.5 ,um.
FIG. 13. Ciliated resorptive cells in the nephridial gland of Nucella /ap/7/us display many mitochondria, some
iysosomes and many apical small vesicles (arrow). Ci, cilium; Db, dense body; Mv, microvilli. Bar = 2 ¡.im.
FIG. 14. Carbohydrate cytochemistry. The cytoplasm of a ciliated resorptive cell shows abundance of perio-
date-reactive particles, mostly glycogen. Lu, lumen; V, apical vesicle. Bar = 1 .5 |im.
FIG. 15. In the nephridial gland, a dense body (Db) in the apical region of a ciliated resorptive cell react pos-
itively for sulphated carbohydrates. Lu, lumen. Bar = 4 цт.
cell activity rather, than to differences be-
tween cell types.
The apical canal system observed in the re-
sorptive cells of both kidney dorsal wall and
nephridial gland cells of Nucella lapillus is
also observed in the pericardial and kidney
cells of Scrobicularia plana (Andrews & Jen-
nings, 1993) and the left kidney of archaeo-
gastropods (Andrews, 1985), as well as in
the digestive gland of Nucella (Dimitriadis &
Andrews, submitted for publication), Lasaea
(McQuiston, 1969), Cardium (Owen, 1970),
194
DIMITRIADIS& ANDREWS
Mytilus (Owen, 1972), and Rissoa (Wigham,
1976), where it was regarded as structure re-
lated to the delivery of nutrients to the endo-
cytotic system. In the light of the recent data,
the elements of the canal system should be
regarded as endosomes, that is, the cell com-
partments, where the endocytosed material
enters the lysosomal pathway (Alberts et al.,
1994).
In the excretory cells of N. lapillus, perio-
date-reactive fibrous material is located inside
the excretory vacuoles, while in certain cases
this intravacuolar material reacts positively for
acid phosphatase. Carbohydrates, among
other substances, have been identified in the
excretory vacuoles in the right kidney of dioto-
cardians (Delhaye, 1976). The presence of a
significant amount of carbohydrates and hy-
drolases inside the excretory vacuoles proba-
bly indicates that the role of these vacuoles
should be regarded as multiple and that
they are implicated in more functions than
the transport and release of the excretory ma-
terial.
In the kidney dorsal wall of N. lapillus. clus-
ters of periodate-reactive particles, mostly
glycogen, are present in the cytoplasm of the
resorptive cells. It is well known that carbohy-
drates are the major source of energy in gas-
tropods molluscs (Livingston & De Zwaan,
1983) and that carbohydrates are largely
stored as glycogen, especially in connective
tissue. The excretory system of prosobranch
gastropods, which forms urine by a filtration
mechanism, has the capacity to resorb valu-
able organic solutes, such as glucose, from
the primary urine, limiting markedly the
amount that is lost to the exterior (Andrews &
Taylor, 1990). Glucose resorption has been
demonstrated in the abalone Haliotis rufes-
cens (Harrison, 1962) and the mesogastro-
pod Littorina littorea (Taylor & Andrews,
1987), while glucose influx to both dorsal wall
and nephridial gland of the kidney appeared
to reflect net inward transport by a saturable
Na'^-dependent carrier mechanism believed
to be responsible for reabsorption in vivo
(Taylor & Andrews, 1 987). On the other hand,
in the herbivorous mesogastropod Littorina lit-
torea, the nephridial gland cells were re-
garded as sites implicated in resorption of or-
ganic solutes on the basis of well-developed
apical microvilli with coated vesicles at their
bases (Andrews & Taylor, 1990). Also in Litto-
rina. Delhaye (1974b) demonstrated the ca-
pacity of the nephridial gland cells to remove
Indian ink, iron saccharide and ferritin from
the blood, like the resorptive cells of the dor-
sal wall, while their ability to accumulate pH]
glucose has been shown in vitro (Taylor & An-
drews, 1990). In N. lapillus. where the mor-
phological features of the nephridial gland
cells are very similar to those in Littorina, an
equivalent absorptive function of the cells is
very possible. Thus, the presence of large
amounts of periodate-reactive particles in the
resorptive cells of kidney folds and nephridial
gland is a strong indication that these cells ab-
sorb carbohydrates and store, at least a part
of them, in their cytoplasm.
The "small mucous cells" observed in the
examined tissues of N. lapillus exist also in
the digestive gland of the same gastropod,
where they also reacted positively for com-
plex carbohydrates (Dimithadis and Andrews,
submitted for publication). Probably this cell
type is related to the mucoid cells described in
the nephridial gland of other monotocardian
prosobranch gastropods (Andrews, 1981), as
well as in the kidneys of freshwater mesogas-
tropods (Andrews, 1988), where they were
specialised as cells related to the resorptive
mechanism of the epithelium. The secretion
product of these mucoid cells is not stained by
alcian blue and is probably a mucoprotein or
neutral mucopolysaccharide.
In conclusion, by using light and electron mi-
croscopic observations in combination with
cytochemical characterization, the present
study provides information about the structure
and function of the kidney and nephridial gland
cells. Additional information, especially by the
use of histochemistry in cryosections and im-
munocytochemistry, is necessary for the bet-
ter understanding of the fine structure and
physiology of the studied interesting cell types.
ACKNOWLEDGEMENTS
This work was undertaken during the sab-
batical leave of the first of the authors at the
Royal Holloway and Bedford New College.
The authors gratefully thank the staff of the
Electron Microscope Unit of the College for
their help and invaluable assistance in the
preparation of some of the material.
LITERATURE CITED
ALBERTS, В., D. BRAY, J, LEWIS, M. RAFF. K.
ROBERTS & J. D. WATSON, 1994, Molecular bi-
ology of the cell. Third edition. Garland Publish-
ing, Inc, New York and London, Pp. 61 8-631 .
ULTRASTRUCTURAL AND CYTOCHEMICAL STUDY
195
ANDREWS, E. В., 1981, Osmoregulation and ex-
cretion in prosobranch gastropods. Part 2: Struc-
ture in relation to function. Journal of Molluscan
Studies 47:248-289.
ANDREWS, E. В., 1985, Structure and function in
the excretory system of archaeogastropods and
their significance in the evolution of gastropods.
Philosophical Transactions of the Royal Society
of London. В 310: 383-406.
ANDREWS, E. В., 1988, Excretory systems of mol-
luscs. Pp. 381-448, E. R. TRUEMAN & M. R.
CLARKE, eds. In: The Mollusca. Vol. II Form and
function. Academic Press, Orlando.
ANDREWS, E. B. & K. H. JENNINGS, 1993, The
anatomical and ultrastructural basis of primary
urine formation in bivalve molluscs. Journal of
Molluscan Studies. 59: 223-257.
ANDREWS, E. B. & P. M. TAYLOR, 1990, Reab-
sorption of organic solutes in some marine and
freshwater prosobranch gastropods. Journal of
Molluscan Studies. 56: 147-162.
BARKA, T & P. G. ANDERSON, 1962, Histochemi-
cal methods for acid phosphatase using hexazo-
nium parasolinin as coupler. Journal of Histo-
chemistry and Cytochemistry. 10: 141-153.
DELHAYE, W., 1974a, Histophysiologie comparée
des organes excréteurs chez quelques Neritacea
(Mollusca-Prosobranchia). Archives de Biologie.
85: 235-262.
DELHAYE, W., 1974b, Histophysiologie comparée
du rein chez les mesogasteropodes Archaeotae-
nioglossa et Littorinoidea (Mollusca-Prosobran-
chia). Archives de Biologie. 85: 461 -507.
DELHAYE. W., 1974c, Contribution a l'étude des
glandes pedieuses de Pomatias elegans (Mol-
lusque. Gasteropode. Prosobranche). Annales
de Sciences Naturelle (Zoologie), 16: 97-110.
DELHAYE, W., 1975, Histophysiologie comparée
du rein chez les mesogasteropodes Rissoidea et
Cerithoidea (Mollusca-Prosobranchia). Archives
de Biologie. 86: 355-375.
DELHAYE, W., 1976, Histophysiology comparée
des organes rénaux chez les Archaeogas-
teropodes (Mollusca-Prosobranchia). Cahiers de
Biologie Manne. 17: 305-322.
DIMITRIADIS, V. K. & E. B. ANDREWS, Ultrastruc-
tural and cytochemical study of the digestive
gland cells of the marine prosobranch mollusc
Nucella lapillus (L.) in relation to function. Sub-
mitted for publication.
FRETTER, V. & A. GRAHAM, 1962, British proso-
branch molluscs. London, Ray Society.
HARRISON, F. M., 1962, Some excretory pro-
cesses in the abalone, Haliotis rufescens. Jour-
nal of Experimental Biology 39: 1 79-1 92.
LEWIS, P. R. & D. P. KNIGHT 1992, Cytochemical
staining method for electron microscopy. Pp.
174-178, in A. M. GLAVERT, ed.. In: Practical
methods in electron microscopy. Elsevier, Am-
sterdam, London, New York, Tokyo.
LITTLE, С & E. B. ANDREWS, 1977, Some as-
pects of excretion and osmoregulation in as-
simineid snails. Journal of Molluscan Studies. 43:
263-285.
LIVINGSTONE, D. R. & A. DE ZWAAN, 1983, Car-
bohydrate matabolism of gastropods. Pp.
1 77-242, in P. w. HOCHCHKA, ed.. The Mollusca 1.
Metabolic biochemistry and molecular biome-
chanics. New York and London, Academic Press.
McQUISTON, R. W., 1969, Cyclic activity in the di-
gestive diverticula of Lasaea rubra (Montagu) (Bi-
valvia: Eulamellibranchia). Proceedings of the
Malacological Society of London. 38: 483-492.
OWEN, G., 1 970, The fine structure of the digestive
tubules of the marine bivalve Cardium edule.
Philosophical Transaction of the Royal Society of
London. В 258: 245-260.
OWEN, G., 1972, Lysosomes, peroxisomes and bi-
valves. Science Progress. 60: 299-318.
SANNES R L., S. SPICER & T KATSUYAMA,
1979, Ultrastructural localization of sulfated and
complex carbohydrates with a modified iron di-
amine procedure. Journal of Histochemistry and
Cytochemistry 27: 1 1 08- 1 1 1 1 .
SPICER, S. S., J. H. HARDIN & M. E. SETSER,
1978, Ultrastructural visualization of sulfated
complex carbohydrates in blood and epithelial
cells with the high iron diamine procedure. Histo-
chemical Journal. 10: 435-452.
TAYLOR, R M. & E. B. ANDREWS, 1987, Glucose
reabsorption by the prosobranch gastropod Litto-
hna littorea (L). Journal of Expehmental Marine
Biology and Ecology 1 08: 99- 1 1 1 .
TAKAGI M., T R. PARMLEY S. S. SPICER, F. R.
DENYS & M. E. SETSER, 1982, Ultrastructural
localization of acidic glycoconjugates with the low
iron diamine method. Journal of Histochemistry
and Cytochemistry 30: 471 -476.
THIÉRY J. P., 1 967, Mise en evidence des polysac-
charides sour coupes fines en microscopie élec-
tronique. Journal of Microscopie. 6: 381 -1 01 .
WIGHAM, G. D., 1976, Feeding and digestion in the
marine prosobranch Rissoa parva (da Costa).
Journal of Molluscan Studies. 42: 74-94.
Revised ms. accepted 1 February 1999
MALACOLOGIA, 1999, 41(1): 197-208
ENERGETICS OF THE RED SLUG ARION RUFUS (GASTROPODA) AND OF THE
GASTROPOD COMMUNITY IN A BEECH FOREST ON LIMESTONE
Anne Theenhaus & Matthias Schaefer
Institut für Zoologie and Anthropologie, Abteilung Ökologie. Berliner Strasse 28,
D-37073 Göttingen, Germany; theenhau@mail.uni-mainz.de
ABSTRACT
The energy budget of the red slug Arion rufus (L.) in a mull beech forest on limestone in Lower
Saxony, Germany, was calculated. Somatic growth, reproduction, secretion of mucus, respira-
tion, excretion of faecal materials, and assimilation efficiency were measured experimentally.
The consumption was calculated from the excretion of faecal materials, using the assimilation ef-
ficiency.
Mean biomass of the population of A. rufus in the forest studied was 0.2 g d wt m~^. The av-
erage density of slugs that grew to maturity and reproduced was one slug per 3 m^. The annual
consumption of the population of A. rufus in the beech forest studied was equivalent to 361 .9 kJ
m"^, the investment into somatic growth to 52.5 kJ m"^, the reproduction to 3.6 kJ m"^, the res-
piration to 27,6 kJ m^^, and the excretion of faecal materials to 79.7 kJ m"^. About 105.3 kJ m"^
were invested into the secretion of mucus, which was 65% of the production. The considerable
investment of gastropods in locomotion probably limits the range of conditions under which pop-
ulations can survive. Ecosystem characteristics and slug qualities that allow the high slug bio-
mass in the beech forest ecosystem were discussed.
The population of A. rufus and the whole gastropod community consumed 0. 1 5% and 2.4% of
the annual leaf litter input, 3.3% and 26% of the annual production of green plant material, and
0.24% and 1.7% of the annual net primary production (NPP), respectively. The direct contribu-
tion of the gastropod community to decomposition processes therefore was small. The con-
sumption of green plant material by the gastropod community was similar to that annually in-
gested by phytophages (Curculionidae and lepidopteran larvae) in the forest studied.
Key words: Arion rufus. gastropod community, energy budget, mucus, beech forest.
INTRODUCTION
In the present investigation, we calculate
the energy budget of the red slug Arion rufus
(L.) in order to understand life history strate-
gies of this slug species in the forest ecosys-
tem. We are especially interested in the role of
mucus in the energy flux of A. rufus. In previ-
ous studies, the energy budget of A. rufus and
A. ater\r\ beech forests was calculated (Stern,
1 969; Jensen, 1 975), however the authors did
not quantify the secretion of mucus. Since
mucus is a major component of gastropod
production (Denny, 1980; Davieset al., 1990),
the omission of the mucus term caused a con-
siderable underestimation of production.
Arion rufus feeds on dead organic material,
living plants, fungi and carcasses and there-
fore contributes directly to decomposition
processes (Jennings & Barkham, 1976;
1979a). In the present investigation, special
attention was given to the consumption data
of the population of A. rufus, because with
these data the impact of A. rufus on its food
and the direct contribution to decomposition
processes could be evaluated. Furthermore,
we extrapolated the consumption data of A.
rufus to the gastropod community in order to
assess the role of gastropods in the forest
ecosystem. Schaefer (1990) constructed the
energy budget of the heterotrophic subsystem
of the forest investigated in the present study.
He found that the contribution of the gastro-
pod community to the energy and material
turnover was low. However, the calculations
were made without data on A. rufus, because
biomass data of A. rufus were not available.
Presumably, this omission resulted in a dras-
tic underestimation of the role of the gastro-
pod community in nutrient cycling in the forest
ecosystem studied. Arion rufus has a lifespan
of little more than a year (Laviolette, 1950;
197
198
THEENHAUS & SCHAEFER
Smith, 1966), and since the body weight of
mature specimens is high, the population of
A. rufus builds up a high biomass during a
short time period. Therefore, we assume that
the energy flow through the population of A.
rufus is high and that A. rufus makes a signif-
icant direct contribution to plant decomposi-
tion processes.
As a result of the high production of bio-
mass of the population of A. rufus in the forest
ecosystem studied, we hypothesize that this
slug species turns high amounts of living
plants and dead organic material into faecal
material, which is then colonised by a rich
community of microfauna (Theenhaus &
Scheu, 1996a) and mesofauna (Theenhaus,
1997). Furthermore, it is assumed that in the
forest ecosystem studied the population of A.
rufus secretes high amounts of slug mucus.
Since slug mucus contains high nutrient con-
centrations, A. rufus directly modulates the
availability of resources for microorganisms
through the secretion of mucus (Theenhaus &
Scheu, 1996b). We assume that the creation
of habitats for other species through the ex-
cretion of faecal material and the secretion of
slug mucus is of quantitative importance in
the forest ecosystem studied. Arion rufus
might therefore be called an ecosystem engi-
neer (Lawton, 1994).
MATERIALSAND METHODS
Study Site
The Göttinger Wald beech forest is situated
on a plateau of Muschelkalk at about 420 m
above sea level in southern Lower Saxony,
Germany. The forest, which is approximately
110-125 years old, has a uniform canopy
layer, which consists almost exclusively of
beech trees {Fagus sylvatica). A shrub layer is
not developed. The herb layer is dominated
by Allium ursinum and Mercurlalls perennis.
In the years 1995 and 1996, in which the ex-
periments were conducted, mean annual tem-
perature in the litter layer was 9.0°C and
7.1 °C, respectively, and mean annual precipi-
tation was 700 mm and 567 mm, respectively.
Energy Budget of A. rufus
In the present study, we first calculated the
energy budget of a "model specimen" of A.
rufus, which is an animal that grows to matu-
rity, deposits one egg clutch, and then dies.
Based on this model specimen, the energy
budget of the population of A. rufus in the for-
est studied was estimated. The energy bud-
get was calculated using the equation given
by Petrusewicz (1967), which was modified
to include a mucus term (e.g., Edwards &
Welsh, 1982; Davies et al., 1990): С - Pg + P^
-f- P^ + R + F, where С = consumption, P =
somatic growth, P^ = reproduction, P^ =
secretion of mucus, R = respiration, and F
= faecal materials.
Defined-area traps similar to those of Fer-
guson et al. (1989) were constructed in coop-
eration with С Döring to estimate slug popu-
lations in the beech forest. The traps
consisted of a square PVC frame (1 m^ area,
30 cm high). This frame was placed onto the
forest soil to enclose slugs, which were sup-
posed to be caught during the following trap-
ping pehod. A PVC lid could be screwed
tightly on a metal rim, which was attached to
the inner edge of the frame. Twelve round
holes in the lid, which were sealed with gauze
(meshsize 1 mm) created a light regime (twi-
light) inside the traps being favourable for
slugs. To prevent slug movements in or out of
the traps, outside the traps soil was piled up
(20 cm high) and pressed tightly to the frame.
A glass jar (diameter 5 cm, 11 cm high; Bar-
ber, 1931) was imbedded into the soil inside
the traps and filled with beer (3 cm high),
which served as baiting agent. Duhng periods
of drought, the ground inside the traps was
moistened regularly. In the years 1995 and
1996, the population of A. rufus was esti-
mated using 12 traps, which were arranged in
four groups (distance 10 m) of three traps
each (distance 3-5 m; block-design; Sokal &
Rohlf, 1 995). Every two weeks, the glass jars
were emptied and refilled with fresh beer.
Trapped slugs were identified, and their ash-
free dry weight was determined. When no
slugs were captured in the traps for four
weeks, it was assumed that all slugs inside
the traps were caught, and therefore the trap-
ping period was finished. Traps were re-
arranged, and a new trapping period started.
To quantify the production of eggs, in au-
tumn 1994 and 1995 mature specimens of A.
rufus were collected in the beech forest and
transferred into the laboratory. Each egg
clutch was divided into three portions and in-
cubated at 5, 10 and 20""C, respectively. The
time between egg deposition and hatching of
juveniles was measured.
The COg production of slugs was measured
ENERGY BUDGET OF ARION RUFUS
199
titrimetrically (1 N KOH as a CO2 absorbent) at
5, 10 and 15'C for 24 h (day/night cycle of
12/12 h). Measurements were done in exper-
imental chambers, which consisted of per-
spex tubes (diameter 6 cm, 15 cm high).
Slugs were acclimated at the respective tem-
perature for 14 days prior to measurements.
To quantify the excretion of faecal materi-
als, specimens of A. rufus were collected in
the beech forest in summer following periods
of drought, in summer following wet periods,
and in autumn (three parallel measurements).
Still in the forest, each slug was placed into a
separate round, perspex vessel (12 cm diam-
eter, 9 cm height), which contained food
(leaves of Allium ursinum, Mercurialis peren-
nis. and Sambucus nigra). The vessels were
then transported into a temperature controlled
room, in which temperature and day/night
cycle were close to those of the respective
time of year. The fresh weight of each slug
and the dry weight of faecal materials, which
were excreted during the incubation by each
slug, were determined 24 h after the collection
of slugs.
To quantify the secretion of mucus, speci-
mens of A. rufus (fresh weight between 0.3
and 15.6 g) were incubated in pre-weighed
perspex vessels (see above) at defined tem-
perature and humidity conditions (5, 10, 15
and 20°C in combination with 100% RH and
10'C in combination with 75% and 55% RH,
respectively) for 2 h. The fresh weight of
mucus, which was secreted during the time of
incubation, was determined by reweighing the
vessels after the incubation and calculating
the difference in weight. Slugs were accli-
mated at the respective experimental condi-
tions (temperature, RH) for 12 h prior to use.
The assimilation efficiency of juvenile A.
rufus (fresh weight between 0.8 g and 2.2 g)
feeding on leaves of M. perennis and A.
ursinum and of mature A. rufus (fresh weight
between 7 g and 1 5 g) feeding on leaves of M.
perennis and Sambucus nigra was deter-
mined gravimetrically. Before measurement,
slugs were fed with one of the respective herbs
for three days. Then the slugs were transferred
to perspex vessels (see above), which con-
tained 7 g f wt of one of the respective herbs of
known water content, five juvenile slugs or one
mature slug per vessel (10°C, day/night cycle
of 14/10; juvenile slugs: six parallel measure-
ments; mature slugs: 15 parallel measure-
ments). After three days of incubation, the
dry weight of faecal materials, which was ex-
creted during the incubation, and the dry
weight of herb materials were determined. The
dry weight of herbs, which were eaten by slugs
during the three days of incubation, was cal-
culated as the difference before and after feed-
ing. Carbon content of faecal materials and
herbs was determined using an elemental
analyser (Carlo Erba Co., Milan). In order to
determine the change in weight of herbs with-
out the influence of slugs, nine vessels without
slugs but with herbs (three vessels with M.
perennis, A. ursinum and S. nigra, respec-
tively) were incubated for three days. The as-
similation efficiency of slugs was calculated
with the formula:
d wt food - cc food - d wt faecal materials
- cc faecal materials
d wt food ' cc food
100,
where a = assimilation efficiency, cc = carbon
content.
The conversion of data into calorific values
was accomplished with conversion factors
from the literature: 1 g d wt tissue of A. rufus
and of egg mass was equated with 20.1 kJ
(Jensen, 1975; Bless, 1978); 1 g d wt mucus
was equated with 18.8 kJ (Calow, 1974;
Richardson, 1975); 1 I CO2 was equated with
21 .8 kJ. This conversion factor applies to ani-
mals with a respiratory quotient of 0.95
(Southwood, 1991), which was found for the
slug Ariolimax columblanus by Denny (1980).
One g d wt faecal materials was equated with
21.4 kJ, since 1 g carbon corresponds to 46
kJ (Humphreys, 1979), and the mean carbon
content of slug faecal materials is 46.5%. One
g d wt green plant material was equated with
18.8 kJ (Schaefer, 1990, 1991).
RESULTS
In both years (1995 and 1996), traps were
moved three times (four catching periods in
each year; Table 1 ). To get a conservative es-
timate of the mean yearly biomass of the pop-
ulation of A. rufus in the forest studied, we as-
sumed that during the time when no trapping
occurred the dry weight of A. rufus was zero.
As a result, the mean dry weight of the pop-
ulation of A. rufus was 0.2 g m"^ (0.15 and
0.25 g d wt m"^ in 1995 and 1996, respec-
tively).
Mean fresh weight of A. rufus after the de-
position of eggs was 9.4 g and 8.4 g in the year
1994 and 1995, respectively (Table 2). Juve-
nile slugs hatched 1 34.0 (n - 39, SD = 8.7) and
200
THEENHAUS & SCHAEFER
TABLE 1 . Ash-free d wt [mg] of specimens of Arion rufus caught with 1 2 defined-area traps in
the Göttinger Wald in the years 1995 and 1996.
Catching period
Ash-free d wt per specimen
Mean
(SD)
23.03.-19.05.95 14, 25, 214, 227, 235, 241, 316, 475, 491, 827, 2095
20.05.-28.06.95 66, 91 , 94, 1 55. 742, 975, 1 030, 1 083
29.06.-11.08.95 960
12.08.-05.10.95 407,1267,1762
07.05.-02.07.96 1 37, 300, 41 2, 581 , 790, 1 466, 1 690
03.07.-12.08.96 231, 939, 1100, 1990
13.08.-10.10.96 812, 1280, 1560, 1650, 1916, 1920, 2009
11.10.-03.12.96 1,2,6,17
*306
(228)
529
(438)
960
(0)
1145
(559)
768
(550)
1065
(626)
1562
(396)
7
(6)
"without the mature specimen of 2095 mg
TABLE 2. Mean fresh weight (f wt) of Arion rufus and egg clutches after the deposition of eggs
in the year 1994 and 1995, number of slugs and egg clutches examined (n), standard devia-
tion (SD), minimum and maximum values and mean number of eggs per clutch.
number of
year
f wt (g)
n
SD
minimum
maximum
eggs per clutch
A. rufus
1994
9.40
11
3.7
4.6
17.7
1995
8.40
32
2.8
4.5
19.3
egg clutch
1994
2.57
11
1.2
1.31
5.13
58
1995
2.26
32
1.0
0.68
5.94
63
TABLE 3. Regression lines between dry weight of specimens of Arion rufus and COg pro
ductionat5, 10 and 15'C. R: COj production [ml COg h ^]: W:dwt of /4. rufus
ber of slugs examined: inter.: intercept: exp.: exponent.
2
n: num-
Temperature
Regression equation
n
r2
p inter.
SD inter.
SD exp.
5°C
10°C
1 5=C
R = 0.17xW^°^
R = 0.36 X W° ^2
R = 0.49xW^°^
14
15
15
0.59
0.99
0.89
< 0.001
< 0.001
< 0.001
0.041
0.019
0.016
0.246
0.026
0.101
TABLE 4. Regression lines between the dry weight of specimens of Arion rufus and the excretion in
summer following dry and wet periods and in autumn (three parallel measurements in each season).
F: d wt cast materials (mg 24 h ^): W: d wt A. rufus (g): n: number of slugs collected.
Regression equation range of n range of R""
P (Slope)
Summer, following dry periods F = 10.6 x W
Summer, following wet periods F = 29.2 x W°
Autumn F = 19.4xW°
52-73
0.47-0.65
< 0.001
68-79
0.68-0.79
< 0.001
58-77
0.36-0.64
< 0.001
66.8 days following the deposition of eggs (n =
69, SD = 6.1 ) at 5°C and 1 0"C, respectively. At
20°C slugs did not hatch.
The COg production of A. rufus increased
with increasing temperature. The relationship
between the dry weight of specimens and the
respiration rate was almost linear (Table 3).
There was a positive correlation between
the log dry weight of A. rufus and the log ex-
cretion of faecal materials (Table 4). In sum-
mer following periods of drought, the faecal
weight excreted per unit weight of slug de-
creased with increasing weight of slugs. In
summer following wet periods, and in autumn,
the exponent of the regression equations
ranged between 0.88 and 1.00 indicating an
almost linear relationship between the excre-
tion and slug weight.
ENERGY BUDGET OF ARION RUFUS
201
The weight of mucus secreted increased
with the weight of the slug following the equa-
tion S = a X W'^, where S = d wt mucus, W = d
vj\A. rufus. < b< 1 (Fig. 1).
The assimilation efficiency of juvenile slugs
was 78.7% (n = 6; SD = 8.1) and 88.5% (n =
6; SD = 5.6) when feeding on M. perennis and
A. ursinum, respectively. The assimilation effi-
ciency of mature slugs was lower than that of
juvenile slugs, being 73.1% (n = 15; SD = 5.9)
and 71 .8% (n = 1 5; SD = 7.2) when feeding on
M. perennis and S. nigra, respectively.
ENERGY BUDGETS
The Energy Budget of Arion rufus
Monthly weight classes of A. rufus were cal-
culated using literature data (Kunkel, 1916;
Abeloos, 1944; Laviolette, 1950; Frömming,
1 954; Smith, 1 966) and present data. As a re-
sult, the model specimen deposits one egg
clutch at the beginning of September (Table
5). Mean temperature in the litter layer in
September and October is 10"C, and there-
fore it is assumed that the juveniles hatch
two months following the deposition of egg
clutches (beginning of November). The model
specimen weighs 9 g in July, deposits one egg
clutch in September, and then dies. Based on
these data, the energy budget of the model
specimen was calculated as follows; mean
dry weight of slugs, which deposited egg
clutches in the laboratory was 1.35 g (27.2
kJ). In the laboratory, slugs never deposited
more than one egg clutch, and therefore it
was assumed that the model specimen de-
posits only one egg clutch. Average d wt of
egg clutches that were deposited in the labo-
ratory 1 994 and 1 995 was 0.54 g (1 0.9 kJ). To
calculate the amount of mucus secreted, it
was assumed that during a period of 24 h
slugs are active for 6 h (Newell, 1971) and
that during slug activity relative humidity is
100%. Therefore, the model specimen se-
cretes 4.58 g d wt mucus (85.9 kJ) during its
life (Table 5). Considering the monthly aver-
age temperature in the litter layer, the model
specimen respires 2.0 I COj (43.5 kJ) during
its life (Table 5). The regression line between
dry weight of A. rufus and weight of faecal ma-
terials excreted was similar in summer follow-
ing wet periods and in autumn (Table 2).
Therefore, data of both seasons were pooled,
resulting in the regression equation F = 19.3 x
W° ^^. For June, July and August the excretion
of faecal materials was calculated from the re-
gression equation of data from summer fol-
lowing periods of drought, which is F = 10.6 x
W°^\ As a result, during its life the model
specimen excretes 3.2 g d wt faecal materials
(68.5 kJ; Table 5). The consumption rate of
the model specimen was calculated from the
excretion of faecal materials using the assim-
ilation efficiency of 75%. This value is closer
to that of mature slugs (72%) than to that of ju-
veniles (84%), because the contribution of the
model specimen to total consumption rate is
higher in the mature than in the juvenile stage.
As a result, the model specimen consumes
5.96 g carbon during its life, which is equiva-
lent to 1 3.2 g dry plant material and 248.7 kJ.
Summing up, the energy budget of the
model specimen in the Göttinger Wald is;
248.7 = 27.2 + 1 0.9 + 85.9 + 43.5 + 68.5 [kJ]
(C = Pg + P. + P, + R + F)
The Energy Budget of the
Population of Arion rufus
Using data of Table 1 , the number of mature
A. rufus per m^ was determined. Slugs with a
dry weight of 0.75 g and more were regarded
as mature, and only those catching periods
were considered in which slugs deposit eggs
(August-October). As a result, the number of
mature specimens per 12 m^ was two and
seven in the year 1995 and 1996, respec-
tively, the average number being one mature
slug per 3 m^. Based on this abundance, the
energy budget of the population of A. rufus
was calculated. It is assumed that each ma-
ture slug lays one egg clutch in autumn, which
consists of 61 eggs. In November, 61 juve-
niles hatch, of which one specimen survives
until July. In September, this specimen de-
posits an egg clutch and then dies. Based on
these data, the monthly abundance of A. rufus
in the Göttinger Wald was calculated using a
survivorship curve. Survivorship curves of
slug populations in forest ecosystems are not
known, but those of laboratory cultures
(Szabó & Szabó, 1929) and of slug popula-
tions on permanent pasture (South, 1989) re-
vealed a more or less constant mortality rate
throughout life. Therefore, a survivorship
curve with a constant mortality rate was used,
which is N, = N^e""^, where Nq = number of ju-
veniles hatched, N, = number of slugs at time
t, r = death rate, and t = time (Slobodkin,
1962). Using the abundance data, monthly
secretion of mucus, respiration and excretion
202
THEENHAUS & SCHAEFER
— 5°C, 100% RH
•■-10°C, 100% RH
--15°C, 100% RH
— 20°C, 100% RH
-•10°C, 75% RH
■ ■■10°C, 55% RH
0.5 1 1.5 2 2.5 3
dry weight Arion rufus (g)
FIG. 1 . Weight of mucus secreted by A. rufus at different temperatures and relative fiumidities as a function
of animal dry weight. Regression equations are for 5'C and 100% RH: S = 3.55 x W°^^ (n = 49, R^ = 0.81,
p (slope) < 0.001 ), for 1 0X and 1 00% RH: S = 3.08 x W° ^^ (n = 53, R^ = 0.83, p (slope) < 0.001 ), for 1 5°C
and 1 00% RH: S = 3.47 x W° ^^ (n = 45, R^ = 0.52, p (slope) < 0.001 ), for 20°C and 1 00% RH: 1 .58 x W° ^^
(n = 34, R^ = 0.54, p (slope) < 0.001 ), for 1 0°C and 75% RH: S = 0.97 x W° ^^ (n = 60, R^ = 0.79, p (slope) <
0.001), and for 1 0X and 55% RH: S = 0.81 x W°^^ (n = 54, R^ = 0.66, p (slope) < 0.001), where RH = rela-
tive humidity, S = d wt mucus [mg h" ^], W = d wt Л. rufus [g], n = number of slugs examined.
TABLE 5. Mean monthly temperature in the litter layer and relative humidity (RH) 2 m above ground
(Göttinger Wald, mean data of the years 1991-1995). Mean monthly fresh weight (F wt) and dry weight (D
wt) of the model specimen of Arion rufus. Monthly secretion of mucus (Mucus), respiration and excretion of
faecal materials (F) of the model specimen.
Temp
Temp.
RH
[°C]
[°C]
[%]
Fwt
D wt
Mucus
Respiration
F
Month
rounded off
[g]
[g]
[mg d wt]
[ml COJ
[mg d wt]
September
11.2
10
85
с^слгл ctano
Cyy OLdLJC
Optnhpr
8.2
10
89
(^c\r\ ctanp
\_/V-il\JUd
cyy oidyc?
November
4.4
5
92
0.05
0.01
20
0.7
7
December
2.0
5
92
0.05
0.01
21
0.7
7
January
1.0
5
87
1
0.15
160
17.1
97
February
0.3
5
85
1
0.15
144
15.5
87
March
2.7
5
78
3
0.46
368
55.8
284
April
5.7
5
72
3
0.46
356
54.0
275
May
9.1
10
71
6
0.92
528
247.7
552
June
11.4
10
81
6
0.92
511
239.8
300
July
14.3
15
78
9
1.38
857
514.8
412
August
14.7
15
75
9
1.38
857
514.8
412
September
11.2
10
85
9
1.38
756
348.5
789
Sum
4578
2009.4
3222
Sum [kJ]
85.9
43.5
68.5
ENERGY BUDGET OF AfílON RUFUS
203
TABLE 6. Number of Arion rufus. monthly secretion of mucus, respiration and excretion per 3 m
Month
A. rufus
[Number]
Mucus
[mg d wt]
Respiration
[ml COg]
Faecal Material
[mg d wt]
September
egg stage
October
P^nn QtPriP
cyy oiayC
November
61
1220 43
427
December
36.5
767 26
256
January
21.9
3504 374
2124
February
13.1
1886 203
1140
March
7.8
2870 435
2215
April
4.7
1 673 254
1293
May
2.8
1478 694
1546
June
1.7
869 408
510
July
1
857 515
412
August
1
857 515
412
September
1
756 349
789
Sum
16737 3816
11124
Sum [kJ]
315.8 82.8
239.2
was calculated (Table 6). The consumption
rate of the population was calculated using an
assimilation efficiency of 80%. As a result, the
annual consumption rate of the population is
8.7 g carbon m"^, which is equivalent to 19.3
g dry plant material and 361 .9 kJ. The annual
investment of the population of A. rufus into
the somatic growth was calculated as sum of
dry weight of slugs, that die each month,
which is 2.6 g d wt m"^ (52.5 kJ).
Summing up, the energy budget of the pop-
ulation of A. rufus in the Göttinger Wald ¡s:
361.9 = 52.5 + 3.6 + 105.3 + 27.6 + 79.7
[kJ m-2 yr-^j
DISCUSSION
Slug populations are difficult to estimate,
because, particularly during periods of
drought, slugs hide underground, where they
can stay for up to three months (Kunkel,
1916). For the determination of slugs in the
field, Ferguson et al. (1989) constructed de-
fined-area traps, which consisted of an iron
ring covering 0.1 m^. A wet sacking was
placed into the traps to encourage the slugs to
remain above ground. In the present investi-
gation, defined-area traps similar to those of
Ferguson et al. (1989) were constructed.
These traps cover a larger area (1 m^) than
those of Ferguson et al. (1989), and a glass
jar with beer is imbedded into the soil inside
the traps, because beer is supposed to be a
good baiting medium for slugs. However, dur-
ing periods of drought, only low number of
slugs were caught by the traps, presumably
because the slugs hid underground. Obvi-
ously, during dry seasons the slug number is
underestimated by the defined-area trap
method. Furthermore, the defined-area traps
are not suitable for the determination of the
age distribution of slug populations, because
in the present investigation the number of ju-
venile slugs was much lower than one would
expect from the number of adult specimens
(Table 1). Presumably juvenile slugs are not
baited by the beer as efficiently as mature
slugs. As a consequence of these shortcom-
ings of traps, in the present investigation the
energy budget of the population of A. rufus
was calculated using data on the abundance
of mature slugs only, and the monthly popula-
tion structure of A. rufus derived from a sur-
vivorship curve. However, the abundance of
mature specimens was calculated from the
average value of two abundance values only,
which were strongly dissimilar (two and seven
mature A. rufus per 12 m^ in the year 1995
and 1996, respectively). Furthermore, repro-
duction does occur at various times of the
year and therefore our model is too simple. Fi-
nally, it was assumed that the mortality rate of
A. rufus remains constant throughout life. This
assumption was based on observations on
other slug species made in the laboratory and
on permanent pasture (Szabó & Szabó, 1929;
South, 1989). However, respective data on A.
rufus in the forest ecosystem are not avail-
able, and therefore the survivorship curve
used in the present investigation is just a
broad estimate. Therefore, when evaluating
the energy budget of the population of A.
rufus it should be kept in mind that this budget
204
THEENHAUS & SCHAEFER
is based on a broad estimate of slug abun-
dance.
Mean biomass of tine population of A. rufus
was 0.2 g d wt m~^. Corsmann (1990) deter-
mined the biomass of A. rufus in the forest in-
vestigated in the present study by hand sort-
ing and found a very similar biomass (0.19 g
d wt m"^). This accordance shows that the de-
fined-area trap of the present study is a suit-
able tool for the determination of slug biomass
in forest ecosystems. Biomass data of A. ater
in different beech forests are comparable to
data of A. rufus found in the present investi-
gation. Jensen (1975) estimated 0.27 g d wt
m""^ in a forest in Denmark, and Jennings &
Barkham (1 976) calculated 1 .5 g fresh weight
m"^ in Great Britain, which is equivalent to a
dry weight of 0.2 g m"^. The similar biomass
of A. rufus and A. ater in different beech
forests suggests that slug biomass is limited
by the same factor in different forests. Jen-
nings & Barkham (1979b) suggested that on
windy nights and during drought or sub-zero
temperatures, intraspecific competition for
food may occur, because slugs remain low on
the ground or beneath the litter layer where
green food is unavailable. However, another
limiting factor of slug biomass may be the
number of hiding places in the forest soil. Hid-
ing places are essential for the survival during
drought or harsh weather conditions. In differ-
ent forest ecosystems, the number of hiding
places might be similar, and therefore the reg-
ulation of slug density through the number of
hiding places would result in a similar slug
biomass in different beech forests. This hy-
pothesis is supported by the fact, that slugs
usually keep the same hiding place for a long
pehod of time ("homing"; Frömming, 1954;
Cook, 1979).
McNeill & Lawton (1970) calculated a re-
gression line between annual production and
respiration of short-lived poikiiotherms being
log R = 1.17 log P -i- 0.14 [kcal]. The insertion
of the production of A. rufus (P = P^ -i- P^ -i- P^)
into this equation results in a respiration value
which is 60 times of that actually calculated in
the present investigation. However, in their
calculations, McNeill & Lawton (1970) explic-
itly omitted data on animals characterized by
high mucus secretion, and when P^ was ex-
cluded from our calculations the respiration
value of A. rufus would be one fourth of that
expected by McNeill and Lawton's equation.
The very high net population production effi-
ciency may be a consequence of life history
characteristics of A. rufus. This slug species
builds up a very high biomass within just one
year, reproduces and then dies. Therefore, no
standing crop of older individuals has to be
maintained. However, in the present investi-
gation the respiration of A. rufus was extrapo-
lated from laboratory measurements at three
temperatures to the field. This extrapolation of
course is too simple, because in the field
slugs are exposed to a complex environment
with fluctuating biotic and abiotic conditions.
In a further experiment, the respiration rate of
A. rufus should be measured in the field.
In the present investigation, the secretion of
mucus constitutes an important component of
the energy budget; however, the determina-
tion of this parameter implies some inaccura-
cies. First, handling of slugs in the laboratory
might have stimulated mucus production dur-
ing incubation, causing an artificial high value
of P^. However, preceeding experiments
showed that the amount of mucus secreted
during the first hour of incubation in the labo-
ratory was not significantly different from that
secreted during the second and third hour.
Second, the secretion of mucus during copu-
lation of slugs was not considered in the pres-
ent calculation. Since copulation is very ex-
pensive in mucus and since slugs may
copulate several times, the amount of mucus
lost due to copulation might cause a signifi-
cant increase in P^. Third, when determining
the energy content of mucus of A. rufus in the
present investigation, the mean of calorific
values of the mucus of two snail species were
used (23.9 and 1 3.9 kJ g"^ dw mucus for Lym-
naea stagnalis and Cepaea nemoralis, re-
spectively). Obviously, such calorific values
vary widely interspecifically, and had another
value been used the outcome of the energy
budget would have been different. Fourth, it
was assumed that during a period of 24 h
slugs are active for 6 h. However, during cold
seasons, slug activity may be very low, result-
ing in low mucus secretion. If between No-
vember and April (months of an average tem-
perature of 5"C; Table 5) zero mucus
secretion is assumed, the term P^ of the en-
ergy budget of the population of A. rufus
would decrease by 35%.
The exponent b in the relationship between
the weight of mucus secreted by A. rufus and
animal dry weight was about 1 at favourable
(100% RH, 5-1 5"C), and about 2/3 at un-
favourable conditions (20"C or low RH, Fig.
1). This dichotomy may reflect the different
shapes of slugs during moving, being more or
less flat at favourable and more or less spher-
ENERGY BUDGET OF ARION RUFUS
205
ical (rather elliptic) at unfavourable abiotic
conditions. In general, doubling in weight of a
flat body results in doubling of the surface (b
= 1 ), whereas the surface of a spherical body
doubles when its weight increases by a factor
of 3 (b = 2/3).
Denny (1980) stated that gastropod crawl-
ing is the most costly form of locomotion in the
animal kingdom. Actually, in the present inves-
tigation, the secretion of mucus of the model
specimen (85.9 kJ) comprises 69% of total
production and therefore is an important com-
ponent of the energy budget. In many calcula-
tions of the energy flow through molluscs, the
secretion of mucus is ignored (Paine, 1965;
Stern, 1969; Hughes, 1970; Mason, 1971;
Jensen, 1975; Jennings & Barkham, 1976;
Streit, 1976; Andreassen, 1981; Phillipson &
Abel, 1 983; Workman, 1 983). The secretion of
mucus is considered in the calculations of
Carefoot (1 967), Paine (1 971 ), Kofoed (1 975),
Richardson (1975), Otto (1976), Edwards &
Welsh (1982), Horn (1986), Davies et al.
(1 990), and Santini et al. (1 995). Calow (1 974)
calculated for freswater gastropods that
13-23% of the assimilated energy is lost via
mucus. The respective value for Cepaea
nemoralis is 12% (Richardson, 1975), for
llyanassa obsoleta 80% (Edwards & Welsh,
1982), and for Patella vulgata 52% (Davies et
al., 1990). In the present investigation, the
model specimen of A. rufus invested 51% of
the assimilated energy into the secretion of
mucus. Most obviously, the considerable in-
vestment of gastropods in locomotion limits
the range of conditions under which popula-
tions can survive. In the light of this high in-
vestment of energy into mucus, the question
arises as to which ecosystem characteristics
and slug qualities allow the relative high slug
biomass in the beech forest ecosystem? First,
the dense herb layer of the beech forest pro-
vides plenty of food for population growth, at
least during favourable weather conditions.
Second, since slugs are omnivorous, their
food intake is independent of single food
items, and therefore food limitation becomes
rare. Third, the high assimilation efficiency of
gastropods allows an exceptionally high effi-
ciency in food utilization. These factors might
contribute to the high biomass of gastropods in
the beech forest ecosystem and make up for
the high cost of gastropod crawling.
In this and the following section, data ob-
tained for A. rufus are used to draw broad
conclusions about the energetics of the gas-
tropod community in the forest ecosystem
studied. For example, it is assumed that the
biomass/excretion ratio of all gastropod spe-
cies in the forest ecosystem studied is identi-
cal to that of A. rufus. This of course is a broad
generalization, because there may be marked
differences in the ecological efficiencies of dif-
ferent pulmonales (Lamotte & Stern, 1987).
Therefore, the calculated values of excretion
and consumption of the gastropod community
should be regarded as a broad estimate. The
production of faecal materials of the popula-
tion of A. rufus in the forest ecosystem stud-
ied is equivalent to 1.7 g carbon m"^ уг\
which is 3.7 g d wt faecal materials m"^ yr"\
The biomass of A. rufus makes up 19% of
total biomass of gastropods in the forest stud-
ied (Corsmann, 1 990). Assuming that the bio-
mass/excretion ratio of all gastropod species
in the forest studied is identical to that of A.
rufus, the gastropod community deposits 1 9 g
m~^ yr~^ as faeces. In comparison, the excre-
tion of faecal materials of lumbhcids (more
than 10 kg m"^ yr"^ Scheu, 1987) is much
higher, which shows that the relative contribu-
tion of the gastropod community to the biotur-
bation in the forest soil is negligible.
The population of A. rufus consumes 1 9.3 g
d wt food m"^ yr~^ in the forest studied. The
input of canopy litter fall in this forest is 331 g
d wt m"^ yr"^ (leaves and bud scales; Schae-
fer, 1990). About 2.6% of food material of A.
rufus is leaf litter (Corsmann, 1990), and
therefore the population of A. rufus has an
annual consumption of 0.50 g d wt leaf litter
m"^, which represents 0.15% of the leaf litter
input per year. Corsmann (1990) determined
the biomass of the most abundant gastropod
species of the beech forest studied. Further-
more, the author examined the food material
in the gut of these gastropod species and de-
termined the relative amount of certain food
material (e.g., leaf litter, green plant material,
fungi) in their gut. Considering these data and
assuming that the biomass/excretion ratio of
all gastropod species in the forest studied is
identical to that of A. rufus, the gastropod
community consumes 8.1 g d wt leaf litter m'^
yr"\ which is 2.4% of the annual leaf litter
input. This value compares with those ob-
tained by Phillipson (1983) and Phillipson &
Abel (1983) in a beech forest. The authors
calculated that the gastropod community con-
sumes 1.8-2.9% of leaf litter input per year.
Much higher values were calculated by Jen-
nings & Barkham (1976, 1979a). These au-
thors found that the population of A. rufus
consumes 1.54% and that the slug commu-
206
THEENHAUS & SCHAEFER
nity consumes 8.4% of the annual leaf litter
input of a beech forest, and therefore makes
a significant direct contribution to plant de-
composition processes. In contrast, data of
the present investigation suggest that the di-
rect contribution of the gastropod community
to decomposition processes is low.
Mean above- and below-ground net pri-
mary production (NPP) of the herb layer in the
forest studied is 100 g d wt m"^ yr"'' (Schaefer,
1990). Green herb material makes up 17% of
the food material of A. rufus (Corsmann,
1 990), and therefore the population of A. rufus
consumes 3.3 g d wt green herb material m"^
yr"\ which represent 3.3% of the annual pro-
duction of green plant material. The gastro-
pod community has a consumption of 26.1 g
d wt green herb material m"^ yr"\ which rep-
resent 26% of the annual production of green
plant material. This phytomass compares well
with that which is annually ingested by phy-
tophages (Curculionidae and lepidopteran
larvae) in the forest studied (Schaefer, 1990),
and therefore the relative consumption of
green plant material by the gastropod com-
munity is high.
The NPP of higher plants in the forest stud-
ied is 2078 g d wt m"^ year"^ (Schaefer, 1 990),
and the population of A. rufus consumes
0.24% of the annual NPP. The gastropod
community consumes 34.5 g d wt higher
plants m year"\ which is 1.7% of the NPP of
higher plants. Obviously, the gastropod com-
munity consumes a very small part of the phy-
tomass in the forest ecosystem and therefore
most probably the feeding pressure of gas-
tropods on plants is low. This is partly due to
the fact that gastropods are omnivorous and
that therefore the strength of prédation dif-
fuses to different trophic unities (Polis, 1994).
It is stated by many authors (e.g., Hairston et
al., 1960; Fretwell, 1987; Oksanen et al.,
1981) that herbivores of forest ecosystems
consume only a small part of the phytomass,
because their biomass is controlled by preda-
tors (top-down control). The low feeding pres-
sure of gastropods on higher plants, however,
most probably is not caused by top-down
forces, because gastropod biomass is limited
by abiotic factors (e.g., drought; Jennings &
Barkham, 1979b).
The energy budget of the present investi-
gation is based on a series of assumptions,
which are strongly subjective. For example,
the population structure of A. rufus was calcu-
lated using data on the abundance of mature
slugs and on the mean number of eggs per
clutch and a survivorship curve. "Mature
slugs" were defined as specimens with a dry
weight of 0.75 g and more, and had another
weight been taken the abundance of mature
specimens would have been different. The
survivorship curve used is a broad estimate
taken from the literature, and it therefore rep-
resents another shortcoming of the present
investigation. Furthermore, the conversion of
data into calorific values was accomplished
with conversion factors from the literature,
and therefore the calorific energy budget
should be evaluated with caution. In addition,
the monthly amount of respiration, secretion
of mucus and excretion of faecal materials of
A. rufus was calculated using mean monthly
temperature and air humidity values of the for-
est under study. Of course, the abiotic char-
acteristics of the forest ecosystem are incom-
pletely simulated by this approach. However,
the energy budgets of populations in general
embody many assumptions about population
structure and generally lack true seasonal
changes in budget terms. The energy budget
of the present investigation should be re-
garded as a broad estimate.
ACKNOWLEDGEMENTS
We are grateful to Thomas Theenhaus,
Claus Döring, Klaus Dornieden, Alexander
Sührig. and Ulrich Strothmann for collecting
large numbers of slugs and for their help in
setting up the defined-area traps. We are also
indebted to Sonja Migge for her help in raising
slugs. We thank two anonymous reviewers for
helpful comments on the manuscript. This
work was supported by a grant of the
Friedrich-Ebert-Stiftung (Anne Theenhaus).
LITERATURE CITED
ABELOOS, M., 1944, Recherches expérimentales
sur la croissance. La croissance des mollusques
arionidés. Bulletin Biologique de la France et de
la Belgique. 78:215-256.
ANDREASSEN, E. M., 1981, Population dynamics
of Arianta arbustorum and Cepaea horîensis in
western Norway. Fauna Norvegica. A, 2: 1 -13.
BARBER, H. S., 1 931 , Traps for cave inhabiting in-
sects. Journal of the Elisha Mitchell Scientific So-
ciety. 46: 259-266.
BLESS. R., 1978, Organische Substanz und Ener-
giegehalt von Pflanzen. Schnecken und Käfern
ENERGY BUDGET OF ARION RUFUS
207
aus Waldstandorten des Naturparks Kottenforst-
Ville bei Bonn. Pedobiologia. 18: 39-47.
CALOW, P., 1974, Some observations on locomo-
tory strategies and their metabolic effects in two
species of freshwater gastropods, Ancylus fluvi-
atilis Müll, and Planorbis contortus Linn. Oecolo-
gia. 16: 149-161.
CAREFOOT, T. H., 1967, Growth and nutrition of
three species of opisthobranch molluscs. Com-
parative Biochemistry and Piiysiology. 21 : 627-
652.
COOK, A., 1979, Homing by the slug Umax pseu-
doflavus. Animal Betiaviour. 27: 545-552.
CORSMANN, M., 1990, Die Schneckengemein-
schaft (Gastropoda) eines Laubwaldes: Popula-
tionsdynamik, Verteilungsmuster und Nahrungs-
biologie. Berichte des Forschungszentrums
Waidökosysteme. A, 58: 1 -208.
DAVES, M. S., S. J. HAWKINS & H. D. JONES,
1990, Mucus production and physiological ener-
getics in Pateiia vulgata L. Journai of Moiiuscan
Studies. 56: 499-503.
DENNY, M., 1980, Locomotion: the cost of gastro-
pod crawling. Science. 208: 1288-1290.
EDWARDS, S. F. & B. L. WELSH. 1 982, Trophic dy-
namics of a mud snail {llyanassa obsoleta (Say))
population on an intertidal mudflat. Estuarine.
Coastal and Shelf Science. 1 4: 663-686.
FERGUSON, С M., В. I. P. BARRATT & P. A.
JONES, 1989, A new technique for estimating
density of the field slug (Dereceras reticulatum
(Müller)). Pp. 331-336, in: i. Henderson, Slugs
and snails in world agriculture. Proceedings of
the British Crop Protection Council. 41 . Laven-
ham Press Limited, Guildford.
FRETWELL, S. D., 1 987, Food chain dynamics: The
central theory of ecology? Oikos. 50: 291 -301 .
FRÖMMING, E., 1954, Biologie der mitteleuropäi-
schen Landgastropoden. Duncker und Hum-
boldt, Berlin.
HAIRSTON, N. G., F. E. SMITH & L. B. SLOBOD-
KIN, 1960, Community structure, population con-
trol and competition. The American Naturalist. 94:
421-425.
HORN, P. L., 1986, Energetics of Chiton pelliser-
pentis (Ouoy & Gaimard, 1835) and the impor-
tance of mucus in its energy budget. Journal of
Experimental Biology and Ecology. 101: 119-
141.
HUGHES, R. N., 1 970, An energy budget for a tidal-
flat population of the bivalve Scrobiculaha plana
(da Costa). Journal of Animal Ecology. 39:
357-379.
HUMPHREYS, W. F., 1979, Production and respi-
ration in animal populations. Journal of Animal
Ecology 48: 427-453.
JENNINGS, T. J. & J. P. BARKHAM, 1976, Ouanti-
tative study of feeding in woodland by the slug
Arion ater. Oikos. 27: 168-173.
JENNINGS. T. J. & J. P. BARKHAM, 1979a, Litter
decomposition by slugs in mixed deciduous
woodland. Holarctic Ecology, 2: 21 -29.
JENNINGS, T. J. & J. R BARKHAM, 1979b, Niche
separation in woodland slugs. Oikos. 33:
127-131.
JENSEN, T. F., 1975, A tentative energy budget for
a summer population of Arion ater L. (Gastro-
poda: Pulmonata). Natura Jutlandica. 18: 10-20.
KOFOED, L. H., 1975, The feeding biology of Hy-
drobia ventrosa (Montagu). II. Allocation of the
components of the carbon-budget and the signif-
icance of the secretion of dissolved organic ma-
terial. Journal of Experimental Marine Biology
and Ecology 19: 243-256.
KÜNKEL, K., 1916, Zur Biologie der Lungen-
schnecken. 441 pp. Winters, Heidelberg.
LAMOTTE, M. & G. STERN, 1987, Les bilans én-
ergétiques chez les mollusques pulmones. Halio-
tis. 16: 103-128.
LAVIOLETTE, P., 1 950, Sur un retard de la maturité
genitale observe chez Arion rufus. Bulletin Bi-
mensuel de la Société Linneenne de Lyon. 19:
52-56.
LAWTON, J. H., 1994, What do species do in
ecosystems? Oikos, 71 : 367-374.
MASON, С. F., 1971 , Respiration rates and popula-
tion metobolism of woodland snails. Oecologia,
7: 80-94.
McNeill, s. & J. Н. LAWTON, 1970, Annual pro-
duction and respiration in animal populations.
Nature. 225: 472-474.
NEWELL, R R, 1971, Molluscs. Pp. 128-149, in: J.
PHiLLiPSON, Methods of study in quantitative soil
ecology. Blackwell, Oxford.
OKSANEN, L., S. D. FRETWELL, J. ARRUDA & P.
NIEMELÄ, 1981 , Exploitation ecosystems in gra-
dients of primary productivity. The American Nat-
uralist, 118:240-261.
OTTO, C, 1976, Production of Ancylus fluviatilis
MÜLLER (Gastropoda) in a south Swedish
stream. Polish Archives of Hydrobiology 23:
421-429.
PAINE, R. T, 1965, Natural history, limiting factors
and energetics of the opisthobranch Navanax in-
ermis. Ecology 46: 603-619.
PAINE, R. T., 1971, Energy flow in a natural popu-
lation of the herbivorous gastropod Tegula fune-
bralis. Limnology and Oceanography 16: 86-98.
PETRUSEWICZ, K., 1967, Suggested list of more
important concepts in productivity studies. Pp.
51 -82, in: K. PETRUSEWICZ, Secondary productiv-
ity in terresthal ecosystems. Vol. 1 . Warsaw &
Cracow, Warsaw.
PHILLIPSON, J., 1983, Slug numbers, biomass
and respiratory metabolism in a beech wood-
land— Wytham Woods, Oxford. Oecologia, 60:
38-45.
PHILLIPSON, J. & R. ABEL, 1983. Snail numbers,
biomass and respiratory metabolism in a beech
woodland — Wytham Woods, Oxford. Oecologia,
57: 333-338.
POLIS, G. A., 1994, Food webs, trophic cascades
and community structure. Australian Journal of
Ecology 19: 121-136.
208
THEENHAUS & SCHAEFER
RICHARDSON, A. M. M., 1975, Energy flux in a
natural population of the land snail, Cepaea
nemoralis L. Oecologia. 19: 141-164.
SANTINI, G., G. CHELAZZI & P. DELIA SANTINA,
1995, Size-related functional and energetic con-
straints in the foraging of the limpet Patella vul-
gata (Mollusca, Gastropoda). Functional Ecol-
ogy. 9: 55^ -558.
SCHAEFER, M., 1990, The soil fauna of a beech
forest on limestone: trophic structure and energy
budget. Oecologia. 82: 128-136.
SCHAEFER, M., 1991, The animal community: di-
versity and resources. Pp. 51-120, in: E. röhrig
& M. ULRICH, Temperate deciduous forests (Eco-
systems of the World IV). Elsevier, Amsterdam.
SCHEU, S., 1987, The role of substrate feeding
earthworms (Lumbricidae) for bioturbation in a
beechwood soil. Oecologia. 72: 192-196.
SLOBODKIN, L. В., 1962, Growth and regulation of
animal populations, 184 pp. Holt, Rinehart and
Winston, New York.
SMITH, B. J., 1966, Maturation of the reproductive
tract of Anon ater (Pulmonata: Ahonidae). Mala-
cologia. 4: 325-349.
SOKAL. R. R. & F. J. ROHLF, 1995, Biometry 3rd
ed, 887 pp. Freeman and Co, New York.
SOUTH, A., 1989, The effect of weather and other
factors on numbers of slugs on permanent pas-
ture. Pp. 355-360, in: I. HENDERSON, Slugs and
snails in world agriculture. Proceedings of the
British Crop Protection Council, 41. Lavenham
Press Limited, Guildford.
SOUTHWOOD, T R. E., 1991 , Ecological methods.
2nd ed, 524 pp. Chapman and Hall, London.
STERN, G., 1969, Bilan énergétique de la limace
Arion rufus (mollusque pulmoné) en période de
croissance. Comptes Rendus de lAcademie des
Sciences. Paris. 269: 1015-1018.
STREIT В., 1976, Energy flow in four different field
populations of Ancylus fluviatilis (Gastropoda-
Basommatophora). Oecologia. 22: 261-273.
SZABÓ, I. M. & SZABÓ, 1929, Lebensdauer,
Wachstum und Altern. Studiert bei der Nacktsch-
neckenart Agriolimax agrestis L. Biología Gener-
alis. 5: 95-U8.
THEENHAUS, A., 1997, Die rote Wegschnecke
Ahon rufus (Gastropoda) — Energiebudget und
Funktion im Zersetzersubsystem eines Kalk-
buchenwaldes. Berichte des Forschungszen-
trums Waldökosysteme Göttingen, A, 148: 1-
167.
THEENHAUS, A. & S. SCHEU, 1996a, Succes-
sional changes in microbial biomass and respira-
tion in faecal material of the slug Ahon rufus
(Gastropoda). So/7 Biology and Biochemistry, 28:
569-577.
THEENHAUS, A. & S. SCHEU, 1996b, The influ-
ence of slug [Anon rufus) mucus and cast mater-
ial addition on microbial biomass, respiration and
nutrient cycling in beech leaf litter. Biology and
Fertility of Soils, 23: 80-85.
WORKMAN, C, 1983, Comparisons of energy par-
titioning in contrasting age-structured popula-
tions of the limpet Patella vulgata L. Journal of
Expehmental l\/lahne Biology and Ecology. 68:
81-103.
Revised ms accepted 1 February 1999
MALACOLOGIA, 1999, 41(1): 209-230
SPECIES OF MACOMA (BIVALVIA: TELLINIDAE) FROM THE PACIFIC COAST OF
RUSSIA, PREVIOUSLY DESCRIBED AS ABRINA (BIVALVIA: SEMELIDAE)
Gennady M. Kamenev^ & Victor A. Nadtochy^
ABSTRACT
Abrina cuneipyga Scarlato, 1981 ; A. sachalinica Scarlato, 1981 ; A. shiashkotanika Scarlato,
1981 ; and A. tatarica Scarlato, 1981, described from the Pacific coast of Russia belong instead
in the genus Macoma Leach, 1 81 9. For one species, a new combination is suggested, Macoma
cuneipyga (Scarlato, 1 981 ). Abrina sachalinica. A. tatarica and A. stiiastikotanika are synonyms
of Macoma /oi/en/ (Jensen, 1905), Macoma calcárea (Gmelin, 1791), and Macoma sp., respec-
tively. The main morphological characteristic on the basis of which these species were previously
included in Abrina Habe, 1952, was the presence of an internal ligament in an oblique resilifer
posterior to the cardinal teeth. Studies of the common northwestern Pacific M. loveni. Macoma
balthica (Linné, 1758), Macoma crassula (Deshayes. 1855), Macoma lama Bartsch, 1921, Ma-
coma incongrua (Martens, 1865), and M. calcárea (Gmelin, 1791) show the presence of a simi-
lar morphological characteristic in young specimens. In Macoma. this characteristic is reduced
as the molluscs grow, but it is preserved in adult Abrina. The results of a comparative analysis of
species of genera Abrina and Macoma, and expanded descriptions of M. cuneipyga. M. loveni,
and Macoma sp.. are presented.
Key words: Macoma, Abrina, northwestern Pacific, systematics, morphology, distribution.
INTRODUCTION
Scarlato (1981), In his monograph on fauna
and distribution of bivalve molluscs of the
northwestern Pacific, gave a description with
very schematic pictures of the type specimens
of four new species of the genus Abrina Habe,
1952 (Fam. Semelidae Stoliczka, 1870): Ab-
rina cuneipyga Scarlato, 1981 ; A. sachalinica
Scarlato, 1981; A. shiashkotanika Scarlato,
1981; and A. tatarica Scarlato, 1981. Previ-
ously, representatives of this genus were not
mentioned in lists of the Russian molluscan
fauna and were found only in subtropical lati-
tudes of the western Pacific (Gould, 1861;
Kuroda, 1951; Habe, 1952, 1958, 1977,
1981).
A study of the type specimens of Abrina
species from the Pacific coast of Russia, as
well as material of Macoma species from the
Arctic seas and the northwestern Pacific, and
of Abrina species from Japan, have shown
that the species described by Scarlato (1 981 )
should instead be included in the genus Ma-
coma Leach, 1819 (Fam. Tellinidae Blainville,
1814).
For one of the four species described by
Scarlato (1 981 ) as Abrina, a new combination
is suggested, Macoma cuneipyga (Scarlato,
1981). Abrina sachalinica, A. tatarica, and A.
shiashkotanika are considered to be syn-
onyms of M. loveni, M. calcárea and Macoma
sp., respectively. The goal of this paper is to
present the results of a comparative analysis
of species of genera Abrina and Macoma and
to give an expanded description of M.
cuneipyga, M. loveni and Macoma sp., sup-
plemented by new data on their shell morphol-
ogy, ecology, and geographical distribution.
MATERIALSAND METHODS
In this study we have used the material col-
lected by expeditions of the following re-
search institutes:
PRIFO expeditions on the shelf zone of the
northern Sea of Okhotsk (R/V "8-452", 1977;
R/V "8-461", 1979), eastern coast of Sakhalin
Island, Sea of Okhotsk (R/V "8-452", 1977),
western coast of Kamchatka, Sea of Okhotsk
(R/V "Professor Levanidov", 1996), eastern
coast of Kamchatka, Pacific Ocean (R/V "Mys
Tikhiy", 1985), and Bering Sea (R/V "Mys
Tikhiy", 1984). A joint IMB-PRIFO expedition
on the shelf and bathyal zones of the Kuril Is-
lands (R/V "Tikhookeansky", 1987).
A joint FERHI-SakhRIFO expedition under
Institute of Marine Biology, Russian Academy of Sciences. Vladivostok 690041 , Russia; inmarbio@mail.primorye.ru
^Pacific Research Institute of Fistiehes and Oceanography, Vladivostok 690600. Russia
209
210
KAMENEV & NADTOCHY
the supervision of ENL Company on the shelf
zone of the northeastern Sakhalin Island. Sea
of Okhotsk, in the vicinity of an oil-drilling site
(R/V "Pavel Gardienko", 1997).
Expeditions of 1MB on the shelf zone of the
Tatar Strait. Sea of Japan (R/V "Atna", 1974;
R/V "Berill", 1977) and Peter the Great Bay,
Sea of Japan (R/V "Ametist", 1983). A joint
IMB-PIBOC expedition on the shelf zone of
the Sea of Japan (R/V "Akademik Oparin",
1995).
The material from the western Kamchatka.
Sea of Okhotsk, southeastern Kamchatka,
Pacific Ocean, and the Bering Sea was fixed
and stored in 4% formaldehyde in PRIFO. All
the other material was fixed in 70% ethanol
and stored dry in 1MB.
We have also used collections of the follow-
ing taxa: M. balthica from the Bering Sea
(MIMB); M. calcárea, M. crassula and M. in-
congrua from the Sea of Japan (MIMB); M.
lama from the Sea of Okhotsk (MIMB); M.
/oi /ел/ from the Arctic seas, the Sea of Japan,
the Sea of Okhotsk, and the North Atlantic
(USNM, ZIN); Abrina lunella (Gould, 1861)
and A. kinoshitai {Kuroda & Habe, 1958) from
the coastal waters of Japan (Drs. T. Kurozumi
and E. Tsuchida, NHMI, NSMT); /4. cuneipyga.
A. sachalinica. A. shiashkotanlka. and A. tatar-
/ca from the Pacific seas of Russia (ZIN). Ma-
coma from the Arctic and Pacific seas of Rus-
sia were fixed and stored in 70% ethanol. All
other material was stored dry.
For the collection material stored in ZIN, in-
ventory numbers are given for holotypes; for
other specimens, their catalogue numbers are
given.
Shell Measurements
For measurements, we have chosen those
parameters most often used in diagnosis, de-
scription, and comparative analysis of species
of the genus Macoma (Dunnill & Ellis, 1969;
Coan, 1971; Scarlato, 1981; Kamenev. 1989,
1990; Kafanov et al., 1997).
Figure 1 shows the position of our shell
morphology measurements. Shell length (L),
height (H), width of each valve (W) (not
shown), anterior end length (A), maximal dis-
tance from posterior shell margin to top of pal-
liai sinus (LI), and minimal distance from top
of palliai sinus to anterior adductor muscle
scar (L2) were measured for each valve. The
ratios of these parameters to shell length
(H/L, W/L, A/L, L1/L, L2/L, respectively) were
determined. Shell measurements were made
FIG. 1 . Placement of shell measurements: L — shell
length; H — height; A — anterior end length; LI —
maximal distance from posterior shell margin to top
of palliai sinus; L2 — minimal distance from top of
palliai sinus to anterior adductor muscle scar.
using a calipers and an ocular micrometer
with an accuracy of 0.1 mm.
We made measurements of:
(1) 51 specimens and 2 left valves of M.
loveni from Spitzbergen (USNM
108789, 2 spec); Laptev Sea (ZIN
106, 3 spec); Gulf of St. Laurence,
North Atlantic (USNM 95638, 1 spec,
1 left valve); Gulf of Maine, North At-
lantic (USNM 159769, 1 spec); Bering
Sea (our material, 1 spec); Kronotsky
Bay, eastern Kamchatka, Pacific
Ocean (our material, 4 spec); Sea of
Okhotsk (ZIN 9853. holotype of A.
sachalinica: ZIN 57, 2 spec; USNM
204814, 1 spec; our material, 5 spec);
Kuril Islands (our material, 19 spec);
Sea of Japan (ZIN 65, 3 spec: ZIN 67,
1 spec: USNM 2048015. 2 spec: our
material 5 spec, 1 left valve).
(2) 22 specimens, 2 hght, 7 left valves of
M. cuneipyga from Kuril Islands (ZIN
9800, holotype of A. cuneipyga: ZIN
10, 1 left valve; our material, 18 spec,
2 right, 6 left valves); Sea of Okhotsk
(our material, 3 spec).
(3) 3 specimens of Macoma sp. from Shi-
ashkotan Island, Middle Kuril Islands
(ZIN 9867, holotype of A. shi-
ashkotanlka: 2 spec. (ZIN 1) from the
type locality).
(4) 3 specimens and one right valve of M.
calcárea from the Sea of Japan (ZIN
9900, holotype of A. tatarica: A. tatar-
STATUS OF ABRINA SPECIES FROM RUSSIA
211
ica (ZIN 5), 1 spec); Sea of Okhotsk
(A. tatahca (ZIN 6), 1 spec, 1 left
valve).
(5) 2 specimens, 5 right, 2 left valves of A.
lunella from Sagami Bay, Japan
(NSMT 45451).
(6) 1 specimen, 1 right, 7 left valves of A.
kinoshitai from Esu-zaki, Wakayama
Prefecture, Japan (NSMT 51909, 1
spec); Funakashi Bay, Iwate Prefec-
ture, Japan (material of Drs. T Kuro-
zumi and E. Tsuchida (NHMI, no num-
ber); 1 left, 1 right valves), Otsuchi Bay,
Iwate Prefecture, Japan (material of
Drs. T Kurozumi and E. Tsuchida
(NHMI, no number), 6 left valves).
Statistics
Statistical analysis of the material used a
package of statistical programs for Windows
STADIA 6.0 (Kulaichev, 1996) and STAT-
GRAPHICS.
The calculated indices (H/L; W/L; A/L; L1/L
and L2/L) are less susceptible to change com-
pared to other measured parameters. A com-
parison by pairs of different parameters of the
samples using parametric and nonparametric
tests was conducted using these indices. All
data was tested with a Kolmogorov test for
their fit to a normal distribution. For the left
valves of M. loveni and M. cuneipyga, the
original data on L2/L and the log^Q of H/L cor-
responded to the norm; for the right valves of
these species, original data on H/L, A/L and
L2/L also corresponded to the norm. For this
reason, we used the Student (T) parametric
test for a comparison by pairs of different
valves of M. loveni and M. cuneipyga for
these indices.
The distribution of the original data by other
indices was different from the norm. Fifteen
transformations were used in an attempt to
bring the obtained data to the norm; however,
none were successful. Therefore, in the com-
parative analyses of the samples of other
indices (W/L, A/L, and L1/L for left valves
and W/L and A/L for right valves), we used
the nonparametric criterion of Kolmogorov-
Smirnov (two sample test).
We used a discriminant analysis to test the
validity of the hypothesis of division of all stud-
ied specimens into two groups (M /oi/en/and
M. cuneipyga).
In linear and stepwise multiple discriminant
analyses of the samples of M. loveni and M.
cuneipyga. we used original data on the
parameters H, L, W, LI, L2. Data on these
parameters corresponded to the norm; 51
specimens of M. loveni and 22 specimens of
M. cuneipyga were used for the analysis.
Throughout this study, statistical signifi-
cance was defined as P < 0.05.
The following abbreviations are used in the
paper: ENL -commercial company "Exon
Neftegas Limited"; FERHI-Far East Re-
search Hydrometeorological Institute, Vladi-
vostok; 1MB -Institute of Marine Biology,
Russian Academy of Sciences, Vladivostok;
MIMB- Museum of the Institute of Marine
Biology, Vladivostok; NHMI -Natural His-
tory Museum and Institute, Chiba; NSMT-
National Science Museum, Tokyo; PIBOC-
Pacific Institute of Bio-organic Chemistry,
Russian Academy of Sciences, Vladivo-
stok; PRIFO- Pacific Research Institute of
Fisheries and Oceanography, Vladivostok;
SakhRIFO -Sakhalin Research Institute of
Fisheries and Oceanography, Yuzhno-Sakha-
linsk; USNM- United States National Mu-
seum of Natural History, Smithsonian In-
stitute, Washington, D. C.; ZIN -Zoological
Institute, Russian Academy of Sciences, St.
Petersburg.
COMPARATIVE ANALYSIS
In Tables 1 and 2 the results of a morpho-
metric analysis of the common Japanese
spec. Abrina lunella and A. kinoshitai are pre-
sented (Figs. 2-11). On the basis of the re-
sults of these studies, as well as descriptions,
pictures and photos of Abrina (Gould, 1861;
Kuroda, 1 951 ; Habe, 1 952, 1 958, 1 961 , 1 964,
1 977, 1 981 ; Ito, 1 967; Kuroda et al., 1 971 ; Ito
et al., 1986; Ito, 1989; Tsuchida & Kurozumi,
1995) and Macoma species (Golikov & Scar-
lato, 1967; Keen, 1969; Coan, 1971 ; Bernard,
1979; Lubinsky, 1980; Scarlato, 1981), Table
3 has been constructed giving the main diag-
nostic characteristics of Abrina and Macoma.
Table 3 shows that the genera Abrina and
Macoma are similar in most morphological
characteristics. The main distinguishing char-
acteristic of Abrina is the presence of a well-
developed internal ligament in a resilifer, a
narrow groove posteria to the cardinal teeth.
In Macoma. an internal ligament is absent.
The results of studies different age groups
of the well-identified and widely distributed
northwestern Pacific M. loveni, M. balthica. M.
calcárea. M. crassula. M. incongrua and M.
lama have shown that young individuals of
these species have a rather large, well-devel-
212
KAMENEV & NADTOCHY
FIGS. 2-11. Shells of Abrina species. 2-8. Abrina lunella (Gould, 1 861 ) (NSMT 45451 ). Sagami Bay, Japan.
2-6: Largest specimen, shell length 16.1 mm. 7, 8: Hinge of the right valve, valve length 13.0 mm (bar = 1
mm). 9-11. Abrina kinoshitai (Kuroda & Habe, 1958) (NSMT 51909), Esu-zaki, Wakayama Prefecture,
Japan, shell length 14.8 mm.
oped internal ligament (Figs. 12-21). In its
form, position, and size it is similar to the in-
ternal ligament of the genus Abrina. It is situ-
ated in an oblique resilifer posteria to the car-
dinal teeth and almost reaches the ventral
edge of the hinge plate (Figs. 12, 13). With
age, its relative size is reduced (Figs. 14, 21),
and in adult specimens of M. crassula. M.
balthica, M. calcárea, M. incongrua and M.
lama, it disappears. In adult specimens of
M. loveni, it can just be noticed on the hinge
plate ventral to the beaks.
Also, the form of the resilifer changes with
age. In young specimens of M. crassula, M.
balthica, M. calcárea, M. incongrua, and M.
lama (shell length, <5-6 mm) and M. loveni (<
10 mm), the form and size of the resilifer rela-
tive to the width of the hinge plate (Figs. 15,
16, 18, 20) is similar to those of adult Abrina
lunella and A. kinoshitai. In larger specimens
(> 8-10 mm), the resilifer is much shorter than
the width of the hinge plate and ovate (Fig.
19). With the exception of M. loveni. in adult
specimens of all investigated species it is ab-
sent. In adult specimens of M. loveni, it is pre-
served in the form of ovate pit on hinge plate
ventral to the beaks posterior to the cardinal
teeth (Fig. 17).
STATUS OF ABRINA SPECIES FROM RUSSIA
213
TABLE 1 . Abrina lunella (Gould, 1 861 ). Shell measurements (mm), indices and summary statistics of all char-
acteristics (NSMT 45451): L — shell length; H — height; W — width; A — anterior end length; LI —maximal dis-
tance from posterior margin to the top of palliai sinus; L2 — minimal distance from the top of palliai sinus to
anterior adductor muscle scar. Numerator indicates shell measurements and indices for the left valve,
denominator — for the right valve. NM — not measured.
Statistics
L
H
W
A
L1
L2
H/L
W/L
A/L
L1/L
L2/L
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
7.8
6.0
1.7
4.8
6.2
0.8
0.77
0.22
0.62
0.79
0.10
10.8
7.7
2.1
6.6
7.2
1.5
0.71
0.19
0.61
0.67
0.14
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
11.1
7.9
2.4
7.0
7.6
1.8
0.71
0.22
0.63
0.68
0.16
11.4
8.0
2.1
7.2
8.0
1.3
0.70
0.18
0.63
0.70
0.11
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
11.9
8.2
2.5
7.6
7.7
2.3
0.69
0.21
0.64
0.65
0.19
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
13.0
10.0
3.0
8.0
8.3
2.3
0.69
0.21
0.64
0.65
0.19
14.1
10.2
2.8
9.1
9.5
2.4
0.72
0.20
0.65
0.67
0.17
14.0
10.3
3.2
9.1
9.0
3.0
0.74
0.23
0.65
0.64
0.21
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
15.0
10.5
3.3
9.0
10.0
2.5
0.70
0.22
0.60
0.67
0.17
16.1
12.5
3.5
9.8
10.7
3.0
0.78
0.22
0.61
0.66
0.19
16.1
12.5
4.0
9.8
11.4
2.4
0.78
0.25
0.61
0.71
0.15
Mean
13.10
9.60
2.63
8.18
8.85
2.05
0.728
0.198
0.625
0.675
0.153
12.70
9.34
2.87
7.90
8.60
2.17
0.737
0.226
0.624
0.683
0.166
SD
2.46
2.23
0.67
1.52
1.56
0.79
0.036
0.017
0.019
0.017
0.035
2.76
2.13
0.74
1.67
1.71
0.70
0.037
0.013
0.017
0.053
0.035
SE
1.23
1.12
0.34
0.76
0.78
0.40
0.018
0.009
0.010
0.009
0.018
1.05
0.81
0.28
0.63
0.65
0.26
0.010
0.005
0.006
0.020
0.013
Min
10.8
7.7
2.1
6.6
7.2
1.3
0.70
0.18
0.61
0.66
0.11
7.8
6.0
1.7
4.8
6.2
0.8
0.69
0.21
0.60
0.64
0.10
Max
16.1
12.5
3.5
9.8
10.7
3.0
0.78
0.22
0.65
0.70
0.19
16.1
12.5
4.0
9.8
11.4
3.0
0.78
0.25
0.65
0.79
0.21
Thus, a well-developed internal ligament
lodged in oblique resilifer in representatives of
the genus Macoma is a juvenile characteristic
that is preserved in Abrina during its entire
life.
A study of the type material of Abrina
species from the seas of Russia has shown
that the main character for their description as
new species of Abrina is the presence of an
internal ligament posterior to the cardinal
teeth (Fig. 22) and having different degrees of
development. It is most distinct in A. sfii-
ashl<otanil<a and A. cuneipyga. However, the
type specimens of all the species of Abrina
described by Scarlato (1981) have a less de-
veloped internal ligament compared with that
of Abrina species inhabiting the coastal wa-
ters of Japan. The size of internal ligament of
the holotypes of these species corresponds to
their shell size and species, similar to the
species of Macoma studied. In other taxo-
nomic characteristics, they are also identical
to the genus Macoma. All the new species de-
scribed by Scarlato (1981) are of a relatively
small size. Therefore, we consider that Scar-
lato (1981) described new species of Abrina
on the basis of a study of young specimens of
Macoma that had internal ligaments. We con-
sider that all species of Abrina described from
the Pacific coast of Russia should be included
in the genus Macoma.
SYSTEMATICS
Family Tellinidae Blainville, 1814
Genus Macoma Leach, 1819
Type species (by monotype): Macoma te-
ñera Leach, 1819; = Teilina calcárea Gmelin,
1791
Diagnosis
Shell small to large in size (20 to 100 mm),
medium in thickness to heavy, moderately in-
flated, ovate-triangular, rounded-triangular,
ovate or round, white, chalky, smooth with
214
KAMENEV & NADTOCHY
TABLE 2. Abrina kinoshitai {Kuroäa & Habe, 1958). Shell measurements (mm), indexes and summary sta-
tistics of all characteristics: L — shell length; H — height; W — width; A — anterior end length; L1 —maximal
distance from posterior margin to the top of palliai sinus; L2 — minimal distance from the top of palliai sinus
to anterior adductor muscle scar. Numerator indicates shell measurements and indices for the left valve,
denominator — for the right valve. NM — not measured.
Statistics L
H
W
A
LI
L2
H/L
W/L
A/L
L1/L
L2/L
Otsuchi Bay (39°22'N,
142°00'E), Iwate Prefecture, Japan, depth 75 m.
sandy
mud
7.5
4.7
1.1
4.8
5.3
0.6
0.63
0.15
0.64
0.71
0.08
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
8Л
5Л
L3
5.2
5.8
0.9
0.61
0.16
0.63
0.70
0.11
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
8Л
5^3
13
5.4
6.2
0.9
0.62
0.15
0.64
0.73
0.11
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
93
5.9
1.5
5.9
7.0
1.0
0.62
0.16
0.62
0.74
0.11
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
10.0
6.3
1.6
6.5
7.2
1.1
0.63
0.16
0.65
0.72
0.11
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
12.8
8.0
2.2
8.3
9.3
1.6
0.63
0.17
0.65
0.73
0.13
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
Funakochi Bay (39 23'N, 142°
OO'E), Iwate Prefecture.
Japan, depth 70
m, mud and sand
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
12.0
7.7
1.9
8.0
8.1
2.1
0.64
0.16
0.67
0.68
0.18
12.7
8.0
2.2
8.3
9.3
1.5
0.63
0.17
0.65
0.73
0.12
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
Esu-Zaki, Wakayama Prefectu
re, Japan (NSMT 51 909)
14.8
9.3
2.4
9.8
11.1
2.1
0.63
0.16
0.66
0.75
0.14
14.8
9.3
2.4
9.8
10.2
3.0
0.63
0.16
0.66
0.69
0.20
Mean 10.51
6.58
1.70
6.78
7.65
1.21
0.625
0.160
0.643
0.726
0.114
13.40
8.50
2.15
8.90
9.15
2.55
0.635
0.160
0.665
0.685
0.190
SD 2.61
1.66
0.50
1.81
2.04
0.49
0.008
0.008
0.013
0.016
0.018
1.98
1.13
0.35
1.27
1.49
0.64
0.007
0.007
0.007
0.014
SE 0.92
0.59
0.18
0.64
0.72
0.17
0.003
0.003
0.005
0.006
0.006
1.40
0.80
0.25
0.90
1.05
0.45
0.005
0.005
0.005
0.010
Min 7.5
4.7
1.1
4.8
5.3
0.6
0.61
0.15
0.62
0.70
0.08
12.0
7.7
1.9
8.0
8.1
2.1
0.63
0.16
0.67
0.69
0.2
Max 14.8
9^
2.4
9.8
11.1
2.1
0.63
0.17
0.66
0.75
0.14
14.8
9.3
2.4
9.8
10.2
3.0
0.64
0.16
0.67
0.69
0.20
TABLE 3. Diagnostic characteristics of the genera Abrina and Macoma.
Abrina Habe, 1952. Type species; Abra
kanamarui Kuroúa. 1951 ; = Macoma
lunella Gou\ä, 1861
Macoma Leach. 1819. Type species:
Macoma teñera Leach, 1819; = Tellina
calcárea Gmelin, 1 791
Shell form and
proportions
Shell surface
Periostracum
Hinge
Ligament
Palliai sinuses
Ovate or triangular-ovate, moderately
inflated; anterior end longer than poste-
rior; posterior end with radial ridge
along posterodorsal margin, twisted to
the right
Smooth, with faint growth checkmarks
Thin, colorless, silky or shiny
Weak, two cardinal teeth in each valve,
lateral teeth absent
Both external and internal; external —
seated on a nymph not projecting
above dorsal margin; internal — lodged
in oblique groove behind cardinal teeth
Long, often of different length and form in
right and left valves, partly confluent
with palliai line
Ovate to subtrigonal, moderately inflated;
equivalve or with left valve some what
larger; equilateral to longer anteriorly;
posterior end with radial ridge along pos-
terodorsal margin, twisted to the right
Smooth, with faint growth checkmarks
Thin, dark to colorless, silky or shiny
Weak, two cardinal teeth in each valve, lat-
eral teeth absent
External, seated on a nymph not projecting
above dorsal margin
Long, often of different length and form in
right and left valves, partly confluent with
palliai line
STATUS OF ABRINA SPECIES FROM RUSSIA
215
FIGS. 12-21. Hinge of /Wacoma species. 12-17. Macoma loveni (Jensen, 1905). 12: Cardinal teeth andin-
ternal ligament of right valve (ZIN 106), Laptev Sea (76'"38'N, 118°20'E), 46 m, valve length 7.1 mm (bar =
1 mm). 13: Cardinal teeth and internal ligament of left valve, Iturup Island (44°23'4"N, 147"^37'8"E), 230 m,
valve length 8.7 mm (bar = 1 mm). 14: Cardinal teeth, external and internal ligament of right valve (ZIN 65),
western coast of Sakhalin Island, Tatar Strait, Sea of Japan, 84-86 m, valve length 13.4 mm (bar = 1 mm).
15: Cardinal teeth and resilifer of right valve, Iturup Island (44°23'4"N, 147'^'37'8"E), 230 m, valve length 8.6
mm (bar = 100 цт). 16: Cardinal teeth and resilifer of right valve (USNM 159769), Gulf of fVlaine, North At-
lantic, valve length 5.0 mm (bar = 1 00 |.im). 1 7: Cardinal teeth and ligament pit of right valve, Shikotan Island
(43"45'N, 146 55'E), 100 m, valve length 15.6 mm (bar = 1 mm). 18, 19. Macoma balthica (Linné, 1758)
(MIMB 12/14742), Signalny Island, Olyutorskiy Bay, Bering Sea, intertidal zone. 18: Cardinal teeth and re-
silifer of left valve, valve length 4.5 mm (bar = 100 (.im). 19: Cardinal teeth and ligament pit of right valve,
valve length 8.2 mm (bar = 1 00 |im). 20. Macoma calcárea (Gmelin, 1 791 ), cardinal teeth and resilifer of right
valve. Peter the Great Bay, Sea of Japan, 16 m, valve length 4.2 mm (bar = 1 mm). 21 . Macoma nipponica
(Tokunaga, 1906), cardinal teeth and ligament of right valve, Peter the Great Bay, Sea of Japan, 12 m, length
6.3 mm (bar = 100 |jm).
216
KAMENEV & NADTOCHY
TABLE 4. Macoma cuneipyga (Scarlato, 1981). Summary statistics of the shell measure-
ments (mm) and indexes: L — shell length; H — height: W — width; A — anterior end length;
LI —maximal distance from posterior margin to the top of palliai sinus; L2 — minimal dis-
tance from the top of palliai sinus to anterior adductor muscle scar. Numerator indicates the
summary statistics for the left valve, denominator — for the right valve.
Characteristics
Mean
SD
SE
Min
Max
п
L
11.29
5.52
1.03
3.1
20.0
29
10.43
5.66
1.16
3.1
19.7
24
H
7.93
3.75
0.70
2.4
14.0
29
7.50
3.96
0.81
2.4
14.0
24
W
2.28
1.13
0.21
03.
4.1
29
2.05
1.09
0.22
0.6
3.8
24
A
8.29
4.19
0.78
1Л
15.3
29
7.73
4.35
0.89
1.8
15.3
24
LI
7.22
3.41
0.63
2.2
12.9
29
5.80
3.07
0.63
1.9
11.0
24
L2
1.70
0.94
0.17
QA
3¿
29
2.50
1.54
0.31
0.7
5.5
24
H/L
0.714
0.038
0.007
0.65
0.80
29
0.730
0.031
0.006
0.69
0.80
24
W/L
0.201
0.015
0.003
0.17
0.24
29
0.197
0.013
0.003
0.17
0.22
24
A/L
0.721
0.010
0.056
0.58
0.82
29
0.723
0.013
0.061
0.58
0.82
24
L1/L
0.653
0.040
0.007
0.59
0.75
29
0.565
0.037
0.008
0.46
0.65
24
L2/L
0.146
0.027
0.005
0.09
0.19
29
0.233
0.043
0.009
0.18
0.30
24
faint growth lines, equivalve or with left valve
somewhat larger, equilateral to longer anteri-
orly. Posterior end with a radial ridge along
posterodorsal margin, twisted to the right,
usually slightly gaping. Periostracum thin,
dark or colorless, silky or shiny. Bevelled es-
cutcheon present in some species. Hinge
weak, two cardinal teeth in each valve, lateral
teeth absent. Ligament external, seated on a
nymph that does not project above dorsal
margin. In young specimens, ligament both
external and internal; internal ligament small,
lodged in lanceolate resilifer posterior to car-
dinal teeth. Palliai sinus long, often of different
length and form in each valve, partly confluent
with palliai line.
/Wacoma силе/руда (Scarlato, 1981)
(Figs. 22-37, Table 4)
Abrina cuneipyga Scarlato, 1 981 : 371 -372,
figs. 192, 193; Kafanov, 1991: 79.
Type material and Locality
Holotype (ZIN 9800), Onekotan Island, Mid-
dle Kuril Islands, 150 m, silty sand, bottom
water temperature of 0.7°C, Coll. N. S. Spi-
rina, 18-VII-1954 (R/V "Lebed").
Other Material Examined
1 lot (ZIN 6) from Green Island (1 spec); 1
lot from Shikotan Island (5 spec); 5 lots from
Kunashir Island (5 spec); 5 lots from Iturup Is-
land (ZIN 2, 1 spec; 4 lots of our material, 7
spec); 3 lots from Onekotan Island (ZIN 1 0, 1
spec; 2 lots of our material, 4 spec); 1 lot
(ZIN 11) from Makanrushi Island (1 spec); 2
lots from Fourth Kuril Strait (5 spec); 1 lot
from Paramushir Island (1 spec); 3 lots from
the western coast of Kamchatka, Sea of
Okhotsk (3 spec); 1 lot from northeastern
coast of Sakhalin Island, Sea of Okhotsk (1
spec). Total of 35 specimens.
Description
(Expanded from that of Scarlato, 1981)-
Exterior: Shell small (to 20.0 mm in length,
west coast of Kamchatka, Sea of Okhotsk),
ovate, of moderate height (H/L = 0.65-0.80),
inequilateral, moderately inflated (W/L of left
valve 0.17-0.24; W/L of right valve 0.17-
0.22), thin, white under periostracum, smooth
with faint growth lines, inequivalve (left valve
slightly longer, more inflated, less twisted to
right than right valve); posterior end distinctly
twisted, slightly gaping. Periostracum non-
STATUS OF ABRINA SPECIES FROM RUSSIA
217
EL IL PCT ACT
FIG. 22. The hinge plate of holotypes of Abrina species from Pacific seas of Russia aided by camera lucida:
A — Abrina sachalinica (left valve, length 15.0 mm); B — Abrina cuneipyga (left valve, 13.0 mm); C — Abrina
tatarica (right valve, posterior cardinal tooth is broken, 9.4 mm); D — Abrina sliiaslil<otanika (right valve, 11 .8
mm). ACT; anterior cardinal tooth; EL: external ligament; IL; internal ligament; PCT: posterior cardinal tooth.
Bar = 1 mm.
polished, thin, adherent, sometimes peeling
off near beaks, gray, yellowish or pink-brown,
thrown into numerous small wrinkles, more
conspicuous at shell margins. Beaks small,
somewhat projecting above dorsal margin,
rounded, not opisthogyrate, well posterior to
midline (A/L = 0.58-0.82). Anterior end ob-
tusely rounded, expanded vertically. Posterior
end pointed, angular, with a faint radial ridge
extending from posterior portion of beaks to
transition of posterior margin to ventral mar-
gin. Anterodorsal margin near beaks slightly
convex, forming small smooth obtuse angle,
extending almost horizontally, smoothly tran-
sitioning to rounded anterior margin. Ventral
margin slightly curved. Posterodorsal margin
very short, straight, steeply extending ven-
trally, forming a hardly noticeable angle at
transition to posterior margin, better ex-
pressed in left valve. Posterior margin
straight, abruptly transitioning to ventral mar-
gin, forming slightly smoothed acute angle in
left valve, more distinctly rounded in right
valve. External ligament short (1/2-2/3 of
posterodorsal margin length).
Interior: Hinge plate very narrow, not arcu-
ate in area of cardinal teeth. Hinge weak, with
two cardinal teeth in each valve. In left valve,
anterior tooth wide, deeply bifid, long, reach-
ing edge of hinge plate; posterior tooth nar-
row, lamellate, shorter, not reaching edge of
hinge plate. In right valve, anterior and poste-
rior teeth of almost same length and width
(anterior slightly longer), reaching edge of
hinge plate; posterior tooth sometimes slightly
bifid. Internal ligament in adult specimens
(> 15 mm) weakly developed, short (< 1/2 of
hinge plate width), lodged in a small pit on
hinge plate under beaks posterior to cardinal
teeth; internal ligament in young specimens
(< 8 mm) well developed, almost reaching
edge of hinge plate, lodged in lanceolate re-
silifer, which extends obliquely posterior to
beaks. Anterior adductor muscle scar large,
ovate, angular, vertically extended; posterior
adductor scar large, rounded, shorter and
wider than anterior scar. Palliai sinuses dis-
tinct, of different length and shape in each
valve. In left valve, palliai sinus long, with wide
rounded top, reaching past midline (L1/L =
0.59-0.75); ventral sinus branch confluent
with palliai line for more than 1/2 of its length.
In right valve, palliai sinus shorter (L1/L =
0.46-0.65) with a narrow, rounded top; ven-
tral sinus branch confluent with palliai line for
more than 1/2 of its length. Shell polished,
within sometimes with faint radial striae espe-
cially noticeable along ventral margin.
218
KAMENEV & NADTOCHY
FIGS. 23-37. Macoma cuneipyga (Scarlato, 1981). 23-27: Holotype of Abrina cuneipyga (ZIN 9800),
Onekotan Island, 1 50 m, shell length 1 3.0 mm. 28: Cardinal teeth and resilifer of left valve of specimen from
the type locality (ZIN 1 0), length 1 3.5 mm (bar = 1 mm). 29: Cardinal teeth and ligament of right valve. South
Kuril Strait (44' 1 2'N, 1 46' 32'E), 1 03 m, valve length 3.6 mm (bar = 1 mm). 30: Cardinal teeth and resilifer of
right valve, Iturup Island (44' 55'1 "N, 1 48''02'4"E), 1 97 m, valve length 6.6 mm (bar = 1 00 |.im). 31 : Cardinal
teeth and ligament of right valve, Iturup Island (44 55'1"N, 148 02'4"E), 197 m, valve length 11.1 mm (bar =
100 |.tm). 32-35: Fourth Kuril Strait, 200 m, left valve length 17.7 mm, right valve length 17.5 mm. 36, 37:
Kunashirsky Strait, Kunashir Island (43'58'5"N, 145 27'2"E), 200 m, left valve length 15.0 mm.
Variability
Shell shape and proportions change little
with age. In young specimens (< 8 mm) in con-
trast to adults, the shell is more rounded,
higher (H/L = 0.73-0.80) and angular: the an-
terodorsal margin is horizontal, then bends
(forming a smooth obtuse angle) and is
STATUS OF ABRINA SPECIES FROM RUSSIA
219
Sakhalin Is. .^ ^ ^ Paramushir Is.
#^
о
Onekotan Is.
Ogvlturupb
Sea of ps
Japan 1^^
Kimashirls. Pacific ОсвЕП
FIG. 38. Distribution of Macoma cuneipyga (♦ — type locality).
Steeply turned ventrally; the posterodorsal
margin at the transition to posterior margin
forms a distinct smooth angle; the posterior
margin at the transition to ventral margin forms
a pointed acute angle; the beaks one placed
less posteriorly and only slightly rounded; the
ridge on the posterior end of the shell is better
expressed; in left valve, the anterior tooth is
non-bifid and only slightly sulcate; the internal
ligament is very well developed, reaching
edge of hinge plate and lodged in a lanceolate
resilifer; the palliai sinus in left valve is longer
and reaches closer to the anterior adductor
muscle scar (L1/L - 0.65-0.75; L2/L = 0.11-
0.16). In adult specimens, the degree of con-
fluence of the palliai sinus with palliai line
varies. In left valve, ventral sinus branch con-
fluent for 1 /4 to 1 /2 of its length with palliai line;
in right valve, sometimes completely conflu-
ent.
Distribution and habitat (Figure 38)
Green Island (43°19'3"N, 146^23'7"E); Shi-
kotan Island (43°39'N, 147°08'E); South Kuril
Strait, Kunashir Island (from 43°48'1"N,
147°30'5"E to 44°12'N, 146°32'E); Blizky Is-
land, Kunashirsky Strait, Kunashir Island (43°
58'5"N, 145°27'2"E); Rok Bay, Iturup Island
(44"00'N, 147°44'E); Kasatka Bay, Iturup Is-
land (44'=527"N, 147°46.1'E); Iturup Island,
coast of the Sea of Okhotsk (45°2176"N,
148°187E); Drakon Cape, Iturup Island (44°
55'N, 148°02'4"E); Onekotan Island (49°16'N,
155°387"E); Makanrushi Island, Fourth Kuril
Strait (from 49°32'3"N, 155°47'2"E to 49°46'
6"N, 155°46'E); Dym Island, Paramushir Is-
land (49°36'1"N, 156°06'E); west coast of
Kamchatka, Sea of Okhotsk (52°45'N, 155°
42'E; 54°00"N, 155°30'E); northeastern coast
of Sakhalin Island, Sea of Okhotsk (143°45'
15"N,52°29'27"E).
This species was registered off the South
Kuril Islands at a depth from 53 m (South Kuril
Strait (43°51'3"N, 146°05'5"E)) to 300 m (Rok
Bay, Iturup Island, and Kunashir Island) on
sand and silty sand, sometimes with some ad-
mixture of gravel and small stones; near the
North Kuril Islands from 1 50 to 590 m on sand
with a admixture of gravel, small stones and
220
KAMENEV & NADTOCHY
TABLE 5. Macoma loveni (Jensen, 1905). Summary statistics of the shell measurements
(mm) and indexes: L — shell length; H — height; W — width; A — anterior end length; L1 —
maximal distance from posterior margin to the top of palliai sinus; L2 — minimal distance
from the top of palliai sinus to anterior adductor muscle scar. Numerator indicates the sum-
mary statistics for the left valve, denominator — for the right valve.
Characteristics
Mean
SD
SE
Min
Max
n
L
15.56
5.95
0.82
3.9
32.5
53
15.36
5.86
0.81
3.9
32.0
52
H
10.90
4.04
0.56
3.0
23.0
53
10.88
3.97
0.56
3.0
23.0
51
W
3.12
1.30
0.18
0.8
7.0
53
2.89
1.12
0.16
0.8
6.1
51
A
10.52
3.96
0.54
2.4
21.5
53
10.48
3.90
0.55
2.4
21.5
51
L1
9.76
3.47
0.48
2.8
18.0
53
8.11
3.00
0.42
2.2
15.6
52
L2
2.24
1.06
0.15
0.4
5.0
53
3.77
1.60
0.22
1.0
7.0
52
H/L
0.705
0.024
0.003
0.65
0.77
53
0.706
0.025
0.003
0.66
0.77
51
W/L
0.199
0.016
0.002
0.17
0.23
53
0.187
0.016
0.002
0.15
0.22
51
A/L
0.675
0.035
0.005
0.62
0.80
53
0.667
0.073
0.010
0.22
0.80
51
L1/L
0.639
0.047
0.006
0.55
0.88
53
0.530
0.034
0.005
0.46
0.61
52
L2/L
0.139
0.025
0.003
0.08
0.20
53
0.243
0.031
0.004
0.18
0.32
52
debris; near the western coast of Kamchatka
at a depth of 50-53 m, on silty sand at a bot-
tom water temperature of 0.83° to 1.42°C.
Near the Kuril Islands, it was observed at a
bottom water temperature of 0.7-5.4°C (Scar-
lato, 1981).
Comparison
M. cuneipyga is distinguished from all other
species of Macoma by its short, pointed pos-
terior end. in shape and proportions, this
species is closest to M. loveni, from which it is
distinguished in having a very narrow, pointed
gaping posterior end, beaks less projecting
above the dorsal margin and less posteriorly
placed, and a narrower hinge plate that is not
arcuate in the area of the cardinal teeth.
Remarks
Scarlato (1 981 ) described this species from
of 22 specimens (10 samples) (Scarlato by
mistake mentioned 12 samples: specimens in
ZIN 1 and 3, 7 and 9 were taken from 2 sam-
ples, not from 4), of which the holotype is the
largest (shell length, 1 3.0 mm). We have stud-
ied all of this material. In addition to the holo-
type, only 4 specimens (samples 2, 6, 10 and
11) could be surely identified as M. cunei-
pyga. All others are young specimens of the
genus Macoma that are not easily identified
and are different from the holotype.
In our material, we have found 31 speci-
mens of different ages. The studies show that
the typical characteristics of this species are
not much prone to age or individual variation.
A well-developed internal ligament is a juve-
nile characteristic. We consider that A.
cuneipyga is a separate species of Macoma.
for which a new combination of M. cuneipyga
is suggested.
Macoma loveni {Jensen, 1905)
(Figs. 12-17, 22, 39-51, Table 5)
Tellina (Macoma) loveni Jensen, 1905: 45,
pi. 1, figs. 5a-h (Steenstrup MS) (cited from
Coan, 1971); Coan, 1971: 31-32, fig. 19, pi.
8, figs. 42, 43, synonymy.
Macoma loveni (Jensen, 1905), Coan,
1971: 31-32, fig. 19. pi. 8, figs. 42, 43. syn-
onymy; Bernard, 1979: 49, fig. 81; Lubinsky
1 980: 42, pi. 9, figs. 2,5,8,11; Scarlato, 1 981
362-363, fig. 362, synonymy; Bernard, 1983
45; Romejko & Kamenev, 1985: 94; Baxter
STATUS OF ABRINA SPECIES FROM RUSSIA
221
FIGS. 39-53. Macoma loveni {Jensen, 1905). 39-43; Holotype of Abrina sachalinica (ZIN 9853), Terpeniya
Bay, Sakhalin Island, Sea of Okhotsk (48'22'N, 145^1 7'E), 220 m, shell length 15.0 mm. 44, 45: ZIN 106,
Laptev Sea (76'38'N, 118"20'E), 46 m, left valve length 11.2 mm. 46, 47: Yury Island (43''11'N, 146°14'E),
300 m, left valve length 22.0 mm. 48, 49: Paramushir Island (49'45'9"N, 1 55"58'5"E), 1 35 m, left valve length
17.5 mm. 50, 51: Northern Sea of Okhotsk (53°34'N, 139°34'E), 130 m, left valve length 13.3 mm. 52, 53:
Northern Sea of Okhotsk (58 •16'5"N, 142°54'3"E), 134 m, left valve length 26.1 mm.
222
KAMENEV & NADTOCHY
1987: 25; Kafanov, 1991: 72; Feder et al.,
1994: 160; Kamenev, 1995: 6.
Abrina sachalinica Scarlato, 1981 : 372, fig.
194; Kafanov, 1991:79.
Material Examined
Abrina sachalinica— holotype (ZIN 9853)
and 3 specimens of this species from the type
locality, Terpeniya Bay, Sakhalin Island, Sea of
Okhotsk (48°22'N, 145°17'E), 220 m, silty
sand. Coll. L. G. Nazvich, 30-VI-1951 (R/V
"Vityaz"); 1 lot (ZIN 4) from Hokkaido Island,
southern Sea of Okhotsk (1 spec); 1 lot (ZIN
5) from west coast of Kamchatka, Sea of
Okhotsk (1 spec); /W. loveni - 1 lot (USNM
1 08789) from Spitzbergen (2 spec); 1 lot (ZIN
94) from Barents Sea (3 spec); 1 lot (ZIN 1 01 )
from Franz-Joseph Land, Barents Sea (4
spec); 3 lots (ZIN 107) from Novaya Zemlya,
Kara Sea (17 spec); 1 lot (ZIN 106) from the
Laptev Sea (4 spec); 1 lot (ZIN 98) from the
East Siberian Sea (7 spec); 2 lots from
Vrangel Island, Chukchi Sea (2 spec); 1 lot
(USNM 95638) from Murray Bay, Gulf of St.
Lawrence, Atlantic Ocean (2 spec); 1 lot
(USNM 159769) from Gulf of Maine, Atlantic
Ocean (1 spec); 1 lot (ZIN 57) from Terpeniya
Bay, Sakhalin Island, Sea of Okhotsk (2
spec); 3 lots from Shikotan Island (ZIN 82, 1
spec; 2 lots of our material, 3spec.); 1 lot (ZIN
67) from Tatar Strait, Sea of Japan (1 spec); 1
lot (ZIN 65) from the western coast of Sakhalin
Island, Tatar Strait, Sea of Japan (5 spec); 1
lot (USNH 204814) from Aniva Bay (1 spec);
1 lot (USNM 204815) from coast of Japan
("coast Yesso", 37°9N), Sea of Japan (2 spec);
2 lots from Peter the Great Bay, Sea of Japan
(2 spec); 1 lot from east coast of Sakhalin Is-
land, Sea of Okhotsk (1 spec); 3 lots from the
northern Sea of Japan (5 spec); 1 lot from
Blizky Island, Kunashirsky Strait (3 spec); 2
lots from Yury Island (2 spec); 4 lots from Itu-
rup Island (10 spec); 1 lot from Paramushir Is-
land (1 spec); 1 lot from Kronotsky Bay, east-
ern coast of Kamchatka, Pacific Ocean (4
spec);1 lot from Bering Sea (1 spec). Total of
92 specimens.
Description
(Expanded from that of Coan, 1971;
Bernard, 1979; and Scarlato, 1981) - Exte-
rior: Shell small (to 37.0 mm, Terpeniya Bay,
Sakhalin Island, Sea of Okhotsk), ovate, of
moderate height (H/L = 0.65-0.77), inequilat-
eral, moderately inflated (W/L of left valve
0.1 7-0.23; W/L of right valve 0.1 5-0.22), thin,
white under periostracum, smooth, with faint
growth lines, sometimes inequivalve (left
valve sometimes slightly longer, more in-
flated than right valve); posterior end slightly
twisted to right, without a gape. Periostracum
thin, gray, grayish-olive and from light to
bright brown with an iridescent sheen, dehis-
cent, non-polished, easily peeled off near
beaks, thrown into small wrinkles, better
expressed at shell margins. Beaks small,
somewhat projecting above dorsal margin,
rounded, not opisthogyrate, well posterior to
midline (A/L = 0.62-0.80). Anterior end ob-
tusely rounded, expanded vertically. Posterior
end slightly truncate, with faint radial ridge ex-
tending from posterior portion of beaks to
transition of posterior margin to ventral mar-
gin. Anterodorsal margin slightly convex, ex-
tending almost horizontally, smoothly transi-
tioning to rounded anterior margin. Ventral
margin slightly curved. Posterodorsal margin
slightly concave, smoothly extending ven-
trally, forming a very smooth angle at tran-
sition to posterior margin. Posterior margin
slightly convex, vertically extending ventrally,
forming a smooth angle at transition to ventral
margin. External ligament short (2/3 of pos-
terodorsal margin length).
Interior: Hinge plate narrow, arcuate in area
of cardinal teeth. Hinge weak, with two cardi-
nal teeth in each valve. In left valve, anterior
tooth wide, bifid, long, reaching edge of hinge
plate; posterior tooth narrow, lamellate,
shorter, not reaching edge of hinge plate. In
right valve, anterior and posterior teeth of al-
most same length and width (posterior slightly
longer), almost reaching edge of hinge plate;
posterior tooth bifid. Internal ligament in adult
specimens (> 1 5 mm) weakly developed, short
(< 1/2 width of hinge plate), lodged in a small
pit on hinge plate under beaks behind cardinal
teeth; internal ligament in young specimens
(to 5-10 mm) well developed, reaching almost
to edge of hinge plate, lodged in a lanceolate
resilifer, which extends obliquely posterior to
beaks. Anterior adductor muscle scar large,
ovate-angular, vertically extended; posterior
adductor scar large, rounded, shorter and
wider than anterior scar. Palliai sinus distinct,
of different length and shape in each valve. In
left valve, palliai sinus long, with wide, rounded
top, reaching past midline (L1/L = 0.55-0.88),
approaching anterior adductor muscle scar
STATUS OF ABRINA SPECIES FROM RUSSIA
223
(L2/L = 0.08-0.20): ventral sinus branch con-
fluent with palliai line for more than 1/3 of its
length. In right valve, pallia! sinus shorter (L1/L
= 0.46-0.61), with narrow, rounded top stop-
ping just short of midline; ventral sinus branch
confluent with palliai line for more than 1/3 of
its length. Shell interior polished, with faint ra-
dial striae especially noticeable along ventral
margin.
Variability
Shell size, shape and proportions, as well
as the inner shell morphology of M. loveni
from the Arctic Ocean and the Pacific Ocean
differ slightly. This species is much larger
(maximal shell size: L = 37.0; H = 25.5; W =
25.5 mm, Terpeniya Bay, Sakhalin Island, Sea
of Okhotsk; Scarlato, 1 981 ) in the Pacific than
in other parts of its distribution. In the Arctic
and the North Atlantic oceans, M. loveni at-
tains 20 mm (Coan, 1971; Bernard, 1979).
Specimens (especially young) from the Pa-
cific Ocean also often have a more elongate
shell, with the beaks placed less posteriorly.
The shell shape of M. loveni from the Pa-
cific varies from ovate to elongate-ovate.
Specimens from the Bering Sea and the
northern Sea of Okhotsk have a more ovate
shell compared to specimens from the Kuril
Islands and the Sea of Japan (Figs. 46-51).
The shape of the posterior end varies from
angular and distinctly truncate to almost
rounded. The length and width of the posterior
end also varies.
The degree of confluence of the palliai
sinus with the palliai line varies in both valves.
In some specimens, the ventral branch of the
palliai sinus in left valve is not confluent with
palliai line. Usually the ventral sinus branch in
the left valve is confluent with the palliai line
for 1/4-1/5 of its length. In the right valve, the
ventral sinus branch is frequently confluent for
1/3 of its length.
In specimens from the Pacific Ocean with a
shell length up to 10-12 mm, the resilium is
large, reaching the edge of hinge plate and
lodged in a lanceolate resilifer (Figs. 13, 15).
In M. loveni from the Arctic Ocean and the
North Atlantic Ocean, a similar internal liga-
ment was observed only in specimens up to
7-8 mm (Figs. 12, 16). Moreover, in speci-
mens from the Arctic and Atlantic oceans, the
hinge plate is relatively wider and more arcu-
ate in the area of the cardinal teeth. In some
specimens from throughout the distribution.
the anterior tooth of the left valve is slightly
bifid.
Distribution and Habitat in the Northwestern
Pacific (Fig. 54)
In the Pacific Ocean, M. loveni occurs in the
Sea of Japan -near Japan (37°9N), Possyet
Bay, Peter the Great Bay (42°33'N,
131 °22'7"E) ; from Olga Bay to Syurkum Cape,
near Moneron Island, near Sakhalin Island
(from Kholmsk City to Kitoushi Cape
(50^=1 0'0"N, 140°57'1"E)); in the Sea of
Okhotsk -near Sakhalin Island (Aniva Bay,
Terpeniya Bay, east coast) and in the northern
part of the sea (58^^16'5"N, 142°54'3"E), near
western Kamchatka (53°35'N, 155°11'5"E);
near the Kuril Islands -in Kunashir Strait
(Blizky Island, 43"58'5"N, 145°27'2"E), near
Yury Island (43''07'N, 1 46°1 2'E; 43°1 1 'N, 1 46°
14'E), near Shikotan Island (43°29'N, 147°
OO'E), near Iturup Island (Pacific and Sea of
Okhotsk coasts), near Paramushir Island
(Dym Island, 49°45'9"N, 155°58'5"E); near
eastern Kamchatka in Kronotsky Bay, Pacific
Ocean; in the Bering Sea (61°34'8"N, 179°
59'7"E): near Bering Island, the Commander
Islands.
In the Sea of Japan, this species was ob-
tained at a depth from 40 m (western coast of
Sakhalin Island, Krasnogorsk Village (48°
277"N, 1 41 °56'6"E) to 850 m (coast of Japan,
USNM 20481 5) on sand, silty sand, sandy silt,
often with admixture of gravel and shell debris
at a bottom temperature of 0.4-5.7°C; in the
Sea of Okhotsk from 16 m (Aniva Bay,
Sakhalin Island) to 220 m (Terpeniya Bay,
Sakhalin Island) on silt and silty sand with
some admixture of gravel at a bottom temper-
ature from -1.7°C(55°34'N, 139°34'E, depth
130 m) to 3.6°C (Aniva Bay, Sakhalin Island,
depth 16 m); near the Kuril Islands from 75 m
(Pacific coast of Iturup Island) to 1,000 m
(Yury Island, 43°07'N, 146°12'E) on sand,
silty sand, sandy silt and silt often with admix-
ture of gravel at a bottom temperature from
1 .3°C (Shikotan Island, depth 1 00 m) to 6.4°C
(Pacific coast of Iturup Island, 75 m); in Kro-
notsky Bay (eastern cost of Kamchatka, Pa-
cific Ocean) -from 134 to 150 m on silt and
sandy silt with admixture of gravel and small
stones at a bottom temperature of 1 .0°C
(53°33'N, 160°09'E, depth 150 m); near the
Commander Islands at a depth of 60-80 m on
sand with some admixture of gravel; in the
Bering Sea at 134 m on sandy silt with some
224
KAMENEV & NADTOCHY
■ 60
Bering Sea
Bering Is.
Kronotslo' Bay t^ , ^.
o**
ф ^ Paramushir Is.
>\Ä\^Iturup_Is.
Kunashiris. Pacific Ocean
FIG. 54. Distribution of Macoma loveniin the northwestern Pacific (♦ — sampling site of holotype of Abrina
sachalinica).
admixture of gravel at a bottom temperature
of 2.56°C.
Comparison
In shell shape and proportions, as well as in
palliai sinus size and form, this species is
closest to M. cuneipyga. However, M. lovenl
differs in having a wide, truncate posterior end
without a gape, with more projecting beaks,
and a wider hinge plate that is arcuate in the
area of the cardinal teeth. A comparison of
these species with reference to other charac-
teristics (Table 6) shows that the in left valves
differ significantly in A/L and L1/L indices, and
their right valves in all indices with the excep-
tion of L2/L. Thus, in addition to the charac-
teristics mentioned above, the beaks of M.
loveni compared to M. cuneipyga are placed
less posteriorly, the palliai sinus in the left
valve is more elongate, and in the right valve
is shorter proportional to shell length. More-
over, as compared with M. loveni. the right
valve of /W. cuneipyga is shorter and more in-
flated than the left valve. This is because M.
cuneipyga is more gaping posteriorly and be-
cause the posterior end of right valve is more
twisted to the right and more rounded. In M.
loveni, the shape of posterior end of right and
left valves is similar.
Discriminant analysis of all lots containing
M. loveni and M. cuneipyga showed that
these species differ significantly (Table 7) in a
complex of parameters (group centroids for
M. cuneipyga and M. loveni 1.60979 and
-0.694421, respectively). Of 73 specimens
analysed, 65 were accurately classified
(89.04%) (1 specimen was mistakenly classi-
fied as M. cuneipyga and 7 as M. loveni). The
most significant characteristics for dividing all
specimens into two species were shell length
(L) and shell anterior end length (A) for right
STATUS OF ABRINA SPECIES FROM RUSSIA
225
TABLE 6. Results of comparison by pairs of mean values of indices of left and righit valves of Macoma
cuneipyga and Macoma loveni using Student (T) and Kolmogorov-Smirnov (S) tests: L — shell lengthi;
H — height; W — width: A — anterior end length; LI —maximal distance from posterior margin to the
top of palliai sinus: L2 — minimal distance from the top of palliai sinus to anterior adductor muscle
scar. P — probability that the index values in M. cuneipyga and M. loveni are drawn from the same
population.
Left valves
Right valves
Indices
Test value
P
Test
Test value
P
Test
H/L
1.10
0.272
T
3.30
0.002
T
W/L
0.259
0.162
S
0.471
0.001
S
A/L
0.581
<0.001
s
0.593
<0.001
s
L1/L
0.344
0.024
s
3.84
<0.001
T
L2/L
1.16
0.246
T
0.94
0.649
T
TABLE 7. Results of discriminant analysis for M. cuneipyga (22 spec.) and M. loveni (5^ spec): L — shell
length: H — height: A-anterior end length: LI —maximal distance from posterior margin to the top of pal-
liai sinus. P — probability that characteristic values in M cuneipyga ano M. /oi/en/are drawn from the same
population.
Discriminant
Canonical
Wilks
Significant
Standardized
Valves
Function
Correlation
Lambda
P
characteristics
coefficient
Both
1
0.73126
0.465254
<0.001
Right
1
0.62539
0.608883
<0.001
L
A
-1.20894
1.612
Left
1
0.63291
0.599421
<0.001
H
A
L1
4.80796
-6.96378
2.562
valves, and height (H), anterior end length
(A), and distance from the shell posterior mar-
gin to the top of palliai sinus (LI) for left
valves.
Remarks
is short in contrast to the width of the hinge
plate (Fig. 41 ). We have found resilium of sim-
ilar shape and size in the specimens of M.
loveni of the same size as the holotype of A.
sachalinica (Fig. 14). Therefore, we consider
that A. sachalinica is a synonym of /W. loveni.
Scarlato (1 981 ) described A. sachalinica on
the basis of a study of 13 specimens (7 sam-
ples), from which the holotype is the largest
(shell length, 1 5.0 mm). We have studied all of
this material. In addition to the holotype, only
5 specimens (samples ZIN 1 , 4 and 5) could
be included in this species. Other specimens
are young specimens (< 8 mm) of Macoma of
uncertain identity. The studies of M. loveni
from the Arctic Ocean and the northwestern
Pacific, as well as its descriptions, photos and
pictures (Golikov & Scarlato, 1967; Coan,
1971; Bernard, 1979; Lubinsky, 1980; Scar-
lato, 1 981 ), show that in shell morphology the
holotype of A. sachalinica is identical to spec-
imens of M. loveni. The holotype of A.
sachalinica has a small internal ligament that
Macoma calcárea (Gmelin, 1791)
(Figs. 20, 22, 55-59, Table 8)
For complete synonymy, see Coan, 1971:
20-21; Scarlato, 1981:356.
Abrina tatarica Scarlato, 1981: 373-374,
fig. 197; Kafanov, 1991:79.
Material Examined
A. /atenea -holotype (ZIN 9900), southern
coast of Sakhalin Island near Kholmsk town
(47°02'N, 142"00'5E), Sea of Japan, 46 m,
sand with some admixture of shell debris, bot-
tom water temperature of 3.9°C, Coll. V. A.
Skalkin, 8-X-1949 (R/V "Toporok"); 1 lot (ZIN
226
KAMENEV & NADTOCHY
TABLE 8. Macoma calcárea (Gmelin, 1791). Shell measurements (mm) and indices of specimens identi-
fied as Abrina tatarica (ZIN 1. 5, 6): L-shell lengtfi; H -height; W- width; A — anterior end length; LI —
maximal distance from posterior margin to the top of palliai sinus; L2 — minimal distance from the top of
palliai sinus to anterior adductor muscle scar. Numerator indicates shell measurements and indices for the
left valve, denominator — for the right valve. NM — not measured. Holotype of Abrina tatarica in italics.
L
H
W
A
L1
L2
H/L
W/L
A/L
L1/L
L2/L
7.0
4.6
1.0
4.7
5.1
0.4
0.66
0.14
0.67
0.73
0.057
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
NM
9.4
6.2
1.4
6.2
6.8
0.7
0.66
0.15
0.66
0.72
0.074
9.4
6.2
1.4
6.2
5.8
1.6
0.66
0.15
0.66
0.62
0.17
9.7
6.3
1.6
6.0
7.0
0.5
0.65
0.16
0.62
0.72
0.052
9.7
6.3
1.1
6.0
6.0
1.5
0.65
0.11
0.62
0.62
0.15
10.5
6.8
1.6
6.6
8.0
0.7
0.65
0.15
0.63
0.76
0.067
10.5
6.8
1.6
6.6
6.5
1.5
0.65
0.15
0.63
0.62
0.14
FIGS. 55-63. Shells of Macoma calcárea (Gmelin, 1791) and Масогла sp. 55-58. Holotype of Abrina tatar-
ica (ZIN 9900). southern coast of Sakhalin Island, Sea of Japan (47"02'N, 142 00'5"E), 46 m, shell length
9.4 mm. 59. Cardinal teeth and ligament of left valve of specimen identified as Abnna tatarica (ZIN 6), Ter-
peniya Bay, Sakhalin Island, Sea of Okhotsk (49 03'1 "N, 1 14 1 7'3"E), 1 6 m, valve length 1 0.5 mm (bar = 1
mm). 60-63. Holotype of Abrina shiashkotanika (ZIN 9867), Shiashkotan Island, 2270 m, left valve length
1 1 .9 mm, right valve length 1 1 .8 mm.
5) from Kievka Bay, Sea of Japan (1 spec); 1
lot (ZIN 6) from Terpeniya Bay, Sakhalin Is-
land, Sea of Okhiotsk (1 spec, 1 left valve); M.
calcárea -^0 lots from Peter the Great Bay,
Sea of Japan (64 spec). Total of 67 speci-
mens, 1 left valve.
Remarks
Abrina tatarica was described from a study
of 16 specimens (6 samples) (Scarlato,
1 981 ). We have studied this material and con-
sider that in addition to the holotype only 2
STATUS OF ABRINA SPECIES FROM RUSSIA
227
TABLE 9. Macoma sp. Shell measurements (mm) and indices: L — shell length; H — height; W — width;
A — anterior end length; LI —maximal distance from posterior margin to the top of palliai sinus; L2 — min-
imal distance from the top of palliai sinus to anterior adductor muscle scar. Numerator indicates shell
measurements and indices for the left valve, denominator — for the right valve. NM — not measured.
Holotype of Abrina shishkotanika in italics.
L
H
W
A
L1
L2
H/L
W/L
A/L
L1/L
L2/L
9.5
6.1
L5
5^
5.6
L6
0.64
0.16
0.62
0.59
0.17
9.5
6.1
1.5
5.9
4.9
2.0
0.64
0.16
0.62
0.52
0.21
10.0
6.9
1.5
6.2
6.5
1.5
0.69
0.15
0.62
0.65
0.15
10.0
6.8
1.5
6.2
NM
NM
0.68
0.15
0.62
NM
NM
11.9
7.6
1.8
7.4
8.0
2.0
0.64
0.15
0.62
0.67
0.17
11.8
7.6
1.6
7.4
6.8
3.0
0.64
0.14
0.63
0.58
0.25
specimens (samples ZIN 5 and 6) can be re-
ferred to this species. The other specimens
differ significantly from holotype and are
young specimens of different species of Ma-
coma. Scarlato (1981) stated that the holo-
type is the largest specimen (shell length, 9.4
mm). However, the largest is a specimen from
a sample ZIN 6 (shell length, 10.5 mm).
On the basis shell morphology, the holotype
of A. tatarica is identical to young specimens
of M. calcárea from Peter the Great Bay, Sea
of Japan. The internal ligament of the holo-
type is small, very short (about 1/3 of hinge
plate width) and lodged in a small pit on the
hinge plate ventral to the beaks posterior to
the cardinal teeth (Fig. 22). We have found ré-
silia of similar shape and size in specimens of
M. calcárea of the same size as the holotype
of A. tatarica. Kafanov et al. (1997) presented
the results of a morphometric analysis of a
large sample (150 spec.) of specimens of dif-
ferent ages (L = 14.3-39.8 mm) of M. cal-
carea from the Bering Sea. A comparative
analysis of these data with ours shows that
the values of all the indices of A. tatarica are
within the range of variation of similar indices
(H/L = 0.657-0.773; W/L - 0.125-0.192; A/L
= 0.503-0.662; L1/L = 0.566-0.801 ) of M. cal-
carea from the Bering Sea. Therefore, we
consider that A. tatarica is a synonym of M.
calcárea.
Macoma sp.
(Figs. 22, 60-63, Table 9)
Abrina shiashkotanika Scarlato, 1981 : 373,
figs. 1 95, 1 96; Kafanov, 1 991 : 79.
Type material and locality
Holotype (ZIN 9867), Shiashkotan Island,
Middle Kuril Islands, 2,270 m, sand. Coll. L.
G. Nazvich, 7-8-VII-1951 (R/V "Vityaz").
Other Material Examined
Two specimens (ZIN 1) from the type local-
ity.
Description
(Expanded from that of Scarlato, 1981)-
Exterior: Shell small (to 11.9 mm, holotype),
elongate-ovate, of moderate height (H/L =
0.64-0.69), inequilateral, flattened (W/L of left
valve 0.15-0.16; W/L of right valve 0.14-
0.16), thin, smooth, with faint growth lines, in-
equivalve (left valve slightly longer, more in-
flated than right valve); posterior end distinctly
twisted to right. Periostracum thin, adherent,
grey, thrown into small wrinkles more con-
spicuous at shell margins. Beaks small, not
very prominent above dorsal margin, slightly
rounded, not opisthogyrate, posterior to mid-
line (A/L = 0.62-0.63). Anterior end obtusely
rounded, not expanded vertically. Posterior
end slightly truncate, with faint (better ex-
pressed in right valve) radial ridge extending
posteriorly from beaks to transition of poste-
rior margin to ventral margin. Anterodorsal
margin straight, extending almost horizontally
parallel to ventral margin, smoothly transi-
tioning to rounded anterior margin. Ventral
margin slightly curved. Posterodorsal margin
slightly concave, smoothly extending ven-
trally, forming a smooth angle at transition to
posterior margin. Posterior margin slightly
convex, almost vertical, extending ventrally
and forming a strongly smooth angle at tran-
sition to ventral margin. External ligament
short (approximately 1/2 of posterodorsal
margin length.)
Interior: Hinge plate narrow, not arcuate in
area of cardinal teeth. Hinge weak, with two
cardinal teeth in each valve. In left valve, an-
terior tooth bifid, wide, long, reaching edge of
hinge plate; posterior tooth narrow, lamellate,
228
KAMENEV & NADTOCHY
shorter, not reaching edge of hinge plate. In
right valve, anterior and posterior teeth of al-
most same length and width (posterior slightly
longer); posterior tooth sometimes bifid. Inter-
nal ligament large, reaching edge of hinge
plate, lodged in lanceolate resilifer, which ex-
tends obliquely posterior to beaks. Anterior
adductor muscle large, ovate, extended verti-
cally; posterior adductor scar large, rounded,
shorter and wider than anterior scar. Palliai si-
nuses indistinct, of different length and shape
in each valve. In left valve, palliai sinus long,
with narrow rounded top, reaching past mid-
line (L1/L = 0.59-0.67), approaching anterior
adductor muscle scar (L2/L = 0.15-0.17);
ventral sinus branch confluent for 1/2 of its
length with palliai line. In right valve, palliai
sinus shorter (L1/L = 0.52-0.58; L2/L =
0.21 -0.25), with a narrower rounded top; ven-
tral sinus branch half confluent with palliai
line. Shell inside polished, with faint radial stri-
aes.
Distribution and Habitat
Known only from type locality.
Comparison
In shell form and proportions, this species is
similar to M. loveni and Macoma moesta (De-
shayes, 1855). However, it has a narrower,
more elongate shell, with beaks placed less
posteriorly, and a narrower posterior end
compared to M. loveni and M. moesta of the
same size.
Remarks
Abrina shiashkotanika was described
based on the study of only three specimens (1
sample) (Scarlato, 1981), and of which the
holotype is the largest (shell length, 1 1 .9 mm).
We have studied these specimens and con-
sider them to belong to one species. In the
description of this species, Scarlato (1981)
mentioned that the anterior tooth of the right
valve is deeply bifid. Our studies have shown
that in the right valve of the holotype both
teeth were non-bifid, but in the other two
specimens the posterior tooth is deeply bifid.
Probably, Scarlato (1981) made a mistake
and described the anterior tooth of the left
valve. Of the three specimens studied, only
the smallest had a deeply bifid anterior tooth
in the left valve.
In describing this species, Scarlato (1981)
had only a small amount of material, with all
specimens mainly smaller than 10 mm and in
poor condition. We think that these are young
specimens of a Macoma, for which it is difficult
to make an accurate species identification.
Possibly, as a result of further studies, addi-
tional material will be obtained representing
all age groups of this species. It is not incon-
ceivable that this species will be considered
synonymous with a known species of Ma-
coma. At present, taking into consideration
the peculiarities of shell morphology and the
depth from which this species was collected
(2,270 m), as well as the lack of any additional
material, we assume it to be a separate
species of Macoma.
ACKNOWLEDGMENTS
We are very grateful to Ms. M. B. Ivanova
(1MB, Vladivostok) for help and consultation
during our work; to Mrs. N. V. Kameneva
(1MB, Vladivostok) for great help during work
on this manuscript; to Professor O. G. Kus-
sakin. Academician of the Russian Academy
of Sciences (1MB, Vladivostok) for consulta-
tion during our work and comments on the
manuscript; to Mr. K. A. Lutaenko (1MB, Vladi-
vostok), Ms. R. N. Germon (USNM, Washing-
ton) and T. N. Belan (FERHI, Vladivostok) for
providing at our disposal the specimens of M.
loveni and M. cunelpyga: \o Dr. B. I. Sirenko,
Mr. A. V. Martynov and all collaborators of Ma-
rine Research Laboratory (ZIN, St. Peters-
burg) for sending to us the specimens of
species of Abrina and help during work with
collection of bivalve molluscs of ZIN; to Mr. A.
Yu. Voronkov (ZIN, St. -Petersburg) for pro-
viding the additional information on the dis-
tribution of M. loveni: to Dr. E. V. Coan (De-
partment of Invertebrate Zoology, California
Academy of Sciences, San Francisco) for the
consultations and comments on the manu-
script; to Dr. H. Saito (NSMT, Tokyo) for send-
ing to us the specimens of A. lunella and A. ki-
noshltalAo Drs. T Kurozumi and E. Tsuchida
(NHMI, Chiba) for sending to us the speci-
mens of A. kinoshital and reprints of neces-
sary papers; to Dr. K. Amano (Joetsu Univer-
sity of Education, Joetsu) for sending to us the
reprints of scientific papers necessary for our
work; to Mr. E. V. Jakush (PRIFO, Vladivos-
tok) for help in work with scanning micro-
scope; to Mr. A. A. Omelyanenko (1MB, Vladi-
vostok) for making photographs; to Mrs. A. V.
STATUS OF ABRINA SPECIES FROM RUSSIA
229
Vysotskaya and Ms. T. N. Koznova (1MB,
Vladivostok) for translating the manuscript
into English.
This research was supported by Grant 98-
04-48279 from the Russian Foundation for
Basic Research.
LITERATURE CITED
BAXTER, R., 1987, Mollusks of Alaska: a listing of
all mollusks, freshwater, terrestrial, and marine
reported from the State of Alaska, with locations
of the species types, maximum sizes and marine
depths inhabited. Shells and Sea Life: Bayside,
California, 161 pp.
BERNARD, F. R., 1979, Bivalve mollusks of the
western Beaufort Sea. Conthbution in Science.
Natural History Museum of Los Angeles County
313: 1-80.
BERNARD, F R., 1983, Catalogue of the living Bi-
valvia of the eastern Pacific Ocean: Bering Strait
to Cape Horn. Canadian Special Publications of
Fisheries and Aquatic Sciences. 61 : 1 -1 02.
COAN, E. v., 1971, The northwest American
Tellinidae. The Veligen 14 (Suppl.): 1-63.
DUNNILL, R. M. & D. V. ELLIS, 1969, Recent
species of the genus Macoma (Pelecypoda) in
British Columbia. National Museum of Canada.
Natural History Paper 45: 1 -34.
FEDER, H. M., N. R. FOSTER, S. С. JEWETT T J.
WEINGARTNER & R. BAXTER, 1994, Mollusks
in the northeastern Chukchi Sea. Arctic. Al:
145-163.
GOLIKOV, A. N. & O. A. SCARLATO, 1967, Mol-
luscs of Possiet Bay (Sea of Japan) and their
ecology Pp. 5-154, in: ya. i. starobogatov, ed..
Molluscs and their role in biocenoses and in the
formation of faunas. Proceedings of the Zoologi-
cal Institute, 42. "Nauka" Publ. House, Leningrad
[in Russian].
GOULD, A. A., 1861, Descriptions of shells col-
lected by the North Pacific Exploring Expedition.
Proceedings of the Boston Society Natural His-
tory 8: 14-40.
HABE, T., 1952, Genera of Japanese shells. No. 3.
Pelecypoda. 187-280 pp. [in Japanese].
HABE, T., 1 958, Report on the mollusca chiefly col-
lected by the S. S. Soyo-Maru of the imperial fish-
eries experimental station on the continental
shelf bordering Japan during the years 1922-
1930. Part 4. Lamellibranchia (2). Publications of
the Seto Mahne Biological Laboratory 7: 19-52.
HABE, T., 1 961 , Coloured illustrations of the shells
of Japan. Vol. II. Hoikusha, Osaka, 182 pp.
HABE, T, 1964, Shells of the western Pacific in
color. Vol. II. Hoikusha, Osaka, 233 pp.
HABE, T., 1977, Systematics of Mollusca in Japan.
Bivalvia and Scaphopoda. Tokyo, 372 pp. [in
Japanese].
HABE, T., 1 981 , Bivalvia. A catalogue of molluscs of
Wakayma Prefecture, the Province of Kii. I. Bi-
valvia, Scaphopoda and Cephalopoda. Publica-
tions of the Sato Marine Biological Laboratory,
Special Publication Series. 7(1): 25-224.
ITO, K., 1 967, A catalogue of the marine molluscan
shell-fish collected on the coast of and off Tajima,
Hyogo Prefecture. Bulletin of the Japan Sea Re-
gional Fisheries Research Laboratory, 1 8: 39-91
[in Japanese, with English abstract].
ITO, K., 1989, Distribution of molluscan shells in the
coastal areas of Chuetsu, Kaetsu and Sado Is-
land, Niigata Prefecture, Japan. Bulletin of the
Japan Sea Regional Fisheries Research Labora-
tory. 39: 37-133 [in Japanese, with English ab-
stract].
ITO, K., Y. MATANO, Y. YAMADA, & S. IGARASHI,
1986, Shell species ought S/S Rokko-Maru off
the coast Ishikawa Prefecture. Bulletin of the
Ishikawa Pre feet и ral Fisheries Expehmental Sta-
tion. 4: 1-179 [in Japanese, with English ab-
stract].
KAFANOV, A. I., 1991, Shelf and continental slope
bivalve molluscs of the northern Pacific Ocean: a
check-list. Far Eastern Branch of the Academy of
Sciences of the USSR, Vladivostok, 200 pp. [in
Russian, with English summary].
KAFANOV, A. I., D. D. DANILIN & A. V. MO-
SHENKO, 1997, Morphometric analysis of taxo-
nomic characters in bivalve mollusks of the genus
Macoma. Biologya Morya [Marine Biology], 23:
345-355 [in Russian, with English summary].
KAMENEV, G. M., 1989, Contribution to the study
of bivalve mollusks of Kraternaya Bight. Biologya
Morya [Marine Biology], 3: 88-93 [in Russian,
with English summary].
KAMENEV, G. M., 1990, A new species of the
genus Macoma (Bivalvia: Tellinidae) from the
Kraternaya Bay (Yankitch Island, Kuril Islands).
Zoologicheskiy Zhurnal [Zoological Journal], 69:
127-132 [in Russian, with English summary].
KAMENEV, G. M., 1995, Species composition and
distribution of bivalve mollusks on the Comman-
der Islands shelf. Malacological Review. 28: 1 -
23.
KEEN, A. M., 1969, Superfamily Tellinacea. Pp.
613-643, in R. с MOORE, ed.. Treatise on inver-
tebrate paleontology. Pt. N, Vol. 1 . Mollusca 6, Bi-
valvia. Geological Society of America and Uni-
versity of Kansas.
KULAICHEV, A. P., 1996, Methods and means of
data analysis in Windows. STADIA 6.0. Computer
science and computers, Moscow, 255 pp. [in
Russian].
KURODA, T, 1 951 , Descriptions of a new genus of
a marine gastropod, Kanamurua. gen. п., and a
new species of a bivalve. Abra kanamarui. sp. п.,
dedicated to Mr. T. Kanamaru on his 60"" birthday.
Venus, 16: 68-72 [in Japanese].
KURODA, T, T HABE, & К. OYAMA, 1971, Sea
shells of Sagami Bay. Maruzeh, Tokyo. 741 pp.
[in Japanese] + 489 pp. [in English].
LUBINSKY, I., 1 980, Marine bivalve molluscs of the
Canadian central and eastern Arctic: faunal com-
position and zoogeography. Canadian Bulletin of
Fisheries and Aquatic Sciences. 207: 1-111.
ROMEJKO, L. V. & G. M. KAMENEV, 1985, Bivalve
230
KAMENEV & NADTOCHY
molluscs of the sublittoral zone of Moneron Is-
land Pp. 85-101, in: o. G. KUSSAKIN, ed., Ben-
thos of the shelf of Moneron Island. Far East Sci-
ence Center, Academy of Sciences of the USSR,
Vladivostok [in Russian].
SCARLATO, O. A., 1981 , Bivalve mollusks of tem-
perate waters of the northwestern Pacific.
■■Nauka" Publ. House, Leningrad, 480 pp. [m
Russian].
TSUCHIDA, E. & T. KUROZUMI, 1995, Fauna of
marine mollusks of sea around Otsuchi Bay,
Iwate Prefecture (5) BIVALVIA-2. Otsuchi Marine
Research Center Report, 20: 13-42 [in Japan-
ese].
Revised ms. accepted 19 January 1999
MALACOLOGIA, 1999, 41(1): 231-252
TERRESTRIAL GASTROPOD RICHNESS OF CARBONATE CLIFF AND
ASSOCIATED HABITATS IN THE GREAT LAKES REGION OF NORTH AMERICA
Jeffrey С Nekola
Department of Natural and Applied Sciences. University of Wisconsin - Green Bay.
Green Bay. Wisconsin 54311, U.S.A.; nekolaj@gbms01.uwgb.edu
ABSTRACT
The richness of terrestrial gastropod communities in 19 different habitat types in a 1,300 x
1 ,000 km region in the Great Lakes region of North America was analyzed using 349 0.01 -0.1
ha samples. Sites supporting high-richness faunas (24 or more taxa) were limited in the study
region to areas south of 45' N. Only weakly significant longitudinal gradients in richness were ob-
served, while a significant latitudinal gradient was present. When only wooded carbonate out-
crops were analyzed, a significant negative correlation between richness and latitude was pres-
ent only between 44'N and 45'N. Highly significant differences in richness between habitats
were also observed. Carbonate cliffs harbored the richest faunas, possessing a mean greater
than 20. Approximately 25% of these sites contained 24 or more taxa, with a maximum richness
of 34 being recorded. Algific talus slopes and lakeshore carbonate ledges were also found to
commonly harbor faunas of 1 7 or more taxa. All of these sites are characterized by shaded, ver-
tical exposures of carbonate bedrock. Only two of the habitats (old fields and open dunes) were
found to never support a dozen or more co-occurring taxa. Based on these analyses, carbonate
cliffs and related habitats in the Great Lakes region should be included among the most impor-
tant habitats on a global scale for molluscan biodiversity
Keywords: terrestrial gastropods, community ecology biodiversity, conservation, North Amer-
ica, Niagaran Escarpment, cliff ecology.
INTRODUCTION
One of the more important components of
community structure is richness, or the num-
ber of co-occurring species (Peet, 1974).
While Solem & Climo (1 985) suggest that land
snail community richness rarely exceeds 1 2, a
number of other studies have documented
much higher rates of sympatry. Tropical forest
ecosystems have the highest reported rich-
nesses at various sample grains, with up to 40
taxa being reported from individual sites in the
Greater Antilles (Solem, 1984), 45 species
from a 400 m^ area in southwestern Came-
roon (de Winter & Gittenberger, 1998), 50
species from a <4 ha site near Amboni Cave in
eastern Tanzania (Emberton et al., 1997), 52
species from a 4 ha area near Manombo,
Madagascar (Emberton, 1995), and 56 spe-
cies from the 4.2 ha Waipipi Scenic Reserve in
New Zealand (Solem et al., 1 981 ). Communi-
ties with high land snail richness have also
been reported from the temperate zone. Up to
24 species have been recorded from 0.01 ha
areas in the Italian Alps (Bishop, 1980), 26
species 0.09 ha regions in British Columbia
coniferous forests (Cameron, 1986), 27
species from 9.1 ha Ekholmen Island in Swe-
den (Nilsson et al., 1 988), 39 species from ap-
proximately 1 ha samples in SW Sweden
(Waiden, 1981), and 44 species from an ap-
proximately 4 ha site on Pine Mountain in Har-
lan County, Kentucky (Emberton, 1995). How-
ever, such sites are uncommon enough that
Tattersfield (1996) concludes, based upon his
review of the international literature, that sites
with 24 or more sympatric terrestrial gastropod
taxa in small to moderate sample sizes (ap-
prox. <10 ha) are of global conservation im-
portance.
Reconnaissance of a dozen eastern Wis-
consin limestone and dolomite cliff sites for
glacial relict snails (Nekola et al., 1996) docu-
mented three sites that possessed 24 or more
co-occurring species. Previous surveys made
from similar habitats in northeastern Iowa
(Frest, 1982, 1987; Prest & Pay, 1981) docu-
mented at least five additional sites that also
equalled or exceeded this level of sympatry. If
these surveys are reflective of such sites as a
whole, carbonate cliffs and associated habi-
tats in central North America could be consid-
231
232
NEKOLA
о = sites with 23 or fewer taxa
• = sites with 24 or more taxa
FIG. 1. Distribution of high-richness terrestrial gastropod communities within Great Lakes region.
ered among the richest global terrestrial gas-
tropod communities, particularly at small
scales. Unfortunately, given the preliminary
nature of these investigations, it was not pos-
sible to determine whether these sites were
simply outliers, or whether other habitats in
this landscape harbored similarly high num-
bers of species.
The purpose of this paper is to investigate
the terrestrial gastropod richness of carbon-
ate cliffs and other habitats in a 1 ,300 x 1 ,000
km region centered on the North American
Great Lakes. Through this it will be possible to
better estimate the frequency of high diversity
assemblages from various habitats, to identify
potential geographic gradients in species rich-
ness over this extent, and to compare the
richness of carbonate cliff habitats to others in
this landscape.
MATERIALS AND METHODS
Study Sites
A total of 349 areas were surveyed for their
terresthal gastropod faunas (Fig. 1, Appendix
I). Sites were chosen for survey if they repre-
sented typical examples of their respective
habitat and (except for anthropogenic habi-
tats) were undisturbed. They ranged from
north-central Iowa through Wisconsin, the
Upper Peninsula of Michigan, and southern
Ontario (including Manitoulin Island and the
Bruce Peninsula), to central New York State
(an extent of approximately 1,300 km), and
from northeastern Minnesota and the Keewe-
naw Peninsula in Michigan to southern Illinois
(an extent of approximately 1 ,000 km). The
bulk of collections were made along or adja-
cent to the Niagaran Escarpment, a narrow
zone of exposed Siluhan-age carbonates ex-
tending from Rochester, New York, to West
Union, Iowa. Sampling was most intensive
(252 sites) from Drummond Island, Michigan,
through northeastern Iowa, where an effort
was made to sample along the Escarpment
from all areas supporting carbonate bedrock
outcrops.
Collections were made from nineteen dis-
tinct habitat types: carbonate cliffs (114 sites),
igneous cliffs (72), rocky woodlands (21),
lakeshore carbonate ledges (19), fens (19),
algific talus slopes (16), tamarack wetlands
(16), lakeshore alluvial banks (12), upland
woods (8), lowland woods (8), white cedar
wetlands (8), calcareous meadows (7), cob-
ble beaches (7), alvars (6), carbonate glades
TERRESTRIAL GASTROPOD RICHNESS IN THE GREAT LAKES
233
(5), igneous shorelines (4), old fields (3),
shale cliffs (3), and open dunes (1).
Carbonate cliffs represent 3-30 m tall,
wooded limestone or dolomite outcrops that
typically support moss or fern-covered
ledges. Igneous cliffs are wooded, 2-20 m tall
basalt, serpentine, or granite outcrops and as-
sociated open talus slopes located on the
Precambrian Shield of northern Wisconsin,
the Upper Peninsula of Michigan, and north-
eastern Minnesota. Rocky woodlands are up-
land tracts with abundant bedrock or glacial-
erratic boulders. Lakeshore carbonate ledges
are <3 m tall, wooded limestone or dolomite
outcrops that are within 3 km of the Lake
Michigan or Lake Huron shore. Fens are peat-
land areas formed at locations of ground-
water discharge that maintain higher soil
moisture and a cooler soil temperatures than
is otherwise found in the surrounding land-
scape (Nekola, 1994). Sampling was only
conducted from sites in which Sphagnum
mosses were either uncommon or lacking. Al-
gific talus slopes are associated with me-
chanical karst systems harboring year-round
ice reservoirs. Air and water drainage from
these ice caves through loose carbonate talus
has created an unique buffered microclimate
where soil temperatures rarely range lower
than -10°C in winter or exceed 10°C in the
summer, and have a more constant soil mois-
ture as compared to surrounding forest soils.
Such sites have been shown (Frest, 1991) to
support populations of the glacial relict snails
Catinella gélida (F. С. Baker, 1927), Discus
macclintockii {F. С. Baker, 1928), Henderso-
nia occulta (Say, 1831), and Vertigo hubrichti
Pilsbry, 1934. Tamarack wetlands represent
almost pure Larix laricina (DuRoi) K. Koch,
stands that are open and support abundant
AInus rugosa (DuRoi) Spreng, and Carex
growth. Collections were limited to areas that
lacked Sphagnum cover. Such sites appear
restricted to regions with thin soils over car-
bonate bedrock. Lakeshore alluvial banks
represent steep wooded banks along the
Lake Michigan shore that are developed
into unconsolidated lacustrine material. Up-
land woods represent wooded tracts devel-
oped on soils lacking large rocky debris. Low-
land woods represent deciduous forests
found in floodplains or depressions. White
cedar wetlands represent forested peatlands,
dominated by Thuja occidentalis L., that
are associated with groundwater seepage.
Surficial soil chemistry can vary from acidic
(where Sphagnum moss is abundant) to neu-
tral or alkaline (where Sphagnum is largely
absent). Litter collections were limited to the
latter class of sites. Calcareous meadows are
open or very sparsely forested wet meadows
found on carbonate-rich mineral (rather than
organic) substrate. Cobble beaches are con-
stantly wet shoreline grassland habitats de-
veloped on flat limestone or dolomite pave-
ment with little or no soil development except
in bedrock fracture planes. Alvars are similar
to cobble beaches except that they are found
in upland locations and become xeric by mid-
summer. Carbonate glades are xeric grass-
land communities with thin soils overlying
limestone, dolomite, or calcareous shales. Ig-
neous shoreline sites occur along the Lake
Superior coast in the Keewenaw Peninsula
where basalts or basalt-derived conglomerate
sequences are exposed. They are largely
treeless, have only limited soil development,
and support a number of western and arctic
disjunct vascular plants. Old fields represent
early successional grasslands that develop
following agricultural abandonment. Shale
cliffs represent wooded cliffs or banks devel-
oped into shale exposures that are often kept
wet through constant groundwater seepage.
Open dunes are xeric grasslands found in
sandy soils along the Great Lakes shore.
Field Sampling
Documentation of the terrestrial gastropod
communities from each site was accom-
plished by hand collection of larger shells and
litter sampling for smaller taxa from represen-
tative 100-1,000 m^ regions within sites. As
suggested by Emberton et al. (1996), sample
collection was concentrated at places of high
micro-mollusc density, with a constant volume
of soil litter (approximately 4 liters) being col-
lected from each site. For woodland sites, lit-
ter collection was concentrated: (1) at places
with an abundance of larger shells; (2) along
the base of rocks or trees; (3) on soil covered
ledges; and/or (4) at cold air vents on the cliff
face or in the associated talus. For open sites,
collections consisted of: (1) small blocks (ap-
prox. 1 25 cm^) of turf; and/or (2) loose soil and
leaf litter accumulations under or adjacent to
shrubs, cobbles and/or boulders.
The location of each sample was marked
on USGS (or equivalent) 7.5 minute topo-
graphic maps. The latitude-longitude coordi-
nates for each was then determined through
digitization of these maps using the ATLAS
DRAW software package. Conversion of loca-
234
NEKOLA
tions into UTM Zone 1 6 coordinates was com-
pleted using ARCINFO.
Laboratory Procedures
Samples were slowly and completely dried
in either a low-temperature soil oven (approx.
бО-Эб^С) or in full sun in a greenhouse. Dried
samples were then soaked in water for 3-24
hours, and subjected to careful but vigorous
water disaggregation through a standard
sieve series (ASTME 3/8" (9.5 mm), 10 (2.0
mm), 20 (0.85), and 40 (0.425 mm) mesh
screens). Sieved sample fractions were then
dried and passed again through the same
sieve series. These dry, resorted fractions
were then hand picked against a neutral-
brown background. All shells and shell frag-
ments were removed.
All recovered, identifiable shells were as-
signed to species (or subspecies) using the
author's reference collection and the Hubricht
Collection at the Field Museum of Natural His-
tory. From this, the total number of taxa per
site was determined. All specimens have
been catalogued and are housed in collec-
tions maintained at the University of Wiscon-
sin-Green Bay.
Statistical Analyses
The frequency of high richness (24 or more
taxa) sites was calculated across all habitats,
and for wooded carbonate outcrops (carbon-
ate cliffs, algific talus slopes, lakeshore car-
bonate ledges) only, within each of the in-
cluded states or provinces (Illinois, Iowa,
Minnesota, Michigan, southern Ontario, and
New York). Testing for statistical differences in
the ratio of high vs. normal or low richness
sites was conducted via the Pearson chi-
square and likelihood ratio tests. The likeli-
hood ratio test was calculated as some of the
predicted values were sparse (< 5), complicat-
ing interpretation of Pearson's chi-square sta-
tistic. The asymptotic distribution of the likeli-
hood ratio test, however, is trustworthy when
the number of observations (349 and 149, re-
spectively) equal or exceed the number of
cells (1 4) by a factor of ten (Zar, 1 984). Based
on apparent differences in the ratio of high-di-
versity sites between northern and southern
sections of the study area, these tests were
repeated following exclusion of sites from Min-
nesota, Michigan and Manitoulin Island.
The relationship between geographic posi-
tion and richness was graphically represented
by plotting site richness vs. UTM N-S or UTM
E-W coordinates for (1) all habitats, and (2)
for wooded carbonate outcrop (carbonate
cliff, algific slope, and carbonate lakeshore
ledge) sites only. The central tendencies in
these relationships were indicated though lo-
cally weighted scatterplot smoothing (Cleve-
land, 1979). The statistical significance of
these relationships, and amount of variance in
richness accounted for by geographic posi-
tion, was estimated using least-squares re-
gression. Cartesian UTM coordinates were
analyzed to preclude biases originating from
use of polar-coordinate latitude and longitude
coordinates.
For the N-S relationships, locally weighted
scatterplot smoothing indicated that the re-
sponse of richness might not be constant.
Tests for such differences in response were
conducted by splitting the data sets into dif-
ferent N-S position regions, and repeating
regression analyses separately for each. The
p-values and r^ for each of these models were
recorded.
The central tendency in site richness
among habitat types was graphically repre-
sented via a box plot with habitats being
sorted along the horizontal axis from the high-
est to lowest means. In box plots the central
line represents the median of the sample, the
margins of the box represent the interquartile
distances, and the fences represent 1 .5 times
the interquartile distances. For data having a
Gaussian distribution, approximately 99.3%
of the data will fall inside of the fences (Velle-
man & Hoaglin, 1 981 ). Outliers falling outside
of the fences are shown with asterisks. Test-
ing for significant differences in the average
richness between habitats was conducted
using ANOVA.
RESULTS
Regional Patterns
Forty of 349 sites harbored 24 or more ter-
restrial gastropod taxa within 0.01-0.1 ha
samples (Fig. 1). Seven sites (4 Iowa, 1 Illi-
nois, 1 Ontario, 1 Wisconsin) harbored 30 or
more taxa, with a maximum richness of 34
being observed from a Brown County, Wis-
consin, site. Eighty-five percent of high rich-
ness sites were found on carbonate cliff habi-
tats. The only non-carbonate cliff habitats that
possessed high terrestrial gastropod richness
were three algific talus slopes (all with imbed-
ded carbonate cliffs) and single white cedar
TERRESTRIAL GASTROPOD RICHNESS IN THE GREAT LAKES
235
TABLE 1. Ratio of high richness (24 or more taxa) to medium and low richness (23 or fewer taxa) sites in
states and provinces within study region.
All sites
Wooded с
#high
arbonate outcrops
State or Province
#high
# medium-low
# medium-low
Illinois
Iowa
Michigan
Minnesota
New York
Ontario
Wisconsin
4
8
1
1
5
21
5
25
74
39
5
17
144
4
7
1
1
5
19
5
16
18
2
14
57
All states and provinces
-
All sites
Wooded carbonate outcropi
Pearson Statistic
p-value
Likelihood Ratio Statistic
p-value
30.763
< 0.00005
34.2086
< 0.00005
6.2791
0.2800
7.3618
0.1951
Minnesota, Michigan and Manitoulin Island s
ites excluded
Ail sites
Pearson Statistic
p-value:
Likelihood Ratio Statistic
p-value:
12.2529
0.0156
10.4536
0.0334
wetland, tamarack wetland, and rocky wood-
land sites.
Approximately 11% of all sampled sites had
24 or more taxa (Table 1). The frequency of
these high richness sites in the seven states
or provinces varied between 0% and 44% of
all sites. These differences were significant
(Pearson chi-square and likelihood ratio p <
0.00005). It appeared possible that this differ-
ence may be attributed to the much lower fre-
quency of high-richness sites in Minnesota,
Upper Peninsula of Michigan, and Manitoulin
Island. However, differences in the frequency
of high-richness sites was found to remain
marginally significant (Pearson chi-square p =
0.01 56; likelihood ratio p = 0.0334), even after
removal of the most northern regions from
analysis. This marginal significance is appar-
ently related to a lowered frequency of high-
richness sites in Wisconsin.
Approximately 25% of all wooded carbon-
ate outcrop sites harbored high richness com-
munities. The frequency of these in the five
states or provinces ranged between 5% and
44% (Table 1), and occurred over the entire
extent of the sample region (Fig. 1). While
carbonate cliff sites of high-richness ap-
peared scarce in the Upper Peninsula of
Michigan and Manitoulin Island, Pearson's
chi-square (p = 0.2800) and the likelihood
ratio (p = 0.1951) tests demonstrated that at
the state or province scale, these differences
were non-significant.
Geographic Gradients
Only a marginally significant (Fig. 2; p =
0.031 ) and weak {r^ = 0.01 3) trend was found
between richness and E-W UTM position
across all habitats. This relationship was
found to not be significant when only wooded
carbonate outcrops were analyzed (Fig. 3; p =
0.106). The relationship between richness
and N-S UTM location, however, was found to
be stronger and more significant both for all
habitats (p < 0.0005; л^ - 0. 1 88) as well as for
wooded carbonate outcrop sites only (p =
0.003; r^ - 0.059), with northerly sites pos-
sessing lower richness than southerly sites.
The shape of the locally weighted scatter-
plot smoothing lines for the N-S relationships,
in conjunction with additional regression
analyses, demonstrate that this pattern is not
constant over the study region. Across all
habitats, only a weak {r^ = 0.063) but statisti-
cally significant (p = 0.001) relationship was
observed south of 5,000 km while north of this
position this same relationship was more sig-
nificant (p < 0.0005) and over 4V^ times
stronger {r^ = 0.289; Fig. 2). When only
236
NEKOLA
E-W Variation
35
30
25 \-
20
15
10
5
N-S Variation
"T 1 r
■
All sites
r2 - 0.188
p < 0.0005
.."...■
Sites <5000km N:
'..'..11 .".■
n = 181
- •
r2 - 0063
""fl^v •.;••■ "
p - 0,001
.-•■VT^ .
Srtes >5000km N
... •.•.■-• .;'
n = 168
1-2 - 0.289
p < 0.0005
500 1000 1500
E-W UTM Coofdrate (1000 meter uits)
4000 4200 4400 4600 4800 5000 5200 6400
N-S 1ЛМ Coordinate (1000 meter uiüs)
FIG. 2. Relationship of terrestrial gastropod richness to E-W an(j N-S UTM location across all 19 habitat
types. A locally weighted scatterplot smoothing line has been fitted to each relationship.
E-W Variation
35
30
25 h
20
N-
-s
Variation
All sites
r2 - 0059
p - 0003
' ~7
-—
-^^Щ,-
Sites <49(X)km N:
n = 58
r2 - 0,010
p - 0.456
-
Sites from 4900-5000 km N
n = 44
r2 - 0,211
p - 0002
-
Sites >5000km N.
n = 47
r2 - 0002
p - 0742
500 1000 1500
E-W UTM Cocrdirete (1000 meter uils)
N-S UTM Coordinate (1000 meter irits)
FIG. 3. Relationship of terrestrial gastropod richness to E-W and N-S UTM location for wooded carbonate
outcrop sites (carbonate cliffs, algific talus slopes, and lakeshore carbonate ledges). A locally weighted scat-
terplot smoothing line has been fitted to each relationship.
wooded carbonate outcrops were considered,
no relationship was apparent south of 4900
km (roughly 44" N; p = 0.456) and north of
5,000 km (approx. 45" N; p = 0.742). How-
ever, a significant (p = 0.002) and moderately
strong (r = 0.221) relationship was apparent
between 4,900 and 5,000 km (Fig. 3).
Habitat Patterns
Comparison of site richness values demon-
strate striking differences among the 19 sam-
pled habitat types (Fig. 4). Carbonate cliffs
were the richest habitats sampled, possess-
ing a mean score approaching 21 . Algific talus
slopes and lakeshore carbonate ledges fol-
lowed, having mean richness scores exceed-
ing 1 7. Both carbonate cliffs and algific slopes
had upper data fences that exceeded 30
species per site. Rocky woodlands, carbonate
glades, calcareous meadows, white cedar
wetlands and fens had mean richness scores
ranging from 1 5.3 to 1 3.9. Igneous shorelines,
tamarack wetlands, lakeshore alluvial banks,
lowland woods and cobble beaches had
mean richness scores ranging from 12 to
1 0.6. Igneous cliffs, alvars, shale cliffs, upland
woods, old fields, and open dunes all had
mean richness scores of less than 1 0. ANOVA
showed these differences to be highly signifi-
cant (p < 0.0005), with almost 50% of ob-
served variance in richness being accounted
for by habitat type.
DISCUSSION
Regional Species Richness Patterns
Although Solem & Climo (1985) stated that
land snail community richness rarely exceeds
TERRESTRIAL GASTROPOD RICHNESS IN THE GREAT LAKES
П — \ — \ — I I \ I — \ — I — I — \ — I — г
237
г2 = 0.473
p < 0.0005
Habitat Type
FIG. 4. Box-plot diagrams of terrestrial gastropod richness across 1 9 habitat types. Habitats are sorted along
the horizontal axis from highest to lowest mean scores.
1 2 taxa, fully 232 of the sites inventoried (66%
of the total) equalled or exceeded this level.
Sites with 1 2 or more taxa were found in 1 7 of
the 19 sampled habitats. Only old field and
open dune habitats never equalled or ex-
ceeded this richness level. It is not clear
whether Solem & Climo were unnecessarily
pessimistic about terrestrial gastropod com-
munity richness, or if the Niagaran Escarp-
ment in the Great Lakes region possesses
uniquely rich community assemblages. While
it seems likely that the former is true, it should
be mentioned that sites with a dozen or more
co-occurring terrestrial gastropod taxa may
be less frequent in other landscapes. For ex-
ample, in this study only 34% of northern Wis-
consin, western Upper Peninsula and north-
eastern Minnesota sites had 12 or more
co-occurring taxa. Burch (1956) reported
maximum mean richness of nine taxa per site
in the eastern piedmont and coastal plain of
Virginia. Clarke et al. (1968) found no more
than nine co-occurring taxa in New Brunswick
forests. In their survey of 1 89 sites (many with
carbonate substrates) in the Black Hills of
South Dakota, Prest & Johannes (1993) re-
port only seven (less than 4%) that harbor a
dozen or more taxa. Cowie et al. (1 995) found
that no more than 1 2 taxa coexisted within ap-
proximately 100 m^ samples on Hawaiian
vegetated lava flows. It will be necessary to
expand these analyses to additional land-
scapes with a greater diversity of geological
substrates and ecological histories to deter-
mine whether the terrestrial gastropod com-
munities of the Great Lakes are uniquely rich,
or if our definition of what constitutes a
species-rich community must be expanded.
Little or no variation in richness was re-
corded over most of the region. Only a very
weak longitudinal trends were identified, and
only north of UTM 5,000 were strong latitudi-
nal trends observed. When only wooded car-
bonate outcrops (carbonate cliffs, algific talus
slopes, and lakeshore carbonate ledges)
were considered, the significant latitudinal
trend in richness was restricted to sites falling
between UTM 4,900 and 5,000 km N (or
roughly 44- N to 45" N). Similarly, the occur-
rence frequency of high richness sites (using
Tattersfield's criteria of 24 or more taxa) was
found to only weakly differ between Illinois,
Iowa, southern Ontario, New York, and Wis-
consin. However, in Minnesota, the Upper
Peninsula of Michigan, and Manitoulin Island
this ratio was over ten times lower across all
238
NEKOLA
habitats, and at least five times lower on
wooded carbonate outcrops.
A number of factors could be responsible
for the significantly lower richness levels ob-
served in the northern reaches of the study
area. At least some of this decrease in rich-
ness may be due to lower-Ca and pH soils as-
sociated with igneous (rather than carbonate)
bedrock on northern sites. However, this can
not explain the significant reductions in rich-
ness observed on the northern-most wooded
carbonate outcrop sites in the Upper Penin-
sula and Manitoulin Island. Perhaps the low
richness values are related to the greater iso-
lation of these sites, because they are sepa-
rated from other carbonate outcrops by the
waters of Green Bay and Georgian Bay as
well as the acidic soils of the Precambrian
Shield. Additional research will be necessary
to tease apart the differential roles played by
contemporaneous and historical processes in
determining regional terrestrial gastropod
richness patterns.
Habitat-Specific Species Richness Patterns
Significant differences were observed
among the 16 sampled habitats, with carbon-
ate cliffs possessing the highest average num-
ber of taxa per site. Over one-half of such sites
harbored 21 or more species. Other habitats
found to harbor rich assemblages of species
included algific talus slopes, lakeshore car-
bonate ledges, rocky woodlands, carbonate
glades, calcareous open meadows, white
cedar wetlands, and fens. All of these habitats
are associated with calcareous substrata, ei-
ther in the form of exposed bedrock, boulders,
talus, wet marl, calcareous alluvium or nutri-
ent-rich peat. The lowest richness habitats
were, in general, associated with more acidic
substrata such as igneous outcrops, sand
dunes, or exposed alluvium. However, this
pattern is not without exception as low-rich-
ness cobble beach and alvar faunas are de-
veloped on carbonate outcrops.
Carbonate Cliffs as Terrestrial Gastropod
Diversity Hot Spots
Wooded carbonate cliffs, on average, sup-
port the highest number of terrestrial gastro-
pod taxa within any habitat in the study re-
gion. The richest 5% of these support 29 or
more taxa, with a maximum of 34 taxa being
recorded. Such sites appear to be among the
richest reported globally from 1 ha or smaller
quadrats. Waiden (1981) observed up to 39
taxa from 1 ha quadrats in wooded talus
slopes in Sweden, while Tattersfield (1996)
identified up to 33 taxa per one-sixth hectare
samples from Kenyan rain forest. Other pub-
lished reports of terrestrial gastropod richness
from 0.1 ha or less quadrats (e.g., Schmid,
1966; Bishop, 1980;Nilssonetal., 1988;Getz
& Uetz, 1994; Cowie et al., 1995; de Winter &
Gittenberger, 1998) have reported no more
than 45 co-occurring taxa. Maximum richness
in Great Lakes carbonate cliff sites is also
within 25% of the richest known North Ameri-
can site (at Pine Mountain, Kentucky; Ember-
ton, 1995).
Further research will be necessary to deter-
mine if the richness levels of carbonate cliffs
in the Great Lakes region are unique, or if sim-
ilar levels are present in other landscapes.
Research from other regions (e.g.. New South
Wales, Australia; Stanisic, 1997; Germany:
Schmid, 1966; Scotland; Cameron & Green-
wood, 1 991 ; Sweden: Waiden 1 981 ) indicates
that maximum terrestrial gastropod richness
frequently occurs on wooded carbonate out-
crops. Based on this current and previous re-
search it seems likely that carbonate cliffs will
be found to be among the most important
habitats for molluscan biodiversity on a global
scale.
ACKNOWLEDGEMENTS
Useful comments on earlier versions of this
manuscript were provided by Douglas Larson,
John Slapcinsky, George Davis, and two
anonymous reviewers. Donna Boelk, Patrick
Comer, Gary Fewless, John Gerrath, Mike
Grimm, Doug Larson, and Mary Standish
helped identify potential survey sites, and as-
sisted in field collection. Matt Barthel, Candice
Kasprzak, Pete Massart, Chela Moore, Eric
North, and Tamara Smith processed many
soil litter samples, and assisted in field collec-
tion. Assistance in litter sample processing
was also provided by students participating in
the Land Snail Ecology Practicum at the Uni-
versity of Wisconsin -Green Bay. Funding
was provided by the Door County Office of the
Wisconsin Chapter of The Nature Conser-
vancy, a Louis Almon grant (administered by
the Wisconsin Academy of Sciences, Arts,
and Letters), three Cofrin Arboretum grants
(administered by the Cofrin Arboretum Com-
mittee at the University of Wisconsin -Green
Bay), the U.S. Fish and Wildlife Service, and
TERRESTRIAL GASTROPOD RICHNESS IN THE GREAT LAKES
239
the Small Grants Program of the Michigan
Department of Natural Resources. Funding
for the survey of Minnesota sites was re-
ceived from the Minnesota Nongame Wildlife
Tax Checkoff and Minnesota State Park Na-
ture Store Sales through the Minnesota De-
partment of Natural Resources Natural Her-
itage and Nongame Research Program.
LITERATURE CITED
BISHOP, M. J., 1980, The Mollusca of acid wood-
land in the Italian province of Novara. Journal of
Conchology. 30: 181-188.
BURCH, J. В., 1956, Distribution of land snails in
plant associations in eastern Virginia. The Nau-
tilus, 70: 60-64.
CAMERON, R. A. D., 1986, Environment and diver-
sities of forest snail faunas from coastal British
Columbia. Malacologia. 27: 341-355.
CAMERON, R. A. D. & J. J. D. GREENWOOD,
1 991 , Some montane and forest molluscan fauna
from eastern Scotland: effects of altitude, distur-
bance, and isolation. Proceedings of the Tenth In-
ternational l\/lalacological Congress (Tübingen
1989): 437-442.
CLARKE, A. H., JR., J. R KELSALL & G. R.
PARKER, 1968, The land snail fauna of Fundy
National Park, New Brunswick. National!\Auseum
of Canada Contributions in Zoology. 223: 5-22.
CLEVELAND, W. S., 1979. Robust locally weight
regression and smoothing scatterplots. Journal of
the American Statistical Association, 70: 548-
554.
COWIE, R. H., G. M. NISHIDA, Y. BASSET & S. M.
GON, III., 1995, Patterns of land snail distribution
in a montane habitat on the island of Hawaii.
Malacologia. 36: 155-169.
DE WINTER, A. J. & E. GITTENBERGER, 1998,
The terrestrial gastropod fauna of a square kilo-
meter patch of rainforest in southwestern
Cameroon: high species richness, low abun-
dance, and seasonal fluctuations. Malacologia,
40:231-250.
EMBERTON, K. C, 1995. Land-snail community
morphologies of the highest-diversity sites of
Madagascar, North America, and New Zealand,
with recommended alternatives to height-diame-
ter plots. Malacologia, 36: 43-66.
EMBERTON, K. C, T A. PEARCE, P. F. KASIGWA,
R TATTERSFIELD & Z. HABIBU, 1997, High di-
versity and regional endemism in terrestrial gas-
tropods of eastern Tanzania. Biodiversity and
Conservation. 6: 1123-1136.
EMBERTON, K. C, T. A. PEARCE & R. RAN-
DALANA, 1996, Quantitatively sampling land-
snail species richness in Madagascan rain-
forests. Malacologia, 38: 203-212.
FREST T J., 1982. Final report. Project SE-1-4
(Iowa Pleistocene Snail). Iowa Conservation
Commission. Des Moines.
FREST T J., 1987, Final report. Project SE- 1-8
(Iowa Pleistocene Snail). Iowa Department of
Natural Resources. Des Moines.
FREST. T. J.. 1991. Summary status reports on
eight species of candidate terrestrial gastropods
from the Driftless Area (Paleozoic Plateau),
upper Midwest. Final Report. Contract #301-
01366, USFWS Region 3. Ft. Snelling, Min-
nesota.
FREST T J. & L. R FAY. 1981 . Vertigo mermacen-
sis (Pulmonata: Pupillidae) from the Niagaran
Escarpment. Iowa. The Nautilus, 95: 33-37.
FREST T J. & E. J. JOHANNES, 1993, Land snail
survey of the Black Hills National Forest. South
Dakota and Wyoming. Final Report, Contract
#43-67TO-2-0054, USDA Forest Service. Black
Hills National Forest and USDI Fish & Wildlife
Service. South Dakota State Office.
GETZ, L. L. & G. W. UETZ. 1994. Species diversity
of terrestrial snails in the southern Appalachian
mountains. U.S.A. Malacological Review, 27:
61-74.
NEKOLA. J. C, 1994. Environment and vascular
flora of northeastern Iowa fen communities.
Rhodora,96: 121-169.
NEKOLA, J. C, T A. SMITH & T J. FREST 1996.
Terrestrial gastropod faunas of Door Peninsula
natural areas. Report to Wisconsin Chapter. The
Nature Conservancy. Madison.
NILSSON. S. G.. J. BENGTSSON & S. AS. 1988.
Habitat diversity or area per se? Species rich-
ness of woody plants, carabid beetles and land
snails on islands. Journal of Animal Ecology. 57:
685-704.
РЕЕТ. R. К,. 1 974, The measurement of species di-
versity. Annual Review of Ecology and Systemat-
ics, 5: 285-307.
SCHMID, G., 1966. Die mollusken des Spitzbergs.
Der Spitzberg bei Tübingen, Natur und Land-
schaftsschutzgebiete, Baden-Würtemburg. 3:
596-701.
SOLEM, A.. 1984, A world model of land snail di-
versity and abundance. Pp. 2-66, in: a. solem &
A. 0. VANBRUGGEN, eds.. World-wide snails: bio-
geographical studies on non-marine Mollusca.
E. J. BhII/W. Backhuys. Leiden.
SOLEM. A. & R M. CLIMO. 1985. Structure and
habitat correlations of sympatric New Zealand
land snail species. Malacologia. 26: 1 -30.
SOLEM, A., F. M. CLIMO & D. J. ROSCOE. 1981,
Sympatric species diversity of New Zealand land
snails. New Zealand Journal of Zoology 8:
453-485.
STANISIC, J.. 1997. An area of exceptional land
snail diversity: the Macleay Valley north-eastern
New South Wales. Memoirs of the Museum of
Victoria, 56:441-448.
TATTERSFIELD, R. 1996. Local patterns of land
snail diversity in a Kenyan rain forest. Malacolo-
gia, 38: 161-180.
VELLEMAN. R R & D. С HOAGLIN. 1981. Appli-
cations, basics, and computing of exploratory
data analysis. Addison-Wesley Press. Reading,
Massachusetts.
WALDÉN, H. W., 1981, Communities and diversity
240
NEKOLA
of land molluscs in Scandinavian woodlands. I.
High diversity communities in taiuses and boul-
der slopes in SW Sweden. Journal of Conchol-
ogy. 30: 35^ -372.
ZAR, J. H., 1984, Biostatistical analysis. Prentice-
Hall, Inc., Englewood Cliffs, New Jersey.
Revised ms. accepted 18 March 1999
APPENDIX I
Name, location, habitat type, and terrestrial gastropod richness of sample sites
Site Name Location Habitat Type Richness
ILLINOIS
Calhoun County
Franklin Hill
90°36'38"W, 39°3'57"N
Carbonate Cliff
28
Jackson County
Kings Ferry Bluff Base
89°26'16"W, 37°36'36"N
Carbonate Cliff
27
Kings Ferry Bluff Crest
89°26'15"W, 37°36'36"N
Carbonate Cliff
19
Madison County
Cliffton Terrace
90°13'36"W, 38"54'51"N
Carbonate Cliff
20
Monroe County
Fults Reserve
90''11'15"W, 38=9'19"N
Carbonate Cliff
20
Fountain Gap
90°15'33"W, 38°22'36"N
Carbonate Cliff
30
Pike County
Shewhart Bluff
9Г6'48"\Л/, 39"39'N
Carbonate Cliff
21
Randolph County
Prairie du Rocher
90°11'56"W, 38 6'28"N
Carbonate Cliff
20
Chester
89°53'6"W, 37"56'42"N
Carbonate Cliff
25
Bremer County
Brayton-Horsley
Buchanan County
Rowley West
Chickasaw County
Stapelton Church
Clayton County
Bixby East
Buck Creek 4
Buck Creek Tributary 1
Buck Creek Tributary 2
Buck Creek Tributary 3
Buck Creek Tributary 4
Buck Creek Tributary 5
Elkader South
South Cedar 2
South Cedar 3
Delaware County
Backbone West
Elk River East
Des Moines County
Iowa Ammunition Plant
Dubuque County
Roosevelt Road
IOWA
92°6'28"W, 42°48'35"N
91°54'39"W, 42°22'15"N
92"6'14"W, 43°1'35"N
91°23'56"W, 42-40'28"N
91°11'23"W, 42'51'35"N
91°10'55"W, 42 51'47"N
91°10'55"W, 42= 51'56"N
91°10'55"W, 42 52'5"N
91°10'58"W, 42 52'1"N
91°11'2"W, 42 51'52"N
91°23'45"W, 42 50'23"N
91°14'38"W, 42 49'58"N
91°14'23"W, 42M9'51"N
91°33'43"W, 42^37'5"N
91°17'27"W, 42°37'42"N
91°17'16"W, 40"46'27"N
90"44'30"W. 42' 32'55"N
Fen
Fen
Fen
14
20
17
Algific Talus Slope
16
Algific Talus Slope
13
Algific Talus Slope
21
Algific Talus Slope
10
Algific Talus Slope
25
Algific Talus Slope
31
Algific Talus Slope
23
Algific Talus Slope
11
Algific Talus Slope
20
Carbonate Cliff
30
Carbonate Cliff
26
Algific Talus Slope
32
Rocky Woods
26
Carbonate Glade
17
TERRESTRIAL GASTROPOD RICHNESS IN THE GREAT LAKES
241
Fayette County
Brush Creek Canyon 1
91°41'27"W, 42°47'2"N
Algific Talus Slope
11
Brush Creek Canyon 2
91°4r20"W, 42°46'46"N
Carbonate Cliff
27
Floyd County
Juniper Hill Shale Glade
92"59'2"W, 43°3'10"N
Carbonate Glade
12
Franklin County
Hampton East Glade
93'^8'13'Ж 42°43'41"N
Carbonate Glade
14
Howard County
Larkin Bridge East
92°5'8"W, 43- 29'32"N
Algific Talus Slope
21
Jackson County
Hamilton Glade
90°34'8"W, 42"4'23"N
Carbonate Glade
14
Lytle Creek 6
90°45'21"W, 42 = 15'34"N
Algific Talus Slope
20
Lytle Creek 16
90°45'28"W, 42^-1 5'40"N
Algific Talus Slope
16
Maquoketa Caves
90°46'22"W, 42°7'3"N
Carbonate Cliff
23
Pine Creek
90°50'41"W, 42"8'27"N
Carbonate Cliff
31
Jones County
Canton Glade
90°59'52"W, 42' 10'46"N
Carbonate Glade
18
Pictured Rocks
91 °6' 18"W, 42^12 '28"N
Carbonate Cliff
23
Linn County
Dark Hollow
91°30'W, 4r53'54"N
Carbonate Cliff
20
Paris Fen
9r35'41"W, 42°13'39"N
Fen
11
Winneshiek County
Bluffton West 2
91°55'16"W, 43°24'10"N
Algific Talus Slope
17
Heritage Farm
91°47'59"W, 43"22'55"N
MICHIGAN
Algific Talus Slope
20
Chippewa County
Bass Cove Cobble Beach
83"32'45"W, 45°55'10"N
Cobble Beach
11
Bass Cove Upland
83"32'46"W, 45"55'24"N
Rocky Woods
11
Hill Lake East
84°30'51"W, 46°7'48"N
Carbonate Cliff
19
Huron Bay
83°45'17"W, 45°57'12"N
Cobble Beach
11
Maple Hill
84°46'55"W, 46"9'34"N
Carbonate Cliff
25
Marble Head Center
83°28'28"W, 45°59'3"N
Carbonate Cliff
18
Marble Head North
83=28'30"W, 45°59'17"N
Carbonate Cliff
19
Marble Head South
83=28 '35"W, 45°58'46"N
Carbonate Cliff
22
Maxton Plains Center 1
83°39'48"W, 46°4'26"N
Alvar
8
Maxton Plains Center 2
83"39'24"W, 46°4'44"N
Tamarack Wetland
8
Poe Point
83°38'30"W, 46°6'10"N
Lakeshore Carbonate
Ledge
14
Prentiss Bay
84 13'49"W, 45°59'25"N
Tamarack Wetland
13
Scott Bay
83°40'1"W, 46°3'31"N
Fen
12
Scott Quarry
84°50'4"W, 46°10'43"N
Carbonate Cliff
23
Seastone Point
83°45'34"W, 46°1'21"N
Carbonate Cliff
22
Spencers Mountain Lower
84 56'39"W, 46"11'33"N
Rocky Woods
19
Spencers Mountain Upper
84"56'43"W, 46°11'29"N
Carbonate Cliff
16
Tourist Road
83°43'48"W, 46°1'44"N
Carbonate Cliff
18
Delta County
Ansel's Point
86°34'26"W, 45°48'12"N
Carbonate Cliff
23
Burnt Bluff
86°42'39"W, 45°41'11"N
Carbonate Cliff
19
Cooks Ridge
86"29'27"W, 45"57'59"N
Rocky Woods
20
Fayette St. Park
86"39'46"W, 45°43'40"N
Carbonate Cliff
18
Garden Bluff
86^37 '48"W, 45°46'47"N
Carbonate Cliff
17
Garden Corners
86"32'4"W, 45 53'23"N
Tamarack Wetland
11
Garden Peninsula Alvar
86°38'38"W, 45°39'48"N
Alvar
5
Goully Harbor
86°36'47"W, 45°46'42"N
Carbonate Cliff
21
242
NEKOLA
Jacks Bluff
Kregg Bay Northwest 1
Kregg Bay Northwest 2
016 West
Pt. Detour 1
Pt. Detour 2
Pt. Detour 3
South River Bay
Gogebic County
Bessemer NE
Copper Mountain
IVlt. Zion Park
Keewenaw County
Agate Point
Brockway Mountain
Cliff Range North 1
Cliff Range North 2
Cliff Range South
Cliffton
Copper Harbor Shore
Copper Harbor Marina
Dans Point
Delaware Gap
Eagle Harbor
Grand Marais Harbor
Mt. Bohemia
Luce County
McLeod Hill
Mackinac County
Dinkey Line Road
Gamble Road
Greene Cedar
Kenneth Road
Martineau Creek
McCann High School
Point St. Ignace
Round Lake 1
Round Lake 2
Summerby Swamp
Townhall Road
Ontonagon County
Adventure Mountain
Cloud Peak
Cranberry River Hill
Cuyahoga Peak
Miscowawbic Peak
Norwich Mountain
Rodgers Cedar Swamp
Summit Peak Lower
Summit Peak Upper
School Craft County
Manistique North
Merwin Creek
Seul Choix Point 1
Seul Choix Point 2
86°3r33"W, 45 5riO"N
Lakeshore Alluvial
Bank
21
86°32'10"W, 45°43'13"N
Alvar
9
86°31'59"W, 45°42'39"N
Cobble Beach
1
86'=43'1"W, 45'40'17"N
Rocky Woods
14
86^37 '47"W, 45^38 '3"N
Alvar
9
86°36'21"W, 45°36'25"N
Cobble Beach
15
86°36'25"W, 45°36'16"N
Lakeshore Carbonate
Ledge
19
86°37'19"W, 45°45'22"N
Carbonate Cliff
20
90°2'24"W, 46°29'31"N
Igneous Cliff
16
90°5'16"W, 46°36'5"N
Igneous Cliff
11
90°10'8"W, 46°28'37"N
Igneous Cliff
2
88°1 '4"W, 47°28'38"N
Igneous Lakeshore
14
87°58'15"W, 47°27'48"N
Igneous Cliff
6
88°15'11"W, 47°23'55"N
Igneous Cliff
11
88°15'9"W, 47°23'55"N
Igneous Cliff
13
88°19'50"W, 47°2r36"N
Igneous Cliff
12
88°19'23"W, 47°21'54"N
White Cedar Wetland
12
87°54'15"W, 47°28'36"N
Igneous Lakeshore
10
87°54'14"W, 47°28'19"N
White Cedar Wetland
11
87°58'37"W, 47°28'47"N
Igneous Lakeshore
15
88°6'57"W, 47 25'29"N
Igneous Cliff
13
88°8'54"W, 47°27'39"N
Igneous Lakeshore
9
88°7'9"W, 47°27'37"N
Igneous Cliff
15
88°0'51"W, 47°23'34"N
Igneous Cliff
12
85°15'37"W, 46°15'17"N
Rocky Woods
23
85°14'16"W, 46°10'49"N
White Cedar Wetland
7
84°45'W, 46°7'42"N
Rocky Woods
21
84°51'51"W, 46°1'41"N
White Cedar Wetland
14
84°50'34"W, 46°5'50"N
Carbonate Cliff
19
84°43'11"W. 45°59'8"N
Fen
14
84°43'36"W, 45=51 '48"N
Rocky Woods
12
84°42'34"W, 45°57'8"N
Rocky Woods
18
84°52'31"W, 45^57' 18"N
Fen
3
84°52'31"W, 45"57'11"N
White Cedar Wetland
13
84°47'43"W, 45^58' 15"N
Fen
19
85°10'28"W, 46"8'19"N
White Cedar Wetland
22
89°4'51"W, 46^46' 17"N
Igneous Cliff
18
89°43'58"W, 46°48'51"N
Igneous Cliff
16
89°26'28"W, 46=42'47"N
Igneous Cliff
8
89°41'59"W, 46°48'56"N
Igneous Cliff
17
89°48'47"W, 46°47'N
Igneous Cliff
16
89°22'54"W, 46°39'37"N
Igneous Cliff
12
89°44'3"W. 46°33'7"N
White Cedar Wetland
7
8946'27"W, 46°44'55"N
Igneous Cliff
3
89°46'25"W, 46°44'48"N
Igneous Cliff
5
86°16'W, 46°1'37"N
Carbonate Cliff
17
86°5'40"W, 46°2'30"N
Rocky Woods
13
85°54'50"W, 45=55' 12"N
Cobble Beach
19
85°54'53"W, 45°55'14"N
Rocky Woods
11
TERRESTRIAL GASTROPOD RICHNESS IN THE GREAT LAKES
MINNESOTA
243
Cook County
Caribou Lake E
90°40'1"W, 47M2'23"N
Igneous Cliff
9
Caribou Lake N
90°40'40"W, 47 42'46"N
Igneous Cliff
6
Carlton Peak
90°51'22"W, 47''35'9"N
Igneous Cliff
2
Cascade River Cedars
90°31 '32"W, 47 '44'36"N
Igneous Cliff
14
Cascade River Cliff
90°32'15"W, 47M3'28"N
Igneous Cliff
12
Iceland Fen
90°7'3"W, 47^47'44"N
Tamarack Wetland
11
John Lake
90°3'29"W, 48^3'56"N
Igneous Cliff
9
Lake Cliff
90°58'23"W, 47 29'40"N
Igneous Cliff
5
Lutsen Mountains
90°41'54"W, 47 '40'49"N
Igneous Cliff
4
McFarland Lake Cliff
90°5'25"W, 48 ■3'34"N
Igneous Cliff
6
McFarland Lake Talus
90°5'18"W, 48°3'36"N
Igneous Cliff
6
Moose Mt. Cliff
90°44'9"W, 47^'39'7"N
Igneous Cliff
11
Mt. Josephine Cliff
89°39'14"W, 47°58'58"N
Igneous Cliff
14
Mt. Josephine Talus
89°39'14"W, 47"58'50"N
Igneous Cliff
11
Oberg Mountain
90°46'38"W, 47^37'44"N
Igneous Cliff
5
Pine Lake
90°6'7"W, 48°3'3"N
Igneous Cliff
12
Pine River Road
90°18'4"W, 47°54'59"N
Igneous Cliff
10
Poplar River
90°43'15"W, 47°39'55"N
Igneous Cliff
13
Poplar River Overlook
90°43'34"W, 47°40'13"N
Igneous Cliff
6
Port of Entry Cliff
89^37' 12"W,47°59'51"N
Igneous Cliff
9
Port of Entry Talus
89°37'6"W, 47=59'47"N
Igneous Cliff
9
Portage Brook
90°1 '24"W, 48°0'2"N
Igneous Cliff
6
Sawbill Road N Cliff
90°51 '37"W, 47''35'50"N
Igneous Cliff
10
South Fowl Lake
90°0'28"W, 48^2'27"N
Igneous Cliff
11
Sugarloaf Cove
90°58'50"W, 47^29' 12"N
Igneous Cliff
7
Temperance River Road
90°53'3"W, 47°34'6"N
Igneous Cliff
8
Temperance River Upland
90^52 '22"W, 47^34 '22"N
Igneous Cliff
14
Timber Creek
90°15'19"W, 47^'53'53"N
Igneous Cliff
10
LaKe County
Day Hill
91°22'59"W, 47=11'36"N
Igneous Cliff
10
Finland Forest
91 °5'25"W, 47^33' 13"N
Igneous Cliff
7
Goldeneye Lake Cliff
91°3'55"W, 47°35'53"N
Igneous Cliff
6
Goldeneye Lake Talus
91°3'48"W, 47°35'50"N
Igneous Cliff
7
Manitou River Falls
9r4'27"W, 47"26'44"N
Igneous Cliff
12
Sawmill Creek
91 °9'43"W, 47^24' 15"N
Igneous Cliff
3
Water Tanks
91°17'45"W, 47°17'42"N
Igneous Cliff
7
St. Louis County
Chester Bowl
92°5'55"W, 46°48'48"N
Igneous Cliff
10
Hawk Ridge Sanctuary
92°1 '55"W, 46°50'49"N
Igneous Cliff
13
Skyline Drive West Cliff
92°10'6"W, 46°45'47"N
Igneous Cliff
5
Skyline Drive West Talus
92"10'6"W, 46°45'43"N
NEW YORK
Igneous Cliff
14
Cayuga County
Fillmore Glen State Park
76°23'52"W, 42°41 '45"N
Shale Cliff
3
Madison County
Cazenova Gorge 1
75°50'41"W, 42°58'49"N
Carbonate Cliff
25
Cazenova Gorge 2
75°50'49"W, 42°58'49"N
Carbonate Cliff
20
Niagara County
Gasport Ravine
78°35'2"W, 43"10'52"N
Carbonate Cliff
9
Schuyler County
Watkins Glen State Park
76=53'24"W, 42°22'16"N
Shale Cliff
14
244
NEKOLA
Wyoming County
Letchworth State Park
78"1'26"W, 42''35'41"N
ONTARIO
Shale Cliff
10
Bruce County
Grotto Trail
81°31'19"W, 45°14'38"N
Carbonate Cliff
19
Lions Head
81°13'4"W, 45 0'28"N
Carbonate Cliff
20
Overhanging Point
81°31'51"W, 45°14'40"N
Carbonate Cliff
17
Grey County
Inglis Fails
80°56'2"W, 44°3r50"N
Carbonate Cliff
28
Metcalfe Rock
80°26'31"W, 44°25'3"N
Carbonate Cliff
22
Skinners Bluff
80°59'31"W, 44°47'36"N
Carbonate Cliff
30
Halten County
Crawford Lake
79°56'27"W, 43°28'27"N
Carbonate Cliff
29
Royal Municipality of Hamilton-Wentworth
Dundas Park
79°58'48"W, 43°14'18"N
Carbonate Cliff
13
Manitoulin District
Burnt Island
82°56'9"W, 45°49'26"N
Cobble Beach
7
Cooks Dock Lower
82°47'20"W, 45°52'50"N
Carbonate Cliff
12
Cooks Dock Upper
82"47'20"W, 45°52'48"N
Carbonate Cliff
15
Cup & Saucer East
82°6'10"W, 45°51'21"N
Carbonate Cliff
18
Cup & Saucer North
82°5'59"W, 45°51'9"N
Carbonate Cliff
18
Janet Head
82°29'16"W, 45 56'43"N
Carbonate Cliff
22
McLean Park Lowland
81°54'28"W, 45"41'31"N
Lowland Woods
8
McLean Park Upland
81°54'25"W, 45"41'44"N
Rocky Woods
12
Mississagi Lighthouse
83°13'19"W, 45°53'36"N
Lakeshore Carbonate
Ledge
13
Niagara County
Beamers Falls
79°34'1"W, 43°11'16"N
Carbonate Cliff
20
Niagara Whorlpool
79°4'1"W, 43°7'24"N
Carbonate Cliff
20
Peel County
Devils Pulpit
79°59'27"W, 43"48'4"N
Carbonate Cliff
26
Simcoe County
Glen Huron
80°11 '41 "W, 44^21' 1"N
Carbonate Cliff
24
Wellington County
Guelph Jail
80°10'44"W, 43 32'51"N
WISCONSIN
Carbonate Cliff
8
Ashland County
Beaverdam Lake
90°48'33"W, 46°19'26"N
Igneous Cliff
3
Loon Lake Bluff
90-38'49"W, 46 20'52"N
Igneous Cliff
12
St. Peter's Dome
90°54'39"W, 46 21 '5"N
Igneous Cliff
9
Bayfield County
Rainbow Lake Wilderness
9r20'13"W. 46°26'35"N
Upland Woods
7
Brown County
Bayshore Park
87°47'59"W, 44°38'12"N
Carbonate Cliff
17
Benderville Wayside
87°50'31"W, 44^36'47"N
Carbonate Cliff
27
Blueberry Marsh
87°53'35"W, 44 31'46"N
Tamarack Wetland
1
Celtis Site
87°50'52"W, 44 36'35"N
Carbonate Cliff
34
Edgewater Villas
87-^49 '8"W, 44 37'42"N
Carbonate Cliff
14
Escarpment Glade
87^^49 '4"W, 44 37'9"N
Alvar
12
Fonferik Glen
87°58'15"W, 44 25'34"N
Carbonate Cliff
25
Gibson Alvar
87°50'52"W, 44°35'26"N
Alvar
12
TERRESTRIAL GASTROPOD RICHNESS IN THE GREAT LAKES
245
Gravel Pit Road
Greenleaf Cliff
Greenleaf Talus
Hilly Haven
Iron Fence Wayside
Lily Lake County Park 1
Lily Lake County Park 2
Neshota County Park 1
Neshota County Park 2
Scray's Hill
UWGB Upland Woods
UWGB Cedar Swamp
UWGB Escarpment
Calumet County
Calumet County Park
Charlesburg Ledge
East River Road
High Cliff State Park 1
High Cliff State Park 2
High Cliff State Park 3
High Cliff State Park 4
Kiel Marsh
Stockbridge
Dodge County
Ledge County Park
Mayville South
Messner Ledge South
Neda Mine
Door County
Bjorklunden
Boyer Bluff 1
Boyer Bluff 2
Brussels Hill North
Brussels Hill Radio Tower
Carlsville Road
Cave Point Shore
Cave Point Uplands
Cherry Escarpment
Door Bluff Park
Ellison Bay Park
Fifield Tract
Prey Tract
Glidden Drive
Hemlock Road 1
Hemlock Road 2
H utter Tract
Kangaroo Lake
Kuchar Fen
LaSalle Park
Little Harbor
McKnight Escarpment
87°47'59"W, 44 36'7"N
88°4'1"W, 44"20'20"N
88°44"W, 44'20'16"N
88°2'43"W, 44°20'59"N
87^49 '40"W, 44=37' 14"N
87°51'3"W, 44^25 '19"N
87^51 '3"W,44"25'22"N
87^48'22"W, 44°24'5"N
87°48'17"W, 44°24'6"N
88"1'37"W, 44°23'N
87°55'22"W, 44'31'33"N
87°55'29"W, 44°31'33"N
87°54'21"W, 44°31'48"N
88°19'11"W, 44 6'45"N
88°12'3"W, 43°58'10"N
88=3'42"W, 44°8'21"N
88°17'56"W, 44°9'11"N
88°16'44"W. 44°10'13"N
88°17'52"W, 44°9'18"N
88^17'49"W, 44"9'14"N
88°3'34"W, 43°53'52"N
88°17'17"W, 44°3'10"N
88°35'2"W, 43°28'11"N
88"32'24"W, 43°27'27"N
88°35'41"W, 43"37'55"N
88°32'5"W, 43°25'23"N
87°7'58"W, 45°1 '54"N
86°56'2"W, 45°25'10"N
86"55'55"W, 45"25'6"N
87°35'45"W, 44"46'13"N
87^35 '27"W, 4444 '47"N
87^22' 1"W, 44^57 '4"N
87°10'33"W, 44"55'38"N
87°10'44"W, 44°55'51"N
87°9'25"W, 45°11'11"N
87°3'53"W, 45°17'48"N
87°5'38"W, 45^15'20"N
87°3'36"W, 45°6'N
87°3'45"W, 45"14'34"N
87°12'36"W, 44°52'53"N
86°52'12"W, 45°20'50"N
86°52'4"W, 45^20 '42"N
87°22'55"W, 44"56'5"N
87°10'4"W, 45"3'12"N
87°10'51"W, 45°5'12"N
87^21 '50"W,44°41'23"N
87°24'7"W, 44"54'45"N
87°19'47"W, 45°0'43"N
Calcareous Meadow
19
Carbonate Cliff
29
Carbonate Cliff
21
Carbonate Cliff
26
Carbonate Cliff
20
Tamarack Wetland
13
Lowland Woods
20
Fen
20
Rocky Woods
Carbonate Cliff
18
23
Upland Woods
White Cedar Wetland
3
26
Carbonate Cliff
25
Carbonate Cliff
20
Carbonate Cliff
26
Tamarack Wetland
12
Carbonate Cliff
24
Carbonate Cliff
26
Rocky Woods
Upland Woods
Tamarack Wetland
14
7
24
Carbonate Cliff
28
Carbonate Cliff
22
Carbonate Cliff
23
Carbonate Cliff
25
Carbonate Cliff
19
Lakeshore Carbonate
21
Ledge
Carbonate Cliff
17
Carbonate Cliff
16
Carbonate Cliff
20
Carbonate Cliff
19
Carbonate Cliff
12
Lakeshore Carbonate
9
Ledge
Rocky Woods
Carbonate Cliff
12
17
Carbonate Cliff
18
Carbonate Cliff
26
Lakeshore Carbonate
17
Ledge
Lakeshore Carbonate
15
Ledge
Lakeshore Carbonate
13
Ledge
Lakeshore Carbonate
23
Ledge
Rocky Woods
Carbonate Cliff
20
24
Rocky Woods
Fen
13
11
Lakeshore Alluvial
15
Bank
Carbonate Cliff
18
Carbonate Cliff
15
246
NEKOLA
Meridian Park
87 9'57"W, 45°0'22"N
Lakeshore Carbonate
Ledge
15
Monument Point South
87°21 '35"W, 44°58'35"N
Carbonate Cliff
15
Moonlight Bay
87°4'1"W, 45°4'54"N
Lakeshore Alluvial
Bank
4
Mountain Park
86^54 '10"W, 45^23 '16"N
Carbonate Cliff
24
Mud Lake
87^4'44"W, 45°5'44"N
Lakeshore Carbonate
Ledge
13
Newport State Park 1
86°59'13"W, 45°14'52"N
Lakeshore Carbonate
Ledge
20
Newport State Park 2
86°59'23"W, 45°15'3"N
Lakeshore Carbonate
Ledge
21
Newport State Park 3
86°59'49"W, 45°15'13"N
Lakeshore Carbonate
Ledge
21
Peninsula State Park 1
87°13'8"W, 45°9'42"N
Carbonate Cliff
18
Peninsula State Park 2
87°12'7"W, 45°9'45"N
Carbonate Cliff
10
Port de Mort
86°59'31"W, 45°17'50"N
Carbonate Cliff
22
Potawatomie State Park 1
87°25'29"W, 44°52'38"N
Carbonate Cliff
20
Potawatomie State Park 2
87°24'46"W, 44°51 '40"N
Carbonate Cliff
18
Potawatomie SW
87°26'31"W, 44°51'45"N
Carbonate Cliff
21
Red River Bluff
87°41'6"W, 44°45'43"N
Carbonate Cliff
12
Ridges Sand Swale
87°7'8"W, 45°4'12"N
Calcareous Meadow
10
Rock Island 1
86°49'44"W, 45°25'20"N
Carbonate Cliff
26
Rock Island 2
86°49'40"W, 45"24'53"N
Upland Woods
15
Rock Island 3
86°49'4"W, 45°25'37"N
Carbonate Cliff
19
Rock Island 4
86°49'8"W, 45°25'31"N
Carbonate Cliff
18
Rock Island 5
86°49'37"W, 45°25'19"N
Carbonate Cliff
28
Rock Island 6
86°49'44"W, 45"25'6"N
Carbonate Cliff
11
Rocky Point
87°19'11"W, 44°46'23"N
Lakeshore Alluvial
Bank
13
Sand Dune Park
86°53'52"W, 45°20'14"N
Lakeshore Alluvial
Bank
14
Shivering Sands
87°15'57"W, 44°52'30"N
Lakeshore Carbonate
Ledge
22
Ski Slope, E-facing
87°14'5"W, 45°6'10"N
Carbonate Cliff
23
Ski Slope, N-facing
87°14'16"W, 45"6'18"N
Carbonate Cliff
15
South Shore Road
86"53'6"W, 45"20'10"N
Lakeshore Carbonate
Ledge
21
Standish
87°21'57"W, 45°57'36"N
Carbonate Cliff
22
Thorpe Pond
87°13'40"W, 45°4'20"N
Fen
10
Toft Point 1
87°5'20"W, 45"3'23"N
Upland Woods
13
Toft Point 2
87°6'3"W, 45°4'50"N
Fen
12
Toft Point 3
87°5'52"W, 45°4'43"N
Tamarack Wetland
11
Toft Point 4
87°5'59"W, 45"4'58"N
Lowland Woods
13
Toft Point 5
87°5'52"W, 45"4'22"N
Lakeshore Carbonate
Ledge
14
Toft Point 6
87°5'5"W, 45°4'40"N
Lakeshore Carbonate
Ledge
19
Toft Point 7
87°5'2"W, 45°3'59"N
Cobble Beach
10
Toft Point 8
87'^5'45"W, 45'4'14"N
Rocky Woods
6
Ulman Woods
87°32'3"W, 44"44'58"N
Rocky Woods
11
Werkheiser Escarpment
87°20'27"W, 44°59'39"N
Carbonate Cliff
15
Wilson Escarpment
87°15'43"W, 45°5'3"N
Carbonate Cliff
5
ouglas County
Flanagan Lookout 1
9r56'9"W, 46"35'8"N
Igneous Cliff
5
Flanagan Lookout 2
9r56'6"W, 46^^35'7"N
Igneous Cliff
7
Pattison State Park 1
92°7'26"W, 46°32'14"N
Igneous Cliff
9
TERRESTRIAL GASTROPOD RICHNESS IN THE GREAT LAKES
247
Pattison State Park 2
92°7'26"W, 46°32'15"N
Igneous Cliff
8
Pattison State Park 3
92°7'26"W, 46^32' 16"N
Igneous Cliff
7
South Range
91°58'26"W, 46°36'15"N
Old Field
4
Fond du Lac County
Ledge Bar
88°21'3"W, 43°52'4"N
Carbonate Cliff
27
Messner Ledge
88"35'31"W, 43'38'13"N
Carbonate Cliff
23
Messner Ledge North
88°35'24"W, 43 38'25"N
Carbonate Cliff
23
Oakfield Brick Yard
88°33'10"W, 43°40'27"N
Carbonate Cliff
26
Oakfield Ledge
88°34'55"W, 43^38 '55"N
Carbonate Cliff
23
Peebles
88°22'40"W, 43"48'32"N
Carbonate Cliff
21
ShIey Pond
88°33'39"W, 43°40'8"N
Carbonate Cliff
25
Iron County
Hurley Visitor Center
90°12'14"W, 46"28'13"N
Igneous Cliff
11
Lake Lavina Bluff 1
90°10'55"W, 46"26'6"N
Igneous Cliff
11
Lake Lavina Bluff 2
90°10'55"W, 46°26'6"N
Igneous Cliff
8
Whitecap Mountain
90°23'49"W, 46^24' 15"N
Igneous Cliff
15
Kewaunee County
Kewaunee Fish Hatchery
87'=33'50"W, 44^^28'45"N
Lakeshore Carbonate
Ledge
19
Lipsky Swamp 1
87°37'15"W, 44°28'57"N
Tamarack Wetland
11
Lipsky Swamp 2
87°37'13"W, 44°28'50"N
Lowland Woods
7
Little Scarboro Creek 1
87°37'26"W. 44°30'39"N
Rocky Woods
18
Little Scarboro Creek 2
87°37'22"W, 44''30'45"N
Cacareous Meadow
16
Mud Lake
87°39'45"W, 44"39'37"N
Tamarack Wetland
19
Stony Creek Woods
87°22'51"W, 44''39'51"N
Lakeshore Alluvial
Bank
16
Thiry Daems
87°42'14"W, 44'='36'8"N
Calcareous Meadow
19
Tisch Mills
87°38'21"W, 44°20'50"N
Tamarack Wetland
12
Manitowoc County
Cato Falls County Park 1
87°50'34"W, 44"5'31"N
Carbonate Cliff
16
Cato Falls County Park 2
87°50'49"W, 44°5'29"N
Lakeshore Alluvial
Bank
9
Cooperstown Swamp
87°53'12"W, 44°16'16"N
Tamarack Wetland
11
Frelich Road Swamp
87°52'8"W, 44°17'40"N
Lowland Woods
21
Kingfisher Farm 1
87°42'3"W, 43°57'43"N
Lakeshore Alluvial
Bank
20
Kingfisher Farm 2
87°42'17"W, 43°57'47"N
Lakeshore Alluvial
Bank
6
Kingfisher Farm 3
87°42'3"W, 43°57'51"N
Lakeshore Alluvial
Bank
6
Kingfisher Farm 4
87°42'14"W, 43°57'46"N
Upland Woods
2
Kingfisher Farm 5
87°42'10"W, 43°57'51"N
Lowland Woods
3
Kingfisher Farm 6
87°42'21"W, 43°57'48"N
Lowland Woods
9
Kingfisher Farm 7
87°42'25"W, 43°57'49"N
Fen
10
Maribel Caves
87°46'11"W, 44°17'9"N
Carbonate Cliff
17
Point Beach 1
87°31'11"W, 44°12'5"N
Upland Woods
11
Point Beach 2
87°30'39"W, 44°11'52"N
Open Dune
4
SLC Bog 1
87°54'5"W, 43°59'31"N
Calcareous Meadow
7
SLC Bog 2
87°54'3"W, 43°59'21"N
Lowland Woods
5
SLC Bog 3
87°53'50"W, 43^59 '3r'N
Old Field
9
Tamarack Road
88°0'55"W, 44°12'12"N
Tamarack Wetland
2
Marinette County
Kimlark Lake
87°50'56"W, 45°39'37"N
Calcareous Meadow
11
Niagara East Bluff 1
87°56'41"W, 45°45'24"N
Igneous Cliff
10
Niagara East Bluff 2
87°56'41"W, 45°45'23"N
Igneous Cliff
15
Pound Roadside
88°1'1"W, 45°7'29"N
Calcareous Meadow
21
Spur Lake 1
88°14'5"W, 45°43'5"N
Fen
15
248
NEKOLA
Spur Lake 2
88 13'58"W, 45°42'59"N
Fen
12
Ozaukee County
Cedarburg Bog 1
88°0'34"W, 43°23'9"N
Fen
10
Cedarburg Bog 2
88°1'4"W, 43°22'59"N
Tamarack Wetland
16
Harrington Beach 1
87"48'10"W, 43"29'26"N
Old Field
6
Harrington Beach 2
87°47'56"W, 43^29' 16"N
Upland Woods
2
Harrington Beach 3
87=4741 "W,43"29'42"N
Lakeshore Carbonate
Ledge
10
Sauk County
Devils Lake
89°43'57"W, 43°24'30"N
Igneous Cliff
17
Sawyer County
Pipestone Falls
91°14'12"W, 45°51'19"N
Igneous Cliff
11
Shawano County
Porter Road
88°30'10"W, 44°44'16"N
Carbonate Cliff
19
Sheboygan County
Evergreen Park
87°44'29"W, 43°46'56"N
Lakeshore Alluvial
Bank
4
Waters Edge
87°46'46"W, 43"34'53"N
Lakeshore Alluvial
Bank
7
Mehles Springs
88"1'16"W, 43°51'38"N
Tamarack Wetland
9
Walworth County
Bluff Creek Fen
88°40'54"W, 42°48'2"N
Fen
19
Washington County
Allenton Fen
88°18'25"W, 43°22'41"N
Fen
19
Waushara County
Bass Lake
89"16'58"W, 44°0'15"N
Fen
17
TERRESTRIAL GASTROPOD RICHNESS IN THE GREAT LAKES
249
0)
Q.
X3
X Ç
û_ ^
< 03
о
CD
Q.
CM r^ m
-.- ^ ■— со
íD '^ -.- о
CM CD -чГ T-
CO CD CM "* 1-
CM ^ со ^ CM со
1- LD Ю со CM Ln CM
■* -.-CM
■.- CD -r- ■.-
■.-00 ■.- ^
1- r-~. 1^
со un со
1- LT) CM со
CM ^ ■■- со ■>- со
CM t- CM
r- ■.- CM -.-■.- ^ T-
CO CM CM N- ■■- со 1-
CM 'ît CM CM CM
CM ^ CM un
LO CM ■>-■-- -.-
■sf 00 r- ■>- CM
Ю 00 ■.- "*
^ CM "* ■.-
1- о -.- CM LO T-
^ 'i 00 CD
^-.СУ)
ч- т- ^ (-\| '^
03 со
со —
■3 03
am
■2 -^
If
-9 g'
1_- >, CD
03 03 —I
CD " — 'Í-'
о ' ~
03 03
ce е g
■s .? .?
a=í
03 -с
Э, X is
СЗ) ~ о
с 03 03
см
(Л -S '^ s
03 03 >. с
"О р 03 о
■ о 5 W 03
с 03 ^^— I
с ^ 03 ~^
е ■ 03 ^
g ш 03 03
■Э i5 -2 -5
f-, "оз "оз "оз
03 то та ш
оооо
оз 00 ^
00 ~
со
00 >;
03 ■^ „9
ф
со
' см
'00 "О
л -03 03
LL t/3 03
m 5 £
^ о
t о -о -^
Р Q "5 -
9=.^ о '^
lili
00 CD
оз 00
00 ^
if
о Ü
-г- со
^ CF3
" 00 -^
00 ;р:
00 _^
CD 00
i; ■^ °а
03 та
та я
со
та 03
й- й- -5
03 t:;
s -с
та 9
-с -с
03 03 _
с: та -с
■а с5 о _
та 03 о о о о 03 52 52 -ïï
OOOOOOQQQQ
Û с та
3 5 03
"3 ^ ^
■;= W со
та -> ^
çq с с
Я О О
íÁ '^ '^
■!2 3 гз
Q LU Uj
= to Cl
:ZJ 3 -—
¿ ^ о
-5 CL-fc
со со со
^ ^ -3
^ ^ "3
с с с
о о сз
CJ U Ü
3 :з э
ÜJ Lu Uj
>^ Л
с5?^
та-2
>.]9
с/) о
03 О)
Оз с
с ^
^ р со с
ел in
та та
CDU
^J ^-' С~ Atn
о -с с оз
■S -Э -S '"-
Q. Q. о. 5
о о о о
о о о w
ел t/) с/3 о
та та та "5
CD CDU
'CD S
та ^^
03 ;^
03 si
Q. Q. со
я -S я
о. о. Q.
о о о
со со со
та та та
CD CD CD
250
NEKOLA
Ют-С\]Г-~ОСМО-^1Г)СОГ^Г^-^1Т>'^иПСОт-СМСООО CÛLOCT)-^
LnŒ)C\JC\JC\JCJ) N-LOCD ■■- C\JN-
ir> Ю
CM O) -^ •>- C\J •>- '^
CT> -îi --- со
Ю -^ со -^
со un со
ID 's)- со •^ CM
т- in ■■- со
•>- см '^
со 00 1- CD см
СМ->-ЮиПС£)-.-СМ ^
in -^ -.-■■- un
t~^ ^ СО г^ г^ см со
СО Г^ CVI
CD
Q.
со
ее
с --
СХ! см
стз го
&<
W Ü
et! —
— со
;— -СГ) со
С^ CD со
го^ ™
СП SÛ CÛ
CD
см со ]-
■^ 'í о
см 00 CD Ч>
>, ф *-'
mew
- -^ CD г- CD
CD ■.- (M ^
^ CD
^ -^ >>, . ■>-
5 -'-о;;;
ел
Я^2.
m ^ QJ
со .СП .сч
с .с с
CÛ
00 üi - с:
00 р CÛ U. х^
Í5 ПЗ m § Е
3 с
со J3
=1 "I О)
Q о -^
■ • С: -3
:- о <^
-2 о. 52 ci'
CÜ с\з стз
i^ is 5s
-с -с -с
(Ь СТЗ s
« 5 s
. ï СТЗ СТЗ ^
з стз га СП Q)
о о
V) V)
"б "б
о о
о и
ш "as
>,.У
^^
СП g
СП 5
Cl сл
in со
:з э
CJ о
со со
о о
о о
го Р Su
Q- -^ 41^ о 00 >•
— ,- <D
с
>. ел
го 3
ел S5
со
(Л .=
ф CÛ
СП >
s R^Ö ^ 2 х-ё w
■5 о
Î2 у с: о V
- 3
3 w
О) :ï:
о
со
сь
с с с с с
о о о о ^
ТЗ TD ТЗ ТЭ ТЗ
-С -С
II
h-
"О
о
О
о
о
о
О
о
1Л
с
со
СО
со
со
со
со
со
СО
CL)
CD
0)
ÜI
О)
Q)
сь
Q)
•^ —
Î5555555
со ^
■-- °° ÎÎ
~ CD -^
ел -
го ^ =)
ел ^ о
i;rcnS
о с о
с с сь
г; 5 "53
■5 СП СП
•с сь О)
с i. i>
?^ о о
со со со
Ш qj сь
5:г:г
с с сь --:
3" CÛ
ш^- 00 <
го S
Ф д
— ' с
^■§
СП £:
СЬ О
О СЬ
со О.
^ СП
CD Г^
CD Ю
00 ОО
s^^
ф
^ ^ о — ' го ^:^
S^
y.î^
^ с сь
■5) W
со СП
СП СП
^1
сь ■=
^- СП
xS52
i< СП
OQ,
СП -5
Q- Q-
го ^
СП СП 3
сер
СП СП s
:5 TD «о
■Sí -2 э
о -и? i
TD со с
О)^ сЗ
о о 3
о. cu О-
TERRESTRIAL GASTROPOD RICHNESS IN THE GREAT LAKES 251
T-i-r^COCDI^i- T- 4-C\JC\J---
-.- <M r^ CM
CO Cnr^ -^ 'ТГ^ CO T- i-OJi- -1- OJCM-'-'--'-
1- (MCOCO CD T-^x-
■>- CO CO CO
■I- ^ ■■- CM CM
■>- CO о -^ Ю
St rC r^ . „ ^ о en •^ —-^ <^ „
■■с -ï С) ^ § s .=1 SÉ -Q^^ :Q:S^§.aíí^?íiSS^§g--KDíWmb^-0
il|l|ll;,llililimiisaa|asâaa»slt.ll|?ss??
llialiiíiillllllllilllllllllliüiülttttf
О- О, CL CL ce со со со со со со со со со со со со со Н= к: fi: р: fi: h^ fi: i> :^> ii i:, i» ii ::^ Г::^ i^ Ь. i> ::í
о 2
CD CD
CD -^
о
СО E -^ ■^
5
Ш ce
.inné,
C.B.
(Say
m (Cl
5
e
S^
H,"^ « :з
3
to :з f
c:
Ibilabn
lealbat
î barba
2
252
NEKOLA
со со
C\J 1- Ю CD
т- CM •.-
т- т- а>
eg -^ -^ -^ о
■^ со
00 Я ^ 00
г^ ^ гг
о тз —
о Г) о ^-.
с со 1^ о
•^ с 00 m
см^^' га ^.$2
со CÛ
О)
Q.
СО
о т- о О)
С35СГ) О)'^
■р <Л -g CJ)
:р -§ $ ¿ ш <» 0^ го о р îii е
c о CL ab ^
vovj4jvj0000C3"JZ(
0)C!lD)D)C3iD)D5.D).C)¿S
о о
о о о
ел to
о стз
пз ^
ел со
О) ш
"О тз
о о
с с
N N
1^ см
■^ 00
тз стз
■■£ -с
с ел
ел 5í
5 s Й,
no - О)
rr с о
íP rP
СМОО-^ЮСОГ^СОСТ)
"О i5
с }S ^ ■■=
■s; 0) TD с
-S 1 f, Ф '^
s [Есгос5^<1>го1>о
Ъ -ni55goa3EÍF?^
ce Рс2.о°-0-СГО=ё5,
■■-CNJCO'^LncDr^OOCT)-^
ф
О)
тз
^^
"О
с
ф
со
ф
со со
ф
Q.
о о ^
со
' U- О
со 1- ^'
со I
о 55
п >
^ л^
Ф ф ф
О) ^ х:
7S ся W
-Я Ф Ф
с -g js: ^
О (Tí m
"5Í -2ь- -^
О 03 СО
ф
"О
го ,
О'
ф
го
с
^ о
го-е
> го
< Ü
о
MALACOLOGIA, 1999, 41(1): 253-269
TERRESTRIAL GASTROPOD RICHNESS PATTERNS IN WISCONSIN
CARBONATE CLIFF COMMUNITIES
Jeffrey С Nekola^ & Tamara M. Smith^
ABSTRACT
The patterns of terrestrial gastropod richness within two species-rich carbonate cliff habitats in
eastern Wisconsin were analyzed at two differing sample scales. Up to 23 taxa were found in
1 m^ quadrats, and 21 taxa in 0.04 m^ quadrats. These observations are among the highest re-
ported globally for 1 ha or smaller samples. At the 1 m^ scale, samples collected within 5 m of
bedrock outcrops had higher richness than more distant sites. At this scale, only soil pH (not Ca,
Mg, N, P, K, percent organic matter, vascular plant species richness, or surface and 20 cm depth
soil temperatures) was found to significantly correlate with species richness. At the 0.04 m^
scale, the richest sites were restricted to areas within 0.5 m of cliff bases. Comparison of maxi-
mum richness levels across varying spatial scales demonstrate that up to a third of the total
fauna may co-exist in <0.04 m^ regions (alpha diversity), up to half of the fauna may coexist in
<100 m^ regions (beta diversity), while the remainder of the taxa (gamma diversity) occurs be-
tween regions separated by at least 10 km.
Key words: terrestrial gastropods, species richness, diversity patterns, conservation, North
America, Niagaran Escarpment.
INTRODUCTION
While a number of studies have docu-
mented richness in terrestrial gastropod
communities at relatively large (> 100 m^)
sample scales (e.g., Paul, 1975; Solem et al.,
1 981 ; Waiden, 1 981 ; Cameron, 1 986; Nilsson
et al., 1988; Emberton, 1995; Tattersfield,
1996; Emberton et al., 1997), fewer have ana-
lyzed terrestrial gastropod community struc-
ture at smaller scales within sites. The re-
search that has been conducted at this scale
has demonstrated terrestrial gastropod com-
munity structure can change markedly over
limited (e.g., <100 m) spatial extents. For in-
stance. Berry (1966) demonstrated significant
changes in faunal composition between moss-
covered and moss-free segments on a single
limestone cliff. Agócsy (1968) reported sub-
stantial differences in species composition
and abundance between adjacent limestone
and sandstone outcrops. Cameron (1978) re-
ported significant shifts in the faunas found on
adjacent vertical and horizontal surfaces.
Kralka (1986) demonstrated that over 60% of
terrestrial gastropod species clustered signifi-
cantly within individual boreal forest stands.
Small-scale patterns in terrestrial gastropod
distribution are also suggested by the control
of soil chemistry on community structure (Out-
eiro et al., 1993; Hermida et al., 1995), as soil
chemistry is known to be highly variable over
1 -2 cm distances (Burrough, 1986).
Unfortunately, little is known about small-
scale community composition and diversity
patterns within the richest known global sites.
At the Waipipi Scenic Reserve in New
Zealand, Solem et al. (1 981 ) did not quantita-
tively subsample the fauna, but rather docu-
mented total gastropod diversity over the
entire 4.2-ha reserve. Emberton (1995) docu-
mented total species richness within a 4-
hectare region near Manombo, Madagascar
without measuring diversity from subsamples
within this site. Similarly, Emberton et al.
(1997) only documented total richness from
entire 4-hectare regions in eastern Tanzania.
Diversity gradients at Pine Mountain, Ken-
tucky, the richest North American terrestrial
gastropod site, have not been documented
(Emberton 1 995). The smallest quadrats sam-
pled by Tattersfield (1 996) from the Kakamega
Forest Reserve in Kenya were 40 x 40 meters
in size, while the minimum quadrat size
sampled by deWinter & Gittenberger (1998)
in southwestern Cameroon was 20 x 20 me-
ters. While Schmid (1 966) did measure terres-
trial gastropod richness from individual 1 m^
^Department of Natural and Applied Sciences, University of Wisconsin-Green Bay, Green Bay, Wisconsin 54311, U.S.A.;
nekolaj@gbms01 .uwgb.edu
^Department of Natural Resources, Cornell University. Ithiaca. New Yorl< 14853, U.S.A.
253
254
NEKOLA& SMITH
ОгееЯ Bay-,
Hwy.54.
New Franken
FIG. 1 . Location of the Celtis and UWGB Escarpment sites in northeastern Brown County, Wisconsin, USA.
quadrats near Tübingen, Germany, these
were not part of a larger, systematic sampling
regime and were not used to document diver-
sity gradients within habitats.
Previous analyses (Nekola, 1999) have
shown that carbonate cliffs from the Great
Lakes region in North America are among
the richest terrestrial gastropod communi-
ties reported from less than 1 ha scales. How-
ever, like other high diversity communities,
nothing was known of: (1) diversity gradients
within these sites; (2) the environmental fac-
tors within sites that correlate with high rich-
ness microsites, and (3) the scales of organi-
zation for faunal diversity within and between
sites. This paper attempts to address these
questions by analyzing terrestrial gastropod
diversity patterns at two sampling scales (1
m^ and 0.04 m^) within two high-richness car-
bonate cliff sites in northeastern Wisconsin,
U.S.A.
MATERIALSAND METHODS
Study Sites
Two wooded carbonate cliffs in Brown
County, Wisconsin, with high levels of terres-
trial gastropod species richness were chosen
for study (Fig. 1). Both occur along the Nia-
garan Escarpment, a 1 ,300 km band of out-
cropping Silurian-age limestones and dolo-
mites that can be roughly divided into five
200-300 km long regions (northeastern Iowa;
eastern Wisconsin through the Garden Penin-
sula of Michigan; eastern Upper Peninsula of
Michigan though Manitoulin Island; Bruce
Peninsula though south-central Ontario; and
southeastern Ontario to western New York
State) which are separated by low areas with
little or no bedrock exposure. Within the east-
ern Wisconsin region, the Niagaran Escarp-
ment is naturally divided into 31 isolated 2-8
km sections of exposed bedrock that emerge
above Pleistocene tills and alluvium.
The Celtis site (87^^50'52"W, 44°36'35"N) is
situated within a 7 km-long Niagaran Escarp-
ment section near the settlement of Ben-
derville. Its canopy is dominated by old-growth
sugar maple {Acer saccharum Marsh.), white
cedar ( Thuja occidentalis L.), paper birch {Be-
tula papyrifera Marsh.) and hackberry {Celtis
occidentalis L.). The bedrock outcrop at the
Celtis site is divided into a 3-6 m primary up-
land cliff and a 2-5 m secondary cliff associ-
ated with a large bedrock block displaced
downslope approximately 10 m (Fig. 2). Cool
SMALL-SCALE RICHNESS PATTERNS WITHIN CARBONATE CLIFFS
4
255
35 meters
FIG. 2. Schematic profile of tine bedrock outcrop at the Celtis site, with location of 1 m^ quadrats along each
transect. Vertical exaggeration is approximately 4x the linear extent.
air seepage from bedrock fissures and talus
occurs throughout the growing season. The
presence of large displaced talus blocks, ex-
panded bedrock joints, and an extensive sub-
tending talus indicate that this area was sub-
jected to intense periglacial erosion during the
late Pleistocene (Stieglitz et al., 1 980). This lo-
cation harbors the single most diverse terres-
trial gastropod assemblage known from the
Great Lakes region, with 34 taxa (Nekola,
1 999). Included in this fauna are the glacial re-
licts Catinella gélida (F С. Baker, 1927), Hen-
derson ia occulta (Say, 1831), Vallon ia gracili-
costa albula (Sterki, 1893), and Vertigo
hubrichti PWsbry, 1934 (Nekola et al., 1996)
The University of Wisconsin -Green Bay
(UWGB) Escarpment site (87"54'21"W, 44°
31'48"N) is located within a 4-km Niagaran
Escarpment section located near the Bay Set-
tlement community. This site consists of a sin-
gle 2-5 m tall upland cliff within a white cedar
and box elder (Acer negundo L.) canopy.
Large displaced talus blocks are absent. This
site was extensively modified by small-scale
quarrying for lime and building material ap-
proximately a century ago (Stieglitz et al.,
1980). As these activities were carried out
with hand tools and were spatially limited, this
site consists of a mixture of modified and un-
modified cliff segments. Unmodified seg-
ments appear essentially identical to the up-
land outcrop at the Celtis site. A total of 25
terrestrial gastropod taxa have been located
here, including the glacial relicts Vallonia gra-
cilicosta albula, Vertigo hubrichti, and Vertigo
n. sp. {'V. iowaensis" o\ Frest, 1991).
Data Sets
A representative and relatively undisturbed
section of exposed cliff was identified at each
site. From a random starting point within
these sections, five transects were laid out
perpendicular to the cliff face at 5-m intervals.
Along each transect both 1 m^ and 0.04 m^
samples were sampled:
1 ruf Quadrats: At the Celtis Site, seven
quadrats, each separated by 5-m distances,
were collected along each transect (Fig. 2) for
a total of 35. At the UWGB Escarpment, three
quadrats, separated by 5-m distances, were
collected from each transect for a total of 15.
Only three samples per transect were
gathered from the UWGB Escarpment site to
avoid highly disturbed forest and recreational
trails occurring at greater distances from the
cliff base.
The vascular plants growing on or over
each quadrat were recorded and their species
richness calculated. Soil temperatures at the
ground surface and at 20 cm depth were mea-
sured using a thermocouple thermometer. A
100 gm dry weight soil sample was collected
by subsampling the corners and center of
each quadrat. These samples were sent to
the Wisconsin State Soils Lab at the Univer-
sity of Wisconsin -Madison for analysis of
256
NEKOLA& SMITH
percent organic matter, N, P, K, S, Ca, Mg,
and pH, using methods outlined in Dahnke
(1988).
Terrestrial gastropod assemblages were
determined by collecting a total of 4-5
deciliters of soil litter from tfie corners and
center of each quadrat. Samples were slowly
and completely dried in either a low-tempera-
ture soil oven (approx. 80-95°C) or in full sun
in a greenhouse. Dried samples were then
soaked in water for 3-24 h, and subjected to
careful but vigorous washing through a stan-
dard sieve series (ASTME 3/8" (9.5 mm), 10
(2.0 mm), 20 (0.85), and 40 (0.425 mm) mesh
screens). The washed fractions were re-dried
and then re-sifted through the original sieve
series. The dry, resorted fractions were hand
picked against a neutral-brown background
using a small sable brush. All shells and shell
fragments were removed.
Recovered, identifiable shells were as-
signed to species (or subspecies) using the
author's reference collection and the Hubricht
Collection at the Field Museum of Natural His-
tory. From this, species composition and rich-
ness per quadrat was calculated. All speci-
mens are housed in collections maintained at
the University of Wisconsin -Green Bay.
0.04 nf Quadrats: 20 x 20 cm quadrats were
collected adjacent to 1 m^ quadrats along
each of the five established transect lines.
Quadrats were sampled at distances of 0, 0.5,
1 .0, 1 .5, and 2.0 meters from cliff bases. The
primary and secondary cliffs at the Celtis site
(located at positions 1 and 5, respectively, on
Fig. 2), and the primary cliff at the UWGB
Escarpment were analyzed in this fashion for
a total of 75 observations. For each quadrat,
transect position and distance from the cliff
base were recorded, and a 2-3 deciliter soil
litter sample collected. These litter samples
were subjected to the same laboratory
procedures described for the 1 m^ samples to
determine terrestrial gastropod composition
and richness.
Comparison of Maximum Richness Levels:
The maximum richness of 0.04 m^ and 1 m^
samples from each site were compared with
known richness values from a series of nested
samples of increasing sample grain. These
grains include each site (observed from a 100
m^ quadrat), escarpment section, escarpment
region. Brown County, and the state of Wis-
consin. Richness values at these increasing
scales of observation are based upon species
lists from other sites (summarized in Nekola,
1999), augmented by other published records
(Levi & Levi, 1 950; Teskey, 1 954; Jass, 1 986).
Richness estimations were limited to carbon-
ate cliff habitats for Escarpment sections
and regions, while those for Brown County and
the state of Wisconsin included all habitat
types.
To compare the maximum richness values
from this study with other reported maximum
richness values, a survey was made of the
published literature to identify other datasets
in which both terrestrial gastropod richness
and sample grain were reported. If multiple
examples of such data were found from a sin-
gle paper, only the richest were entered for a
given sample size. Through this process, a
total of 35 records from four continents
(Africa, Australia, Eurasia, and North Amer-
ica) were recorded (Burch, 1956; Schmid,
1966; Agócsy, 1968; Mason, 1970; Berry,
1973; Paul, 1975; Uminski & Focht, 1979;
Bishop, 1980; Solem et al., 1981; Van Es &
Boag, 1981; Waiden, 1981; Nilsson et al.,
1 988; Cameron & Greenwood 1 991 ; Young &
Evans, 1991; Cameron, 1992; Outeiro et al.,
1993; Getz & Uetz, 1994; Cowie et al., 1995;
Emberton, 1995; Wardhaugh, 1995; Tatters-
field, 1996; Emberton et al., 1997; de Winter
& Gittenberger, 1998).
Statistical Analyses
1 nf Quadrats: Analysis of the effect of
quadrat position and site on richness was con-
ducted via ANOVA. These data were graphi-
cally represented using box plots (Velleman &
Hoaglin, 1981). In box plots, the central line
represents the median of the sample, the mar-
gins of the box represent the interquartile dis-
tances, and the fences represent 1 .5 times the
interquartile distances. For data having a
Gaussian distribution, approximately 99.3% of
the data will fall inside of the fences. Outliers
falling outside of the fences are shown with as-
terisks. Identification of the environmental
variables that best predict observed richness
was accomplished through multiple linear re-
gression using a backwards stepwise selec-
tion procedure. Beginning with the most non-
significant, variables were removed from the
model until all remaining p-values fell below
the 0.05 level. Analysis of residuals and indi-
vidual variable distributions indicated that data
transformations were not necessary.
0.04 nf Quadrats: Analysis of the effect of
quadrat position, cliff position, and site
location on richness was conducted via
SMALL-SCALE RICHNESS PATTERNS WITHIN CARBONATE CLIFFS
257
ANOVA. These data were graphically repre-
sented for each of the three sampled cliffs
(Celtis Site primary and secondary, and
UWGB Escarpment primary) using box plots.
A full ANOVA with all interaction terms was not
conducted as a secondary cliff was not
present at the UWGB Escarpment site.
Comparison of Maximum Richness Levels:
The percent of total richness from the five
nested larger sample grains overlying each
maximally-rich 0.04 m^ and 1 m^ quadrat was
calculated for each site. The richness of these
different sample areas was natural-log trans-
formed and regressed against natural log-
transformed estimates of habitat area. Habitat
size estimates for Niagara Escarpment sec-
tions and regions were generated by multiply-
ing average cliff-base habitat width (approx. 5
m) by habitat length. The natural log of maxi-
mum richness vs. the natural log of sample
size was also plotted for the 35 literature rich-
ness records and for the maximum richness
0.04, 1, and 100 m^ quadrats from the Celtis
and UWGB Escarpment sites. While the lim-
ited number of samples prevented use of in-
ferential statistics to test for significant differ-
ences between the maximum species area
curves for carbonate cliffs (n = 3 for each site)
and other habitats, a qualitative assessment
was made.
RESULTS
1 m^ Quadrats
Richness of terrestrial gastropods at the
two sites ranged between and 23 taxa (Ta-
bles 1 , 2), with a mean of 6.6 taxa/ quadrat.
Across both sites, mean richness was 9.9
from quadrats collected at transect position 1 ,
7.1 at position 2, 7.2 at position 3, 6.8 at posi-
tion 4, 5.8 at position 5, 2.4 at both positions 6
and 7 (Fig. 3). ANOVA of these data demon-
strated that this variation was weakly signifi-
cant (p = 0.04; Table 3). Further ANOVA tests
demonstrated that no significant differences
were present between the mean richness of
transect positions 1 -5 (p = 0.326), or between
the Celtis and UWGB Escarpment sites (p =
0.646). No interaction between transect posi-
tion and site on richness was observed (p =
0.947). Backwards stepwise linear regression
of ten environmental variables on richness
demonstrated that only pH (p < 0.0005) and P
(p = 0.04) were significant predictors (Table
3. 4 5^ 6, 7.
Transect Position
FIG. 3. Box-plot diagram of terrestrial gastropod
richness at 1 m^ scales with increasing distance
from the primary upland cliff.
25
20
15
10
r2 =
1
0284
1
1
1
P <
0.0005
©
•
-
s
о
о
^
/"
о
^..^^
о
©
о
©
í^
в
Q
щ^
G
О
G
^
^о
m
О
о
о
о
о
о
©
о
1 O-
л
_о
i
I
1
70
72
76
рН
FIG. 4. Scatterplot of terrestrial gastropod richness
in 1 m^ quadrats vs. soil pH, with best-fit line.
Twelve of the data points are not apparent in this
graph as they overlap previously plotted values.
4), accounting for almost 30% of observed
richness variation. However, this level of sig-
nificance of P appears to based upon a single
outlier. When this observation was removed
from analysis, the p value for P in a multiple
linear regression of pH and P on richness
dropped to 0.266. The amount of variation in
richness accounted for by pH alone was
found to exceed 28% (Fig. 4).
ü
ш
r^
■.- о T-
ш
CD
LU
Ю
^ CM ■.-
LU
^
со ^
LU
СО
LU
CM
^ en T-
LU
-
O CM T-
-
Q
r^
Q
со
Q
un
Q
^
-
^
Q
со
T- 1Л C\J C\J
cg
Q
0\I
■--
Q
^
1- r^ со t-
Ü
1^
О
Cû
Ü
LO
■■-
Ü -^
H О
1- со
1- CM
C\J
Ü
со
Ü
CM
-
о
-
-
Cû
h-
CÛ
со
Cû
LD
- -
Cû
'i
со
Cû
со
CM CM CM
Cû
Сч]
-
Cû
-
-
<
h-
<
Cû
CM T- ■■-
<
ю
-
<
^
--
<
со
1- со со t-
<
СЧ
<
■■-
со
CM 1- со
CM 1- CM
,^1^ ^ (T>
T- •.- 00 •^
CM CD ■,- .
00-53
T— ^ 03 .^
>^ ro
ra c/)
'СО :рсл
СП
О)
'Î2^c^
T- CM
>,2J 00
_C0 ^-
00 - 00
5^oo
^ en
CD о
^ i:; t
CT3 CÜ-55
О
^ Sr- го со .-^ го со
- ^00
■ï
5 03
^ t о
та о
cj .S о ro
■g JN с Q
о о O) ^
Ü -c CL 5=
>^x:
(D о
< 1£
11
Д -ÎS Q. en стз та та та
to (Л ^- -ïï -iS -¿S
гГ-С: Cj CJ
.i5 та
^ ^^
■3 <^
о .Í2
О Q
СЛ Ci. Q.
3 О CJ
CJ со СЛ
:з та та
LU езо
9- S- S-
о о о
со о о
о о о
ел ел со Cl
та та та э
езооез
со "•'
р со
Is
■5 "D
го О
§ .о
та "55
00 rñ "-'
T- Ш QJ
■ J¿
>,< ce
СО та .
таЯ°°
3 0)-^
^^ g -fc
§ та^
§ $ S
Oí "ï^ _3
- . со
"О i-
со ■,—
со >.
— ф
та с
.с
7Z -^^
Q
Q^ та 3
Ï Q. Q.
С <
11
a D5
5 CD
-S ■5>
о t
со ь.
"5 -^
Э го
- та О - го
та er -о
0) о -^
tr с Е
Д) .о ^
о о
со СП
о о
см г^ со
со со ел
СП (Л
-.- со
см -.- ■.- со 'd-
см см
^ со
1^ со
un со Г-.
■.- со
■— eg
■— ^ со
сч ^
со со
О) со
о о
-.- см т-
^- со 1Л
Tf оо
см см
C4J т- 1- г^ ю
о о
■>- о^ со
см СП г^
ТЗ (Я
_ с
го -Е
s:rs¡ t ^Ь£S
SMALL-SCALE RICHNESS PATTERNS WITHIN CARBONATE CLIFFS
259
TABLE 2. Abundance and richness of terrestrial molluscs in 1 5 1 x 1 m quadrats at the UWGB Escarpment
site
Transect
A
A
A
В
В
В
С
С
С
D
D
D
Е
Е
Е
Quadrat
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
Auguispira altérnala (Say, 1 81 7)
Carychium exile H. С. Lea, 1842
Catinella avara (Say. 1 824)
Cochlicopa lubrica (Müller, 1774)
Cochlicopa lubricella (Porro, 1838)
Discus cronkhitei {Nev\/comb, 1865)
Gastrocopta armífera (Say, 1 821 )
Gastrocopta holzingeri {S\ei\<\. 1889)
Hawaiia miniscula (A. Binney, 1 840)
Helicodiscus shimeki Hubricht, 1962
Strobilops labyrinthica (Say, 1 81 7)
Succinea ovalis Say, 1817
Triodopsis multilineata (Say, 1821 )
Valíanla costata (Müller, 1 774)
Valíanla gracilicosta Reinhardt, 1883
Valíanla pulchella (Müller, 1774)
Vertiga gauldi [к. Binney, 1843)
Vertiga hubrichtl (Pllsbry. 1934)
Vertiga "lawaensis"
Vertiga pygmaea (Draparnaud, 1 801 )
Immature Individuals
Total Individuals
Richness
2
6
1
3
1
3
2
2
6
11
8
15
8
18
10
3
2
6
2
6
5
1
2
1
1
2
3
1
1
1
1
8
3
7
1
5
5
1
4
3
1
3
3
11
3
4
1
1
1
2
1
1
3
1
1
1
1
2
2
1
4
1
8
2
1
3
2
2
1
1
1
1
2
1
6
19
1
1
14
5
16
7
1
8
3
30
1
10
20
1
2
5
2
4
2
2
1
3
2
9
1
1
4
1
7
1
2
4
2
3
5
3
2
2
13
3
22 12 21 42 52 36 37 45 23 25 55 26 57 22 12
546 11 85668878 12 95
TABLE 3. Summary statistics for ANOVA of terrestrial gastropod species richness
in 1 m^ quadrats vs. quadrat distance from base of primary cliff at both the Celtis
and UWGB Escarpment sites.
Source
Sum-of-Squares
df
F-Ratio
P
Distance from base
Error
Squared multiple r:
313.48
835.90
0.273
6
43
2.688
0.026
TABLE 4. Results of backwards stepwise linear
regression of 10 environmental variables on terres-
trial gastropod species richness at 1 m^ grains.
Variables are listed in the order in which they were
removed from the model. The p-values reported are
those immediately prior to removal of that variable
from the model.
Variable
p-value
Mg
0.968
Percent Organic Matter
0.783
Soil Temperature at 20 cm Depth
0.406
К
0.314
Vascular Plant Species Richness
0.293
Ca
0.283
N
0.167
Surface Soil Temperature
0.121
P
0.040
PH
< 0.0005
0.04 m^ Quadrats
Terrestrial gastropod richness varied from
to 21 taxa on all three sampled cliffs (Tables
5-7). At the primary upland cliff at the Celtis
site, average species richness was 1 6.8 at the
cliff base, 9.4 at 0.5 m, 3.2 at 1 m, 2.8 at 1.5
m, and 2.2 at 2 m distances from the cliff. At
the secondary cliff at the Celtis site, average
species richness was 14.4 at the cliff base,
1 1 .2 at 0.5 m, 5.2 at 1 m, 2.8 at 1 .5 m, and 2.8
at 2 m distances from the cliff. At the UWGB
Escarpment site, average species richness
was 8.0 at the cliff base, 8.0 at 0.5 m, 6.0 at 1
m, 5.8 at 1 .5 m, and 6.4 at 2 m distances from
the cliff (Fig. 5). ANOVA of these data (Table
8) demonstrated that distance from the cliff
base and the interaction between this variable
260
NEKOLA& SMITH
ш ю
Ш 'd-
Ш 00
Ш C\J
ш -^
Q m
Q 'í
Q СО
Q c\j
a --
о LO
ü -í-
ü CD
ü CM
ü -
CÛ uo
CQ 'í
m СО
CÛ OJ
CD -^
< uo
< ^
< со
< со
ce
TJ
с
3
X]
<
1Л
ш
о
го
_|
СЛ
■D
m
С
го
Í
га
3
о
OîCDCD-'-LnoO СП т-^^1-1П
i-CMtOr^OJOO-'-CM UOCOr^i-CD 1-1-
со со
LO
со
со со см
Г\] 00 1-С\|1-С0 r^N-co
■^
со со
СЧ1
^ со
CV
1Л ^
CD
со
со CÛ
со 1-
uo 00
со CD
со
О 1-
00 см
со
-
-
см 1-
•^
СМ 00
см см
см
см
со г^
см со со с» С£>
1- см un со
1- г^
СП
1- 1- 00 1- ^ со со
со г^ со '^ LO со
1Л
со
со
^
со
г^ со
1Л 1-
CD иО
со см
00
-
CD см
см
00
г^ со
о т
-
-
^
см 00
un 1-
1-1-1- ю ^
см
О) 1-
CD см
со
1- со 1-
1- 1- со
1- (М íC) in íT) 1-
'í со 1-
1- со CM 1-
CD СО
un
о t--
со
00 CM
^
CD
г- un
'í CM 00
со 1-
г^
CD
■* CD
un 1-
~ 00 '^ i^
^^CM CD 00 -^
^ - 1- CM T- 1- 00
00 ;r~
00
- P?,^
00
со ^ ~ ■-; 1-
Ф
^ ra jï: >-
>> en ra ^
>- CD ~i с
^^:§-§
ce Ш -g-;;
S -5 i5 о
=¡ ^ с -с
D) О-й cj
c стз (тз о
h- >^
tz- о ^
(тзО
^d:
с: Е
Ф С
^ стз
" о
u^ 2 о ^
О О съ .Í2 со
оо Q Q ез
го го л? го
со СЛ ^^
^ -5 Î §
5 2.1^^
с "6 -!ü с
о о о <1)
CJ о -с Q.
со 'í ^
'^ >> -с
íí^ (- го
c$^.i^
--CÛ
5<
Ci. пз
8 ф
с/5 -is -ïï -is -is -is
2 is -Э -S
Q. Q. Q. Q.
О О О О
CJ О О CJ
со со со со
СТЗ СТЗ et! СТЗ
CD ез CD es
СО
ф-21
сс с ^
с s о
■^ 'Л
-с Í5 §
^ СХЗ (D
CDU
~ Ф Ol
со с ■^
1— Ш (Ц
>>< ГО
ГО ^-CÛ
СО стз .
г= с
^ Б -fc
Ç m S
X СО i5
ГО __
со ^.
СТЗ ^
■У 5
X £
^^ m ^^
С 2 с:
Qi та гз
ï О. О.
■ой
i5 to
о", та
11
■i= та
. ^- оо
"О ^
го^
х:
Е >;
■ф Ф
is m
со
о !§
>- Й го
ir О Í2
m
it
та CD
О) о
г- 5" Q) Ф со
"D i^ СЛ
о о - а Ф
О) .О) S го _ с
"t ■£ с Е iS -Я
i^ Ь. N ^ f^ (Г
SMALL-SCALE RICHNESS PATTERNS WITHIN CARBONATE CLIFFS
261
Ü
LU "*
Ш СП
Ш CM
Ш •.-
Q in
Q ^
Û CO
Q СЧ
Q T-
O ю
О ^
Ü CO
Ü CM
Ü ■-
CÛ un
CÛ ^
CD CO
CD СЧ
CÛ ■.-
< in
< M-
< CO
< oo
< T-
■^ о
I- о
CO
T- Ю
CM
^ y-
^
^
CD 00
CM
LD T- un
CO
о
CM
00 C\J
'S- T-
CD in en
CVJ
LO
CO
o in
00 T-
СЧ
СЧ
СЧ T-
^ CO
OJ
CO Ю
^ CO CM T-
■^ -r- T- in
■>- CO Ю -r- ^ T- CO
CD
^ 00
eg CM -^ CM CO
CM
■^ CM •.-■.-
о ^ ^
CM
о CM T- x- ^
■.- СЧ ■.-
CM CM 1- 1- ■.-
'i Ю LO
52
^-l- (^
Ou
? -3?
СП Oí
с со
:^ Ф
О) J
9 ^ Sí ^
i5 о
2|
si
Q LU
í^^í^гo
- . r; CM t-
>^^ 00 со
-О ■,- '^ '^
- ^^>^
;^ >< ce л
^ Í5 и С
5 ф ч ш
is^ is -У
О) Е с: tr
^ S о о
о лз о о
Ci со СП га
со
О- а а
о о сз
о CJ о
р о о
CT)
со -^,-^
00 -^ со
•.- СЧ СЧ
_. 00 со
;^ -^ 1-
ш >. >,
*- со си
W со ел
? с со
3i о -^
О) -о Й
■Soc
о (Ь о
•с Q.Ç
со со .со
■^ ю 00 1- -г- со
■^ СЛСМт-'^т-СМ-г-
со ■>-
со со
со
ю см
со
-
00 LO
ю ■>-
см
•■- 00
2 tC'^
S^2
:¿f ;= со
^00 ш^
9- "=^5
о о "S
cj о ^
g g ^
^ '!Л 9-
^ ■« -U -U со со -с^
LU ез ез CD ез ез ез
со со со
со со со
Q. 00
^ ='
^ го
§9,
^ ■-
с 2Í
^ <л
О) >ч о
с го ;z
Е ^-°
bû сох
о 2 -с:
со CL ÍO
р to to
g CJ CJ
_ CO CO
00 Ф ^ .--■'-■*
(0
■1- CO Ш Ф
5i-^ < ro
ro 2 — m
ел ^^ CO .
- с
CD TD
С ■£
СО с
со с
.о с
X :Е СО
^ ТЗ 1-
й => СО
■^ с ,
^ s го
to Lr
^ а
i-' :i-' со с
езо
.to to
2 -о -s
о о
со ^
о qj
II
со о
Cl О)
о.
-с
ГС
^ .ел
- - - -> о tr
Л> ^ со :з -fc (ь
COi>
3; 5 CL о.
Ф О
:d s; с^
о о "5 2
.оз.слЗ га
tr t с Е
IS
262
NEKOLA& SMITH
ш
ш
ш
ш
ш
Q
Q
Q
Q
^
~
Û
о
с
ф
Ü
E
о.
го
Ü
о
(Л
ш
О
CD
о
^
О
3
ф
^
О
го
¡л
CÛ
ГО
ГО
CÜ
3
сг
-4J
СО
Е
о
о
CÛ
(М
X
о
CÜ
C\J
Ln
(M
<
с
о
<
ел
_2
о
<
Е
го
<
1Л
О)
<
о
CD 1-
00
Ln
CD
CM
сч
^
СО
"*
Ti- Г^ C\J
^ ■■- -,- ■.-
\г> <у> -^ со
со ■>- Гч1
Сч1 ■.- Г^ OJ Ю
C\J CvJ со 04 ■.- 04 CT)
N-
■^ -^ C4J
OvJ -.-со
т- OvJ Г^
•>- CvJ о
04
т- 1-00
со 04 •.-•.-
■>-■>- т- СО
СМ
^ СТ5 т- со
СП
04
CNJ Cvl
04
■■- СО
С^О 04
С4
СМ
СМ
04 04
см
ю
■.- ю ■.-
1- 1Л 04
СО ^ 1-1-
1- СО 1-
0J 00 CÛ
со
04 Г-- со Ю
04
0^ О CT)
т- СО
04 1- СО СО
04
ю LO CT) СО
1- со 1Л N-
04
1- 1- m cvj г-
04
СМ 04 со
04
1- СО СО см СО
04
un ^ со со
Tt со 00
[^ со т-
00 1-
^ ^^^ 0^-.
^ Е
>• га С4 (D ^ о
го Ф 00 — ^-' '^
ел -J 1- о
~ — ; го -
^ X ел
*= <li ^
СТЗ О) §
ев g стз
r^ f^ ^ — -
Ф et!
>' го
га w
. ^ го с
CT) "<М
-со 2 ^о —
со g CT) N.
_- '" ^-00
ш ё ё -'
со с -Q
Q)
СЗ)
со 1^ -о ¡^
с g Ä g s
< 15
стз стз
о о
щ :¿ :¿ to _ _
с: -с -с я -Ь: Ь
стз стз
9- 9-
о о
CJ CJ
о о
стз стз
о о о о Q
<2 о bí
о о й
со со
стз (тз
сзез
стз с ^
о ^ со
C^:2"S
ê 5 8
со à -.s
стз стз CD
CDÎÏ
га i^ >-U
ел 1- (fl
^00 ел ^-
СТЗ ^ ^^
:£ го го 5
с ел с е
с со s 3
-Q го 2 я
^ ^ £ to
'^ Sa 8
о о о :^
i= CJ с стз
ел ел к :^»
st <^
00 О)
-со CL
< с:
Й -с í; и
о Ù
5 с
3 -о
О) о. о 2
СЗ).С
о о
СЗЗ.СЗ)
■£ ■£
о о v£' ;£'-.£' \д' Ä
-о 2
= =1 Û
со о ^'
to С5 го
с: — 0)
Q) Р го
о о о 5
CI).C3).D) со
t -е -5 Е
СЬ Q) 0) с:
^ ^ ^ ^
ßä
SMALL-SCALE RICHNESS PATTERNS WITHIN CARBONATE CLIFFS
263
TABLE 8. Summary statistics for ANOVA of terrestrial gastropod species rich-
ness in 0.04 m^ quadrats vs. quadrat distance from cliff base, cliff position, and
site location.
Source
Sum-of-Squares
df
F-Ratio
P
Site
0.08
1
0.008
0.928
Cliff position
2.88
1
0.296
0.589
Distance from base
406.08
4
10.423
< 0.0005
Distance * Site
271.72
4
6.974
< 0.0005
Distance * Cliff
25.12
4
0.645
0.633
Error
584.04
60
Squared multiple л; 0.699
Primary Cliff. Celtis Site
Secondary Cliff. Celtis Site
Primary Cliff. Campus Site
• 1
JL^
i Ä □ '
" 1
Ьв"
0.0 0.5 1.0 15 20
I frc5m Clitf (m)
00 05 1,0 15 20
Distance from Cliff (m)
0.5 1,0 15 20
Distance from Clrtf (m)
FIG. 5. Box-plot diagram of terrestrial gastropod richness from 0.04 m scales at increasing distances from
three cliff exposures at the Celtis and UWGB Escarpment sites.
and site were both highly significant (p <
0.0005). In all, distance from cliff base and
site accounted for 70% of the observed varia-
tion in richness. Mean richness was not af-
fected by cliff type (phmary or secondary; p =
0.649) and cliff type did not interact with the
rate of richness decrease from the cliff base (p
-0.518).
Comparison of Maximum Richness Levels
At the UWGB Escarpment, a maximum of
1 2 taxa were observed from single 0.04 and 1
m^ quadrats. As this site represents the lone
surviving carbonate cliff community within the
6 km Bay Settlement section, the total rich-
ness of this Escarpment section equals site
richness (25 taxa). For the Celtis site, all
small-scale richness values were higher, with
a maximum richness of 21 taxa occurhng in a
single 0.04 m^ sample, and 23 taxa in a single
1 m^ sample. Carbonate outcrops along the 8
km Benderville Escarpment Section (within
which the Celtis site occurs) support 38 taxa.
All carbonate cliffs along the 350 km Eastern
Wiscosin-Garden Peninsula Escarpment Re-
gion support 62 taxa. The total richness of
Brown County terrestrial gastropods, across
all habitat types, is 65 taxa, while a total of 95
taxa have been documented across all habi-
tats in the state of Wisconsin (Table 9).
The faunas of maximum richness 0.04 m^
quadrats at the Celtis site thus account for up
to 91% of maximum 1 m^ richness, 62% of
site richness, 55% of Escarpment section
richness, and 34% of Escarpment region rich-
ness. Individual 0.04-m^ quadrats also har-
bored up to 32% of the entire county fauna,
and 22% of the entire state fauna. Given their
similar maximum richness, almost identical
results are present for maximum richness 1
m^ quadrats, which can harbor up to 68% of
the entire site fauna, 61% of the Escarpment
section fauna, and 37% of the entire Escarp-
ment region fauna. Individual 1-m^ quadrats
can also harbor up to 35% of the entire county
fauna, and 24% of the entire state fauna. Be-
cause of the lower richness levels at the
UWGB Escarpment Site, these numbers
tended to be lower by almost 30-50% from
Celtis site levels. Regression analysis demon-
strates that a high correlation (p < 0.0005; /^ =
264
NEKOLA& SMITH
TABLE 9. Percent of site, escarpment section, escarpment region, county, and
state terrestrial gastropod faunas contained withiin maximum diversity 0.04 and
1 m^ quadrats.
UWGB Escarpment Site
Percent Overlap
Sample Grain
Richness
0.04 m^ 1 m^
0.04 m^
12
1m2
12
100
Site (100 m^)
25
48 48
Niagara Escarpment
Bay Settlement Section
25
48 48
Eastern Wisconsin Region
62
19 19
Brown County
65
18 18
Wisconsin
95
13 13
Celtis Site
Percent Overlap
Sample Grain
Richness
0.04 m^ 1 m^
0.04 m^
21
1m2
23
91
Site (100 m^)
34
62 68
Niagara Escarpment
Benderville Section
38
55 61
Eastern Wisconsin Region
62
34 37
Brown County
65
32 35
Wisconsin
95
22 24
0.839) exists between natural log-trans-
formed richness and sample area, with the
best-fit line having an intercept of 2.93 and a
slope of 0.063 (Fig. 6). This translates to an
average richness of 18.7 taxa per 1 m^ and
33.4 taxa per hectare.
Comparison of maximum richness from
0.04 and 1 m^ quadrats to other reported
maximum richness values demonstrates the
Celtis site site is among the richest reported
globally from small observational scales (Fig.
7). Maximum richness levels at the Celtis site
compare favorably with the richest reported
1 -400 m^ samples in Germany, Sweden, and
Scotland (Schmid, 1966; Waiden, 1981;
Cameron & Greenwood 1991). The richest
global terrestrial gastropod faunas, collected
over larger areas (400-40,000 m^) in New
Zealand, Madagascar, Tanzania, and Came-
roon appear to fall along the same maximum
species-area curve defined from the Celtis
Site. While maximum richness at the UWGB
Escarpment is lower, these observations still
lie within the upper half of previously reported
maximum richness levels for the given range
of sample scales.
г2 = 0.839
p < 0.0005
у = 293 + ООбЗх
-10 О 10 20
Natural Log Area (Square Meters)
30
FIG. 6. Scatterplot of natural log-transformed ter-
restrial gastropod richness vs. natural log-trans-
formed sample area for Wisconsin carbonate cliff
land snail faunas. The scales of observation in-
clude: maximum-richness 0.04, 1, and 100 m^
quadrats from the Celtis and UWGB Escarpment
sites: the respective escarpment sections for each:
escarpment region: county: and state.
SMALL-SCALE RICHNESS PATTERNS WITHIN CARBONATE CLIFFS
265
3.6
3.0 -
2.4 -
1
о
о «
о о о©о
• о ш
• •
о
о
° о
о
о
@ a®
о
8
о
о
о
о
о 10
ln(Sample Q'än In m^
20
• = Celtis Site @ = UWGB Site О = Samples from other
Global Localities
FIG. 7. Species-area plot of maximum-richness
0.04, 1 and 100 m^ quadrats from the Celtis and
UWGB Escarpment sites compared to other maxi-
mum richness levels reported from other global
sites.
DISCUSSION
Diversity Patterns Within Carbonate
Cliff Communities
The high richness of terrestrial gastropods
within carbonate cliff habitats occurs down to
very limited spatial scales, with up to 62% of
site richness (and up to 22% of total state rich-
ness) being found within single 0.04 m^ areas
along cliff bases. The limitation of high-diver-
sity assemblages to the immediate vicinity of
vertical bedrock outcrops is striking, with rich-
ness decreasing by almost six-fold from to 1
m from cliff bases. This rapid and drastic re-
duction in richness helps explain why, on av-
erage, 0.04 m^ quadrats at cliff bases harbor
more species than adjacent 1 m^ quadrats. As
the 1 m^ samples consisted of pooled sub-
samples taken from the corners and center of
each quadrat, only two subsamples per
quadrat were thus collected from mollusc-rich
microsites. This had the unintended effect of
diluting snail density and richness. However,
the 0.04 m^ cliff-base quadrats, which con-
sisted to a similar total volume of soil litter,
were collected entirely from the richest mi-
crosites, so that no dilution in snail density or
richness occurred.
While diversity at the 0.04 m^ scale was
markedly higher adjacent to cliffs at the undis-
turbed Celtis site, such spatial limitation of
richness was not observed at the UWGB Es-
carpment. In addition, maximum (but not
mean) richness at both 0.04 and 1 m^ scales
at the UWGB Escarpment site was roughly
one-half that recorded at the Celtis Site. It is
probable that these lower maximum diversity
levels, and lack of strong micro-scale diversity
gradients, are related to this site's past quar-
rying history, which may have simplified the
range of microhabitats present along the cliff
base.
The exact mechanisms that lead to high
levels of terrestrial gastropod richness at
micro-scales within these carbonate cliff sites
have not been documented. First, it is not
known how many of these shells originate
from individuals living in the quadrat versus
shells that have been deposited from nearby
areas, such as adjacent vertical rock faces. If
this latter process is important, levels of
micro-scale sympatry could be substantially
lower than the observed shell richness sug-
gests. However, preliminary observations of
living individuals on cliff bases and adjacent
vertical faces suggest that the majority of
shells originate from within quadrats. Addi-
tionally, most species observed from cliff-
base quadrats are represented by at least one
living snail or recently dead shell, while spe-
cies reported from more distant quadrats are
almost always represented by long-dead shell
fragments.
Second, even if shells from high-richness
microsites are locally derived, observed rich-
ness may be exaggerated if shells persist in
the soil for long periods. In this case, species
lists will represent an integration over the per-
sistence-time of shells. This could lead to an
overestimate of microsympatry if local faunas
are in a state of constant flux. Use of radio-
isotope dating on shells could provide a pos-
sible test for shell half-life, which would help
set the temporal scales of integration for such
shell-banks in carbonate cliff soils. However,
the fact that most taxa are represented in
samples by at least one live or recently dead
shell suggests that high levels of sympatry are
likely maintained at both limited temporal and
spatial extents.
The existence of diverse terrestrial gastro-
pod assemblages at very small scales on or
adjacent to carbonate outcrops is also likely
not unique to the two sites chosen for analy-
sis. Oualitative observations of other carbon-
ate cliffs in Illinois, Iowa, New York, Ontario,
Wisconsin, and southwestern England sug-
266
NEKOLA& SMITH
gest that the co-occurrence of 20 or more taxa
at 1 m^ or smaller grains may be typical in
undisturbed sites.
The limitation of high richness terrestrial
gastropod assemblages to the immediate
proximity of cliffs also indicates that these
microhabitats must be afforded special pro-
tection if their biodiversity is to be protected.
Unfortunately, planned and spontaneous
recreational trails in reserves are often routed
through these exact areas as they are aes-
thetically pleasing and provide access to
charismatic natural features such as caves,
fissures, and rock walls. Such trails may place
any high-richness terrestrial gastropod as-
semblages in serious jeopardy. For instance,
the cliff base at Bayshore County Park, 6 km
north of the Celtis Site, has been turned into a
graveled trail which now lacks a terrestrial
gastropod fauna (Nekola, unpublished data).
Environmental Controls on Small Scale
Richness Patterns
The literature regarding the environmental
controls of terrestrial gastropod richness and
abundance is very conflicting. Burch (1955)
suggested that in eastern Virginia snail abun-
dance (and presumably diversity) was related
to soil organic matter, Ca, Mg, and К, but not
pH. Lack of correspondence between terres-
trial gastropod distribution and pH has also
been demonstrated in southwestern Ireland
(Bishop, 1977) and the Italian Alps (Bishop,
1980). However, Gleich & Gilbert (1976),
stated that in central Maine soil moisture, but
not soil Ca, was the most important determi-
nant of snail abundance. Outeiro et al. (1993)
demonstrated that soil texture and pH were
the most important factors effecting terrestrial
gastropod distribution in central Spain. Soil
pH was also identified as an important deter-
minant of terrestrial gastropod density and di-
versity by Waiden (1981) and Gärdenfors
(1992) in southern Sweden, and by Bishop
(1976) in Somerset, England. Getz (1974)
and Getz & Uetz (1994) identified soil mois-
ture, tree diversity, and leaf litter diversity as
major determinants of diversity in the Great
Smoky Mountains. However, Locasciulli &
Boag (1987) demonstrated in Alberta forests
that terrestrial gastropod abundance was not
related to vegetation.
Within eastern Wisconsin carbonate cliff
habitats, only soil pH (and not soil Ca, Mg, N,
P, K, percent organic matter, soil temperature,
or vascular plant richness) was found to cor-
relate significantly with terrestrial gastropod
richness at the 1 m^ scale. This result is con-
sistent with the analysis of Bishop (1980),
who stated that soil pH will only be an impor-
tant environmental correlate of terrestrial gas-
tropod assemblages when soil Ca levels are
high. Interactions between environmental
variables may explain why such a diversity of
factors exist that have (and have not) been
shown to influence terrestrial gastropod rich-
ness across various habitats and regions.
Thus, extrapolation of these results to other
habitats or regions may be risky. These re-
sults do, however, provide insight into the en-
vironmental correlates of terrestrial gastropod
richness within Ca-rich carbonate cliffs.
Spatial Scales of Terrestrial
Gastropod Coexistence
A complete assessment of the scales over
which terrestrial gastropod coexistence is me-
diated is not possible because the current
data are restricted to five discrete spatial
scales (0.04 m^, 1 m^, 100 m^, approx. 3.5 ha,
and approx. 1750 ha). However, given the
wide total range covered, some preliminary
insights into this issue can be made.
Three of the measured scales appear to
harbor the bulk of terrestrial gastropods. Up
to one-third of the total regional fauna may
occur within individual 0.04 m^ regions. Over
50% of the total regional fauna may occur
within 100 m^ regions on individual sites. The
remainder of the fauna (almost 50% of the re-
gional total) is largely found between sites on
different Escarpment Sections (e.g., sites 10
km or more apart, corresponding to 3.5 -i- ha
of cliff base habitat). However, little increase
in richness was observed between 0.04 m^
and 1 m^ extents within sites, and between
sites within the same Escarpment Section
(100 m^-3.5 km of cliff base habitat).
One tentative conclusion that can be drawn
from these results is that alpha diversity
{sensu Whitaker, 1975) in these sites is best
measured at scales no larger than 0.04 m^,
beta diversity {sensu Whitaker, 1975) is best
measured at scales no larger than 100 m^,
and gamma diversity {sensu Cody, 1986) is
best measured between sites at least 10 km
distant from one another. Although Emberton
(1995) states that distinctions between alpha,
beta, and gamma diversity are hazy for ter-
restrial gastropod communities, these data
suggest that such problems in resolution may
be due to poorly chosen observational scales.
SMALL-SCALE RICHNESS PATTERNS WITHIN CARBONATE CLIFFS
267
For instance, if the Celtis and UWGB Escarp-
ment sites had been sampled at typical mala-
cologial sampling scales of 0.1 ha or larger,
distinctions between alpha and beta diversity
would be impossible to make, as sample res-
olution would be at least 1000 times greater
than the scale at which alpha diversity likely
exists.
Additional research will be necessary to
document the mechanisms that allow for
species coexistence at these differing scales.
Ecological theory suggests that alpha diversity
levels may be related to levels of niche parti-
tioning between species (Auerbach & Shmida,
1987). However, competition and prédation
have only rarely been shown to influence ter-
restrial gastropod distribution and abundance
(Cain, 1 983; Cowie & Jones, 1 987; Smallhdge
& Kirby, 1988). The high density of shells (up
to 330 per 0.04 m^ quadrat) further suggests
that resource levels are also high, cautioning
against use of resource-ratio models (e.g.,
Tilman, 1988). Identification of the small-scale
coexistence mechanisms in these habitats
may be of broad ecological importance, as it is
rare for alpha diversity to constitute such a
high proportion of regional diversity, and thus
for the rate of species accumulation with in-
creasing sample size to be so low (Rosen-
zweig, 1995).
At larger scales of observation, habitat het-
erogeneity may be important in determining
levels of species richness (Auerbach &
Shmida, 1987; Rosenzweig 1995). The few
additional species added between site and
escarpment section scales suggests that the
universe of microenvironments found within a
given carbonate cliff site may be very similar
to those present within an entire Escarpment
section. At the largest scales, coexistence of
terrestrial gastropods will likely be mediated
by large environmental gradients (including
climate), differential colonization histories of
habitats (Ricklefs & Schlüter, 1993; Nekola, in
press), and the incomplete dispersal of
species between sites (Auerbach & Shmida,
1987).
It is important to note that the spatial scales
of coexistence for terrestrial gastropod com-
munities, and hence the optimal scales of ob-
servation, may differ between systems and
landscapes. For instance, it is not clear that
alpha diversity will always be confined to such
small scales in habitats that support substan-
tially lower densities of individuals and taxa.
Gamma diversity will also likely vary between
landscapes, as preliminary analyses have
documented for Niagaran Escarpment car-
bonate cliffs in which rates of community
turnover may vary by an order-of-magnitude
(Nekola, unpublished data). Such observa-
tions suggest that additional research will be
necessary within and between a diversity of
habitats and landscapes to determine if any
general rules exist to guide the ecological
sampling of terrestrial gastropod communi-
ties.
ACKNOWLEDGEMENTS
Robert Cameron, Douglas Larson, John
Slapcinsky and an anonymous reviewer pro-
vided useful comments on earlier versions of
this manuscript. Assistance in sample pro-
cessing was provided by Catherine Steele
and students enrolled in the spring 1998 Ter-
restrial Gastropod Ecology Practicum at the
University of Wisconsin -Green Bay. Work at
the Leslie Hubricht collection was made pos-
sible through a Prince Visiting Scholar grant
from the Field Museum of Natural History.
Field research was funded by the Louis Almon
Fund (administered through the Wisconsin
Academy of Sciences, Arts, and Letters) and
a Cofrin Arboretum Grant (administered
through the Cofrin Arboretum committee of
the University of Wisconsin —Green Bay).
LITERATURE CITED
AGÓCSY. P., 1968, Data to quantitative conditions
in the mollusk faunas of two different substrates
in central Hungary. Acta Zoológica Academiae
Scientarium Hungaricae. 14: 1-6
AUERBACH, M. & A. SHMIDA, 1987, Spatial scale
and the determinants of plant species richness.
Trends in Ecology and Evolution. 2: 238-242.
BERRY, A. J,, 1966, Population structure and fluc-
tuations in the snail fauna of a Malayan limestone
hill. Journal of Zoology. London. 150: 11-27.
BERRY, F. G., 1973. Patterns of snail distribution at
Ham Street Woods National Nature Reserve,
East Kent. Journal of Conchology 28: 23-35.
BISHOP M. J., 1976, Woodland Mollusca around
Nettlecombe, Somerset. Field Studies. 4: 457-
464.
BISHOP M. J., 1977, Tne Mollusca of acid wood-
land in West Cork and Kerry. Proceedings of the
Royal Irish Academy 77B: 227-244.
BISHOP M. J., 1980, The Mollusca of acid wood-
land in the Italian province of Novara. Journal of
Conchology 30: 181-188.
BURCH, J. В., 1 955, Some ecological factors of the
soil affecting the distribution and abundance
268
NEKOLA& SMITH
of terrestrial gastropods in eastern Virginia. The
Nautilus. 69: 62-69.
BURCH, J. В.. 1956, Distribution of terrestrial gas-
tropods in plant associations in eastern Virginia.
The Nautilus. 70: 60-64.
BURROUGH, P. A., 1986, Principles of geographi-
cal information systems for land resources as-
sessment. Oxford University Press, New York.
CAIN, A. J., 1983, Ecology and ecogenetics of ter-
restrial molluscan populations. Pp. 597-647, In:
w. D. RUSSELL-HUNTER, ed.. The l\Aollusca. Vol. 6.
Ecology. Academic Press, London.
CAMERON, R. A. D., 1978, Differences in the sites
of activity of coexisting species of land mollusc.
Journal of Conchology 29: 273-278.
CAMERON, R. A. D.. 1986, Environment and diver-
sities of forest snail faunas from coastal British
Columbia. Malacologia. 27: 341-355.
CAMERON, R.A. D., 1992, Land snail faunas of the
Napier and Oscar Ranges, Western Australia: di-
versity, distribution and speciation. Biological
Journal of the Linnean Society. 45: 271 -286.
CAMERON, R. A. D. & J. J. D. GREENWOOD,
1 991 , Some montane and forest molluscan fauna
from eastern Scotland: effects of altitude, distur-
bance, and isolation. Proceedings of the Tenth In-
ternational Malacological Congress (Tübingen
1989), 437-442.
CODY, M. L., 1986, Diversity, rarity, and conserva-
tion in Mediterranean-climate regions. Pp. 122-
152, In: M. E. SOULE, ed.. Conservation Biology.
Sinauer Associates, Sunderland, Massachusetts.
COWIE, R. H. & J. S. JONES, 1987, Ecological in-
teractions between Cepaea nemoralis and Ce-
paea hortensis: competition, invasion, but no
niche displacement. Functional Ecology. 1: 91-
97.
COWIE, R. H., G. M. NISHIDA, Y BASSET & S. M.
GON, III, 1995, Patterns of terrestrial gastropod
distribution in a montane habitat on the island of
Hawaii. Malacologia, 36: 155-169.
DAHNKE, W. C, 1988, Recommended chemical
soil test procedures for the north-central region.
North Dakota Agricultural Experiment Station
Bulletin. 499, North Dakota State University,
Fargo.
DE WINTER, A. J. & E. GITTENBERGER, 1998,
The terrestrial gastropod fauna of a square kilo-
meter patch of rainforest in southwestern
Cameroon: high species richness, low abun-
dance, and seasonal fluctuations. Malacologia.
40:231-250.
EMBERTON, K. C, 1995, Land-snail community
morphologies of the highest-diversity sites of
Madagascar, North America, and New Zealand,
with recommended alternatives to height-diame-
ter plots. Malacologia. 36: 43-66.
EMBERTON, K. C, T A. PEARCE. R F. KASIGWA,
R TATTERSFIELD & Z. HABIBU, 1997, High di-
versity and regional endemism in terrestrial gas-
tropods of eastern Tanzania. Biodiversity and
Conservation, 6: 1123-1136.
FREST, T J., 1991, Summary status reports on
eight species of candidate terresthal gastropods
from the Driftless Area (Paleozoic Plateau),
upper Midwest. Final Report, Contract #301-
01366, USFWS Region 3, Ft. Snelling, Min-
nesota.
GÄRDENFORS, U., 1992, Effects of artificial liming
on terrestrial gastropod populations. Journal of
Applied Ecology 29: 50-54.
GETZ, L. L., 1974, Species diversity of terrestrial
snails in the Great Smoky Mountains. The Nau-
tilus. 88: 6-9.
GETZ, L. L. & G. W. UETZ, 1994, Species diversity
of terrestrial snails in the southern Appalachian
mountains, U.S.A. Malacological Review, 27: 61 -
74.
GLEICH, J. G. & F. F. GILBERT 1976, A survey of
terrestrial gastropods from central Maine. Cana-
dian Journal of Zoology 54: 620-627.
HERMIDA, J., R ONDINA & A. OUTEIRO, 1 995, In-
fluence of soil characteristics on the distribution
of terrestrial gastropods in northwest Spain. Eu-
ropean Journal of Soil Biology. 31 : 29-38.
JASS, J. P., 1986, Supplemental distribution rec-
ords for Wisconsin terrestrial gastropods. Trans-
actions of the Wisconsin Academy of Science,
Arts, and Letters. 74: 105-107.
KRALKA, R. A., 1986, Population characteristics of
terrestrial gastropods in boreal forest habitats.
American Midland Naturalist. 115: 156-164.
LEVI, L. R. & H. W. LEVI, 1 950, New records of ter-
restrial gastropods from Wisconsin. The Nautilus,
63: 131-138.
LOCASCIULLI, O. & D. A. BOAG, 1987, Microdis-
tribution of terrestrial snails (Stylommatophora) in
forest litter. Canadian Field Naturalist. 101: 76-
81.
MASON, С F., 1970, Snail populations, beech litter
production, and the role of snails in litter decom-
position. Oecologia. 5: 215-239.
NEKOLA, J. C, 1999, Terrestrial gastropod rich-
ness of carbonate cliff and associated habitats in
the Great Lakes region of North America. Mala-
cologia. 4^■.23^ -252.
NEKOLA, J. C, in press, Paleorefugia and neo-
refugia: the influence of colonization history on
ecological pattern. Ecology.
NEKOLA, J. C, T A. SMITH & T J, FREST 1996,
Terrestrial gastropod faunas of Door Peninsula
natural areas. Report to Wisconsin Chapter. The
Nature Conservancy. Madison.
NILSSON. S. G., J. BENGTSSON & S. AS, 1988,
Habitat diversity or area per se? Species rich-
ness of woody plants, carabid beetles and terres-
trial gastropods on islands. Journal of Animal
Ecology 57: 685-704.
OUTEIRO, A., D. AGÜERA & С. PAREJO, 1993.
Use of ecological profiles and canonical corre-
spondence analysis in a study of the relationship
of terrestrial gastropods and environmental fac-
tors. Journal of Conchology. 34: 365-375.
PAUL, С R. C. 1975, The ecology of Mollusca in
ancient woodland. I. The fauna of Hayley Wood,
Cambridgeshire. Journal of Conchology. 28:
301-327.
RICKLEFS, R. E. & D. SCHLÜTER, 1993, Species
SMALL-SCALE RICHNESS PATTERNS WITHIN CARBONATE CLIFFS
269
diversity: regional and historical influences. Pp.
350-363, in R. E. RICKLEFS & D. SCHLÜTER, eds..
Species diversity in ecological communities. Uni-
versity of Chicago Press, Chicago.
ROSENZWEIG, M. L., 1995, Species diversity in
space and time. Cambridge University Press,
New York.
SCHMID, G., 1966, Die mollusken des Spitzbergs.
Der Spitzberg bei Tübingen, Natur und Land-
schaftssctiutzgebiete. Baden-Würtemburg, 3:
596-701
SMALLRIDGE, M. A. & G. С KIRBY, 1988, Com-
petitive interactions between the terrestrial gas-
tropods Ttiebia pisana (Müller) and Cernuella vir-
gata (DaCosta) from South Australia. Journal of
Molluscan Studies. 54: 251 -258.
SOLEM, A., F. M. CLIMO & D. J. ROSCOE, 1981,
Sympatric species diversity of New Zealand ter-
restrial gastropods. New Zealand Journal of Zo-
ology. 8: 453-485.
STEIGLITZ, R. D.. J. M. MORAN & J. D. HARRIS,
1980, A relict geomorphological feature adjacent
to the Silurian escarpment in northeastern Wis-
consin. Transactions of the Wisconsin Academy
of Sciences. Arts, and Letters. 68: 202-207.
TATTERSFIELD, P., 1996, Local patterns of terres-
trial gastropod diversity in a Kenyan rain forest.
Malacologia, 38: 1 61 - 1 80.
TESKEY M. C, 1954, The mollusks of Brown
County, Wisconsin. The Nautilus. 68: 24-28.
TILMAN, D., 1988, Plant strategies and the dynam-
ics and structure of plant communities. Princeton
University Press, Princeton.
UMINSKY T & U. FOCHT 1979, Population dy-
namics of some land gastropods in a forest habi-
tat in Poland. Malacologia. 18: 181-184.
VAN ES, J. & D. A. BOAG, 1981, Terrestrial mol-
luscs of central Alberta. Canadian Field-Natural-
ist. 95: 76-79.
VELLEMAN, P. F & D. С HOAGLIN, 1981, Appli-
cations, basics, and computing of exploratory
data analysis. Addison-Wesley Press, Reading,
Massachusetts.
WALDÉN, H. W., 1981, Communities and diversity
of land molluscs in Scandinavian woodlands. I.
High diversity communities in taluses and boul-
der slopes in SW Sweden. Journal of Conchol-
ogy, 30:351-372.
WALDHAUGH, A. A., 1995. The terrestrial mollus-
can fauna of some woodlands in northeast York-
shire, England. Journal of Conchology, 35: 313-
327.
WHITTAKER, R. H., 1975, Communities and
ecosystems. MacMillan Publishing, New York.
YOUNG, M. S. & J. G. EVANS, 1 991 . Modern land
mollusc communities from Flat Holm, South
Glamorgan. Journal of Conchology 34: 63-70.
Revised ms. accepted 2 March 1999
MALACOLOGIA, 1999, 41(1): 271 -281
SAMPLING TERRESTRIAL GASTROPOD COMMUNITIES: USING ESTIMATES OF
SPECIES RICHNESS AND DIVERSITY TO COMPARE TWO METHODS
Karen D. McCoy^
Department of Zoology. University of Guelpti, Ontario, N1G 2Wb Canada
ABSTRACT
Terrestrial gastropods are sometimes sampled for ecological and parasitológica! studies using
simple traps of cardboard placed on the ground. Tfiis method enables the collection of large
numbers of individuals with relatively little effort. However, the violation of assumptions of the trap
method may mean that samples are biased. I examined the reliability of this method by compar-
ing species composition, richness and diversity of gastropod collections from traps to paired
hand-searched plots in Algonquin Park, Ontario, Canada. Ten samples for each method were
made from five replicate sites in three habitat types, white birch, hardwood and logged forest
stands. Of the 18 species found overall, 17 and 16 species were found using hand search and
trap methods respectively. Some species, especially Arion circumscriptus. were consistently
over-represented using traps, while most other species had greater representation in hand-
search collections. In general, hand-search collections had significantly greater species richness
and diversity than trap collections. Further, the similarity in species composition of the two col-
lections seemed to depend, at least in white birch and hardwood habitats, on the level of site di-
versity; as site diversity increased the percent similarity decreased. However, estimated rich-
ness, calculated using a jackknife estimator that accounts for heterogeneity in species detection,
was not different for the two collection methods. There was no difference in average species' de-
tection per sample between collections; the probability of sampling a species using a single sam-
ple within a site was low for both methods (16-34% of species). However, overall detection of
species on a site (average across all samples within a site) tended to be higher for hand-
searched plots, especially in more complex habitats. Thus, as trap sampling does not provide
representative gastropod collections, this method may not be appropriate to use when physical
collections or detailed information on species abundances are required. Conversely, as there is
no difference between methods in estimated richness, traps may be suitable to estimate gastro-
pod community metrics, provided appropriate estimation models are used.
Key words: Gastropoda, land Mollusca, community ecology, species richness, diversity, sam-
pling techniques, richness estimation.
INTRODUCTION
Terrestrial gastropods are often sampled
using simple traps of moistened cardboard,
masonite or tile, particularly for parasitological
studies (e.g.. South, 1965; Gleich & Gilbert,
1976; Kearney & Gilbert, 1978; Boag, 1982;
Samuel et al., 1 985; Strayer et al., 1 986). This
method is relatively simple and time efficient,
and has been considered an effective means
of sampling to estimate relative population
density and compositional changes of gastro-
pod communities, both spatially and tempo-
rally (Boag. 1 982). This method assumes that
(1 ) all species and age classes are equally at-
tracted to the traps at the times when the traps
are typically examined, dawn or dusk, and (2)
that individuals are only attracted vertically
from the litter layers beneath and not horizon-
tally to the trap. However, there are interspe-
cific differences in gastropod activity (Blinn,
1963; Cameron, 1970, 1978; Cain & Cowie,
1978; McCoy & Nudds, 1997), and in the use
of these traps (Boag, 1990). Further, some
gastropod species are thought to have large
home ranges (Blinn, 1963; Thomas, 1944;
Cook, 1979) such that individuals may be at-
tracted to traps for shelter or food from an
area of unknown size; there is now evidence
that such horizontal movement to traps exists
(Boag, 1990; Hawkins et al., 1997). Thus, the
violation of the assumptions of trap sampling
could result in biased gastropod collections
both in terms of the number of individuals per
Vresent address: Laboratoire d'Ecologie, Université Pierre et Marie Curie, C.N.R.S. - UMR 7625, 7 quai Saint Bernard,
75252 Paris, FRANCE: kmccoy@snv.jussieu.fr
271
272
McCOY
species collected and the number of species
represented (species richness).
The extent to which any collection tech-
nique accurately represents the gastropod
community in a given area is difficult to deter-
mine as there are limitations associated with
all techniques (e.g., Bishop, 1977; Emberton
et al., 1996). Bishop (1977) suggested that
the least biased way to sample terrestrial gas-
tropods would use stratified random quadrats
in which litter is either searched, or removed
and searched in the laboratory. Likewise,
Newell (1971) reviewed numerous direct and
indirect methods and found that hand sorting
soil samples, while slow and laborious, pro-
vided more accurate estimates of abundance
of some species than did surface counts
alone. He recommended the use of indirect
methods (such as trap sampling) only for
those animals that could not be reliably esti-
mated by a direct method. Conversely, Van Es
& Boag (1 981 ) compared hand-searched field
samples to samples sorted in the laboratory
after litter was dried and sieved and found that
hand-searched collections contained fewer
individuals than did litter samples and were bi-
ased towards larger specimens. Emberton et
al. (1996) suggested that combining direct,
timed searches with litter-plus-soil samples
was the most efficient means of quantifying
microgastropods (< 5 mm) in tropical rain-
forests. Others have recommended using in-
direct trap sampling, within certain climatic
limits, due to its efficiency in collecting
(Hawkins et al., 1997). Thus, depending on
the nature of ecological question under con-
sideration and the degree of sampling bias in
the technique, a less laborious method such
as traps could provide a viable alternative to
labour-intensive methods such as hand
searching individual quadrats.
Bias in gastropod sampling can be prob-
lematic for numerous reasons. First, terres-
trial gastropod fauna have been collected for
various ecological studies: examining pat-
terns of species distributions and biogeo-
graphical relationships (e.g., Burch, 1956;
Roth & Lindberg, 1 981 ; Van Es & Boag, 1 981 ;
Cameron, 1986; Gascoigne, 1994; Tatters-
field, 1996), evaluating biodiversity and com-
munity changes (e.g., Nilsson et al., 1988;
Niemelä, 1997), and studying population dy-
namics (e.g., Williamson et al., 1977; Uminski
& Focht, 1979). Bias in sampling methods
may mean that ecological inferences drawn
from collected gastropods are incorrect. Like-
wise, terrestrial gastropods are commonly
sampled to estimate the prevalence of para-
sites for whom they act as intermediate hosts.
These collections are typically used to infer
the relative importance of different species in
transmission (e.g., Lankester & Anderson,
1968; Kearney & Gilbert, 1978) and to predict
the risk of infection to vertebrate hosts (e.g.,
Boag, 1985). Nonrepresentative samples of
gastropods can lead to inaccurate assess-
ments of parasite prevalence and misconcep-
tions about transmission dynamics. If surveys
are to provide useful ecological information, a
rigorous approach to sampling is required,
and evaluations of existing procedures need
to be performed to improve sampling protocol.
In this study, I used a stratified, paired-plot
design to directly compare trap and hand-
search collection techniques for sampling ter-
restrial gastropod communities in three differ-
ent habitat types to determine whether traps
provide representative samples of gastropod
species composition, richness and diversity. In
order to partially overcome the size bias inher-
ent in hand-searched collections and to pro-
vide a more representative collection for com-
parison, I combined hand searches in the field
with litter-plus-soil subsamples hand sorted in
the laboratory. Further, to evaluate the effec-
tiveness of trap sampling for different types of
ecological studies, I compared estimates of
species richness calculated using both direct
counts and the jackknife estimator of Burnham
& Overton (1979). The jackknife estimator of
species richness specifically takes into ac-
count heterogeneity in species detectability,
that is, differences among species in their
probability of being sampled given that they
are present at a given sampling location.
MATERIALS AND METHODS
Field Methods
The study took place in Algonquin Park,
Ontario, Canada (45^'35'N, 78°30'W), in June
and July 1 996. As this work was carried out in
connection with a broader study of the trans-
mission of a nematode parasite {Pare-
laphostrongylus tenuis) of white-tailed deer
{Odocoileus virginianus), three habitat types
commonly used by white-tailed deer were
chosen to compare collection methods: white
birch {Betula papyrifera), mixed hardwood
and selectively logged sites (Kohn & Mooty,
1 971 ; Kearney & Gilbert, 1 978). Five sites for
each habitat type were randomly selected
SAMPLING TERRESTRIAL GASTROPOD COMMUNITIES
273
from a list of suitable sites. Suitable sites were
determined by stand content (greater than
60% white birch or hardwood content), acces-
sibility and, for the logged sites, time since
disturbance (logged between 1990 and
1994). The order in which the 15 sites were
examined was determined by lottery.
At each site, two transects, each with five
plot markers, were set up. The plot markers
were spaced 10 m apart and there was 20 m
between the two transect lines. On either side
of the transects, 10 m from each plot marker,
a 1 X 1-m piece of moist cardboard was set
out. Therefore, at each site, a total area of 40
X 40 m was examined using ten plots and ten
paired traps. As traps are typically left out for
a period of time before snails are collected
from them (e.g.. Gleich & Gilbert, 1976; Kear-
ney & Gilbert, 1978), the boards were set out
seven days before the site was scheduled to
be searched. Pieces of woody debris were
placed on the top of the boards to keep them
in place and to help maintain the moisture be-
neath.
All sites were examined between 0600h
and 0900h. At each plot marker, a 1 x 1-m
area was outlined using a wooden frame such
that the marker was in the center of the plot.
Two random subsamples were taken from
within each plot using a corer of 15 cm diam-
eter. These subsamples included leaf litter,
soil to at least a 5 cm depth, and any plants
and debris from within the subsample area.
Each subsample was later handsorted in the
laboratory. Once subsamples were removed,
the remainder of the plot was searched for 1
min for gastropods; all vegetation and possi-
ble shelters were examined and leaf litter and
soil were sifted by hand. Although it is clear
that not all gastropod individuals were found
using this method, the 10-min period was suf-
ficient to thoroughly examine the entire plot
once. After a plot was searched, its corre-
sponding cardboard trap was examined for
gastropods and all specimens on and under
the board, and on the debris on top of the
board were collected.
The subsamples removed from each plot
were examined the same day they were col-
lected. To examine for gastropods, a subsam-
ple was emptied into a large, light-coloured
sorting pan. During a 10-min period, each leaf
or piece of debris was individually examined,
soil was spread and sifted by hand and all
gastropods found were removed. Because
the volume of each subsample was relatively
small (1.0-1.5 L), the 10-min period allowed
an exhaustive examination of all material.
Nevertheless, to ensure greater representa-
tion, particularly of microgastropods, the two
subsamples from a single plot were sorted by
different observers.
All collected gastropods were counted and
identified in the lab. Species identifications
were made using keys by Pilsbry (1946) and
Burch (1962). Identifications were aided by a
zoogeographical study by Oughton (1948)
and were confirmed using voucher specimens
at the Royal Ontario Museum, Canada.
Statistical Methods
Richness for each collection method was
calculated two ways: (1) as a direct count of
the number of species found at a site and, (2)
because it was unlikely that all species pres-
ent at a site were sampled, as estimated by
the jackknife estimator of Burnham & Overton
(1979) using the program CAPTURE (species
richness values can also be calculated using
subprogram SPECRICH from software COM-
DYN available on the Internet -http ;//www.
mbr-pwrc.usgs.gov/comdyn.html).
The jackknife estimator for closed popula-
tions calculates species richness using the
pattern of observed species occurrences
(presence/absence) across replicated sam-
ples within a site (Nichols & Conroy, 1996)
and is based on a capture-recapture model
(model M^) that takes into account hetero-
geneity in the probability of capture among
species. The program CAPTURE has a built-
in selection procedure that chooses the ap-
propriate capture-recapture model for the pre-
sented data. There are several different
models possible, each with different assump-
tions regarding the source of variation in de-
tection probabilities (Boulinier et al., 1998,
provides an overview). Nonetheless, here I
used only the richness estimates provided by
model M^ as this model has been found to be
relatively robust to other factors affecting de-
tection probabilities (e.g., observer influence,
quadrat heterogeneity) (Otis et al., 1978) and
as heterogeneity in species detection is likely
to be the main source of variation in detection
probability when estimating species richness
(Boulinier et al., 1998). Detectability refers to
the probability of sampling at least one indi-
vidual of a given species in a plot (the sam-
pling unit) given that the species is present in
that plot (Boulinier et al., 1998).
It is well recognized that terrestrial gas-
tropods tend to be heterogeneously distrib-
274
McCOY
uted in space (e.g., Cameron, 1986; Ember-
ton et al., 1996). Thus, for estimating species
richness, the assumption of closure (i.e., that
the community being sampled does not
change among replicated sample locations)
(Burnham & Overton, 1979) is likely to be vio-
lated at some spatial scale. For this study,
sampling methods equally covered a small
(40 X 40 m), uniform area of forest. As terres-
trial gastropods should be relatively mobile at
this scale, I felt it was safe to assume that, at
each plot or trap within a site, I sampled from
the same gastropod community.
The program CAPTURE also provides an
estimate of the average instantaneous de-
tectability of species for a given site (or the av-
erage probability of sampling species on a
plot) based on the number of species cap-
tured on each plot within the site. From the
values of estimated species richness, the
overall detectability of species (observed #
species/estimated # species) on a site can
also be calculated. Thus, the average instan-
taneous detectability gives an indication of the
probability of capturing all species at a single
sample location within a site (average proba-
bility per plot), and the overall detectability de-
scribes the probability of capturing all species
across all sampling locations within a site
(probability across ten plots). The jackknife
estimator can be biased if there is a low prob-
ability of detecting species but regardless is
considered to provide a better estimate of
richness than counts alone (Nichols & Con-
roy, 1996).
The diversity at each site was calculated
separately for each collection method using
DIVERS (Krebs, 1989). The Shannon-Wiener
diversity index was considered the most ap-
propriate diversity index to use for gastropod
communities, as it makes no assumptions
about the shape of the underlying distribution
of species-abundance and is relatively insen-
sitive to changes in the dominant species
(Krebs, 1989). This index will likely be biased
to different degrees depending on the species
present in different sites, but it is still appropri-
ate here as I compare the relative diversity of
the two collection methods for the same loca-
tions where the same species are available to
be sampled. As microgastropods from hand-
searched subsamples in the laboratory were
directly included in the list of species for a
given plot without extrapolating the numbers
to the entire area of the plot, I considered tests
for differences in diversity between methods
to be conservative.
If the two collection methods sample differ-
ent species at a given location, we might not
find differences between methods using such
measures as richness or diversity. Thus, I cal-
culated the average similarity in species com-
position of methods on a site using the
% similarity (Renkonen) coefficient for each
plot pair (a hand-searched plot and its cor-
responding cardboard trap). This coefficient
uses the abundance of each species on a plot
and can be calculated using the program SIM-
ILAR (Krebs, 1 989). It can be biased if sample
sizes are small and diversity is high (Wolda,
1 981 ), but should still perform well in this case
as gastropod diversity is relatively low in the
study area considered and sample sizes are
equal. Further, because the effectiveness of
the two sampling methods could change with
varying levels of diversity/richness at different
sampling sites, I regressed the similarity in
species composition of the two methods at a
site against diversity and estimated species
richness. As there were two estimates of di-
versity/richness for a given site (one for each
method), the diversity/richness estimate used
for regressions was the higher of the two esti-
mates.
Calculated estimates for the two collection
methods were statistically compared using
analysis of variance and paired t-tests (Zar,
1984). All tests were performed using SAS
(SAS Institute, 1996) and were considered
significant at the 0.05 level.
RESULTS
Across all three habitat types a total of 18
gastropod species were sampled; 17 and 16
species were found using hand-search and
trap methods, respectively. The average num-
ber of individuals of each species collected
using each method is summarized in Table 1.
One slug species, Arlon circumscriptus, was
consistently collected in greater numbers by
the trap method in all habitats. Six other spe-
cies {Anguispira alternata. Discus cronkhitei.
Zonitoides arboreus. Deroceras laeve. Meso-
don sayanus and Stenotrema fraternum) also
tended to be found in higher numbers in trap
collections, although the difference between
methods changed depending on habitat type;
all other species had similar or greater repre-
sentation using hand searches (Table 1). Two
unique species (Paravitrea multidentata and
Columella edentula) were found only by hand
searching; individuals of С edeníu/a were only
SAMPLING TERRESTRIAL GASTROPOD COMMUNITIES
275
TABLE 1. Number of individuals of each gastropod species collected in white birch, hardwood and logged
sites using the hand-search collection method (Hand) and the trap-collection method (Trap).
Family
White Birch
Hardwood
Logged
Total
Species
Hand
Trap
Hand
Trap
Hand
Trap
Hand
Trap
Arionidae
Arion circumscriptus (Johnston)
Endodontidae
222
349
104
307
57
261
383
917
Discus cronkhitei (Newcomb)
Anguispira altérnala (Say)
Helicodiscus parallelus (Say)
Punctum minutissimum (Lea)
Limacidae
82
42
38
88
3
16
9
20
78
1
1
27
1
26
45
97
26
1
125
1
77
103
263
29
1
2
Dereceras laeve (Müller)
8
12
3
13
2
10
13
35
Philomycidae
Palifera dorsalis (Binney)
Polygyridae
Mesodon sayanus (Pilsbry)
Stenotrema fraternum (Say)
Triodopsis dentifera (Binney)
Pupillidae
9
1
3
2
1
11
1
11
4
6
17
7
3
1
28
11
3
37
9
3
15
34
18
3
Columella edentula (Draparnaud)
Strobiiopsidae
Strobilops labyrinthica (Say)
Succineidae
4
77
5
14
1
2
1
4
93
7
Succinea avails (Say)
Vallonidae
6
5
11
18
9
10
26
33
Zoogenetes harpa (Say)
Zonitidae
10
2
1
11
2
Retlnella binneyana (Morse)
Euconulus fulvus (Müller)
Paravltrea multidentata (Binney)
Zonitoides arboreus (Say)
22
9
6
4
2
9
3
15
12
3
5
24
13
18
3
39
1
1
160
38
42
9
55
4
8
193
TOTAL
540
481
220
472
269
611
1,029
1,564
detected in white birch stands. On logged
sites, one unique species (Triodopsis denti-
fera) was collected by traps and was not found
during hand searches (Table 1 ).
Overall, for both richness and diversity,
hand-searched samples had significantly
greater estimates than did samples from traps
(Fi^4 = 4.75, P - 0.047; F^^^ = 7.04, P =
0.019; Tables 2,3). Richness on a site varied
from 4 to 13 species and diversity from 0.28
to 2.38 for hand searches. For trap collec-
tions, from 1 to 12 species were found on a
site and diversity levels varied from to
1.73. By habitat type, white birch and logged
stands showed greater differences between
methods in both components than hardwood
stands, which showed very little (Table 2).
Overall, there was no difference among habi-
tat types in either richness or diversity, but
there was significant variation in both mea-
sures among different sites within the same
habitat (Table 3).
Unlike for observed richness, there was no
significant difference in species richness of
the two collections as estimated by the jack-
knife (F^^4 - 0.27, P = 0.61; Tables 2,3).
There was also no difference in average in-
stantaneous detectability between the two
methods (hand: 0.28 ± 0.03, trap: 0.23 ±
0.03; t = 1 .1 1 , P - 0.28). Over all habitat types
and methods, estimates of average instanta-
neous detectability were relatively low (be-
tween 16 and 34%), suggesting only a small
proportion of the community is sampled using
one plot or trap on a site. However, the over-
all detectability at a site was higher, ranging
between 10% and 100%, with averages of
80% and 70% for hand and trap methods re-
spectively (Table 2). As there was a significant
interaction in overall detectability between
habitat and collection technique (Table 3), the
effect of technique for this variable was exam-
ined for each habitat type independently. The
overall detectability of species was higher for
hand searches than for traps for two of the
three habitats (Table 2); however, this differ-
276
McCOY
TABLE 2. Summary of community estimates for two gastropod collection methods across three habitat
types. Hand refers to the average estimate value (± standard error) based on collections sampled by
quadrat based plot searches and trap refers to the average estimate value {± standard error) based on
collections sampled using cardboard traps.
White birch
Hardwood
Logged
All sites
Richness
Hand
8.8 ± 1.83
8.0 ± 1.05
10 ± 0.63
8.9 ± 0.71
Trap
6.4 ± 1.03
7.8 ± 1.07
7.4 ± 1.89
7.2 ± 0.76
Diversity
Hand
1.32 ± 0.40
0.99 ± 0.15
1.29 ± 0.20
1.20 ±0.15
Trap
0.60 ± 0.14
0.86 ± 0.17
0.95 ± 0.30
0.80 ± 0.12
Est. Richness
Hand
9.2 = 1.74
15.4 ± 4.45
11 ± 0.84
11.87 ± 1.65
Trap
8.4 = 1.69
8.4 ± 1.03
15 ± 5.55
10.6 ± 2.0
Overall Detectability
Hand
0.86 ± 0.12
0.67 ± 0.34
0.87 X 0.09
0.80 ± 0.22
Trap
0.76 ± 0.18
0.88 ± 0.10
0.47 ± 0.32
0.70 ± 0.27
TABLE 3. Summary of ANOVAs for gastropod community estimates from trap and hand
search methods across three habitat types. Sources were considered significant at
P<0.05.
Source
df
Type III
F value
P value
Richness
Habitat
2
6.47
0.68
0.522
Site(habitat)
12
154.40
2.71
0.039
Technique
1
22.53
4.75
0.047
Diversity
Habitat
2
0.22
0.65
0.537
Site(habitat)
12
5.28
2.57
0.048
Technique
1
1.20
7.04
0.019
Est. Richness
Habitat
2
94.87
1.07
0.369
Site(habitat)
12
697.00
58.08
0.310
Technique
1
12.03
0.27
0.610
Overall Detectability
Habitat
2
0.10
1.44
0.275
Site(habitat)
12
0.69
1.60
0.215
Technique
1
0.071
1.96
0.187
Habitat*technique
2
0.46
6.34
0.013
ence was significant only on logged sites (t =
3.621, P = 0.022).
Sinnilarity in species composition of the two
collections varied from to 86% across sites.
White birch and hardwood sites showed re-
lated levels of similarity [53.30% (± 21.28)
and 52.56% (± 17.75) respectively], but aver-
age similarity was much lower on logged sites
[25.37% (± 25.00)]. Regressions of diversity
and similarity, both across all habitat types
and within each habitat type, were not signifi-
cant at the 0.05 level. However, data from
white birch and hardwood showed similar
trends. When I combined the data from only
these two habitat types, I found that as site di-
versity increased, the two methods had more
dissimilar compositions (Fig. 1 ; % similarity =
(-18.93)diversity + 75.65, F. „ = 6.52, P =
0.034, R^ = 0.449). The regressions of esti-
mated richness and similarity showed similar
trends, but for this measure the regression
using only white birch and hardwood sites
was not significant (F^ g = 2.80, P = 0.133).
For all regressions performed, however, the
power of significance tests were low (between
6% and 30%), indicating a high potential to
commit Type II errors. Further, when the com-
position of collections were highly dissimilar at
a site, hand searches always had higher di-
versity than did trap collections.
DISCUSSION
If we assume that the hand-search collec-
tion method provides a relatively accurate
SAMPLING TERRESTRIAL GASTROPOD COMMUNITIES
277
FIG. 1. Regression of diversity of terrestrial gas-
tropods versus percent similarity in species compo-
sition of hand search and trap collections. Regres-
sion line is based on data from white birch (solid
circles) and mixed hardwood (solid squares) sites.
representation of the gastropod community,
then the trap collection method using card-
board appears to be generally biased in terms
of both observed species richness and diver-
sity. There were evident differences between
methods in observed richness and diversity
for white birch and logged sites, but very little
for hardwood stands. Further, there was a
much lower similarity in species composition
between the two collections in logged sites
compared to the other two habitats. Regres-
sions of diversity and similarity in collection
methods, while only significant when data
from logged sites were removed, implied that
the methods yield relatively similar lists of
species when community diversity is low and
become increasingly dissimilar as diversity in-
creases. Because of high variation in esti-
mates among sites, even within the same
habitat, there was a high probability of com-
mitting a Type II error in all regressions per-
formed; further replicates would be required
to examine the relationship between similarity
in collections and diversity more completely.
Nonetheless, traps seemed to perform best,
in terms of consistency, in hardwood stands
and worst in logged sites.
The observed differences among habitats
in the degree of bias in trap collections could
be related to either direct differences in the
species present in the three habitats or, more
indirectly, could be associated with differ-
ences in habitat spatial complexity. For exam-
ple, there tends to be less understory vegeta-
tion and debris in hardwood stands compared
to white birch or logged sites (pers. obs.).
Where there are fewer potential refuge sites,
gastropods may be more readily attracted to
traps. Conversely, where there are erratic
areas of potential refuge (as in recently
logged sites), the ability to collect using traps
will depend on the placement of the trap and
the performance will be less predictable.
The distance across which gastropods
might be attracted to cardboard traps is
largely unknown. In a study of dispersal of
Agriolimax reticulatus (= Dereceras reticula-
tum) on a fallow field. South (1 965) found that
slugs travelled a mean distance of 1.13 m in
seven days. However, Umax maximus has
been known to move directly towards a food
source from up to 7.5 m away (Cook, 1979).
The slugs Arion circumscriptus and Dero-
ceras laeve. and five snail species (family
Polygyridae: Angulspria alternata. Mesodon
sayanus. Stenotrema fraternum: family En-
dodontidae: Discus cronkhitei: family Zonti-
dae: Zontitoides arboreus), tended to be col-
lected in higher numbers from cardboard
traps. All of these species were, with the ex-
ception of A. alternata, consistently found dur-
ing hand searches and most are relatively
large, obvious gastropods that can be easily
spotted during searches. Such aspects sug-
gest that these species are over-represented
in trap collections. Because most of these
species are robust and all are active in the leaf
litter (McCoy & Nudds, 1997), they may "find"
cardboard traps from greater distances and
aggregate there. Other species, such as Stro-
bilops labyrinthica (family Strobilopsidae),
and some members of families Endodontidae
{Helicodiscus paral lelus. Punctum minutlssi-
mum) and Zonitidae (Paravitrea multidentata.
Euconulus fulvus) appear to be underrepre-
sented in cardboard trap collections. All of
these species are relatively small (< 3.5 mm),
and while some (such as H. parallelus) are
more subterranean in nature, at least one has
been found on tree trunks (pers. obs.) and is,
therefore, relatively mobile.
It is not clear why some species would be
attracted to cardboard traps and others would
not. It was previously shown that different
species show different affinities for traps. For
example, Boag (1990) demonstrated that Dis-
cus cronkhitei used artificial masonite shelters
consistently less often than did Euconulus ful-
278
McCOY
vus. I found the opposite: D. cronkhitei was
sampled relatively more frequently than E. ful-
vus by cardboard traps. This further suggests
that the affinity of different species for traps
can change depending on the trap material
used, in addition, considerable variation has
been found in the frequency of snails adher-
ing to shelters over time and depending on
ambient temperature and moisture conditions
(Boag, 1990; Hawkins et al., 1997). Likewise,
there are substantial differences in activity
and use of different micro-sites by species
found at the same location (Cain & Cowie,
1978: Cameron, 1978). These aspects imply
that the quantitative characteristics of gastro-
pod populations can vary greatly depending
on both the sampling technique and material
used for collection, and on the weather condi-
tions at the time collections are performed.
I estimated species richness using a cap-
ture-recapture model (model M^) that incorpo-
rated variation in detection probabilities
among species in the community. This model
was most frequently selected in bird commu-
nity studies, emphasizing the potential impor-
tance of heterogeneity in species' detection
(Boulinier et al., 1998). While the model se-
lection procedure of CAPTURE did not select
model Mf., more frequently than other capture-
recapture models for the gastropod collection
data, this model did fit more than 80% of the
time for both methods, suggesting that het-
erogeneity in species detection exists among
gastropod species.
Overall species detection at a site (i.e.,
probability of capturing a species at least
once across all sampling locations) tended to
be higher for hand-searched plots compared
to traps, but was typically between 70% and
80%, meaning that most species were seen
by both methods when ten replicates within a
site were used. At the scale of the plot, how-
ever, the average instantaneous detectability
of gastropod species was low. A low probabil-
ity of detecting species on plots means that
estimated richness could be slightly biased.
Nonetheless, previous work suggests that it
should still be closer to the actual number of
species than simple counts alone (Nichols &
Conroy, 1996). For this study, the estimated
richness values found match well with the
number of species recorded for the area. For
example, I observed an average of nine
species per site, but calculated a estimated
average richness of 12 species. As 18 spe-
cies were found overall and 27 species have
been previously reported in the Algonquin re-
gion (Oughton, 1948), I expect that a few
species were never sampled on a site. Thus,
the low probability of detecting species on a
plot underlines the potential problems associ-
ated with using simple counts for determining
species richness, regardless of the sampling
technique used and demonstrates the useful-
ness of species richness estimators. These
types of estimators are now being more fre-
quently employed in community-level studies
(e.g., Derleth et al., 1989; Morrison, 1996;
Boulinier et al., 1998; Nichols et al., 1998a).
The differences I found between the collec-
tion methods can have important implications
for studies that require representative sam-
ples of the gastropod community. For exam-
ple, this study was performed as part of a
larger project examining the transmission of
Parelaphostrongylus tenuis, a common nema-
tode parasite of white-tailed deer {Odocoileus
virginianus) in northeastern North America.
Terrestrial gastropods act as intermediate
hosts for this and other such parasites (Ander-
son & Prestwood, 1 981 ). To assess the poten-
tial risk of transmission of these parasites and
to study the transmission ecology in general,
gastropods are commonly collected using
traps (Kearney & Gilbert, 1 978; Samuel et al.,
1985; Upshall et al., 1986: Robb & Samuel,
1990). If certain species are missed or mis-
represented in samples, calculated estimates
of prevalence may be incorrect. For example,
many of the species I found to be over-repre-
sented in trap collections (e.g., D. cronkhitei,
Z. arboreus anä D. laeve) are typically consid-
ered to be important intermediate hosts of P.
tenuis based on their relative abundance and
prevalence of infection (e.g., Lankester & An-
derson, 1968: Piatt, 1989). Further, the role in
parasite transmission played by many of the
species I found to be commonly underrepre-
sented in trap collections has never been as-
sessed. In addition, the bias in the trap method
appears to be stronger in high diversity sites,
such as white birch stands, and these stands
have been considered to be potential trans-
mission foci (Kearney & Gilbert, 1978). Thus,
due to sampling technique, estimates of
prevalence and our understanding of trans-
mission of parasites such as P. tenuis could be
inaccurate.
Richness, based on simple counts, and di-
versity were significantly different for the two
collection methods, but there was no differ-
ence between methods in estimated richness.
This suggests that for the examination of gas-
tropod community richness alone, collections
SAMPLING TERRESTRIAL GASTROPOD COMMUNITIES
279
from cardboard traps can be as representa-
tive as collections based on hand searches.
Estimates of richness are essential in many
ecological studies, and invertebrates are only
now being considered more in studies of bio-
diversity (e.g., Dobyns, 1997; Niemelä, 1997).
Therefore, while cardboard traps are not reli-
able for physically obtaining a representative
gastropod collection, they may be sufficient
for establishing the pattern of species occur-
rences such that estimates of richness are
similar to what they would be if a more labori-
ous and time consuming method of searching
individual quadrats was performed. Other
studies on invertebrates have also found that
certain species were missed by different col-
lection methods and that estimation tech-
niques were useful for determining overall
richness (Morrison, 1996; Dobyns, 1997).
Nevertheless, estimation techniques are still
limited in that they can not necessarily provide
information about specific species changes
within the community as some species are
never actually detected. However, methods to
make inferences about such community at-
tributes are now being developed (Mines et
al., 1999; Nichols et al., 1998a, b).
Various methods have been developed and
used to sample terrestrial gastropods. How-
ever, as there are problems and biases asso-
ciated with all sampling techniques, the most
appropriate method will depend on the ques-
tion being asked. If the goal is to estimate the
richness of a gastropod community, it is ap-
propriate to use simple cardboard traps to
sample as long as the estimation model used
takes into account potential heterogeneities
inherent in the system. However, if the goal is,
for example, to determine the relative abun-
dance of species over space and time and/or
to estimate the prevalence of a parasite (such
as P. tenuis), traps may not be sufficient; po-
tentially important components of the commu-
nity may be missed or mis-represented in in-
consistent manners. I recommend collecting
gastropods by hand searching quadrats,
using sorted subsamples, whenever this
method can practically be undertaken.
ACKNOWLEDGMENTS
Many thanks to Stephanie Edwards,
Thomas Nudds, Alison Stuart, Lisa Enright,
Heather Hager, Elizabeth Boulding, Ronald
Brooks and Nils Chr. Stenseth for comments
and assistance at various stages of this work.
Thanks also to Thierry Boulinier for running
my data through CAPTURE and for critical
comments. Helpful comments and sugges-
tions for improving this manuscript were also
provided by two anonymous reviewers. The
Ontario Ministry of Natural Resources, Parks
Ontario kindly permitted this work to be car-
ried out in Algonquin Park and the staff at the
Wildlife Research Station provided a wonder-
ful place to work. This work was supported by
a Natural Sciences and Engineering Re-
search Grant to TD. Nudds and by the Envi-
ronmental Youth Corp (EYC).
LITERATURE CITED
ANDERSON, R. С & A. K. PRESTWOOD, 1981,
Lungworms. Tall Timbers l=lesearch Station Ms-
celaneous Publications. 7: 266-317.
BISHOP, M. J., 1977, Approaches to the quantita-
tive description of terrestrial mollusc populations
and habitats. Malacologia. 16: 61-66.
BLINN, W. C, 1963, Ecology of the land snails
Mesodon thyroldus and Mogona profunda. Ecol-
ogy. 44: 498-505.
BOAG, D. A., 1982, Overcoming sampling bias in
studies of terrestrial gastropods. Canadian Jour-
nal of Zoology. 60: 1289-1292.
BOAG, D. A., 1985, Microdistribution of three gen-
era of small terrestrial snails (Stylommatophora:
Pulmonía). Canadian Journal of Zoology. 63:
1089-1095.
BOAG, D. A., 1 990, On the effectiveness of artificial
shelters in the study of population attributes of
small terrestrial gastropods. Canadian Journal of
Zoology 68: 254-262.
BOULINIER, T, J. D. NICHOLS, J. R. SAUER, J. E.
HINES & K. H. POLLOCK, 1998, Estimating
species richness to make inference in community
ecology: the importance of heterogeneity in
species detectability. Ecology 79: 1018-1028.
BURCH, J. В., 1956. Distribution of land snails in
plant associations in eastern Virginia. Nautilus.
70:60-64.
BURCH, J. В., 1 962, ¡How to know tfie eastern land
snails. W. M. Brown Company Publishers,
Dubuque, Iowa. 214 pp.
BURNHAM, K. P. & W. S. OVERTON. 1979, Robust
estimation of population size when capture prob-
abilities vary among animals. Ecology 60:
927-936.
CAIN, A. J. & R. H. COWIE. 1978, Activity of differ-
ent species of land-snail on surfaces of different
inclinations. Journal of Conchology 29: 267-
272.
CAMERON, R. A. D., 1970, The effect of tempera-
ture on the activity of three species of helicid snail
(Mollusca: Gastropoda). Journal of Zoology, Lon-
don. 162:303-315.
CAMERON, R. A. D., 1978, Differences in the sites
280
McCOY
of activity of coexisting species of land mollusc.
Journal of Conchology. 29: 273-278.
CAMERON. R, A. D. 1986, Environment and diver-
sities of forest snail faunas from coastal British
Columbia. Malacologia, 27: 341 -355.
COOK, A., 1979, Homing by the slug Umax pseu-
doflavus. Animal Behaviour, 27: 545-552.
DERLETH, E. L., D. G. McAULEY & T. J. DWYER,
1989, Avian community response to small-scale
habitat disturbance in Maine. Canadian Journal
of Zoology. 67: 385-390.
DOBYNS, J. R. 1997, Effects of sampling intensity
on the collection of spider (Araneae) species and
the estimation of species richness. Environmen-
tal Entomology. 26: 150-162.
EMBERTON, K. C, T. A. PEARCE & R. RAN-
DALANA, 1996, Quantitatively sampling land-
snail species richness in Madagascan rain-
forests. Malacologia. 38: 203-212.
GASCOIGNE, A., 1994, The biogeography of land
snails in the islands of the Gulf of Guinea. Biodi-
versity and Conservation. 3: 794-807.
GLEICH, J. G. & F. F. GILBERT. 1976, A survey of
terrestrial gastropods from central Maine. Cana-
dian Journal of Zoology 54: 620-627.
HAWKINS, J. W., M. W. LANKESTER & R. R. A.
NELSON, 1997, Sampling terrestrial gastropods
using cardboard sheets. Malacologia. 39: 1 -9.
HINES, J. E., T. BOULINIER, J. D. NICHOLS, J. R.
SAUER & К. H. POLLOCK, 1999, COMDYN:
software to study the dynamics of animal com-
munities using a capture-recapture approach.
Bird Study {\r\ press).
KEARNEY, S. R. & F. F. GILBERT, 1978, Terrestrial
gastropods from the Himsworth Game Preserve,
Ontario, and their significance in Parelapfio-
strongylus tenuis transmission. Canadian Jour-
nal of Zoology 56: 688-694.
KOHN, B. E. & J. J. MOOTY 1 971 , Summer habitat
of white-tailed deer in north-central Minnesota.
Journal of Wildlife Management. 35: 476-487.
KREBS, С. J., 1989, Ecological metfiodology
Harper and Row, Publishers, New York.
LANKESTER, M. W. & R. С ANDERSON, 1968,
Gastropods as intermediate hosts of Pneu-
mostrongylus tenuis Dougherty of white-tailed
deer. Canadian Journal of Zoology 46: 373-383.
McCOY K. D. & T. D. NUDDS, 1997, Interspecific
variation in climbing by gastropods: Implications
for the transmission of Parelaptiostrongylus
tenuis. American Midland Naturalist, 137: 320-
328.
MORRISON, L. W., 1996, The ants (Hymenoptera:
Formicidae) of Polynesia revisited: species num-
bers and the importance of sampling intensity.
Ecograpfiy 19: 73-84.
NEWELL, P. F., 1971, Molluscs, pp. 128-149, In: j.
PHiLLiPSON, ed.. Methods of study in quantitative
soil ecology: population, production and energy
flow. Blackwell Scientific Publications, Oxford.
NICHOLS, J. D., T. BOULINIER, J. E. HINES, K. H.
POLLOCK & J. R. SAUER, 1998a, Inference
methods for spatial variation in species richness
and community composition when not all species
are detected. Conservation Biology. 12: 1390-
1398.
NICHOLS, J. D., T. BOULINIER, J. E. HINES, K. H.
POLLOCK & J. R. SAUER, 1998b, Estimating
rates of local species extinction, colonization and
turnover in animal communities. Ecological Appli-
cations. 8: 1213-1225.
NICHOLS, J. D. & M. J. CONROY, 1 996, Estimation
of species richness. In d. r. Wilson, f. r. cole,
J. D. NICHOLS, R. RUDRAN & M. FOSTER, eds.. Mea-
suring and monitoring biological diversity. Stan-
dard methods for mammals. Smithsonian Institu-
tion Press, Washington, D. С
NIEMELÄ, J. 1997, Invertebrates and boreal forest
management. Conservation Biology. 11: 601-
610.
NILSSON, S. G., J. BENGTSSON & S. AS, 1988,
Habitat diversity or area per se? Species rich-
ness of woody plants, carabid beetles and land
snails on islands. Journal of Animal Ecology, 57:
685-704.
OTIS, D. L., K. R BURNHAM, G. С WHITE & D. R.
ANDERSON, 1978, Statistical inference from
capture data on closed animal populations.
Wildlife Monographs. 62.
OUGHTON, J., 1948, A zoogeographical study of
the land snails of Ontario. University of Toronto
Press, Toronto.
PILSBRY H. A., 1946, Land Mollusca of North
America. Vol. I, Part 1. Academy of Natural Sci-
ences of Philadelphia Monographs, 3.
PLATT, T. R. 1989, Gastropod intermediate hosts of
Parelaphostrongylus tenuis (Nematoda: Metas-
trongyloidea) from northwestern Indiana. Journal
of Parasitology 75: 519-523.
ROBB, L. A. & W. M. SAMUEL, 1 990, Gastropod in-
termediate hosts of lungworms (Nematoda: Pro-
tostrongylidae) on a bighorn sheep winter range:
aspects of transmission. Canadian Journal of Zo-
ology 68: 1976-1982.
ROTH, B. & D. R. LINDBERG, 1981, Terrestrial
mollusks of Attu, Aleutian Islands, Alaska. Arctic.
34: 43-47.
SAMUEL, W. M., T. R. PLATT, & S. M. KNISPEL-
KRAUSE, 1985, Gastropod intermediate hosts
and transmission of Parelaphostrongylus odo-
coilei. a muscle-inhabiting nematode of mule
deer, Odocoileus h. hemionus. in Jasper National
Park, Alberta. Canadian Journal of Zoology. 63:
928-932.
SAS INSTITUTE, INC.. 1996. SAS user's manual.
Cary, USA.
SOUTH, A., 1965, Biology and ecology of Agholi-
max reticulatus (Müll.) and other slugs: spatial
distribution. Journal of Animal Ecology 34: 403-
419.
STRAYER, D. H., D. H. PLETSCHER & S. R HAM-
BURG, 1986, The effects of forest disturbance
on land gastropod communities in northern
New England. Canadian Journal of Zoology 64:
2094-2098.
TATTERSFIELD, R, 1996, Local patterns of land
SAMPLING TERRESTRIAL GASTROPOD COMMUNITIES
281
snail diversity in a Kenyan rain forest. Malacolo-
gia, 38: 161-180.
THOMAS, D. C, 1 944, Field sampling for slugs. An-
nals of Applied Biology, 31 : 1 63-1 64.
UMINSKI, T & U. FOCHT 1979, Population dy-
namics of some land gastropods in a forest habi-
tat in Poland. Malacologia. 18: 181-184.
UPSHALL, S. M., M. D. B. BURT & T G. DIL-
WORTH, 1986, Parelaphostrongylus tenuis in
New Brunswick: the parasite in terrestrial gas-
tropods. Journal of Wildlife Diseases, 22: 582-
585.
VAN ES, J. & D. A. BOAG, 1981, Terrestrial mol-
luscs of central Alberta. Canadian Field Natural-
ist, 95: 75-79.
WILLIAMSON, P., R. A. D. CAMERON & M. A.
CARTER, 1977, Population dynamics of the
landsnail Cepaea nemoralis L.: a six-year study.
Journal of Animal Ecology, 46: 1 81 -1 94.
WOLDA, H., 1981, Similarity indices, sample size
and diversity. Oecologia, 50: 296-302.
ZAR, J. H., 1984, Biostatistical analysis, second
edition. Prentice-Hall, Inc., Englewood Cliffs, N.J.
Revised ms. accepted 29 April 1999
MALACOLOGIA, 1999, 41(1): 283-296
POPULATION STRUCTURE IN A SNAIL SPECIES FROM ISOLATED MALAYSIAN
LIMESTONE HILLS, INFERRED FROM RIBOSOMAL DNA SEQUENCES
M. Schilthuizeп^*^ J. J. Vermeulen^ G. W. H. Davison^ & E. Gittenberger2'^
ABSTRACT
We sequenced the first internal transcribed spacer (ITS-1) of the ribosomal DNA in nine pop-
ulations of the vertiginid Gyliotrachela hungerfordiana. which lives on isolated (and threatened)
limestone hills in the Malaysian peninsula. Current data suggest that the species is an obligate
calcicole. The application of a tentative molecular clock suggests a Quaternary divergence for
the G. hungerfordiana populations. A strong positive correlation between genetic and geographic
distance was observed, which, combined with geological data, suggests that the hill populations
may be interconnected by as yet unsampled populations.
Key words: internal transcribed spacer, ITS-1, Gastropoda, Pulmonata, Vertiginidae,
Gyliotrachela, gene flow, Southeast Asia.
INTRODUCTION
Land snails have proverbially poor abilities
for dispersal (e.g., Cowie, 1984; Schilthuizen
& Lombaerts, 1994), which causes them to
show evolutionary patterns at much smaller
spatial scales than many other organisms of
similar size. As a result, strong geographic
structunng of populations is common in snails
(e.g., in Liguus: Hillis et al., 1987). Another
consequence is endemism, which is seen, for
example, in the Mediterranean clausiliid
genus Albinaria, of which almost 30 species
are endemic to the island of Crete, with distri-
bution areas of sometimes only one kilometer
across (Gittenberger, 1991; Welter-Schultes,
1998).
An impressive situation of high endemism
and geographic structuring of land snails in a
strongly fragmented habitat exists in peninsu-
lar Malaysia. Here, limestone is exposed in
the form of "tower karst" and other karstifica-
tions, limited to about three hundred hills,
scattered over the peninsula. These hills are
often very small, the largest with a diameter of
a few kilometers, but most measuring only a
few hundred meters across. In spite of their
small size, the hills are a prominent feature of
the landscape, because they usually stand
isolated, are riddled with caves and are
bounded by precipitous cliffs.
For more than a century, malacologists
have been interested in the rich malacofauna
that the hills support (de Morgan, 1885). High
numbers of species are found, and the mor-
phologies of some Diplommatinidae fore-
shadow the bizarre and extravagant forms
found in this group in Borneo (Vermeulen,
1993, 1994; Gittenberger, 1995). But espe-
cially fascinating is the staggering degree of
endemism in these calcicolous snails.
Tweedie (1961) gave an overview of six taxa
containing many obligate caldcóles {Diplom-
matina, Opisthostoma. Vertiginidae, Discarte-
mon, Oophana, and Sinoennea). He listed the
presence of 106 species on 28 hills or hill-
clusters, of which 70 are endemic to only one
locality. Some calcicolous species, however,
are widespread and occur on almost all hills
without a trace of morphological differentia-
tion (e.g. Gyliotrachela hungerfordiana and
some Alycaeus species).
Geologically, the hills form the exposed
parts of a number of larger paleozoic lime-
stone deposits, which are elsewhere overlain
by non-calciferous alluvial deposits (Gale,
1986; Crowther, 1986). Some hills may thus
have been connected in the past, while others
have always been separate. Consequently,
the hills form virtual "islands" for obligately
calcicolous land snails, which they may reach
by incidental dispersal. Alternatively, the pop-
^ Department of Genetics. Wageningen Agricultural University. Dreijenlaan 2. 6703 HA Wageningen, Ttie Netherlands;
Menno.Schilthuizen@fungen.el.wau.nl
^National Museum of Natural History "Naturalis". P.O. Box 9517. 2300 RA Leiden, The Netherlands
%WF Malaysia. Locked Bag No. 911. Jalan Sultan P.O.. 46990 Petaling Jaya. Selangor, Malaysia
''institute of Evolutionary and Ecological Sciences, Leiden University, Leiden, The Netherlands
'correspondence
283
284
SCHILTHUIZENETAL.
ulations on the hills may be relicts from a time
when the hills were part of large continuous
plateaus, which were subsequently frag-
mented.
In this paper, we examine a relatively wide-
spread representative of the peninsular
Malaysian hill malacofauna, using molecular
and geological data, to answer the following
questions; (1) what pattern of phylogeo-
graphic relationships exists among the popu-
lations of this widespread species, and (2)
how has the population structure been
shaped, that is, what are the relative influ-
ences of dispersal and habitat fragmentation
over geological time?
By analyzing the variance in a noncoding
nuclear DNA marker, we attempt to differenti-
ate between various alternative population
structures. In the case of ancient vicariance,
we expect to find genetic distances that reflect
the age of fragmentation of the limestone hills,
while dispersal would result in genetic dis-
tances more or less related to geographic dis-
tance. Under the latter hypothesis (dispersal),
indications of the type and frequency of dis-
persal may be gleaned from the degree of
correlation between genetic and geographic
distance; if dispersal is randomly oriented
(i.e., corresponding to an island model of pop-
ulation structure; Wright, 1931), stochasticity
would result in a poor fit, while dispersal
occurring mainly among neighboring hills (i.e.,
corresponding to a stepping-stone model;
Kimura, 1953) would be revealed by a strong
correlation (Kimura & Weiss, 1964).
Sadly, there are other motives for working
on this fauna. The hills of peninsular Malaysia
are disappearing and becoming depauperate
at an alarming rate. Forest clearing has de-
stroyed the vegetation on some hills; and in
the densely populated areas near Ipoh and
Kuantan, many hills are being removed by
quarrying. The true rate of species loss can
only be guessed at, but the extinction of at
least one endemic snail species, Opisthos-
toma sciaphilum, from Bukit Ranching, has
been documented (Schilthuizen et al., un-
publ.).
FIG. 1. Gyliotrachela hungerfordiana (von Möilen-
dorff). Scale bar = 1 mm.
FIG. 2. Gyliotrachela frequens van Benthem
Jutting. Scale bar = 1 mm.
species G. frequens (Fig. 2) was selected to
serve as an outgroup in the phylogenetic
analysis.
MATERIAL AND METHODS
Selection of Taxa
We selected the widespread and morpho-
logically uniform vertiginid Gyliotrachela hun-
gerfordiana for study (Fig. 1). The related
Collecting
In July 1 997, the first author visited 22 lime-
stone hills in the West-Malaysian states of Pa-
hang, Kelantan, Perakand Perils. Living snails
were discovered by eye using two strategies:
(a) close inspection of limestone rock faces,
POPULATION STRUCTURE OF A MALAYSIAN LAND SNAIL
285
either damp or dry, bare or covered in algae,
mosses and lichens; and (b) sifting through
damp and decaying leaf litter on limestone
rocks or at the base of the limestone cliffs. All
snails were put in 100% ethanol on the spot
and kept at ambient temperatures until arrival
in the laboratory for further processing. Identi-
fication of the material was carried out by the
second author while the material remained in
alcohol. Gyliotrachela hungerfordiana was
collected from nine of the 22 localities (Fig. 3):
loc. 5, State of Pahang: Qua Bama (ca. 10 km
W of Kuala Lipis); loc. 8, State of Kelantan;
Gua Musang, southern of the two hills that the
road to Kuala Kerai passes between; loc. 9,
State of Kelantan: rocks 59 km in the direction
of Gua Musang, measured along the road
from Kuala Krai; loc. 16, State of Perak; Bukit
Tambun (ca. 6 km E of Ipoh); loc. 22, State of
Perak; hill directly east of Sungai Siput Ufara
hospital; loc. 23, State of Perlis; hill ca. 1 km S
of Kangar; loc. 24, State of Perlis; 9 km along
the road from Kangar to Kaki Bukit; loc. 25,
State of Perlis; Gua Kelam at Kaki Bukit; loc.
26, State of Perlis, Timah Tasoh (ca. 1 6 km NE
of Kangar). All samples were taken between
27. vi. 1997 and 17.vii.1997. Gyliotrachela fre-
quens was taken only from locality 8. Voucher
specimens have been deposited in the collec-
tion of the National Museum of Natural History
"Naturalis", Leiden.
Molecular Techniques
DNA was isolated from pools of between
one and five complete snails with their shells,
using either a phenol/chloroform extraction as
described previously (Schilthuizen et al.,
1998a) or a sucrose-based protocol (van
Moorsel & van Nes, unpublished), which can
be briefly summarized as follows. Snails were
ground in 200 ц1 of sucrose-buffer (0.1 M Tris;
0.02 M NaCI; 0.2 M sucrose; 0.05 M EDTA)
and centhfuged. The pellet was incubated at
65°C for 60 min in 200 ).il SDS-buffer (0.02 M
Tris; 0.01 M EDTA, 1 .25% SDS), 1 5 |.il of cold
KAc was added, and the mixture was incu-
bated on ice for 60 min and centrifuged. The
DNA was precipitated from the supernatant by
the addition of two volumes of 100% ethanol
and incubation at -20°C for 30 min. The DNA
was dried and treated with 200 ng of RNase.
Full details can be obtained from M.S. on re-
quest. Homogenization was always done with
a sterile, disposable plastic pestle. The DNA
was dissolved in 50 |.il of Tris-EDTA buffer
(phenol protocol) or 30 )il of ddH20 (sucrose
protocol) and stored at -20''C. The first inter-
nal transcribed spacer of the nuclear riboso-
mal DNA was amplified with the SuperTaq
enzyme (HT Biotechnology, Cambridge, Eng-
land) as described previously (Schilthuizen et
al., 1995) and isolated using the "freeze-
squeeze" technique (Tautz & Renz, 1 983). Be-
cause PCR-amplification was at times too
weak for direct sequencing, we resorted to
cloning (PCR-based error is usually not a con-
cern with this methodology; Schilthuizen et al.,
1 998b). After isolation, the fragments were lig-
ated into Promega or Invitrogen T-tailed vec-
tors, following the manufacturer's instructions.
Colonies were screened for the presence of
the correct insert by PCR. Plasmid DNA was
isolated from the bacteria using QIAPrep spin
columns (QIAGEN). One or two clones per
sample were sequenced in both directions on
an ABI automated sequencer.
Alignment
Before alignment, all chromatograms were
checked and reading errors were corrected
blindly where necessary (this never amounted
to more than three corrections in a single
sequence). Vector and primer sequences
were removed. Sequences in the ingroup
were sufficiently similar to allow manual align-
ment. Wherever alignment with the outgroup
was ambiguous, missing data were intro-
duced into the outgroup sequence.
Phylogenetic Analysis
Phylogenetic analyses of the data set were
performed in PAUP3.1 (Swofford, 1993). Gaps
were treated as missing data. Searches for the
most parsimonious trees were carried out with
the branch-and-bound option. Bootstrap repli-
cates were carried out 1 00 times, using heuris-
tic searches. In addition, Bremer (1988) sup-
port was determined. Kimura's 2-parameter
genetic distances (Kimura, 1980) were calcu-
lated with the DNADIST program of the
PHYLIP package (Felsenstein, 1995).
RESULTS
PCR-products ranged in length from 755 to
772 bp, including primers (52 bp), and the
flanking regions of 1 8S (1 46 bp) and 5.8 S (87
bp). These lengths correspond well with other
ITS-1 lengths reported in mollusks (Anderson
286
SCHILTHUIZENETAL.
50 km
Penang
N
Kuan tan
FIG. 3. A map of the northern part of Peninsular Malaysia, with the limestone hills drawn in black (modified
after Gobbett, 1965). The numbers refer to localities where Gyliotrachela hungerfordiana and G. frequens
were collected (see text for further details).
& Adiard, 1994; Schilthuizen et al., 1995;
Armbruster et al., unpubl.).
We obtained sixteen sequences from G.
hungerfordiana and one sequence for the out-
group, G. frequens (Appendix, Table 1). They
have been deposited in GenBank under
accession numbers AF118000-AF118016.
Only small genetic distances were found
among the G. hunderfordiana sequences, the
largest being 0.048 between sequence a from
locality 5 and sequence b from locality 23. A
comparison between pairwise genetic dis-
tances and pairwise geographic distances
between sequences revealed a strongly sig-
nificant (p < 0.005) positive correlation (Fig. 4,
Appendix, Table 2). The phylogenetic analysis
produced 18 most parsimonious trees (length
= 89 steps, Rl = 0.95), which showed two
alternative topologies for three monophyletic
groups of sequences, and otherwise only
minor differences in topology within each of
these three monophyletic groups (Figs. 5, 6).
The fact that duplicate sequences from a sin-
gle locality always formed monophyletic
groups might justify the small sample sizes.
Geographic structuring is apparent in the
trees also, as these show monophyly for the
sequences derived from populations in Perils,
Pahang -i- Kelantan, and Perak.
DISCUSSION
Unfortunately, it is difficult to estimate reli-
ably from the molecular data the time since
divergence. Unlike the situation for mitochon-
POPULATION STRUCTURE OF A MALAYSIAN LAND SNAIL
287
0.05 -[-
0.045 --
0.04 --
0.035 --
0.03 --
0.025 --
0.02 --
■
0.015 --. ■
■ ■
0.01 I"'
0.005 ¿-„
I
■ ■■
Á
100 200 300 400
geographic distance (km)
FIG. 4. The relationship between geographic dis-
tance and Kimura's 2-parameter distance for the
sequences of Gyliotrachela hungerfordiana.
drial DNA, corroborated molecular clocks for
the ITS regions are hardly available yet, and
where they are, they differ by orders of mag-
nitude among taxonomic groups. In the
angiosperm families Cucurbitaceae and
Winteraceae, substitution rates of 3.62 x 10"^
and 3.4 X 10"^ per site per million years (MY)
were calculated, respectively (Jobst et al.,
1998; Suh et al., 1993), while in Chlorophyta,
a rate of 0.8 - 2.0 x 10"^ was estimated
(Bakker et al., 1 995). In animals, rates of sub-
stitution in ITS appear to be somewhat higher.
Schlötterer et al. (1994) give a figure of 1 .2 x
10~^ for Drosophila, and preliminary data for
clausiliid land snails from Greek islands indi-
cate a similar rate (van Moorsel, unpublished
data).
Here, we will adopt a substitution rate of 1 x
10~^ per site per MY as a very rough molecu-
lar clock. Applying this rate to the average
genetic distance between sequences on
either side of the node basal to all G. hunger-
fordiana sequences in the trees, we obtained
an estimated divergence time of 1.8 MYA for
the populations of G. hungerfordiana. It
should be stressed that, given the lack of
agreement in the few calibrated molecular
clocks available, not too much confidence
should be placed on this date. However, it
may be safe to assume a Late Tertiary or
Quaternary origin for G. hungerfordiana.
Given the low degree of genetic divergence
among the G. hungerfordiana populations, it
seems unlikely that vicahance has played an
important role; hills which have been studied
geologically are thought to be older than Late
Tertiary/Quaternary (Gale, 1986). However, in
view of the uncertainty about the calibration of
the ITS-1 molecular clock, this reasoning may
be little meaningful. More importantly, geolog-
ical data indicate that most of the hills from
which the species was sampled have never
been part of one continuous plateau (Paton,
1961). It is for this reason not likely that vic-
ahance events have been important in its dis-
tribution pattern. Rather, the limestone hills on
which it lives now must have been colonized
after dispersal.
Several mechanisms for passive dispersal
in small snails have been suggested, includ-
ing wind and water mediated dispersal. In ref-
erence to Gyliotrachela and similar snails,
Tweedie (1961) has suggested that flooding
may be important in producing dispersal
among hills that are situated close together.
However, the drainage patterns in the penin-
sula preclude any long-range dispersal by this
mechanism. Stagnant water may also provide
means of dispersal, and geological data
(Gale, 1986; Crowther, 1986) indicate that
lacustrine conditions have prevailed around
several limestone hills in the past. But here,
too, dispersal would be across very small dis-
tances. Another possibility is wind-dispersal.
Kirchner et al. (1997) demonstrate how
Truncatellina, a vertiginid very similar in size
to G. hungerfordiana, could be blown over
distances of several kilometers during storms.
Some additional characteristics of dispersal
may be gleaned from Figure 4, which suggests
a linear relationship between geographic and
genetic distance. If dispersal from one hill to
another were infrequent and undirected (i.e., a
population structure corresponding to Wright's
[1 931 ] island model, where all possible pairs of
subpopulations are equally likely to exchange
migrants), such a clear relation would not be
expected. The fact that genetic distance is re-
liably predicted (r^ = 0.77) by geographic dis-
tance, suggests that a structured network of
dispersal connects the hills. This corresponds
to a stepping-stone model (Kimura, 1953).
Under such a model, genetic similarities drop
steeply with increasing numbers of intervening
populations (Kimura & Weiss, 1964). The fact
that we observe a strong relationship with ge-
ographic distance, suggests that the hill popu-
96
hungerfordiana.lod 6#a
32 — hungerfordiana.loc22#a
''^ — hungerfordiana.lod 6#b
hungerfordiana.loc22#b
hungerfordiana.loc23#a
hungerfordiana.loc23#b
dO I
d2
hungerfordiana.loc24#a
hungerfordiana.loc24#b
hungerfordiana.loc25#a
hungerfordiana.loc25#b
hungerfordiana.loc26#a
hungerfordiana.loc5#a
611- hi
M
d2 L
83
— hungerfordiana.loc5#b
hungerfordiana.loc8#a
hungerfordiana.loc8#b
hungerfordiana.loc9#a
frequens
FIG. 5. A representative most parsimonious tree of the Gyliotrachela sequences. Bootstrap percentages and
decay indices have been indicated on the branches.
hungerfordiana.lod 6#a
hungerfordiana.loc22#a
hungerfordiana.lod 6#b
hungerfordiana.loc22#b
hungerfordiana.loc23#a
hungerfordlana.loc24#a
hungerfordiana.loc24#b
hungerfordiana.loc25#a
hungerfordiana.loc25#b
hungerfordiana.loc26#a
hungerfordiana.loc23#b
hungerfordiana.loc5#a
hungerfordiana.loc5#b
hungerfordlana.loc8#a
hungerfordiana.loc8#b
hungerfordiana.loc9#a
frequens
FIG. 6. Strict consensus over all 18 most parsimonious trees.
POPULATION STRUCTURE OF A MALAYSIAN LAND SNAIL
289
lations cannot represent directly adjacent pop-
ulations in a two-dimensional stepping-stone
lattice. Rather, to obtain this result, it is neces-
sary to postulate unsampled populations in
between. Unfortunately, the population genet-
ics of ribosomal DNA are as yet far from clear
(Hillis et al., 1991; Rich et al., 1997), which
makes a quantitative analysis of dispersal pa-
rameters and spatial details of the population
structure impossible. Therefore, it is not possi-
ble to tell whether the hills that separate our
sample sites (e.g., the six or more hills be-
tween sites 8 and 9) will suffice as additional
stepping stones. This might be tested, for in-
stance, by exhaustively sampling the hills in a
given subregion.
ACKNOWLEDGEMENTS
The authors wish to thank Sook-Peng
Phoon, Fenna Schilthuizen, Jan Schilthuizen,
and Kai Foon Phoon for help in the field.
Stephen Gale (University of Sydney, Aus-
tralia), Ian Metcalfe (University of New Eng-
land, Armidale, Australia), and the Geological
Survey (Kuala Lumpur, Malaysia) provided
geological data and references. Tony van
Kampen operated the automated DNA se-
quencer, and the Department of Entomology
(Wageningen Agricultural University) provided
a camera lucida. Coline van Moorsel and Ton
de Winter participated in fruitful discussions
and kindly checked the manuscript. M.S. ac-
knowledges financial support from the Nether-
lands Organisation of Scientific Research.
LITERATURE CITED
ANDERSON, J. J. & R. D. ADLARD, 1994,
Nucleotide sequence of a rDNA internal tran-
scribed spacer supports synonymy of Saccostrea
commercialis and S. glomerata. Journal of
Molluscan Studies. 60: 196-197.
BAKKER, F. T, J. L. OLSEN & W. T STAM, 1995,
Evolution of nuclear rDNA ITS sequences in the
Cladophora albida sericea clade (Chlorophyta).
Journal of l^olecular Evolution. 40: 640-651 .
BREMER, K., 1988, The limits of amino acid
sequence data in angiosperm phylogenetic
reconstruction. Evolution. 42: 795-803.
COWIE, R. H., 1984, Density dispersal, and neigh-
bourhood size in the land snail Theba pisana.
Hered/iy, 52:391-401.
CROWTHER, J., 1986, Chemical erosion in tower
karst terrain, Kinta Valley, Peninsular Malaysia.
Pp. 427-441, in: к. paterson & M. M. sweeting,
eds.. New directions in karst. Geo Books, Nor-
wich, UK.
DE MORGAN, J., 1885, Mollusques terrestres et
fluviátiles du Royaume de Perak et de Pays
voisin (presqu'île Malaise). Bulletin de la Société
Zoologique de France. 10: 353-428.
FELSENSTEIN, J. S., 1995, PHYLIP (Pfiylogeny
Inference Package). Version 3.57c. University of
Washington, Seattle, Washington, U.S.A.
GALE, S. J., 1 986, The hydrological development of
tropical tower karst: an example from Peninsular
Malaysia. Pp. 443-459, in: к. paterson & м. м.
sweeting, eds., Nev\/ directions in karst. Geo
Books, Norwich, UK.
GITTENBERGER, E., 1 991 , What about non-adap-
tive radiation? Biological Journal of the Linnean
Society. 43: 263-272.
GITTENBERGER, E. 1995, On the other hand. . .
Nature. 373: 19.
GOBBETT, D. J. 1965, The formation of limestone
caves in Malaya. Malayan Nature Journal. 19:
4-12.
HILLIS, D. M., D. S. ROSENFELD & M. SANCHEZ,
1987, Allozymic variability and heterozygote
deficiency within and among morphologically
polymorphic populations of Liguus fasciatus
(Mollusca: Pulmonata: Bulimulidae). American
Malacological Bulletin. 5: 153-157.
HILLIS, D. M., С MORITZ, С. A. PORTER & R. J.
BAKER, 1991 , Evidence for biased gene conver-
sion in concerted evolution of ribosomal DNA.
Science. 251:308-310.
JOBST J., K. KING & V. HEMLEBEN, 1998,
Molecular evolution of the internal transcribed
spacers (ITS1 and ITS2) and phylogenetic rela-
tionships among species of the family Cucurbit-
aceae. Molecular Ptiylogenetics and Evolution. 9:
204-219.
KIMURA, M., 1953, "Stepping stone" model of pop-
ulation. Annual Report of ttie National Institute of
Genetics of Japan. 3: 62-63.
KIMURA, M., 1 980, A simple method for estimating
evolutionary rate of base substitution through
comparative studies of nucleotide sequences.
Journal of Molecular Evolution. 16: 1 1 1 - 1 20.
KIMURA, M. & G. H. WEISS, 1964, The stepping-
stone model of population structure and the
decrease of genetic correlation with distance.
Genei/cs, 49:561-576.
KIRCHNER, С, R. KRÄTZNER & F W. WELTER-
SCHULTES, 1997, Flying snails- how far can
Truncatellina (Pulmonata: Vertiginidae) be blown
over the sea? Journal of Molluscan Studies. 63:
479-487.
PATON, J. R., 1 961 , A brief account of the geology
of the limestone hills of Malaya. Bulletin of the
Raffles Museum, 26: 66-75.
RICH, S. M., B. M. ROSENTHAL, S. R. TELFORD,
A. SPIELMAN, D. L. HARTL & F J. AYALA, 1 997,
Heterogeneity of the internal transcribed spacer
(ITS-2) region within individual deer ticks. Insect
Molecular Biology 6: 123-129.
290
SCHILTHUIZEN ETAL.
SCHILTHUIZEN, M., E. GITTENBERGER & A.
GULTYAEV, 1995, Phylogenetic relationships
inferred from the sequence and secondary struc-
ture of ITS1 rRNA in Albinaria and putative
Isabellaria species (Gastropoda, Pulmonata,
Clausiliidae). Molecular Phylogenetics and
Evolution. 4: 457-462.
SCHILTHUIZEN, M., J. HONDA & R. STOUT-
HAMER, 1998b, Parthenogenesis-inducing Wol-
bachla in Trichogramma kaykal (Hymenoptera:
Trichogrammatidae) originates from a single in-
fection. Annals of the Entomological Society of
America. 91:410-414.
SCHILTHUIZEN, M. & M. LOMBAERTS, 1994,
Population structure and levels of gene flow in
the Mediterranean land snail Albinaria corrúgala
(Pulmonata: Clausiliidae). Evolution. 48: 577-
586.
SCHILTHUIZEN, M., G. NORDLANDER, R.
STOUTHAMER & J. J. M. VAN ALPHEN, 1998a,
Morphological and molecular phylogenetics in the
genus Leptopilina (Hymenoptera: Cynipoidea:
Eucoilidae). Systematic Entomology, 23: 253-
264.
SCHLÖTTERER, C, M. T HÄUSER, A. VON HAE-
SELER & D. TAUTZ, 1994, Comparative evolu-
tionary analysis of rDNA ITS regions in Dro-
sophila. Molecular Biology and Evolution. 1 1 :
513-522.
SUH, Y., L. B. THIEN, H. E. REEVE & E. A. ZIM-
MER, 1993, Molecular evolution and phyloge-
netic implications of internal transcribed spacer
sequences of ribosomal DNA in Winteraceae.
American Journal of Botany 80: 1 042-1 055.
SWOFFORD, D. L., 1993, PAUP 3.1. Sinauer
Associates, Sunderland.
TAUTZ, D. & M. RENZ, 1983, An optimized freeze-
squeeze method for the recovery of DNA frag-
ments from agarose gels. Analytical Biochem-
istry 132: 14-19.
TWEEDIE, M. 1961, On certain Mollusca of the
Malayan limestone hills. Bulletin of the Raffles
Museum. 26: 49-65.
VERMEULEN, J. J., 1993, Notes on the non-marine
molluscs of the island of Borneo 5. The genus
Diplommatina (Gastropoda Prosobranchia:
Diplommatinidae). Basteria. 57: 3-69.
VERMEULEN, J. J., 1 994, Notes on the non-marine
molluscs of the island of Borneo 6. The genus
Opisthostoma (Gastropoda Prosobranchia:
Diplomatinidae). Basteha. 58: 75-191.
WELTER-SCHULTES, F. W., 1998, Albinaria in
central and eastern Crete: distribution map of the
species (Pulmonata: Clausiliidae). Journal of
Molluscan Studies. 64: 275-279.
WRIGHT, S., 1931, Evolution in Mendelian popula-
tions. Genetics. 16: 97-159.
Revised ms. accepted 29 April 1999
POPULATION STRUCTURE OF A MALAYSIAN LAND SNAIL
APPENDIX
291
TABLE 1 . Aligned sequences for Gyliotrachela hungerfordiana and G. frequens. The 5' end of the 1 8S region
is at position 146, the 3' end of the 5.8S region is at position 694.
10
1
20
1
30
1
40
1
50
1
iiungerfordiana
.loc5#a
AGCGGTTCAG
TGAGGGCCTC
GGATTGGTCT
CGGTCTGGTG
CGCAAGTGCC
nungerfordiana
.loc5#b
AGCGGTTCAG
TGAGGGCCTC
GGATTGGTCT
CGGTCTGGTG
CGCAAGTGCC
lunger fordiana
. loc8#a
AGCGGTTCAG
TGAGGGCCTC
GGATTGGTCT
CGGTCTGGTG
CGCAAGTGCC
lunger fordiana
.loc8#b
AGCGGTTCAG
TGAGGGCCTC
GGATTGGTCT
CGGTCTGGTG
CGCAAGTGCC
lunger fordiana
.loc9#a
AGCGGTTCAG
TGAGGGCCTC
GGATTGGTCT
CGGTCTGGTG
CGCAAGTGCC
lunger fordiana
locl6#a
AGCGGTTCAG
TGAGGGCCTC
GGATTGGTCT
CGGTCTGGTG
CGCAAGTGCC
lunger fordiana
locl6#b
AGCGGTTCAG
TGAGGGCCTC
GGATTGGTCT
CGGTCTGGTG
CGCAAGTGCC
lunger fordiana
loc22#a
AGCGGTTCAG
TGAGGGCCTC
GGATTGGTCT
CGGTCTGGTG
CGCAAGTGCC
lunger fordiana
loc22#b
AGCGGTTCAG
TGAGGGCCTC
GGATTGGTCT
CGGTCTGGTG
CGCAAGTGCC
lunger fordiana
loc23#a
AGCGGTTCAG
TGAGGGCCTC
GGATTGGTCT
CGGTCTGGTG
CGCAAGTGCC
rtunger fordiana
loc23#b
AGCGGTTCAG
TGAGGGCCTC
GGATTGGTCT
CGGTCTGGTG
CGCAAGTGCC
lunger fordiana
loc24#a
AGCGGCTCAG
TGAGGGCCTC
GGATTGGTCT
CGGTCTGGTG
CGCAAGTGCC
lunger fordiana
loc24#b
AGCGGTTCAG
TGAGGGCCTC
GGATTGGTCT
CGGTCTGGTG
CGCAAGTGCC
lunger fordiana
loc25#a
AGCGGTTCAG
TGAGGGCCTC
GGATTGGTCT
CGGTCTGGTG
CGCAAGTGCC
lunger fordiana
loc25#b
AGCGGTTCAG
TGAGGGCCTC
GGATTGGTCT
CGGTCTGGTG
CGCAAGTGCC
lunger fordiana
loc26#a
AGCGGTTCAG
AGCGGTTCAG
60
1
TGAGGGCCTC
TGAGGGCCTC
70
1
GGATTGGTCT
GGATTGGTCT
80
1
CGGTCTGGTG
CGGTCTGGTG
90
1
CGCAAGTGCC
CGCAAGTGCC
100
1
frequens
lunger fordiana
loc5#a
GGCACCGCTG
GCCGAGAAGA
AGCTCGAACT
CGATCGCTTG
GAGAAAGTAA
lunger fordiana
loc5#b
GGCACCGCTG
GCCGAGAAGA
AGCTCGAACT
CGATCGCTTG
GAGAAAGTAA
lunger fordiana
loc8#a
GGCACCGCTG
GCCGAGAAGA
AGCTCGAACT
CGATCGCTTG
GAGAAAGTAA
lunger fordiana
loc8#b
GGCACCGCTG
GCCGAGAAGA
AGCTCGAACT
CGATCGCTTG
GAGAAAGTAA
lunger fordiana
loc9#a
GGCACCGCTG
GCCGAGAAGA
AGCTCGAACT
CGATCGCTTG
GAGAAAGTAA
lunger fordiana
locl6#a
GGCGCCGCTG
GCCGAGAAGA
AGCTCGAACT
CGATCGCTTG
GAGAAAGTAA
lunger fordiana
locl6#b
GGCACCGCTG
GCCGAGAAGA
AGCTCGAACT
CGATCGCTTG
GAGAAAGTAA
lunger fordiana
loc22#a
GGCACCGCTG
GCCGAGAAGA
AGCTCGAACT
CGATCGCTTG
GAGAAAGTAA
lunger fordiana
loc22#b
GGCATCGCTG
GCCGAGAAGA
AGCTCGAACT
CGATCGCTTG
GAGAAAGTAA
lunger fordiana
loc23#a
GGCACCGCTG
GCCGAGAAGA
AGCTCGAACT
CGATCGCTTG
GAGAAAGTAA
lunger fordiana
loc23#b
GGCACCGCTG
GCCGAGAAGA
AGCTCGAACT
CGATCGCTTG
GAGAAAGTAA
lunger fordiana
loc24#a
GGCACCGCTG
GCCGAGAAGA
AGCTCGAACT
CGATCGCTTG
GAGAAAGTAA
lunger fordiana
loc24#b
GGCACCGCTG
GCCGAGAAGA
AGCTCGAACT
CGATCGCTTG
GAGAAAGTAA
lunger fordiana
loc25#a
GGCACCGCTG
GCCGAGAAGA
AGCTCGAACT
CGATCGCTTG
GAGAAAGTAA
lunger fordiana
loc25#b
GGCACCGCTG
GCCGAGAAGA
AGCTCGAACT
CGATCGCTTG
GAGAAAGTAA
lunger fordiana
loç26#a
GGCACCGCTG
GCCGAGAAGA
AGCTCGAACT
CGATCGCTTG
GAGAAAGTAA
frequens
GGCACCGCTG
110
1
GCCGAGAAGA
120
1
AGCCCGAACT
130
1
CGATCGCTTG
140
1
GAGAAAGTAA
150
1
lunger fordiana
loc5#a
AAGTCGTAAC
AAGGTTTCCG
TAGGTGAACC
TGCGGAAGGA
TCATTAACGG
lunger fordiana
loc5#b
AAGTCGTAAC
AAGGTTTCCG
TAGGTGAACC
TGCGGAAGGA
TCATTAACGG
lunger fordiana
loc8#a
AAGTCGTAAC
AAGGTTTCCG
TAGGTGAACC
TGCGGAAGGA
TCATTAACGG
lunger fordiana
loc8#b
AAGTCGTAAC
AAGGTTTCCG
TAGGTGAACC
TGCGGAAGGA
TCATTAACGG
lunger fordiana
loc9#a
AAGTCGTAAC
AAGGTTTCCG
TAGGTGAACC
TGCGGAAGGA
TCATTAACGG
lunger fordiana
locl6#a
AAGTCGTAAC
AAGGTTTCCG
TAGGTGAACC
TGCGGAAGGA
TCATTATCGG
lunger fordiana
locl6#b
AAGTCGTAAC
AAGGTTTCCG
TAGGTGAACC
TGCGGAAGGA
TCATTATCGG
lunger fordiana
loc22#a
AAGTCGTAAC
AAGGTTTCCG
TAGGTGAACC
TGCGGAAGGA
TCATTATCGG
lunger fordiana
loc22#b
AAGTCGTAAC
AAGGTTTCCG
TAGGTGAACC
TGCGGAAGGA
TCATTATCGG
lunger fordiana
loc23#a
AAGTCGTAAC
AGGGTTTCCG
TAGGTGAACC
TGCGGAAGGA
TCATTATCGG
lunger fordiana
loc2 3#b
AAGTCGTAAC
AAGGTTTCCG
TAGGTGAACC
TGCGGAAGGA
TCATTATCGG
iiunger fordiana
loc24#a
AAGTCGTAAC
AAGGTTTCCG
TAGGTGAACC
TGCGGAAGGA
TCATTATCGG
tiunger fordiana
loc24#b
AAGTCGTAAC
AAGGTTTCCG
TAGGTGAACC
TGCGGAAGGA
TCATTATCGG
lunger fordiana
loc25#a
AAGTCGTAAC
AAGGTTTCCG
TAGGTGAACC
TGCGGAAGGA
TCATTATCGG
lunger fordiana
loc25#b
AAGTCGTAAC
AAGGTTTCCG
TAGGTGAACC
TGCGGAAGGA
TCATTATCGG
lunger fordiana
loc26#a
AAGTCGTAAC
AAGTCGTAAC
AAGGTTTCCG
AAGGTTTCCG
TAGGTGAACC
TAGGTGAACC
TGCGGAAGGA
TGCGGAAGGA
TCATTATCGG
TCATTATCGG
frequens
292
SCHILTHUIZEN ETAL.
TABLE 1. Continued.
160 170
\ L
TATAAT CATCA
TATAAT CATCA
TATAAT CATCA
TATAAT CATCA
TATAAT CATCA
TATAAT CATCA
TATAAT CATCA
TATAAT CATCA
TATAAT CATCA
TATAAT CATCA
TATAAT CATCA
TACAAT CATCA
TACAAT CATCA
TATAAT CATCA
TATAAT CATCA
TATAAT CATCA
TATTATTACA AAATACGTCA
210 220
I L
TGAAATTA — TGCTGAT
TGAAATTA — TGCTGAT
TGAAATTA — TGCTGAT
TAAAATTA — TGCTGGT
TGAAATTA — TGCTGAT
TGAAAATA — TGCTGAT
TGAAAATA — TGCTGAT
TGAAAATA — TGCTGAT
TGAAAATA — TGCTGAT
TGAAAATA — TGCTGAT
TGAAAATA — TGCTGAT
TGAAAATA — TGCTGAT
TGAAAATA — TACTGAT
TGAAAATA — TGCTGAT
TGAAAATA — TGCTGAT
TGAAAATA — TGCTGAT
TG — TATAGA TAATGCTGAT
260 270
I L
TCCCGTT GCCGATCGGG
TCCCGTT GCCAATCGGG
TCCCGTT GCCGATCGGG
TCCCGTT GCCGATCGGG
TCCCGTT GCCGATCGGG
TCCCGTT GCCGATCGGG
TCCCGTT GCCGATCGGG
TCCCGTT GCCGATCGGG
TCCCGTT GCCGATCGGG
TCCCGTT GCCGATCGGG
TCCCGTT GCCGATCGGG
TCCCGTT GCCGATCGGG
TCCCGTT GCCGATCGGG
TCCCGTT GCCGATCGGG
TCCCGTT GCCGATCGGG
TCCCGTT GCCGATCGGG
TCGTCTCATT GCCGATCGGG
180 190 200
\ I L
GGCTGCAGCG GGGCGCGCAG CGGCTTATGA
GGCAGCAGCG GGGCGCGCAG CGGCTTATGA
GGCAGCAGCG GGGCGCGCAG CGGCTTATGA
GGCAGCAGCG GGGCGCGCAG CGGCTTATGA
GGCAGCAGCG GGGCGCGCAG CGGCTTATGA
GGCAGCAGCG GGGCGCGCAG CAGCTTATGA
GGCAGCAGCG GGGCGCGCAG CGGCTTATGA
GGCAGCAGCG GGGCGCGCAG CGGCTTATGA
GGCATCAGCG GGGCGCGCAG CGGCTTATGA
GGCAGCAGCG GGGCACGCAG CGGCTTGTGA
GGCAGCAGCG GGGCGCGCAG CGGCTTGTGA
GGCAGCAGCG GGGCGCGCAG CGGCTTGTGA
GGCAGCAGCG GGGCGCGCAG CGGCTTGTGA
GGCAGCAGCG GGGCGCGCAG CGGCTTGTGA
GGCAGCAGCG GGGCGCGCAG CGGCTTGTGA
GGCAGC -GGCTTGTGA
GGC ATGA
230 240 250
I I L
TGAACGTCTG TC
TGAACGTCTG TC
TGAACGTCTG TC
TGAACGTCTG TC
TGAACGTCTG TC
TGAACGTCTG TC
TGAACGTCTG TC
TGAACGTCTG TC
TGAACGTCTG TC
TGAACGCCTG TC
TGAACGCCTG TC
TGAACGCCTG TC
TGAACGCCTG TC
TGAACGCCTG TC
TGAACGCCTG TC
TGAACGCCTG TC
GGAACGTGTC TCGTCTCGTC TCGTCTCGTC
280 290 300
I I L_
GACCGCAAGA AGCGCCGCCC CGGTCGGTTG
GACCGCAAGA AGCGCCGCCC CGGTCGGTTG
GACCGCAAGA AGCGCCGCCC CGGTCGGTCG
GACCGCAAGA AGCGCCGCCC CGGTCGGTCG
GACCGCAAGA AGCGCCGCCC CGGTCGGTTG
GACCGCAAGA AGCGCCGCCC CGGTCGGTTG
GACCGCAAGA AGCGCCGCCC CGGTCGGTTG
GACCGCGAGA AGCGCCGCCC CGGTCGGTTG
GACCGCAAGA AGCGCCGCCC CGGTCGGTTG
GACCGCAAGA AGCGCCGCCC CGGTCGGTTG
GACCGCAAGA AGCGCCGCCC CGGTCGGTTG
GACCGCAAGA AGCGCCGCCC CGGTCGGTTG
GACCGCAAGA AGCGCCGCCC CGGTCGGTTG
GACCGCAAGA AGCGCCGCCC CGGTCGGTTG
GACCGCAAGG AGCGCCGCCC CGGTCGGTTG
GACCGCAAGA AGCGCCGCCC CGGTCGGTTG
GACCGCAAGA AGCGCCGCCC CGGTCGGTTG
tiungerf ordiana .
bungerf ordiana .
fiunger f ordiana .
hungerf ordiana .
hungerf ordiana .
[lungerf ordiana .
nungerf ordiana .
lungerf ordiana .
hiungerf ordiana .
nungerf ordiana .
fiungerf ordiana .
hungerf ordiana .
[lungerf ordiana .
lunger f ordiana .
lungerf ordiana .
lunqerf ordiana .
loc5#a
loc5#b
loc8#a
loc8#b
loc9#a
locl6#a
locl6#b
loc22#a
loc22#b
loc23#a
loc2 3#b
loc24#a
loc24#b
loc25#a
loc25#b
loc26#a
hungerf ordiana ,
hiungerf ordiana .
[lungerf ordiana .
[lungerf ordiana ,
lungerf ordiana .
hungerf ordiana .
[lungerf ordiana .
[lungerf ordiana .
[lungerf ordiana ,
[lungerf ordiana .
[lungerf ordiana ,
lungerf ordiana .
lungerf ordiana .
lungerf ordiana .
[lunger f ordiana .
[lungerf ordiana .
loc5#a
loc5#b
loc8#a
loc8#b
loc9#a
locl6#a
locl6#b
loc22#a
loc22#b
loc23#a
loc2 3#b
loc24#a
loc24#b
loc25#a
loc25#b
loc26#a
[lungerf ordiana .
lungerf ordiana .
[lungerf ordiana ,
[lungerf ordiana .
[lungerf ordiana .
[lungerf ordiana ,
hungerf ordiana .
[lungerf ordiana ,
[lungerf ordiana .
[lungerf ordiana .
[lungerf ordiana .
[lungerf ordiana .
[lungerf ordiana ,
[lungerf ordiana .
[lungerf ordiana ,
[lungerf ordiana .
f requens
loc5#a
loc5#b
loc8#a
loc8#b
loc9#a
locl6#a
locl6#b
loc22#a
loc22#b
loc2 3#a
loc23#b
loc24#a
loc24#b
loc25#a
loc25#b
loc26#a
POPULATION STRUCTURE OF A MALAYSIAN LAND SNAIL
293
TABLE 1. Continued.
310
320
330
340
_1_
350
tiungerfordiana
rtungerfordiana
fiungerfordiana
hiungerfordiana
fiungerfordiana
riungerfordiana
hunger fordiana
riungerfordiana
lunger fordiana
lunger fordiana
lunger fordiana
lunger fordiana
lunger fordiana
lunger fordiana
lunger fordiana
lunger fordiana
frequens
. loc5#a
. loc5#b
, loc8#a
. loc8#b
.loc9#a
. locl6#a
,locl6#b
,loc22#a
.loc22#b
,loc2 3#a
.loc2 3#b
,loc24#a
,loc24#b
, loc25#a
,loc25#b
. loc26#a
ACCGCTCCCC
ACCGCTCCCC
ACCGCTCCCC
ACCGCTCCCC
ACCGCTCCCC
ACCGCTCCCC
ACCGCTCCCC
ACCGCTCCCC
ACCGCTCCCC
ACCGCTCCCC
ACCGCTCCCC
ACCGCTCCCC
ACCGCTCCCC
ACCGCTCCCC
ACCGCTCCCC
ACCGCTCCCC
ACCGCTCCCC
360
L_
TGTTTCGGGG
TGTTTCGGGG
TGTTTCGGGG
TGTTTCGGGG
TGTTTCGGGG
TGTTTCGGGG
TGTTTCGGGG
TGTTTCGGGG
TGTTTCGGGG
TGTTTCGGGG
TGTTTCGGGG
TGTTTCGGGG
TGTTTCGGGG
TGTTTCGGGG
TGTTTCGGGG
TGTTTCGGGG
TGTTTCGGGG
370
L_
TACCTAGTCT
TACCTAGTCT
TACCTAGTCT
TACCTAGTCT
TACCTAGTCT
TACCTAGTCT
TACCTAGTCT
TACCTAGTCT
TACCTAGTCT
TACCTAGTCT
TACCTAGTCT
TACCTAGTCT
TACCTAGTCT
TACCTAGTCT
TACCTAGTCT
TACCTAGTCT
TACCTAGTCT
380
TATCTCGCAC
TATCTCGCAC
TATCTCGCAC
TATCTCGCAC
TATCTCGCAC
TATCTCGCAC
TATCTCGCAC
TATCTCGCAC
TATCTCGCAC
TATCTCGCAC
TATCTCGCAC
TATCTCGCAC
TATCTCGCAC
TATCTCGCAC
TATCTCGCAC
TATCTCGCAC
CGTCTCGCAC
390
TCAATACGGC
TCAATACGGC
TCAATACGGC
TCAATACGGC
TCAATACGGC
TCAATACGGC
TCAATACGGC
TCAATACGGC
TCAATACGGC
TCAATACGGC
TCAATACGGC
TCAATACGGC
TCAATACGGC
TCAATACGGC
TCAATACGGC
TCAATACGGC
TCAATACGGC
400
(lungerf ordiana ,
hungerf ordiana .
tiungerf ordiana .
hungerf ordiana .
Iiungerf ordiana .
tiungerf ordiana ,
tiungerf ordiana ,
hungerf ordiana .
hungerf ordiana .
hungerf ordiana .
Riungerfordiana .
hungerf ordiana ,
Riungerfordiana .
Riungerfordiana .
Riungerfordiana .
Riungerfordiana .
loc5#a
loc5#b
loc8#a
loc8#b
loc9#a
locl6#a
locl6#b
loc22#a
loc22#b
loc23#a
loc23#b
loc24#a
loc24#b
loc25#a
loc25#b
loc26#a
CCACGGTGAC
CCACGGTGAC
CCACGGTGAC
CCACGGTGAC
CCACGGTGAC
CCACGGTGAC
CCACGGTGAC
CCACGGTGAC
CCACGGTGAC
CCACGGTGAC
CCACGGTGAC
CCACGGTGAC
CCACGGTGAC
CCACGGTGAC
CCACGGTGAC
CCACGGTGAC
CCACGGTGAC
410
L
GGCAAGAGCT
GGCAAGAGCT
GGCAAGAGCT
GGCAAGAGCT
GGCAAGAGCT
GGCAAGAGCT
GGCAAGAGCT
GGCAAGAGCT
GGCAAGAGCT
GGCAAGAGCT
GGCAAGAGCT
GGCAAGAGCT
GGCAAGAGCT
GGCAAGAGCT
GGCAAGAGCT
GGCAAGAGCT
GGCATGAGCT
420
TTCAGCTCGC
TTCAGCTCGC
TTCAGCTCGC
TTCAGCTCGC
TTCAGCTCGC
TTCAGCTCGC
TTCAGCTCGC
TTCAGCTCGC
TTCAGCTCGC
CTCAGCTCGC
TTCAGCTCGC
TTCAGCTCGC
TTCAGCTCGC
TTCAGCTCGC
TTCAGCTCGC
TTCAGCTCGC
CTCAGCTCGC
430
CGGGTCGTCA
CGGGTCGTCA
CGGGTCGTCA
CGGGTCGTCA
CGGGTCGTCA
CGGGTCGTCA
CGGGTCGTCA
CGGGTCGTCA
CGGGTCGTCA
CGGGTCGTCA
CGGGTCGTCA
CGGGTCGTCA
CGGGTCGTCA
CGGGTCGTCA
CGGGTCGTCA
CGGGTCGTCA
CGGGTCGTCA
440
GGTCTAAGGA
GGTCTAAGGA
GGTCTAAGGA
GGTCTAAGGA
GGTCTAAGGA
GGTCTAAGGA
GGTCTAAGGA
GGTCTAAGGA
GGTCTAAGGA
GGTCTAAGGA
GGTCTAAGGA
GGTCTAAGGA
GGTCTAAGGA
GGTCTAAGGA
GGTCTAAGGA
GGTCTAAGGA
GGTCTAAGGA
450
Riungerfordiana
Riungerfordiana
Riungerfordiana
Riungerfordiana
Riungerfordiana
Riungerfordiana
Riungerfordiana
Riungerfordiana
Riungerfordiana
Riungerfordiana
Riungerfordiana
Riungerfordiana
Riungerfordiana
lunger fordiana
Riungerfordiana
Riungerfordiana
,loc5#a
loc5#b
loc8#a
loc8#b
, loc9#a
,locl6#a
,locl6#b
, loc22#a
,loc22#b
, loc23#a
,loc2 3#b
,loc24#a
.loc24#b
. loc25#a
.loc25#b
loc26#a
GCGCTGCTCT
GCGCTGCTCT
GCGCTGCTCT
GCGCTGCTCT
GCGCTGCTCT
GCGCTGCTCC
GCGCTGCTCC
GCGCTGCTCC
GCGCTGCTCC
GCGCTGCTCT
GCGCTGCTCT
GCGCTGCTCT
GCGCTGCTCT
GCGCTGCTCT
GCGCTGCTCT
GCGCTGCTCT
GCGCCGCTCT
GATTGCTCTG
GATTGCTCTG
GATTGCTCTG
GATTGCTCTG
GATTGCTCTG
GACTGCTCTG
GATTGCTCTG
GATTGCTCTG
GATTGCTCTG
GATTGCCCTG
GATTGCTCTG
GATTGCTCTG
GATTGCTCTG
GATTGCTCTG
GATTGCTCTG
GATTGCTCTG
GACTGCTCTA
TGAGCGGCGC
TGAGCGGCGC
TGAGCGGCGC
TGAGCGGCGC
TGAGCGGCGC
TGAGCGGCGC
TGAGCGGCGC
TGAGCGGCGC
TGAGCGGCGC
TGAGCGGCGC
TGAGCGGCGC
TGAGCGGCGC
TGAGCGGCGC
TGAGCGGCGC
TGAGCGGCGC
TGAGCGGCGC
TGAGCGGCGC
CGCCCCGGTG
CGCCCCGGTG
CGCCCCGGTG
CGCCCCGGTG
CGCCCCGGTG
CGCCCCGGTG
CGCCCCGGTG
CGCCCCGGTG
CGCCCCGGTG
CGCCCCGGTG
CGCCCCGGTG
CGCCCCGGTG
CGCCCCGGTG
CGCCCCGGTG
CGCCCCGGTG
CGCCCCGGTG
CGCCCCGGTA
GTTG-TGAGG
GTTG-TGAGG
GTTG-TGTGG
GTTG-TGTGG
GTTG-TGTGG
ATTG-CGTGG
ATTG-CGTGG
ATTG-CGTGG
ATTG-CGTGG
ATTG-TGTGG
ATTG-TGTGG
ATTG-TGTGG
ATTG-TGTGG
ATTG-TGTGG
ATTG-TGTGG
ATTG-TGTGG
GTTGGTGTGG
294
SCHILTHUIZENETAL.
TABLE 1. Continued.
460 470 480
L I L
-ATAATGGAG G GTACCTG
-ATÄATGGAG G GTACCTG
-ATAATGGAG G GTACCTG
-ATAATGGAG G GTACCTG
-ATAATGGAG G GTACCTG
GATAATGGAG G GTACCTG
GATAATGGAG G GTACCTG
GATAATGGAG G GTACCTG
GATAATGGAG G GTACCTG
-ATAACGGAG G GTACCTG
-ATAACGGAG G GTACCTG
-ATAACGGAG G GTACCTG
-ATGACGGAG G GTACCTG
-ATAACGGAG G GTACCTG
-ATAACGGAG G GTACCTG
-ATAACGGAG G GTACCTG
-ATCAAGGAG GCAAGGCCGG AGGGTACCTG
510 520 530
I \ L_
TCGGCGGATC CGGGTGCGAT AGCTCCTGCG
TCGGCGGATC CGGGTGCGAT AGCTCCTGCG
TCGGCGGATC CGGGTGCGAT AGCTCCTGCG
TCGGCGGATC CGGGTGCGAT AGCTCCTGCG
TCGGCGGATC CGGGTGCGAT AGCTCCTGCG
TCGGCGGGTC CGGGTGCGAT AGCTCCTGCG
TCGGCGGGTC CGGGTGCGAT AGCTCCTGCG
TCGGCGGGTC CGGGTGCGAT AGCTCCTGCG
TCGGCGGGTC CGGGTGCGAT AGCTCCTGCG
TCGGCGGGTC CGGGCGCGAG AGCTCCTGCG
TCGGCGGGTC CGGGTGCGAG AGCTCCTGCG
TCGGCGGGTT CGGGTGCGAG AGCTCCTGCG
TCGGCGGGTT CGGGTGCGAG AGCTCCTGCG
TCGGCGGGTT CGGGTGCGAG AGCTCCTGCG
TCGGCGGGTT CGGGTGCGAG AGCTCCTGCG
TCGGCGGGTT CGGGTGCGAG AGCTCCTGCG
TCGGCGGGTC TGGGTGCGAC AGCTCCTGCG
560 570 580
1 I L
GCTTAAAGA? GTCGGCC-GT A TGCT
GCTTAAAGA? GTCGGCC-GT A TGCT
GCTTAAAGA? GTCGGCC-AT A-TATATGCT
GCTTAAAGA? GTCGGCC-AT A-TATATGCT
GCTTAAAGA? GTCGGCCCAT A TGCT
GCTTAAAGA? GTCGGCC-AT G TGCT
GCTTGAAGA? GTCGGCC-AT G TGCT
GCTTAAAGA? GTCGGCC-AT G TGCT
GCTTAAAGA? GTCGGCC-AT G TGCT
GCTTAAAGA? GTCGGCC-AC G TGCT
GCTTAAAGA? GTCGGCC-AT G TGCT
GCTTAAAGA? GTCGGCC-AT G TGCT
GCTTAAAGA? GTCGGCC-AT G TGCT
GCTTAAAGA? GTCGGCC-AA G TGCT
GCTTAAAGA? GTCGGCC-AA G TGCT
GCTTAAAGA? GTCGGCC-AT G TGCT
GCTTAAAGA? GTCGGCC-AT GCTCGCGGCT
490
500
hungerf ordiana .
Iiungerf ordiana ,
bungerf ordiana ,
hungerf ordiana .
hungerf ordiana .
iiungerf ordiana .
Iiungerf ordiana .
lungerf ordiana .
iiungerf ordiana .
iiungerf ordiana .
nungerf ordiana .
iiungerf ordiana .
iiungerf ordiana .
iiungerf ordiana .
iiungerf ordiana .
iiunqerf ordiana .
loc5#a
loc5#b
loc8#a
loc8#b
loc9#a
locl6#a
locl6#b
loc22#a
loc22#b
loc23#a
loc2 3#b
loc24#a
loc24#b
loc25#a
loc25#b
loc26#a
TGCGCTCGAC
TGCGCTCGAC
TGCGCTCGAC
TGCGCTCGAC
TGCGCTCGAC
TGCGCTCGAC
TGCGCTCGAC
TGCGCTCGAC
TGCGCTCGAC
TGCGCTCGAC
TGCGCTCGAC
TGCGCTCGAC
TGCGCTCGAC
TGCGCTCGAC
TGCGCTCGAC
TGCGCTCGAC
TGCGCTCGAC
540
CCGCT-CTGC
CCGCT-CTGC
CCGCT-CTGC
CCGCT-CTGC
CCGCT-CTGC
CCGCT-CTGC
CCGCT-CTGC
CCGCT-CTGC
CCGCT-CTGC
CCGCTGCTGC
CCGCTGCTGC
CCGCTGCTGC
CCGCTGCTGC
CCGCTGCTGC
CCGCTGCTGC
CCGCTGCTGC
CG-CT-CTGC
550
iiungerf ordiana .
iiungerf ordiana ,
iiungerf ordiana .
hungerf ordiana .
hungerf ordiana .
iiungerf ordiana .
iiungerf ordiana .
bungerf ordiana .
iiungerf ordiana .
iiungerf ordiana .
iiungerf ordiana .
nungerf ordiana .
iiungerf ordiana .
iiungerf ordiana .
iiungerf ordiana .
iiungerf ordiana .
loc5#a
loc5#b
loc8#a
loc8#b
loc9#a
locl6#a
locl6#b
loc22#a
loc22#b
loc23#a
loc23#b
loc24#a
loc24#b
loc25#a
loc25#b
loc26#a
GTGCAAACGC
GTGCAAACGC
GTGCAAACGC
GTGCAAACGC
GTGCAAACGC
GTGCAAACGC
GTGCAAACGC
GTGCAAACGC
GTGCAAACGC
GTGCAAACGC
GTGCAAACGC
GTGCAAACGC
GTGCAAACGC
GTGCAAACGC
GTGCAAACGC
GTGCAAACGC
GTGCAAACGC
590
AGGCCGCGA?
AGGCCGCGA?
AGGCCGCGA?
AGGCCGCGA?
AGGCCGCGA?
AGGCCGCGA?
AGGCCGCGA?
AGGCCGCGA?
AGGCCGCGA?
AGGCCGCGA?
AGGCCGCGA?
AGGCCGCGA?
AGGCCGCGA?
AGGCCGCGA?
AGGCCGCGA?
AGGCCGCGA?
AGGCCGCGA?
600
iiungerf ordiana .
iiungerf ordiana .
iiungerf ordiana .
iiungerf ordiana .
hungerf ordiana .
iiungerf ordiana .
Iiungerf ordiana .
iiungerf ordiana .
iiungerf ordiana .
hungerf ordiana .
iiungerf ordiana .
bungerf ordiana .
iiungerf ordiana .
iiungerf ordiana .
iiungerf ordiana .
iiungerf ordiana .
loc5#a
loc5#b
loc8#a
loc8#b
loc9#a
locl6#a
locl6#b
loc22#a
loc22#b
loc23#a
loc23#b
loc24#a
loc24#b
loc25#a
loc25#b
loc26#a
CGAGCA?ACC
CGAGCA?ACC
CG?GCA?ACC
CG?GCA?ACC
CGAGCG?ACC
CGAGCA?ACC
CGAGCA?ACC
CGAGCA?ACC
CGAGCA?ACC
CGAGCA?ACC
CGAGCA?A-C
CGAGCA?ACC
CGAGCA?ACC
CGAGCA?ACC
CGAGCA?ACC
CGAGCA?ACC
-GACCC-GCC
CGCCCGCTCC
CGCCCGCTCC
CGCCCGCTCC
CGCCCGCTCC
CGCCCGCTCC
CGCCCGCTCC
CGCCCGCTCC
CGCCCGCTCC
CGCCCGCTCC
CGCCCGCTCC
CGCCCGCTCC
CGCCCGCTCC
CGCCCGCTCC
CGCCCGCTCC
CGCCCGCTCC
CGCCCGCTCC
CGCCC — T —
POPULATION STRUCTURE OF A MALAYSIAN LAND SNAIL
295
TABLE 1. Continued.
610 620 630 640 650
1 1 1 1 1
lungerf ordiana . loc5#a
-lungerf ordiana . loc5#b
lungerf ordiana . loc8#a
lungerf ordiana . loc8#b
lungerf ordiana. loc9#a
lunger f ordiana . loc 1 6#a
lunger f ordiana . loc 1 6#b
lungerf ordiana . loc22#a
lungerf ordiana. loc22#b
lungerf ordiana. loc23#a
lungerf ordiana . loc2 3#b
lungerf ordiana. loc24#a
lungerf ordiana. loc24#b
lunger f ordiana . loc2 5#a
lungerf ordiana. loc25#b
lungerf ordiana. loc26#a
GTCTTCCT — TATTT-AATT TGTTACGCTT GGTGGCTCGT CTGTCTTATT
GTCTTCCT — TATTT-AATT TGTTACGCTT GGTGGCTCGT CTGTCTTATT
GTCTTCCT — TATTT-AATT TGTTACGCTT GGTGGCTCGT CTGTCTTATT
GTCTTCCT — TATTT-AATT TGTTACGCTT GGTGGCTCGT CTGTCTTATT
GTCTTCCT — TATTT-AATT TGTTACGCTT GGTGGCTCGT CTGTCTTATT
GTCTTCCT — TATTT-AATT TGTTACGCTT GGTGGCTTGT CTGTCTTATT
GTCTTCCT— TATTT-AATT TGTTACGCTT GGTGGCTCGT CTGTCTTATT
GTCTTCCT— TATTT-AATT TGTTACGCTT GGTGGCTCGT CTGTCTTATT
GTCTTCCT— TATTT-AATT TGTTACGCTT GGTGGCTCGT CTGTCTTATT
GCCTTCCT— TATTT-AATT TGTTACGCTT GGTGGCTCGT CTGTCCTATT
GCCTTCCT — CATTT-AATT TGTTACGCTT GGTGGCTCGT CTGTCCTATT
GCCTTCCT — TATTT-AATT TGTTACGCTT GGTGGCTCGT CTGTCCTATT
GCCTTCCT — TATTT-AATT TGTTACGCTT GGTGGCTCGT CTGTCCTATT
(3CCTTCCT — TATTT-AATT TGTTACGCTT GGTGGCTCGT CTGTCCTATT
GCCTTCCT — TATTT-AATT TGTTACGCTT GGTGGCTCGT CTGTCCTATT
GCCTTCCT — TATTT-AATT TGTTACGCTT GGTGGCTCGT CTGTCCTATT
trequens
660 670 680 690 700
1 1 1 I 1
hungerfordiana. loc5#a
hungerfordiana. loc5#b
nungerf ordiana . loc8#a
lungerf ordiana. loc8#b
lungerf ordiana . loc9#a
nungerf ordiana . loc 1 6#a
lunger f ordiana . loc 1 6#b
lungerf ordiana . loc22#a
lungerf ordiana . loc22#b
lungerf ordiana. loc23#a
lunger f ordiana. loc23#b
lungerf ordiana. loc24#a
lungerf ordiana . loc24#b
lungerf ordiana . loc25#a
lungerf ordiana . loc25#b
lunqerf ordiana. loc26#a
TGTCAGTTAC CGAAAAA С AAGATTGCTT GTCGTACAAC
TGTCGGTTAT CGAAAAA С AAGATTGCTT GTCGTACAAC
TGTCCGTTAT CGAAAAA С AAGATTGCTT GTCGTACAAC
TGTCGGTTAT CGAAAAG С AAGATTGCTT GTCGTACAAC
TGTCGGTTAT CGAAAAA С AAGATTGCTT GTCGTACGAC
TGTCGGTTAT CGAAAAA С AAGATTGCTT GTCGTACAAC
TGTCGGTTAT CGAAAAA С AAGATTGCTT GTCGTACAAC
TGTCGGTTAT CGAAAAA С AAGATTGCTT GTCGTACAAC
TGTCGGTTAT CGAAAAA С AAGATTGCTT GTCGTACAAC
TGTCGGTTAT CGAAAAA — A AAAACAAA-C AAGATTGCTT GTCGTACAAC
TGTCGGTTAT CPAAAAAAAA AAAAC7AAAC AAPATTGCTT GTCGT7CAAC
TGTCGGTTAT CGAAAAAAAA AAAA С AAGATTGCTT GTCGTACAAC
TGTCGGTTAT CGAAAAAAAA AAA С AAGATTGCTT GTCGTACAAC
TGTCGGTTAT CGAAAAAAAA A С AAGATTGCTT GTCGTACAAC
TGTCGGTTAT CGAAAAAAAA A С AAGATTGCTT GTCGTACAAC
TGTCGGTTAT CGAAAAA -AAACAAAAC AAGATTGCTT GTCGTACAAC
GGTTAT CGAAAAA-CC AAAACAAAA- TGCTT GTCGTACAAC
710 720 730 740 750
1 1 1 1 1
rfrequens
lungerf ordiana . loc5#a
lungerf ordiana . loc5#b
lungerf ordiana. loc8#a
lungerf ordiana. loc8#b
lungerf ordiana. loc9#a
lunger f ordiana . loc 1 6#a
lungerf ordiana . loc 1 6#b
lungerf ordiana. loc22#a
lungerf ordiana. loc22#b
lungerf ordiana . loc2 3#a
lungerf ordiana. loc23#b
lungerf ordiana. loc24#a
lungerf ordiana. loc24#b
lungerf ordiana. loc25#a
lungerf ordiana. loc25#b
lungerf ordiana . loc26#a
TTTGAGCGGT GGATCACTCG GCTCGTGCGT CGATGAAGAG CGCAACCAGC
TTTGAGCGGT GGATCACTCG GCTCGTGCGT CGATGAAGAG CGCAGCCAGC
TTTGAGCGGT GGATCACTCG GCTCGTGCGT CGATGAAGAG CGCAGCCAGC
TTTGAGCGGT GGATCACTCG GCTCGTGCGT CGATGAAGAG CGCAGCCAGC
TTTGAGCGGT GGATCACTCG GCTCGTGCGT CGATGAAGAG CGCAGCCAGC
TTTGAGCGGT GGATCACTCG GCTCGTGCGT CGATGAAGAG CGCAGCCAGC
TTTGAGCGGT GGATCACTCG GCTCGTGCGT CGATGAAGAG CGCAGCCAGC
TTTGAGCGGT GGATCACTCG GCTCGTGCGT CGATGAAGAG CGCAGCCAGC
TTTGAGCGGT GGATCACTCG GCTCGTGCGT CGATGAAGAG CGCAGCCAGC
TTTGAGCGGT GGATCACTCG GCTCGTGCGT CGATGAAGAG CGCGGCCGGC
TTT ??????? ?????????? ?????????? ?????????? ??????????
TTTGAGCGGT GGATCACTCG GCTCGTGCGT CGATGAAGAG CGCAGCCAGC
TTTGAGCGGT GGATCACTCG GCTCGTGCGT CGATGAAGAG CGCAGCCAGC
TTTGAGCGGT GGATCACTCG GCTCGTGCGT CGATGAAGAG CGCAGCCAGC
TTTGAGCGGT GGATCACTCG GCTCGTGCGT CGATGAAGAG CGCAGCCAGC
TTTGAGCGGT GGATCACTCG GCTCGTGCGT CGATGAAGAG CGCAGCCAGC
TTTGAGCGGT GGATCACTCG GCTCGTGCGT CGATGAAGAG CGCAGCCAGC
freauens
296
SCHILTHUIZENETAL.
TABLE 1. Continued.
76C
1
77C
1
780 790 800
1 1 1
lungerf ordiana . loc5#a
TGCGTGAATT
AATGTGAATT
GCAGAACACA
lungerfordiana. loc5#b
TGCGTGAATT
AATGTGAATT
GCAGAACACA
lungerf ordiana. loc8#a
TGCGTGAATT
AATGTGAATT
GCAGAACACA
lungerf ordiana. loc8#b
TGCGTGAATT
AATGTGAATT
GCAGAACACA
-lungerf ordiana. loc9#a
TGCGTGAATT
AATGTGAATT
GCAGAACACA
lungerf ordiana. locl6#a
TACGTGAATT
AATGTGAATT
GCAGAACACA
lunger f ordiana . loc 16#b
TGCGTGAATT
AATGTGAATT
GCAGAACACA
lungerf ordiana . loc22#a
TGCGTGAATT
AATGTGAATT
GCAGAACACA
lungerf ordiana. loc22#b
TGCGTGAATT
AATGTGAATT
GCAGAACACA
lungerf ordiana. loc23#a
TGCGTGAATT
AATGTGAATT
GCAGAACACA
lungerf ordiana . loc2 3#b
lungerf ordiana . loc24#a
■>'>'>■>■>■>'>'>■>■>
■>■>■>■>■>■>■>■>■>■>
■?->■?'>'>'}'>'>'>'>
TGCGTGAATT
AATGTGAATT
<X.KQ.M^Q.KQ.K
lungerf ordiana . loc24#b
TGCGTGAATT
AATGTGAATT
<X.KGM^Q.KCK
lungerf ordiana. loc25#a
TGCGTGAATT
AATGTGAATT
GCAGAACACA
lungerf ordiana. loc25#b
TGCGTGAATT
AATGTGAATT
GCAGAACACA
lungerf ordiana. loc26#a
TGCGTGAATT
AATGTGAATT
GCAGAACACA
TGCGTGAATT
AATGTGAATT
GCAGAACACA
TABLE 2. Pairwise Kimura's 2-parameter distances among the G. hungerfordiana sequences.
hunger f ordiana
loe5#a
0.0000
hungerfordiana
loc5#b
0.0085
0.0000
hungerfordiana
loo8#a
0.0129
0.0107
0.0000
hungerfordiana
loc8#b
0.0193
0.0150
0.0085
0.0000
hungerfordiana
loo9#a
0.0128
0.0085
0.0064
0.0107
0.0000
hungerfordiana
lool6#a
0.0324
0.0230
0.0259
0.0302
0.0236
0.0000
hungerfordiana
lool6#b
0.0280
0.0236
0.0215
0.0258
0.0193
0.0085
0.0000
hungerfordiana
loc22#a
0.0280
0.0236
0.0215
0.0258
0.0193
0.0085
0.0042
0.0000
hungerfordiana
loc22#b
0.0281
0.0237
0.0216
0.0259
0.0193
0.0085
0.0042
0.0042
O.OOOO
hungerfordiana
loc23#a
0.0394
0.0349
0.0328
0.0372
0.0304
0.0282
0.0238
0.0238
0.0238
0.0000
hungerfordiana
loc23#b
0.0481
0.0436
0.0414
0.0458
0.0391
0.0368
0.0324
0.0324
0.0324
0.0126
0.0000
hungerfordiana
loc24#a
0.0414
0.0369
0.0348
0.0392
0.0324
0.0302
0.0258
0.0253
0.0258
0.0063
0.0147
0.0000
hungerfordiana
loc24#b
0.0459
0.0413
0.0392
0.0436
0.0369
0.0345
0.0302
0.0302
0.0302
0.0105
0.0190
0.0042
0000
hungerfordiana
loc25#a
0.0415
0.0370
0.0348
0.0392
0.0325
0.0302
0.0258
0.0258
0.0259
0.0064
0.0127
0.0042
0084
0.0000
hungerfordiana
loo25#b
0.0437
0.0392
0.0370
0.0414
0.0347
0.0324
0.0280
0.0280
0.0280
0.0085
0.0148
0.0063
0105
0.0021
0.0000
hungerfordiana
loo26#a
0.0405
0.0359
0.0337
0.0382
0.0313
0.0289
0.0244
0.0244
0.0244
0.0043
0.0108
0.0022
0065
0.0022
0.0044
0.0000
MALACOLOGIA, 1999, 41(1): 297-317
ALLOZYME ANALYSES TEST THE TAXONOMIC RELEVANCE OF RIBBING IN
CHINESE ONCOMELANIA (GASTROPODA: RISSOACEA: POMATIOPSIDAE)
George M. Davis\ Yi Zhang^, Xingjiang Xu^ & Xianxiang Yang^
ABSTRACT
The Tropical Medical Research Center of the Institute of Parasitic Diseases, Shanghai, China,
is studying the genetics of Oncomelania snail populations throughout China relative to the trans-
mission of the human blood parasite Schistosoma japonicum by genetically competent popula-
tions of Oncomelania. On the basis of allozymes and C01 gene sequence data, Oncomelania
has diverged significantly on a regional basis. There are three groups of populations we classify
as subspecies: O. h. robertsoniin Yunnan and Sichuan above the Yangtze River Three Gorges
(small, smooth and varix-less shells); O. h. tangio\ Fujien Province, removed from the Yangtze
River drainage (small, smooth squat snails with a double thick varix); and O. /7. hupensis in the
Yangtze River drainage below the Three Gorges (large shells, always with varix, primarily
ribbed). These three clusters of populations differ from each other by Nei's "D" (allozymes) of
0.267 to 0.405. Thus, there is considerable genetic divergence among the subspecies.
There is the residual problem of what, genetically and taxonomically, are large smooth shelled
Oncomelania. with varix, found below the Three Gorges in the Yangtze drainage? It is well known
that in the floodplains of the Yangtze, Oncomelania populations are ribbed; above the effects of
the annual flooding, the snails are smooth. Some taxonomists have called these smooth-shelled
populations with varix O. fausti. The Miao River of Hubei Province offers a natural experiment to
resolve this problem. In a small river there are four population centers that are affected by flood-
ing; they have ribbed shells (RSP); above the flood zone there are several population centers
where the shells are smooth but with pronounced varix (SSP).
Allozyme data (35 loci, 72 alleles; mean samples per locus, 44 to 122) resolve the problem.
Mean Nei's D for RSP = 0.045 ± 0.036; for SSP = 0.024 ± 0.016; comparing RSP and SSP =
0.038 ± 0.035. There are no significant differences among these populations; they are all O. h.
hupensis. Ribbing is genetically controlled by a single gene with multiple alleles (Davis & Ruff,
1973). It thus appears that ribbing is a genetically controlled adaptation for dealing with annual
flooding and survival by water transport. It also appears that terminal varix formation is controlled
by a second gene. We predict that there will be no difference among all populations in the Miao
River for the capacity to transmit the same genetic strain of S. japonicum. The role of unique al-
leles and gene flow are addressed.
Key words: systematics, Oncomelania, China, allozymes, population genetics, ribbing,
Yangtze River, flooding, Hardy-Weinberg equilibrium.
INTRODUCTION
Oncomelania is one of eight Gondwanian
genera of the rissoacean family Pomatiopsi-
dae. The genus is comprised of a morphosta-
tic radiation (defined in Davis, 1994) with only
two species, O. hupensis Gredler, 1881, and
O. minima Bartsch, 1936. The former, an am-
phibious species, is distributed from Burma
(fossil) throughout southern China, Taiwan,
the Philippines, Sulawesi and Japan, with a
series of geographically isolated subspecies;
the latter, an aquatic species, is found in
northern Japan. Oncomelania hupensis is of
particular importance to humans as it is the in-
termediate host for the human blood fluke
Schistosoma japonicum. The Tropical Med-
ical Research Center of the Institute of Para-
sitic Diseases, Shanghai, China, is studying
the coevolution of Oncomelania snail popula-
tions with Schistosoma japonicum. The hy-
pothesis is: "as snail populations diverge
genetically, so must the associated schisto-
somes; genetic divergence of schistosomes
may have an important impact on human
pathology and vaccine development for anti-
^ Academy of Natural Sciences of Philadelphia. 1900 Benjamin Franklin Parkway. Philadelphia, Pennsylvania 19103 U.S.A.;
Davis@acnatsci.org
Chinese National Center of Systematic Medical Malacology, Institute of Parasitic Diseases. Shanghai, China;
Zhangyi@hotmail.com
Hubei Institute of Schistosomiasis Control, Zhuodaoquan, Wuchang, Wuhan. Hubei Province, China
297
298
DAVIS ETAL.
FIG. 1 . General position of the Miao River with respect to The Yangtze River and Wuhan City. Locations for
control populations HB 22 and HB 6 are shown.
Schistosoma vaccines". We have used al-
lozyme electrophoresis and gene sequencing
of the mt C01 and CYTb genes to begin to ad-
dress the hypothesis. On the basis of al-
lozymes and C01 data, Oncomelania has di-
verged significantly on a regional basis. There
are three primary groups we call subspecies:
O. h. robertsoni in Yunnan and Sichuan,
above the Yangtze River Three Gorges, with
small, smooth and varix-less shells; O. h.
tangi of Fujian Province, removed from the
Yangtze River drainage, with small, smooth
squat snails with a double thick varix; O. h. hu-
pensis in the Yangtze River drainage below
the Three Gorges with large shells, always
with varix, primarily ribbed (Davis et al.,
1 995). These three clusters of populations dif-
fer from each other by Nei's "D" (allozymes) of
0.267 to 0.405; by 0.28 (28%) C01 sequence
divergence (over 578 nucleotide positions).
There is great genetic divergence among the
subspecies!
There is the residual problem of what, ge-
netically and taxonomically, are large smooth-
shelled Oncomelania, with varix, found below
the Three Gorges in the Yangtze drainage? It
is well known that in the floodplains of
the Yangtze, Oncomelania populations are
ribbed; above the effects of the annual flood-
ing, the snails are smooth. Some taxonomists
have called these smooth-shelled populations
with varix O. fausti (reviewed in Davis et al.,
1995). The Miao River of Hubei Province
(Figs. 1, 2) offers a natural experiment to re-
FIG. 2. Localities along the Miao River and the re-
lationship of the Miao River to the Yangtze river.
A-C populations have ribbed shells and are below
the ti)ridge. Populations D-G have smooth shells
and are above the bridge.
solve this problem. In this short river, there are
four population centers that are affected by
flooding; they have ribbed shells (RSP);
above the flood zone there are several popu-
lation centers where the shells are smooth but
with pronounced varix (SSP). The purposes
of this paper are to determine the following;
(1) Are the RSP and SSP populations signifi-
ALLOZYME ANALYSES TEST THE TAXONOMIC RELEVANCE OF RIBBING
299
cantly different genetically? (2) Are there indi-
cations of gene flow or allele gradients? (3)
Are there populations with unique alleles? (4)
Are the populations subject to annual flooding
genetically unstable, that is, with polymorphic
loci in considerable Hardy-Weinberg disequi-
librium due to immigration of snails floated
into the river by the annual flooding of the
Yangtze from various other population cen-
ters (in contrast to genetically stable popula-
tions above the effects of the flooding)? (5)
Are the Miao River populations significantly
different from control populations from other
parts of Hubei Province?
MATERIALSAND METHODS
Locality Data
All populations collected in 1994 are from
Hubei Province, except the universal control,
GC, from Anhui Province (Figs. 1, 2). GC,
HB6 and HB22 are control populations. GC is
the universal control population used with all
populations of Oncomelania studied from
throughout China. HB6 is the control popula-
tion for Hubei Province. HB22 is a population
outside the Miao River basin but not distant
from the Miao River. Shells of these popula-
tions are shown in Figure 3.
HB6 (Figs. 1, 31): 28 Sept. 1994; Jian Li
County; Mao Shi Township; Su Hu Canal; Su
Jia Yuan Section of the canal. The snails were
collected from the banks of the Grand Canal
about 1 .5 m from the edge of the water under
dense grass. The grass indicated that the
area was not often flooded; at the edge of the
water there were very narrow beaches of
sand-shell perhaps a foot to 2 ft. wide in
places. The greatest density of snails was in a
narrow band. Above the band the soil was
drier and snails correspondingly fewer. The
slope was about 1 5"; at the upper edge about
1 5 ft. from the water it was about 25". With the
Oncomelania^ere Parafossarulus, and some
dead shells of Semisulcospira. There were
the same millipedes as in previous localities
with Oncomelania. High humidity under the
grass. Snails were in pockets of two to ten per
pocket. We collected about 1,200 snails. The
snails were numerous in this locality; all snails
came from a 1 5.2 m-long stretch of bank. The
canal had high density of Viviparidae [Bel-
lamya quadrata type]; shells of Anodonta
were abundant.
The canal was built in 1959 to 1962; ac-
cording to the Hubei Institute of Schistosomi-
asis Control, in 1962 there were no Oncome-
lania in the canal; they came down from the
three lakes region at the head of the canal.
The shells are longer and more slender than
those from Gui Chi (contrast Fig. 3 I and H),
and they are not as heavily ribbed.
HB11 (A, Figs. 2,3): 1 Oct.; Song Zi County;
Xin Jiang Kou Township; De Shun Village.
Miao River about 0.4 km above the mouth of
the river. Snails were collected from the moist
to wet mud under grasses of small ponds in the
narrow flood plain of the river. The snails were
ribbed and with a thick varix. Snail density was
great. All snails were heavily ribbed.
HB12 (not on Fig. 2): 1 Oct.; Song Zi
County; Cheng Dian Township; Ten Zi Qiao
Village; Miao River 13 km upstream from
HB1 1 . Snails, very few, were collected from a
small stream flowing over a small dam into the
Miao River some 50 m above the bridge
crossing the Miao. Snails came from the wet
mud under thick grass at the edges of the
stream near the dam and rapidly flowing
water. No snails were found on the banks of
the main river that had a rapid flow over cob-
bles. There were too few snails for elec-
trophoresis. Snails had smooth shells.
The bridge (Fig. 2, bridge) has great signif-
icance. For some 60 m up- and downstream
the snails are reputed to have a mixture of
phenotypes from ribbed to smooth. One km
downstream one begins to get ribbed snails
that transmit Schistosoma japonicum. There
is apparently no transmission, at this time,
by smooth snails at the bridge or above it.
The bridge marks the uppermost limit of the
Yangtze flood during the rainy season. The
bridge is about 13 km from the mouth of
the river.
HB1 3 (D, Figs. 2, 3): 1 Oct.; Song Zi County;
Cheng Dian Township; Ten Zi Qiao Village;
Miao River 1 .0 km above the bridge and HB1 2.
Snail population density high; snails came
from behind a natural dike or bank between
the river and the true bank of the river. This
backwater has water flowing through it cutting
through the secondary bank into the river. A lit-
tle upstream, the river bed widens and this
ridge separates the river from the natural river
bank by some 40 m with filled in land with
crops. The snails came from mud under thick
grasses at the edge of the water filling in the
backwater. The snails were smooth.
HB14 (F, Figs. 2, 3): 1 Oct.; Song Zi County;
Cheng Diam Township; Ten Zi Qiao Village;
Miao River 1 .5 km above bridge. At this local-
300
DAVIS ET AL.
FIG. 3. Representative shells from the populations of this study. Figs. A-G are from populations A-G in Fig.
2. Fig. H is a shell from GuiChi; I is from Jian Li; J is from Jing Men.
ALLOZYME ANALYSES TEST THE TAXONOMIC RELEVANCE OF RIBBING
301
Ity two branches of the river bend around and
are separated from each other by only the
road's width of some 14 m. On the south side
is a dam and spillway. Behind the spillway
there is no discernable current. The water is
choked with aquatic plants. Extensive marshy
areas are at the margins. Snails are dense on
the mud under the grass at the water's edge.
HB1 4 is on the exact opposite side of the road
from the north branch. Snails are smooth.
HB15 (E, Figs. 2, 3): 1 Oct.; Song Zi
County: Miao River 1 .5 km above bridge. Op-
posite HB1 4 from the north branch of the river.
Wide marshy areas on each side of the slow
flowing steam harbor dense snail populations.
Snails come from the wet mud under dense
grass cover. Snails are smooth.
HB1 6 (G, Figs. 2, 3): 1 Oct.; Song Zi County;
Cheng Diam Township; Long Tan He Village;
Miao River; 7 km above the bridge and 5 km
from the top of the river; elevation about 47 m.
The bridge elevation is 42 m. Snails numerous
on mud under dense grass cover in the
marshy margin of the river. Marshy border
some 4.6 m wide. Stream banks some 15 m
apart. Free flowing water about 4.6 m wide and
choked with aquatic vegetation. Current very
slow flow. Riverbanks 1 .8 m high on S side; 1 .2
m high on N side. Snails are smooth.
HB1 7 (B, Figs. 2, 3); collected by the schis-
tosomiasis control station about 30 Oct. Song
Zi County; Xinjiang Kou Township; De Shun
Village; Miao River main branch; Lower Miao
River 3 km upstream from HB11. Snails with
ribbed shells.
HB18 (C, Figs. 2, 3); 26 Sept.; Song Zi
County; Lao Cheng Township; Yu Jia Du Vil-
lage. Miao River mainstream. Collected by
Schistosmiasis Control Unit. 2.0 km down-
stream from the bridge. 8.5 km upstream from
HB17. Snails with ribbed shells.
HB22 (Figs. 1 , 3J; Control; not Miao River);
6 Oct.; Jing Men City; 50 km SE of the city;
Sha Yang Agricultural Management; Mia
Liang State Farm; group 1 3. Snails came from
a drainage canal bounded by cotton fields on
one side and a road on the other with rice
fields beyond that. The ditch was about 5.2 m
wide from bank top to bank top; about 1 .2 m
to 1.5 m deep with sides about 40". Snails
were on the mud at the water level of the ditch
up about 0.3 m under grass cover. Density
was high. Shells smooth.
GC (Control) (Fig. 3H): Anhui Province; Gui
Chi City; 117° 20.6' E; 30° 30' N. collected in
1994. Control population. The snails came
from the flood planes of the Yangtze River.
Electrophoresis
Horizontal starch gel electrophoresis of tis-
sue proteins from ten populations was fol-
lowed by staining for the following 35 loci;
AAT-1, AAT-2 (aspartate aminotransferase,
2.6.1.1); AK-1 and 2 (adenosine kinase,
2.7.1.20); АО (aldehyde oxidase, 1.2.3.1);
ACPH-1 and 2 (acid phosphatase, 3.1.3.2);
APH (alkaline phosphatase, 3.1.3.1); CK-1
and 2 (creatine kinase, 2.7.3.2); EST-1, EST-
2, EST-3 (esterase, 3.1.1.1); GDH (glutamate
dehydrogenase, 1.4.1.2); G6PD (glucose-6-
phosphate dehydrogenase, 1.1.1.49); GPI
(glucose-6-phosphate isomerase, 5.3.1.9);
HBD (hydroxybutyrate dehydrogenase,
1.1.1.30); ISDH-1, ISDH-2 (isocitrate dehy-
drogenase, 1.1.1.42); LDH (L-lactate dehy-
drogenase, 1.1.1.27); MDH (malic dehydroge-
nase, 1.1.1.37); ME-1, ME-2 (malic enzyme,
1.1.1.40); MP! (mannose-6-phosphate iso-
merase, 5.3.1.9); NADD-1 and 2 (NADH de-
hydrogenase, 1 .6.99.3); 6PGD-1 and 2 (phos-
phogluconate dehydrogenase, 1.1.1.44);
PGM-1, PGM-2 (phosphoglucomutase,
5.4.2.2); OCT (octopine dehydrogenase,
1 .5.1 .11 ); SDH-1 ; SDH-2 (sorbitol dehydroge-
nase, 1.15.1.1); SOD (SOD (superoxidismu-
tase, 1.15.1.1); XDH (xanthine dehydroge-
nase, 1.1.1.204). Procedures are those of
Ayala et al. (1973) as modified by Dillon &
Davis (1980), Davis (1983), Davis & Fuller
(1981), Davis et al. (1981, 1988), and most re-
cently for Oncomelania, by Davis et al. (1994,
1995).
Genetic parameters were calculated using
BIOSYS-1 (Swofford & Selander, 1981).
Hardy-Weinberg equilibrium was analyzed for
all polymorphic loci. Nei's (1978) genetic dis-
tance and Wright's (1978) modified Rogers
distance were calculated and phenograms
constructed using the UPGMA method. An
unrooted tree based on Wright's D was also
constructed using the FITCH program of
PHYLIP version 3.4 (Felsenstein, 1989). This
phylogentic analysis program does not as-
sume equal rates of evolution. Some 50 repe-
titions of FITCH were run with randomized
input order and optimization by global branch
rearrangement. We used both Nei's D and
Wright's modified Rogers D as explained in
Davis (1994). The former is traditional and
widely used, and as such sets a standard for
comparisons. However, it is not metric, and
closely related populations are rather com-
pacted towards the 0.01 end of the range.
With Wright's D, a metric, the closely com-
302
DAVIS ETAL.
TABLE 1. Shell length and smooth/ribbed characteristics. N = 5 unless stated otherwise.
Population
No.
Whorls
Shell
Length
Length Body
Whorl
No. Ribs
CONTROLS
GUICHI
JINGMEN
JIAN LI
MIAO RIVER
BELOW BRIDGE
6-7 eroded
8.5-9.0
9.0-9.5
9.6 ± 0.3
8.4 ± 0.6
9.1 ± 0.3
5.3 ± 0.2
4.0 ± 0.2
4.3 ± 0.1
14 ± 1
SMOOTH
18 ±2
A
В
С
ABOVE BRIDGE
8.5-9.0
8.5-9.0
9.0-10.0
8.2 ± 0.7
8.1 ± 0.4
9.4 ± 0.2
4.2 ± 0.3
4.1 ± 0.2
4.6 ± 0.3
19±2
18 ± 2
21 ± 4
D
E
F
G
8.5
8.5
8.5-9.0
8.5
7.0 ± 0.5
6.8 (N = 2)
7.0 ± 0.5
7.4 ± 0.5
3.7 ± 0.2
3.6
3.6 ± 0.1
3.8 ± 0.3
SMOOTH
SMOOTH
SMOOTH
SMOOTH
pacted Nei's values are more spread out and
better reflect the actual differences among
populations in terms of a finite limit.
RESULTS
Table 1 provides basic shell data. Shells
above the bridge (Fig. 3, D-G) are of the
same whorl number (possibly excepting pop-
ulation C) as populations below the bridge
(Fig. 3, A-C) but are significantly smaller. The
Gui Chi controls (Fig. 3, H) are significantly
larger than the other ribbed populations
(based on length of body whorl, because shell
length cannot be calculated due to the eroded
condition of the apices of all shells). The Gui
Chi population also has significantly fewer
ribs per whorl than do the other ribbed popu-
lations.
Table 2 lists the genotype frequencies for
the ten populations listing the 16 of 35 loci
(46%) that are either polymorphic or have al-
ternative alleles; these involve 71 alleles. The
Miao River populations are arranged in the
table from downstream (left side) to upstream
(right side). Indices of genetic variability are
given in Table 3. Mean sample sizes per locus
ranged from 44.4 to 122.3. The percentage of
polymorphic loci ranged from 11 .4 to 22.9, the
value not correlated with position along the
river. The highest values were from terminal
populations at the head and mouth of the
river. Heterozygosity was low, ranging from
0.014 to 0.059 (direct count). The mean num-
ber of alleles per locus was low with the aver-
age for all populations = 1 .36 ± 0.09.
Populations are scored for the number of
polymorphic loci (out of 12 such loci) and
number of loci considered to differ signifi-
cantly from Hardy-Weinberg equilibrium
(Hwe) (Table 4). All but population A had some
loci significantly different from HWe (range of
0-4), with the highest number occurring in the
Anhui, Gui Chi population, the middle popula-
tion below the bridge (B), and the uppermost
population above the bridge (G). The actual
loci involved are scored in Table 5, where it is
seen that the EST1, AAT1, ME1, and PGM1
were the loci most often out of HWe (from 4 to
7 populations). The ME1 locus was only out of
HWe in populations above the bridge. Proba-
bilities and fixation indices (F) and direction
and index of deviation (D) are given for all rel-
evant loci for all populations in Table 6. Het-
erozygote deficiency was the predominant
phenomenon. We examined the populations
to see if there were discernible patterns of
gene flow. Table 7 lists cases of alternative al-
leles (in monomorphic loci where populations
have different alleles), unique alleles to On-
comelania. and alleles unique to the Miao
River. Pie diagrams showing the frequencies
of alleles (alternative alleles or polymor-
phisms) are given in Figure 4, with columns
arranged in order from downstream (right
end) to upstream (to the left). There are alter-
native alleles or unique alleles in all but one
Miao River population. Two alternative alleles
are found in population A at the mouth of the
river, while one alternative allele is found in
population D just above the bridge. Of partic-
ular interest are the unique alleles found in
populations E and F, found only some 20 m
ALLOZYME ANALYSES TEST THE TAXONOMIC RELEVANCE OF RIBBING
303
TABLE 2. Allele frequencies for the 16 loci (46%) that are polymorphic or with alternative alleles among the
1 populations. Populations GC, HB6 and HB22 are not in the Miao River drainage. There are 71 alleles total.
For A-G, see Fig 2.
HB11
HB17
HB18
HB13
HB14
HB15
HB16
Locus
GC
HB6
HB22
A
В
С
D
F
E
G
AAT1
(N)
100
80
50
50
50
50
50
70
50
160
A
0.985
0.938
1.000
1.000
0.980
1.000
1.000
1.000
1.000
0.972
В
0.015
0.044
0.020
0.022
D
0.019
0.006
AAT2
(N)
100
80
50
50
50
50
50
70
50
160
A
1.000
0.906
1.000
1.000
0.970
1.000
1.000
1.000
1.000
1.000
В
0.013
0.020
С
0.081
0.010
ACPH2
(N)
60
65
50
50
50
70
50
60
50
110
A
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
D
1.000
AK1
(N)
60
65
50
50
50
70
50
60
50
110
A
1.000
0.923
1.000
1.000
1.000
1.000
1.000
1.000
1.000
В
1.000
E
0.077
APH
(N)
100
80
50
50
50
60
50
70
50
160
A
1.000
1.000
1.000
1.000
0.980
1.000
1.000
1.000
1.000
1.000
В
0.020
CK1
(N)
A
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
D
1.000
EST1
(N)
100
80
50
50
44
50
50
70
50
159
A
0.175
0.788
0.930
0.450
0.602
0.610
0.780
0.679
0.940
0.755
В
0.115
0.087
0.070
0.310
0.205
0.190
0.070
0.050
0.030
0.097
С
0.020
0.240
0.080
0.007
0.020
0.038
D
0.150
0.125
0.114
0.200
0.150
0.257
0.110
E
0.007
F
0.010
EST2
(N)
100
80
50
50
50
50
50
50
100
153
A
0.995
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
0.990
С
0.005
0.010
GPI
(N)
80
30
50
50
50
50
50
70
50
160
A
0.656
0.883
1.000
0.760
0.740
0.770
0.930
0.871
0.980
0.962
В
0.262
0.113
0.060
0.220
0.190
0.060
0.036
0.031
С
0.044
0.100
0.010
0.093
0.020
E
0.038
0.033
0.040
0.040
0.040
0.006
H
0.040
HBD1
(N)
50
6
50
50
50
60
50
10
50
90
A
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
0.444
С
0.556
ME1
(N)
80
30
50
50
50
50
50
70
50
160
A
1.000
1.000
1.000
1.000
1.000
1.000
0.700
0.957
0.570
0.956
В
0.020
0.006
С
0.140
0.029
0.290
0.019
D
0.100
0.014
0.140
0.019
E
0.040
continued)
304
DAVIS ETAL.
TABLE 2.
(Continued)
HB11
HB17
HB18
HB13
HB14
HB15
HB16
Locus
GC
HB6
HB22
A
В
С
1
D
F
E
G
OCT
(N)
86
80
50
47
50
50
50
70
50
152
A
0.547
0.444
0.620
0.649
0.680
0.530
0.940
0.557
0.550
0.651
В
0.064
0.019
0.021
0.010
0.010
С
0.314
0.525
0.230
0.277
0.260
0.360
0.030
0.279
0.120
0.204
D
0.076
0.150
0.053
0.050
0.110
0.030
0.136
0.300
0.135
Е
0.013
F
0.029
0.030
PGM1
(N)
70
20
50
50
50
50
50
70
50
143
A
0.679
1.000
0.530
0.610
0.810
0.530
0.930
0.740
0.790
0.745
В
0.007
0.030
0.010
0.030
0.003
E
0.071
0.080
0.130
0.020
0.060
0.214
0.180
0.105
F
0.229
0.440
0.300
0.060
0.420
0.010
0.043
0.030
0.147
G
0.014
PGM2
(N)
80
30
50
50
50
50
50
70
50
160
A
0.975
0.983
0.980
1.000
0.940
1.000
1.(
300
0.993
1.000
1.000
В
0.025
0.017
0.060
0.007
D
0.020
XDH
(N)
60
65
50
50
50
60
50
60
50
110
A
1.000
1.000
1.000
1.000
1.000
1.(
DOO
1.000
1.000
1.000
D
1.000
TABLE 3.
Populatioi
n genetic
indices.
Mean
Mpon no
Mean heterozygosity
Populations
sample No.
per locus
1 VIC^Cll 1 1 IVJ.
alleles
per locus
% loci
polymorphic
Direct count
Hdywbg
expected
CONTROLS
GC1-Gui Chi
НВ-6 -Jian Li
НВ-22 Jing
66.7 (3.3)
54.7 (3.9)
1.5(0.2)
1.4(0.1)
20
20
0.039 (0.020)
0.030(0.017)
0.061 (0.028)
0.047 (0.020)
Men City
MIAO RIVER
50.7(0.7)
1.2(0.1)
11.4
0.034(0.021)
0.036 (0.022)
HB-11 =
HB-17:
HB-18:
BRIDGE
= A
= B
= C
47.1 (1.9)
49.8 (0.20)
56.6(1.30)
1.3(0.02)
1.4(0.10)
1.3(0.10)
11.4
20
14.3
0.059 (0.029)
0.034(0.016)
0.046 (0.022)
0.060 (0.029)
0.057 (0.025)
0.060 (0.028)
HB-13:
HB-14:
HB-15:
HB-16:
= D
= F
= E
= G
48.6(1.7)
55.8(3.1)
44.4 (2.7)
122.3(4.6)
1.3(0.20)
1.4(0.2)
1.3(0.1)
1.5(0.2)
14.3
17.1
14.3
22.9
0.014(0.008)
0.038(0.019)
0.023(0.015)
0.027(0.014)
0.035(0.018)
0.052 (0.024)
0.048 (0.025)
0.059 (0.025)
from each other but in different branches of
the river (Fig. 2). With perhaps the exception
of allele В in EST-1, there is no discernable
pattern of gene flow or allele frequency gra-
dation among populations. Allele В in the case
cited decreases from the mouth of the river to
in HB15; however, it is present in low fre-
quency in HB-4 and 16.
Nei's and Wright's modified Rogers' genetic
distances are given in Table 8; the relevant
phenograms are given in Figures 5 and 6. The
respective cophenetic values are 0.995 and
0.978. The overall genetic distances involved
(clustering at <0.08, Nei's D) clearly shows
conspecificity (for species concepts involving
Oncomelania: Davis 1994; Davis et al., 1995).
There is no separate cluster of ribbed snails
or separate cluster for smooth snails. The
ribbed control from Anhui Province (GC) clus-
ters most closely with hbbed populations B. С
ALLOZYME ANALYSES TEST THE TAXONOMIC RELEVANCE OF RIBBING
305
TABLE 4. Populations scored for number of polymorphic loci and number of loci significantly
differing from Hardy-Weinberg equilibrium where significance using exact probabilities is set
at 0.06. There are 12 polymorphic loci total.
No. Polymorphic
No. Significantly
% Significantly
Population
loci
different
different
Controls
Gui Chi
7
3
43
HB-6
7
4
57
HB-22
4
1
25
Miao River
Below Bridge
A=HB-11
4
B = HB-17
7
4
57
C = HB-18
5
2
40
Above Bridge
D = HB-13
5
2
40
E = HB-15
5
2
40
F = HB-14
6
3
50
G = HB-16
7
4
57
TABLE 5. Chi-square tests for deviation from Hardy-Weinberg Equilibrium; N = not polymorphic, S significance is
derived from using exact probabilities; S = P < .06. [ ] = number of loci differing significantly from H-W. P values
are given.
Locus:
AAT1 AAT2 AKI APH EST1 EST2 GPI HBD1 ME1 OCT PGM1 PGM2
Population
GuiChi [4] S N
HB-6 [4] S S
HB-22 [1] N N
MIAO RIVER BELOW BRIDGE
A=HB-11 [0] N N
В = HB-17 [4] N S
C = HB-18[2] S N
MIAO RIVER ABOVE BRIDGE
D = HB-13[3] N N
E = HB-15[2] N N
F = HB-14[3] N N
G = HB-16[4] S N
N
N
S
1.000
0.218
N
N
0.831
S
S
S
N
S
N
1.000
N
N
1.000
N
1.000
N
N
1.000
N
N
N
N
0.559
0.579
s
N
N
0.774
N
0.249
N
N
0.754
0.338
N
N
S
S
N
0.269
N
N
0.210
S
1.000
N
N
0.388
N
0.254
N
N
0.267
S
N
N
N
S
N
1.000
N
s
1.000
S
N
N
N
1.000
N
1.000
N
S
0.097
S
N
N
N
s
N
0.588
N
S
0.628
S
1.000
N
N
s
1.000
1.000
N
S
0.110
S
N
and smooth population F. Population A at the
mouth of the river stands slightly apart from all
others. Table 9 provides inter-population com-
parisons of values of Nei's D. There is no sig-
nificant difference between the average D for
all Miao River populations (0.038 ± 0.035)
and the separate average D for populations
above the bridge (0.024 ± 0.01 6) or below the
bridge (0.045 ± 0.036).
The unrooted FITCH tree is given in Figure
7, freed from the constraints of phenogram
construction and based on a metric distance
measure. There are no two distinct clusters of
populations representing smooth and ribbed
shells. Populations A, Jian Li, and D indicate
their unique genetic structure and divergence
from all the other populations. Note that A dif-
fers from D by a Nei's D of 0.1 06 (Table 8), the
greatest divergence among Miao River popu-
lations.
DISCUSSION
Do Smooth and Ribbed Snails Deserve
Separate Taxonomic Rank?
The question appears resolved. Based on
this natural experiment, populations of On-
comelania within the Yangtze River drainage
below the Three Gorges of the Yangtze River
that are smooth (but with varix), and with the
same allometry as snails of ribbed populations
in the lower Yangtze drainage, are one sub-
species, O. hupensis hupensis. Wang et al.
(1998) support this concept on the basis of a
306
DAVIS ETAL.
TABLE 6. Deviation from Hardy-Weinberg and heterozygote deficiency for all popula-
tions. P is based on exact probabilities and significance is set at 0.06. Population A is
at the mouth of the Miao River; G is farthest upstream.
Population
locus
P
Fixation index (F)
D
Controls
Gui Chi
AAT1
0.02
0.662
-0.663
EST1
0.00
0.713
-0.714
GPI
0.22
0.220
-0.225
OCT
0.83
0.137
-0.142
PGM1
0.05
0.318
-0.323
PGM2
0.00
1.000
-1.000
Jian Li
AAT1
0.00
0.790
-0.791
AAT2
0.00
0.927
-0.928
AK
0.00
1.000
-1.000
EST1
0.00
0.369
-0.373
Jing Men
PGM2
0.01
1.000
-1.000
MIAO RIVER
Below Bridge
A(HB-11)
GPI
0.25
-0.134
0.123
PGM1
0.39
0.172
-0.180
B(HB-17)
AAT2
0.03
0.659
-0.662
APH
0.01
1.000
-1.000
EST1
0.00
0.645
-0.649
GPI
0.27
0.205
-0.213
OCT
0.21
0.144
-0.153
PGM1
0.00
0.567
-0.571
C(HB-18)
AAT1
0.01
1.000
-1.000
EST1
0.39
0.166
-0.175
GPI
0.25
0.188
-0.196
OCT
0.27
0.169
-0.177
PGM1
0.01
0.335
-0.342
Above bridge
D(HB-13
EST1
0.00
0.451
-0.456
ME1
0.00
1.000
-1.000
PGM1
0.00
0.848
-0.849
E(HB-15)
ME1
0.00
0.965
-0.965
OCT
0.10
0.122
-0.131
PGM1
0.00
0.708
-0.711
F(HB-14)
EST1
0.00
0.393
-0.660
GPI
0.59
-0.115
0.107
ME1
0.00
1.000
-1.000
OCT
0.63
0.108
-0.115
PGM1
0,010
0.358
-0.362
G(HB-16)
AAT1
0.00
0.659
-0.660
EST1
0.00
0.444
-0.446
HBD1
0.00
1.000
-1.000
ME1
0.00
1.000
-1.000
OCT
0.11
0.235
-0.237
PGM1
0.00
0.458
-0.460
few enzymes (esterases and MDH, slab PAG),
and a few specimens of ribbed and smooth-
shelled Oncome/an/a from three counties (five
populations) in Hube! Province. They found
very little difference among populations (data
not sufficient or scoreable for population ge-
netic analysis).
Ribbing is associated with annual flooding
of the Yangtze River and its tributaries. As is
well known to Chinese field workers attempt-
ing to monitor and control Oncomelania snails
(Liu et al., 1 981 ), snails from any elevation, or
a man-made situation that removes a popula-
tion from the annual floods, attain a smooth
shell but still retain the varix. Molecular ge-
netic data do not support the concept of dif-
ferent taxonomic status for these two shell
types. Accordingly, Katayama fausti Bartsch,
ALLOZYME ANALYSES TEST THE TAXONOMIC RELEVANCE OF RIBBING
307
TABLE 7. Distribution of unique and alternative alleles among the 1 populations.
Alternative
Alleles unique to
Population
alleles
Unique alleles
the Miao River
CONTROLS
Gui Chi
none
none
Jian Li
XDH/d
AAT1/d (0.019)
АК-1/е (0.077)
Jing Men
none
none
MIAO RIVER
Below Bridge
A
AK1/b
CKI/d
none
GPI/h (0.040)
В
none
none
AAT2/b (0.020)
AAT2/C (0.010)
APH/b (0.020)
С
none
none
none
Above Bridge
D
ACPH2/d
none
MEI/e (0.040)
E
none
none
EST1/f (0.010)
F
none
none
ESTI/e (0.007)
PGM2/b (0.007)
G
none
HBD/c (0.556)
AATI/d (0.006)
EST2/C (0.010)
1 925, is a synonynn of O. hupensis hupen-
sis Gredler, 1 881 . {Katayama used to be used
as a genus to include all smooth forms of On-
comelania hupensis.)
Davis & Ruff (1973) employed breeding ex-
periments and showed that ribbing in On-
comelania hupensis from mainland China is
controlled by a single gene with multiple alle-
les. Davis (1979) considered the smooth
shelled condition to be primitive and that rib-
bing is the derived condition. We argue that the
earliest ecology and shell morphology of On-
comelania hupensis is that seen in Yunnan
and Sichuan, areas into which pomatiopsine
snails were introduced through a trajectory
from the Indian Plate into northern Burma and
Yunnan, China, in the Miocene (Davis, 1979).
The habitat involves marshy ecotones in hilly
regions not subjected to annual flooding. It is
noteworthy that these Yunnan and Sichuan
snails, O. hupensis robertsoni, are smooth
and have no varix. The natural course of evo-
lution would have been through dispersal
down the evolving Yangtze River. With the ge-
ological formation of the Three Gorges section
of the Yangtze creating an effective barrier to
passive movement of snails up or down the
Yangtze River, snails living in the flood planes
and on the islands of the Yangtze River,
through mutation, developed ribs and a varix.
The hypothesis is that ribbing contributes
greatly to survival by significantly increasing
shell surface area, facilitating floatation and
dispersal during flooding. Additionally, ribbing
may also increase shell strength as well as
surface area. Greater strength would abet sur-
vival during flotation and vagaries of being
swept into solid objects. The loss of ribbing but
not the varix indicates that there is a different
gene(s) governing varix formation.
Floatation during Yangtze River flooding is
a major source of dispersion for Oncomelania
hupensis hupensis and the schistosomes
they transmit. This phenomenon is apparently
not known outside China. During the floods,
snails are lifted off the islands in the Yangtze
and floodplains and floated by the millions
down the river to be deposited on down-
stream floodplains or swept into canals when
the flood gates are opened. The impact on im-
portation into the canals of Hubei Province
has been documented by Xu & Fang (1988)
and Xu et al. (1 989, 1 993), Yang et al. (1 992).
While snails on the islands either float off or
drown, snails on the floodplains often escape
flooding by climbing tree trunks, often to
heights of more than three meters. With refer-
ence to drowning, Oncomelania hupensis is
an amphibious species. The young stay sub-
merged during their early stages of develop-
ment, often floating upside down, feeding on
the surface of quiet water. As adults, the
snails are found out of but near water on the
banks of irrigation ditches and swamps, on
shaded, moist soil. During drought, the adults
move down into the soil and aestlvate. Adults
308 DAVIS ETAL.
HB16 HB14 HB15 HB13 HB18 HB17 HBll
©■
DA
■ в
DC
■ D
■ E
■ F
DH
О
AAT-1
ААГ-2
ACPH-2
- AK-1
APH
О-Э-О-0-Э-С1-О-
■(3-е-
(3-(3-(3C Э 0-Э
■o-
■o-
CK-1
EST-1
EST-2
GPI
HBD-1
ME-1
OCT
PGM-1
PGM-2
FIG. 4. Pie diagrams showing allele frequencies and alternative alleles for each population. Populations are
in rows with HB 11 (=A) at the mouth of the river and HB 1 6 (=G) at the head of the river. The colors are coded
A, B, etc. for alleles A, B, etc. listed in Table 2. For example, solid orange in Fig. 4, HB11 indicates an alter-
native allele B. The dominant allele is A.
ALLOZYME ANALYSES TEST THE TAXONOMIC RELEVANCE OF RIBBING
309
TABLE 8. Matrices of Nei's 1978 genetic distance (above tine diagonal) and Wright's modified 1978 Rogers
genetic distance (below the diagonal).
JIAN
JING
Population
GC
LI
A
D
F
E
G
В
С
MIN
GC
0.035
0.066
0.039
0.003
0.010
0.012
0.001
0.001
0.005
JIAN LI
0.182
0.103
0.072
0.035
0.042
0.045
0.035
0.038
0.040
A
0.247
0.305
0.106
0.068
0.077
0.077
0.066
0.066
0.069
D
0.193
0.258
0.310
0.036
0.035
0.044
0.036
0.043
0.040
F
0.052
0.181
0.250
0.185
0.006
0.010
0.002
0.004
0.006
E
0.100
0.198
0.265
0.184
0.080
0.014
0.010
0.013
0.009
G
0.108
0.205
0.265
0.203
0.100
0.118
0.011
0.013
0.009
В
0.041
0.181
0.246
0.185
0.048
0.098
0.105
0.004
0.007
С
0.040
0.190
0.245
0.201
0.067
0.113
0.113
0.064
0.004
JING MIN
0.071
0.195
0.253
0.196
0.077
0.093
0.108
0.087
0.063
г GCl -Anhui-Ribs
- HB18 -С -Ribs
г- HB 14 -F-Smooth
- HB17 -В -Ribs
— HB22 - Control. Jing Men - Smooth
— HB 15 -Е- Smooth
— HB16 -G -Smooth
— HB6 - Control. Jian Li - Ribs
— HB 13 -D-Smooth
— HBll -A-Ribs
H 1 1-
H 1 1 1 1 1 1 1-
.10 .09 .08 .07 .06 .05 .04 .03
FIG. 5. Phenogram based on Nei's genetic distance.
cannot withstand continual submersion; they
will drown. Because of this, drowning is a
method used to control schistosomiasis is in
some areas of China, including the Miao
River. During the dry season, a control dam is
closed at times, flooding a large area of graz-
ing land with numerous marsh-edged pools
harboring large populations of Oncomelania.
Our site В is in this section of river. The flood-
ing does reduce the sizes of populations in-
undated (personnel communication; Dr. Yang,
Xian-Xiang, Director, Hubei Institute of Schis-
tosomiasis Control).
After revisiting and collecting snails on Lao
Zhou Island in the Yangtze River in Tong Ling
County, Anhui Province, we now understand
.02 .01
.00
why three populations, including the Lao
Zhou, snails did not group with the clusters we
classified as O. h. hupensis, O. h. tangi, and O.
h. robertsoni (Davis et al., 1995) in the Fitch
tree based on allozyme data from 14 On-
comelania populations. The other populations
were Gui Chi (one of our control populations)
from Anhui, and Jian Li from Hubei Province.
These three localities are heavily flooded and
swept during the annual floods. Snails found in
these locations are not populations, really, but
aggregates of snails imported from diverse
areas and deposited with the receding flood-
waters. In Davis et al. (1 995), these three pop-
ulations were considered as possible "hybrid"
populations, because of alleles shared with
310
DAVIS ETAL.
CK'I - Anhui - Ribs
HB 18 -С -Ribs
HB 14 -F -Smooth
UBI 7 -В -Ribs
HB22 - Control. Jing Men - Smooth
HB 1 5 - E - Smooth
HB 16 -G -Smooth
HB6 - Control. Jian Li - Ribs
HB 13 -D -Smooth
HBll -A- Ribs
40 .36 .32 .28 .24 .20 .16 .12 .08 .04 .00
FIG. 6. Phenogram based on Wright's modified Rogers genetic distance.
TABLE 9. Comparison of Nei's D among popula-
tions.
Populations
Mean and S.D.
No.
Above bridge x Below
Bridge
Below Bridge
Above Bridge
Controls y Below Bridge
Controls X Above Bridge
0.038 (±0.035)
0.045 (±0.036)
0.024 (±0.016)
0.041 (±0.039)
0.032 (±0.023)
12
3
6
6
8
Other populations and subspecies, especially
with the upstream robertsoni subspecies. We
here call such populations "genetically unsta-
ble" populations where true population struc-
ture is not reached and HWe is not attained in
most polymorphic loci.
Genetically stable populations, theoreti-
cally, would be defined as those with normal
panmixia, little or no immigration, and all loci
in HWe, that is, populations that are large and
have had stable population structure for many
years. There are three populations that could
arguably fit these criteria; HB-11, HB-22 with
one locus not in HWe due to very low fre-
quency of one allele, HB-18, and HB-15 with
2 of 5 polymorphic loci not in HWe. The popu-
lation at the head of the river, HB-1 6 should be
genetically stable but equals GuiChi in having
57% polymorphic loci not in Hwe.
Theoretically, population HB-11 at the
mouth of the Miao River should be "unstable"
due to annual flooding, but on the basis of H-
W, this is not the case. The mouth of the Miao
River is not on the Yangtze, but on a loop that
branched off the main channel of the Yangtze.
While the lower Miao River is flooded annu-
ally, it may not receive many introductions of
snails from upstream localities. However, as
clearly seen in Figures 5-7, population A is
clearly divergent from all the others.
The Question of Heterozygote Deficiency
Four reasons could account for heterozy-
gote deficiency such as seen here: (1 ) organ-
isms came from two or more populations with
different allele frequencies (Wahlund effect),
(2) natural selection acts against heterozy-
gotes at some loci, (3) there is inbreeding, or
(4) our scoring is in error and is biased against
heterozygotes (Ayala et al.. 1973). Error in
scoring may result in one, or perhaps two loci
being considered not in HWe, especially
where EST is involved and gels can be difficult
to score. However, we consider that 1 and 3
above are probably mostly responsible, al-
though we cannot discount selection against
heterozygotes.
How often is heterozygote deficiency found
in molluscs overall, and such rissoacean
snails as Oncomelania. in particular? It is im-
portant to place Oncomelania. a rissoacean
snail, in context with other molluscs with re-
gard to what one might expect in terms of het-
erozygosity and heterozygote deficiency. Het-
erozygote deficiency is common among
molluscs, especially bivalves where heterozy-
gosity mostly exceeds 0.250 (see references
below). Numerous authors have discussed
this phenomenon in the Bivalva (for example,
Ayala et al.. 1973; Mitton & Koehn, 1973;
Lassen & Turano, 1978; Wilkins, 1978;
ALLOZYME ANALYSES TEST THE TAXONOMIC RELEVANCE OF RIBBING 31 1
В
F
Jing Men
Jian Li
FIG. 7. An unrooted FITCH tree based on Wright's D. Line lengths are proportional to branch lengths.
Gosling & Wilkins, 1981; Mallet & Haley,
1983; Skibinski et al., 1983; Adamkewicz et
al., 1984; Hedgecock & Okazaki, 1984; Singh
& Green, 1984; Koehn & Gaffney, 1984;
Zouros & Foltz, 1984; Diehl & Koehn, 1985;
Mallet et al., 1985; Hoagland, 1986; Volckaert
& Zouros, 1989; Gaffney et al., 1990; Beau-
mont, 1991; Nirchio et al., 1991; Bricelj &
Krause, 1992; Gaffney et al., 1992; Liu et al.,
1995; Gardner et al.. 1996; Marsden et al.,
1996; Michinina & Rebordinos, 1997). Signifi-
cant deviation from HWe in bivalves has been
attributed to Wahlund effect in limited cases,
and natural selection on young just post set-
tlement in many instances.
In bivalves generally, heterozygosity is
much greater than that found in gastropods.
Selander & Ochman (1983) summarized al-
lozyme data relating to the genetic structure
of about 100 species of gastropods. Het-
erozygosity of amphimictic species averaged
about 0.12 and tanged from 0.004 to 0.198.
Issues of selfing and parthenogenesis involve
both aquatic and terrestrial pulmonates, es-
pecially certain slugs in which there is little or
no heterozygosity, or in which there is consid-
erable heterozygote deficiency do not pertain
to Oncomelania. Foltz et al. (1982 a, b, 1984)
provided data on slugs. There are also strong
indications of self fertilization in certain ba-
sommatophoran freshwater pulmonates of
the genera Biomphalaria and Bulinus. taxa in-
volved in the transmission of schistosomes in-
fecting humans in Africa and South America
(Biomphalaria) (Bandoni et al., 1990; Mimp-
foundi & Greer, 1990a, b; Bandoni et al.,
1995; Doumas et al., 1996; Mukaratirwaetal.,
1996). In stylommatophoran pulmonate land
snails, selection seems to operate on some
loci. Emberton (1993) found over-representa-
tion of some rare alleles in the young of wide-
spread populations of some polygyhds. Fal-
niowski et al. (1998) found only one of many
loci of Bradybaena fructicum not to be in
HWe.
Among the "prosobranch" grade, non-hy-
drobioid gastropods, populations of some
species of some genera have been found to be
in HWe (all loci in HWe); Nassarius (Gooch et
al., 1972; Sanjuan et al., 1997), Patella
312
DAVIS ETAL
(Wilkins, 1977), /We/ano/des (Livshits& Fishel-
son, 1983), and some populations of Littorina
(Rolan-Alvarez et al., 1 995). Some have been
out of HWe. including some Littorina (Rolan-
Alvarez et al., 1995) and Sipiionaria (Johnson
& Black, 1 984), both attributed to the Wahland
effect.
Papers involving hydrobioid snails and al-
lozymes have been few, and most of these
have been concerned primarily with genetic
distances and taxon relationships. Heterozy-
gosity in hydrobioid populations is lower than
the average for snails overall; in Oncomelania
it was 0.052 ± 0.030 (N = 14) (Davis et al.,
1994, 1995). Heterozygosities reported for
two Australian genera were 0.038 ± 0.01 6 (N
= 26 populations) and 0.043 ± 0.018 (N = 48
populations) (Ponder et al., 1996; also see
Ponder et al., 1994, 1995). Heterozygosity for
IHydrobia was equally low, reported as 0.008
- 0.074 (Davis et al., 1988, 1989). Haase
(1993) studied one population each of three
species of ¡Hydrobia sensu lato and found that
for one there were heterozygote deficiencies
at virtually all polymorphic loci, whereas in the
other two populations there was low or no
variability. Haase attributed the heterozygote
deficiencies to selection due to parasite pres-
sure; the cases of low to no heterozygosity to
genetic drift and bottlenecks.
To what do we attribute heterozygote defi-
ciencies in this study? There are two probable
explanations. (1) There appears to be a
Wahland effect in "unstable" populations such
as Gui Chi and Jian Li, localities swept by the
Yangtze River annual flooding. There is a mix-
ture of alleles from snails of diverse localities.
Because there are no diagnostic loci that
serve to separate the subspecies, the use of
multidimensional scaling and a Prim network,
or a FITCH unrooted tree aides one to identify
unstable populations. Such trees are freed
from the constraints imposed in a UPGMA
phenogram. For example, Davis et al. (1995:
fig. 9) present a FITCH tree in which there are
three clusters of populations attributed to
three subspecies, but three populations did
not fit into the clusters; on further investiga-
tion, it was apparent that these three acted as
hybrid populations but were actually geneti-
cally unstable populations.
(2) There have been extensive efforts along
the Miao River to control snails using mollus-
caciding and by flooding to drown snails.
However, the snails have not been eradi-
cated, and we have been able to collect snails
from these same localities over several years.
The life span of Oncomelania hupensis is
three to four years (Davis, 1967; van der
Schalle & Davis, 1968), and a female, fertil-
ized just once, can store viable sperm over
the coarse of her lifetime. Accordingly, as fe-
males produce quantities of eggs it is possible
for one or two females to totally repopulate a
locality. Accordingly, we attribute the het-
erozygote deficiencies in the upper Miao
River, in localities where there should be "sta-
ble" populations, to extensive inbreeding. As
Oncomelania hupensis is under continual as-
sault in efforts to control schistosomiasis, and
habitats for these snails are ever changing
due to land-use changes in China, it may be
difficult to locate genetically stable popula-
tions using allozymes.
We are now testing an alternative to al-
lozymes, COI gene sequences. They serve
very well to clearly demonstrate the diver-
gence of the subspecies of Oncomelania
(Davis et al., 1998) but are much more con-
servative within populations. We have come
to expect variation of to 2 base substitutions
in a sequence length of 648 base pairs (0.00
to 0.31%) in isolated and genetically stable
populations, but 4 to 8 (0.62 to 1.23%) or
more base substitutions in unstable popula-
tions, indicating the import of differing geno-
types with flooding. A study on this phenome-
non is in progress.
The Subspecies Question in Mainland China
As discussed above, smooth shelled snails
in the Yangtze River drainage previously clas-
sified as Oncomelania fausti belong to On-
comelania hupensis hupensis. The genetic
relationships among the three subspecies
found on the mainland are graphically por-
trayed in Figure 8 (data from Davis et al.,
1995) (justification for using the polytypic
species designation given in Davis, 1994, and
Davis et al., 1995). All the populations of this
study fall within the variance around the mean
value for O. hupensis hupensis.
There is one more name that requires some
discussion, Oncomelania hupensis guang-
xiensis. named recently by Liu et al. (1981)
from the Xun Jiang (Xun River) drainage,
which becomes the Xi Jiang flowing into
Guangdong Province from Guangxi Province.
The Xi Jiang does not flow to the Yangtze
River, but to the South China Sea near Macao.
Snails classified as guangxiensis are gener-
ally smooth but with sporadic, irregular ap-
pearance of low ribs (Davis et al.. 1 995: figs. 2,
ALLOZYME ANALYSES TEST THE TAXONOMIC RELEVANCE OF RIBBING 313
O. h. robertsoni
O. h. tangi
O. h. hiipensis
FIG. 8. Graphic relationships, in scale, among the three subspecies of Oncomelania hupensis on the main-
land of China. The standard deviations are marked (curved lines).
4A). The shells have a strong vahx. We place
this nominal subspecies in the synonymy of O.
hupensis hupensis because snails studied
electrophoretically by Davis et al. (1995) are
within the vahance ofO. h. hupensis (Davis et
al., 1995: fig. 9) We hypothesize that these
snails reached Guangxi Autonomous Region
via a canal dug some centuries ago connect-
ing the Xiang Jiang River in Hunan Province to
Guangxi, thus connecting the snail-rich Tong
Ting Lake district of Hunnan to the southern
coast. On the basis of allozymes, the Gui Ping
County Guangxi snails from near the conflu-
ence of the Qian Jiang and Yu Jiang that forms
the Xun Jiang are most closely related to a
population from Hunan (Nei's mD of 0.150
from Yue Yang, Hunan).
There are still some unresolved problems
involving smooth shells with a strong varix.
While it seems clear that Oncomelania hupen-
sis hupensis living along the Yangtze and up
into the lower stretches of rivers flowing into
the Yangtze belong to the same subspecies,
whether they have ribbed or smooth shells,
there are some populations in the hills in east-
ern China of coastal provinces, Jiangsu and
Zhejiang, isolated from the Yangtze River
drainage, that are problematic.
Zhou et al. (1995) studied the allozymes of
34 populations from nine provinces They used
16 loci, of which 5 were esterase loci. In
UPGMA clustering of Nei's (1978) "D", they
also found that Sichuan and Yunnan snails
clustered apart from O. hupensis {sensu Davis
et al., 1995), as did the one population they
had from Fujian (O. hupensis tangi). One pop-
ulation with a smooth shell and varix, from
Anhui) clustered with the ribbed-shelled popu-
lations. Using Fitch-Margoliash least-squared
cluster analysis, all smooth-shelled Oncome-
lania clustered together apart from ribbed-
shelled populations, but the Anhui smooth-
shelled population was basal and distinctly
apart, rather intermediate between the smooth
and ribbed-shelled populations. Zhou et al.
(1995) concluded that there were two taxo-
nomically distinct groups: ribbed and smooth-
shelled types. Aside from the Sichuan, Yun-
nan and Fujian populations, they examined
only three other smooth-shelled populations,
the one from Anhui and two from Jiangsu
provinces. They did not considerthe shell mor-
phological and biogeographical differences
that separate hupensis, robertsoni and tangi.
They did consider that smooth-shelled popu-
lation groups might be separated into sub-
species, that is, those from Fujian, Yunnan,
Sichuan, and the hilly region of Jiangsu
314
DAVIS ETAL.
Province at the extreme eastern edge of
China.
Davis et al. (1994) had studied three popu-
lations of Oncomelania from Zhejiang Prov-
ince, like Jiangsu, a coastal province. Two
populations were ribbed and one smooth.
One ribbed and the smooth population were
very close to each other geographically,
whereas the second ribbed population was a
considerable distance from the others. On the
basis of Nei's (1978) "D", the smooth-shelled
population was highly divergent from the
ribbed snails, a result similar to that later
found by Zhou et al. (1995) for the two
smooth-shelled from hilly areas of Jiangsu
Province. All these smooth-shelled popula-
tions had a strong varix. Davis et al. (1995)
found that one smooth-shelled population
from Zhejiang Province clustered closely with
the Yunnan and Sichuan O. h. robertsoni: it
had the least genetic distance with the
Sichuan populations considering all pair-wise
comparisons among populations. Davis et al.
(1995) classified this Zhejiang population as
roberstoni.
It has yet to be determined how these east-
ern hill-dwelling populations relate to smooth-
shelled Miao River snails studied here. These
few special populations require intense study.
Are tfiey truly genetically divergent from O. h.
hupensis? Are they part of the robertsoni
complex but have independently evolved a
varix? Are they a distinct subspecies?
ACKNOWLEDGEMENTS
Figure 3 was made by Zhang Yi from SEM
photographs made by Shen Bing-Gui. The re-
maining figures were produced by Dr.
Thomas Wilke of ANSP who also did the
PHYLIP analysis. We thank Drs. Walter R.
Hoeh, Kent State University, and David. O. F.
Skibinski of the University College of Swan-
sea for reviewing and criticizing this paper.
The work was funded by U.S.A., N.I.H. grants
All 1373 (Davis), and AI 39461 (Shanghai
T.M.R.C).
LITERATURE CITED
ADAMKEWICZ, S. L., S. R. TRAUB & J. R. WALL,
1984, Genetics of the clam Merceneria mercene-
ria: size and genotype. Malacologia. 25:
525-533.
AYALA, F. J., D. HEDGECOCK, G. ZUMWALT & J.
VALENTINE, 1973, Genetic variation in Tridacna
maxima, an ecological analog of some unsuc-
cessful evolutionary lineages. Evolution. 27:
177-191.
BANDONI, S. M., M. MULVEY, D. K. KOECH & E.
S. LOKER, 1990, Genetic structure of Kenyan
populations of Biomptialaria pfeifferi (Gas-
tropoda: Planorbidae). Journal of Molluscan
Studies. 56: 383-392,
BANDONI, S. M., M. MULVEY & E. S. LOCKER,
1995, Intraspecific and interspecific patterns of
allozyme variation among species of Biom-
ptialaria Preston, 1910 (Gastropoda: Planor-
bidae). Biocfiemical Systematics and Ecology.
23:593-616.
BARTSCH. P.. 1925, Some new intermediate fiosts
of the Asiatic human blood fluke. Journal of the
Wastiington Academy of Sciences. 15:71 -73.
BARTSCH, P., 1936, Molluscan intermediate host
of the Asiatic blood fluke Schistosoma japonicum
and a species confused with them. Smithsonian
t\/liscellaneous Collection. 95: 1 -60.
BEAUMONT A. R., 1991, Genetic studies of labo-
ratory reared mussels, Mytilus edulis: heterozy-
gote deficiencies, heterozygosity, and growth. Bi-
ological Journal of the Linnean Society 44:
273-285.
BRICELJ, V. M. & M. K. KRAUSE. 1992, Resource
allocation and population genetics of the bay
scallop, Argopecten irradians irradians: effects of
age and allozyme heterozygosity on reproductive
output. Marine Biology 113: 253-261.
DAVIS, G. M., 1967, The systematic relationship of
Pomatiopsis lapidaria and Oncomelania hupen-
sis formosana (Prosobranchia: Hydrobiidae).
Malacologia. 6: 1-143.
DAVIS, G. M., 1979, The origin and evolution of the
gastropod family Pomatiopsidae, with emphasis
on the Mekong River Triculinae. Monograph of
the Academy of Natural Sciences of Philadelphia.
20: 1-120.
DAVIS, G. M., 1 983, Relative roles of molecular ge-
netics, anatomy, morphometries and ecology in
assessing relationships among North Amehcan
Unionidae (Bivalvia). In: j. oxford & D. rol-
LiNSON, eds.. Protein polymorphism: adaptive &
taxonomic significance. Systematics Associa-
tion Special Volume, 24: 193-122. Academic
Press.
DAVIS, G. M., 1994. Molecular genetics and taxo-
nomic discrimination. In: м. G. harasewych & s.
TILLER, eds. Molecular techniques and molluscan
phylogeny. Proceedings of a Symposium. 11th In-
ternational Malacological Congress. Siena. Italy.
The Nautilus. 8. Supplement 2: 3-23
DAVIS, G. M., С E. CHEN, X. P. ZHENG. S. H. YU
& L. LI. 1994, Molecular genetic and anatomical
relationships among pomatiopsid (Gastropoda:
Prosobranchia) genera from southern China.
Proceedings of the Academy of Natural Sciences
of Philadelphia. 145: 191-207.
DAVIS, G. M., V. E. FORBES & G. R. LOPEZ. 1988,
Species status of northeastern American Hydro-
bia (Gastropoda: Prosobranchia): ecology mor-
ALLOZYME ANALYSES TEST THE TAXONOMIC RELEVANCE OF RIBBING 31 5
phology and molecular genetics. Proceedings of
the Academy of Natural Sciences of Pfiiladelptiia,
140; 191-246.
DAVIS. G.M.& S.J. H. FULLER, 1981, Genetic re-
lationships among Recent Unionacea (Bivalvia)
ot North America. l\/lalacologia, 20: 217-253.
DAVIS, G. M., W. H. HEARD, S. J. H. FULLER & С
HESTERMAN, 1981, Molecular genetics and
speciation in Elliptio and its relationship to other
taxa of North American Unionidae (Bivalvia). Bio-
logical Journal of tfie Linnean Society, 15: 131 -
150.
DAVIS, G. M., M. McKEE & G. R. LOPEZ, 1989,
The Identity of Hydrobia trúncala (Gastropoda:
Hydrobiinae): comparative anatomy, molecular
genetics, ecology. Proceedings of the Academy
of Natural Sciences of Philadelphia. 141:
333-359.
DAVIS. G. M. & M. D. RUFF. 1973. Oncomelana
hupensis (Gastropoda: Hydrobiidae): hybridiza-
tion, genetics, and transmission of Schistosoma
japonicum. Malacological Review. 6: 181 -1 97.
DAVIS, G. M., T WILKE, С M. SPOLSKY CHI-
PING QIU, D. QIU, M. Y XIA, Y ZHANG & G.
ROSENBERG, 1998, Cytochrome oxidase I-
based phylogenetic relationships among the
Pomatiopsidae, Hydrobiidae, Rissoidae, and
Truncatellidae (Gastropoda: Caenogastropoda:
Rissoacea). Malacologia. 40(1/2): 251-266.
DAVIS, G. M., Y ZHANG, Y H. GUO & С. M. SPOL-
SKY, 1995, Population genetics and systematic
status of Oncomelania hupensis (Gastropoda:
Pomatiopsidae) throughout China. Malacologia.
37: 133-156.
DIEHL, W. J. & R. K. KOEHN, 1985. Multiple-locus
heterozygosity, mortality, and growth in a cohort
of !\/lytilus edulis. Marine Biology 88: 265-271 .
DILLON, R. & G. M. DAVIS, 1980, The Goniobasis
of southern Virginia and northwestern North Car-
olina: genetic & shell morphometric relationships.
Malacologia. 20: 83-98.
DOUMAS, С, F. VIARD. A. F. PERNOT В. DELAY
& P. JARNE. 1996. Inbreeding depression, neu-
tral polymorphism, and copulatory behavior in
freshwater snails: a self-fertilization syndrome.
Evolution. 50: 1908-1918.
EMBERTON, K., 1993, Over-representation of rare
alleles in juveniles and lack of pattern in geo-
graphic distributions of alleles in land snails.
Malacologia. 35(2): 361 -369.
FALNIOWSKI, A., A. KOZIK, M. SZAROWSKA, M.
RAPALA-KOZIK & I. TURYANA, 1998, Morpho-
logical and allozymic polymorphism and differ-
ences among local populations of Bradybaena
fruticum (O. F. Muller, 1777) (Gastropoda: Sty-
lommatophora: Helocoidea). Malacologia. 35:
371-388.
FELSENSTEIN, J., 1989, PHYLIP— Phylogeny in-
ference package (version 3.2). Cladistics. 5:
164-166.
FOLTZ, D. W., H. OCHMAN, J. S. JONES, S. M.
EVANGELISTI & R. SELANDER, 1982a, Genetic
population structure and breeding systems in ar-
ionid slugs (Mollusca: Pulmonata). Biological
Journal of the Linnean Society, 1 7: 225-241 .
FOLTZ, D. W., H. OCHMAN & R. SELANDER,
1984, Genetic diversity and breeding systems in
terrestrial slugs of the families Limacidae and Ar-
ionidae. Malacologia. 25: 593-605.
FOLTZ, D. W., B. M. SCHAITKIN & R. SELANDER.
1982b, Gametic disequilibrium in the self fertiliz-
ing slug Dereceras laeve. Evolution. 36: 80-85.
GAFFNEY P. M., С V. DAVIS & R. O. HAWES,
1992, Assessment of drift and selection in hatch-
ery populations of oysters {Crassostrea vir-
ginica). Aquaculture. 105: 1 -20.
GAFFNEY P. M., T M. SCOTT R. K. KOEHN & W.
J. DIEHL, 1990, Interrelationships of heterozy-
gosity, growth rate and heterozygote deficiencies
in the coot clam, Mulinia lateralis. Genetics. 124:
687-699.
GARDNER, J. P. A., R. F. EYLES & A. PANDE,
1996, Biochemical-genetic variation in a wild and
a cultured population of the greenshell mussel,
Perna canaliculatus. New Zealand Journal of Ma-
rine and Freshwater Research. 30: 435-441 .
GOOCH, J L., B, S. SMITH & D. KNUPP. 1972. Re-
gional survey of gene frequencies in the mud
snail Nassahus obsoletus. Biological Bulletin.
142:36-48.
GOSLING. E. M. & N. P. WILKINS. 1981. Ecologi-
cal genetics of the mussels Mytilus edulis and M.
galloprovincialis on Irish coasts. Marine Ecology
Progress Series. 4: 221 -227.
GREDLER, P. V.. 1881, Zur Concylien-Fauna von
China. III. Stück. Jahrbücher der Deutschen
Malakozoologischen Gesellschaft. 8: 119-132.
HAASE, M., 1993, The genetic differentiation in
three species of the genus Hydrobia and system-
atic implications (Caenogastropoda: Hydrobi-
idae). Malacologia. 35(2): 389-398.
HEDGECOCK, D. & N. B. OKAZAKI, 1984, Genetic
diversity within and between populations of
American oysters (Crassostrea). Malacologia.
25: 535-549.
HOAGLAND, K. E.. 1986. Genetic variation in
seven wood-boring teredinid and pholadid bi-
valves with different patterns of life history and
dispersal. Malacologia. 27: 323-339.
JOHNSON, M. S. & R. BLACK, 1984, The Wahlund
effect and the geographic scale of variation in the
intertidal limpet Siphonaria sp. Mahne Biology
79: 295-302.
KOEHN, R. K. & P. M. GAFFNEY, 1984, Genetic
heterozygosity and growth rate in Mytilus edulis.
Mahne Biology 82: 1-7.
LASSEN, H. H. & F. J. TURANO, 1978, Clinal vari-
ation and heterozygote deficit at the Lap locus in
Mytilus edulis. Mahne Biology 49: 245-254.
LIU, L. L-, K. SOONG & С P. CHEN, 1995, Al-
lozyme variation in the Pacific oyster Crassostrea
gigas along the coast of Taiwan. Zoological Stud-
ies. ЗЛ: 177-182.
LIU, Y Y, T К. LOU, Y X. WANG & W. Z. ZHANG,
1981, Subspecific differentiation of oncomelaniid
snails. Acta Hydrobiologica Sínica. 6: 253-266.
316
DAVIS ET AL.
LIVSHITS, G. & L. FISHELSON, 1983, Biology and
reproduction of the freshwater snail Melanoides
tuberculata Gastropoda Prosobranchia in Israel.
Israel Journal of Zoology. 32: 21 -36.
MALLET, A. L. & L. E. HALEY, 1983, Effects of in-
breeding and spat formation in the American oys-
ter. Aquaculture. 33: 229-235.
MALLET, A. L., E. ZOUROS, K. E. GARTNER
KEPKAY, K. R. FREEMAN & L. M. DICKIE, 1985,
Larval viability and heterozygote deficiency in
populations of marine bivalves: evidence from
pair matings of mussels. Marine Biology. 87:
165-172.
MARSDEN, J. E., A. P. SPIDLE & B. MAY 1996,
Review of genetic studies of Dreissena spp.
American Zoologist. 36: 259-270.
MICHININA, S. R. & L. REBORDINOS, 1997, Ge-
netic differentiation in marine and estuarine nat-
ural populations of Crassostrea angulata. Journal
of the Marine Biological Association of the United
Kingdom. 77: 773-784.
MIMPFOUNDI, R. & G. GREER, 1990a, Allozyme
variation among populations of Biomphalaria
pfeifferi (Krauss, 1848) (Gastropoda: Planor-
bidae) in Cameroon. Journal of Molluscen Stud-
ies, 56: 46^ -468.
MIMPFOUNDI, R. & G. GREER, 1990b, Allozyme
variation among populations of Bulinus forskali
(Ehrenberg, 1831) (Gastropoda: Planorbidae) in
Cameroon, Journal of Molluscan Studies. 56:
363-372.
MITTON, J. B. & R. K. KOEHN, 1973, Population
genetics of marine pelecypods. III. Epistasis be-
tween functionally related isoenzymes of Mytilus
edulis. Genetics, 73: 487-496.
MUKARATIRWA, S., H. R. SIEGISMUND, T. K.
KRISTENSEN & S. K. CHANDIWANA, 1996, Ge-
netic structure and parasite compatability of Buli-
nus globosus (Gastropoda; Planorbidae) from
two areas of different endemicity of Schistosoma
haematobium in Zimbabwe. International Journal
for Parasitology 26: 269-280.
NEI, M., 1978, Estimation of average heterozygos-
ity and genetic distance from a small number of
individuals. Genetics. 89: 583-590.
NIRCHIO, M., J. E. PEREZ & H. CEQUEA, 1991,
Allozyme variation of Lap loci in Crassostrea rhi-
zophorae in relation to temperature and/or salin-
ity. Scientia Marina. 55: 563-567.
PONDER, W. H., D. J. COLGAN, G. A. CLARK,
A. С MILLER & T. TERIZ, 1 994, Microgeographic,
genetic and morphological differentiation of fresh-
watery snails - the Hydrobiidae of Wilson's
Promontory, Victoria, South-eastern Australia.
Australian Journal of Zoology 45: 557-678.
PONDER, W. H., D. J. COLGAN, T. TERZIS, S. A.
CLARK & A. С MILLER, 1996, Three new mor-
phologically and genetically determined species
of hydrobiid gastropods from Dalhousie Springs,
northern South Australia, with the description of a
new genus. Molluscan Research. 17: 49-109.
PONDER, W. H., P. EGGLER & D. J. COLGAN,
1995, Genetic differentiation of aquatic snails
(Gastropoda: Hydrobiidae) from artesian springs
in arid Australia. Biological Journal of the Linnean
Society 56: 553-596.
ROLAN-ALVAREZ, E., G. ALVAREZ & С ZAPATA,
1995, Distinct genetic subdivisions in sympatric
and sibling species of the genus Littorina (Gas-
tropoda: Littorinidae). Heredity, 74: 1 -9.
SANJUAN, A., M. PEREZ-LOSADA & E. ROLAN,
1997, Allozyme evidence for cryptic speciation in
sympatric populations of Nassahus spp. (Mol-
lusca: Gastropoda). Journal of the Marine Biolog-
ical Association of the United Kingdom. 77:
773-784.
SELANDER, R. & H. OCHMAN, 1983, The genetic
structure of populations as illustrated by mol-
luscs, pp. 93-100, in Isozymes: topics in biologi-
cal and medical research. New York, Alan R. Liss.
SINGH, S. M. & R. H. GREEN, 1984, Excess of al-
lozyme homozygosity in marine molluscs and its
possible biological significance. Malacologia. 25:
569-581.
SKIBINSKI, D. O. F., J. A. BREADMORE & T. F.
CROSS, 1983, Aspects of the population genet-
ics of Mytilus (Mytilidae: Mollusca) in the British
Isles. Biological Journal of the Linnean Society.
19: 137-184.
SWOFFORD, D. L. & R. SELANDER, 1 981 , Biosys-
1. A computer program for the analysis of allelic
vahation in genetics. 60 pp. University of Illinois
VAN DER SCHALIE, H. & G. M. DAVIS, 1968, Cul-
turing Oncomelania snails (Prosobranchia: Hy-
drobiidae) for studies of oriental schistosomiasis.
Malacologia. 6: 321 -367.
VOLCKAERT F. & E. ZOUROS. 1989, Allozymic
and physiological variation in the scallop Pla-
copecten magellanicus and a general model for
the effects of heterozygosity on fitness in marine
molluscs. Mahne Biology. 103: 51-62.
WANG, S. H., L. HE & Z. B. KANG, 1998, Relative
relationship study of rib-shell Oncomelania and
smooth-shell Oncomelania in Hubei Province.
Chinese Schistosomiasis Control. 10: 267-269.
WILKINS, N. R, 1977, Genetic vahability in littoral
gastropods: phosphoglucose isomerase and
phosphoglucomutase in Patella vulgata and P.
áspera. Mahne Biology. 40: 151 -155.
WILKINS, N. P., 1978, Length-correlated changes
in heterozygosity at an enzyme locus in the scal-
lop {Pectén maximus L.). Animal Blood Groups
Biochemistry and Genetics. 9: 69-77.
WRIGHT, S., 1978, Variability within & among nat-
ural populations. Chicago: University of Chicago
Press.
XU, X. & T. FANG, 1988, Observation on collecting
Oncomelania hupensis\Nher\ opening the sluices
during flood season in Yangtze River. Kasetsart
Journal (Natural Science). 22: 251 -260.
XU. X., T FANG, X. YANG, С YU, С XIAO, F. LIU
& Z. BAO. 1993, Experiment of prevention of
snail spreading by blocking nets at irrigation
floodgates along the Yangtze River. Chinese
Journal of Zoology 28: 1 2- 1 5.
XU, X., T FANG, С YU, JIANG & J. CAO, 1989,
ALLOZYME ANALYSES TEST THE TAXONOMIC RELEVANCE OF RIBBING
317
Relationship between sluicing for irrigation and
spreading of Oncomelania hupensis in Hubei
Province. Kasetsart Journal (Natural Science)
23:281-286.
YANG, X., X. XU, T FANG & С VU, 1992, Experi-
mental observation on the prevention of snail
spreading by reverse siphoning. Chinese Journal
of Zoology. 27: 12-14.
ZHOU, X., L. SUN, Q. HONG, XU OIU, A LU Z
WU & T K. KRISTENSEN, 1995. Population ge-
netics of Oncomelania spp. in mainland China. I.
Genetic variations among populations of On-
comelania spp. Chinese Schistosomiasis Con-
trol, 7: 67-71 .
ZOUROS, E. & D. W. FOLTZ, 1984, Possible ex-
planations of heterozygote deficiency in bivalve
populations. Malacologia, 25(2): 583-591.
Revised ms. accepted 1 May 1999
. VOL. 41, NO. 1 1999
MALACOLOGIA
Infernational Journal of Malacology
Revista Internacional de Malacologia
Journal International de Malacologie
Международный Журнал Малакологии
Internationale Malakologische Zeitschrift
Publication dates
Vol. 31, No. 1 29 Dec. 1989
Vol. 31, No. 2 28 May 1990
Vol. 32, No. 1 30 Nov. 1990
Vol. 32, No. 2 7 Jun. 1991
Vol. 33, No. 1-2 6 Sep. 1991
Vol. 34, No. 1-2 9 Sep. 1992
Vol. 35, No. 1 14 Jul. 1993
Vol. 35, No. 2 2 Dec. 1993
Vol. 36, No. 1-2 8 Jan. 1995
Vol. 37, No. 1 13 Nov. 1995
Vol. 37, No. 2 8 Mar. 1996
Vol.38, No. 1-2 17 Dec. 1996
Vol. 39, No. 1-2 13 May 1998
Vol.40. No. 1-2 17 Dec. 1998
VOL. 41, NO. 1 MALACOLOGIA 1999
CONTENTS
R. ARAUJO, M. A. RAMOS, & R. MOLINET
Growth Pattern and Dynamics of a Southern Peripheral Population of
Pisidium amnicum (Müller, 1774) Bivalvia: Sphaehidae) in Spain 119
ALEXIA BLANC, GUY PINCZON DU SEL, & JACOUES DAGUZAN
Relationships Between Length of Prey/Predator for the Most Important Prey
of the Cuttlefish Sepia officinalis L. (Mollusca: Cephalopoda) 139
GEORGE M. DAVIS, Yl ZHANG, XINGJIANG XU, & XIANXIANG YANG
Allozyme Analyses Test the Taxonomic Relevance of Ribbing in Chinese
Oncomelania (Gastropoda: Rissoacea: Pomatiopsidae) 297
VASILIS K. DIMITRIADIS & ELIZABETH B. ANDREWS
Ultrastructural and Cytochemical Study of the Kidney and Nephridial Gland
Cells of the Marine Prosobranch Mollusc Nucella lapillus (L.) in Relation
to Function 1 87
MICHAELA. HOGGARTH
Descriptions of Some of the Glochidia of the Unionidae (Mollusca:Bivalvia) 1
GENNADY M. KAMENEV & VICTOR A. NADTOCHY
Species of Macoma (Bivalvia: Tellinidae) from the Pacific Coast of Russia,
Previously Described as Abrina (Bivalvia: Semelidae) 209
ALEXANDRA LEITÄO, PIERRE BOUDRY JEAN-PHILIPPE LABAT, &
CATHERINE THIRIOT-QUIÉVREUX
Comparative Karyological Study of Cupped Oyster Species 175
KAREN D. MCCOY
Sampling Terrestrial Gastropod Communities: Using Estimates of Species
Richness and Diversity to Compare Two Methods 271
BENITO MUÑOZ, ARTURO ALMODOVAR , & JOSÉ R. ARREBOLA
Ganula gadirana n. sp., A New Hygromiidae from Southern Spain
(Pulmonata: Helicoidea) 163
JEFFREY С NEKOLA
Terrestrial Gastropod Richness of Carbonate Cliff and Associated Habitats in
the Great Lakes Region of North America 231
JEFFREY С NEKOLA & TAMARA M. SMITH
Terrestrial Gastropod Richness Patterns in Wisconsin Carbonate Cliff
Communities 253
M. SCHILTHUIZEN, J. J. VERNEULEN, G. W. H. DAVISON, & E. GITTENBERGER
Population Structure in a Snail Species from Isolated Malaysian Limestone
Hills, Inferred from DNA Sequences 283
ANNE THEENHAUS & MATTHIAS SCHAEFER
Energetics of the Red Slug Arion rufus (Gastropoda) and of the Gastropod
Community in a Beech Forest on Limestone 197
SUSAN A. TOCZYLOWSKI, R. DOUGLAS HUNTER, & LISA M. ARMES
The Role of Substratum Stability in Determining Zebra Mussel Load
on Unionids 151
GEERATJ. VERMEIJ
The Accumulation of Taxonomic Knowledge: The History of Species
Descriptions of Some Predatory Gastropods 147
WHY NOT SUBSCRIBE TO MALACOLOGIA
ORDER FORM
Your name and address
Send U.S. $36.00 for a personal subscription (one volume) or U.S. $55.00 for an
institutional subscription. Make checks payable to "MALACOLOGIA."
Address: Malacologia
Department of Malacology
Academy of Natural Sciences
1900 Benjamin Franklin Parkway
Philadelphia, PA 19103-1195, U.S.A.
AWARDS FOR STUDY AT
The Academy of Natural Sciences of Pfiiladelphia
The Academy of Natural Sciences of Philadelphia, through its Jessup and
McHenry funds, makes available each year a limited number of awards to support
students pursuing natural history studies at the Academy. These awards are primar-
ily intended to assist predoctoral and immediate postdoctoral students. Awards usu-
ally include a stipend to help defray living expenses, and support for travel to and
from the Academy. Application deadlines are 1 March and 1 October each year.
Further information may be obtained by writing to: Chairman, Jessup-McHenry
Award Committee, Academy of Natural Sciences of Philadelphia, 1900 Benjamin
Franklin Parkway, Philadelphia, Pennsylvania 19103-1195, U.S.A.
41 NO. 1
MALACOLGIA
1999
INSTRUCTIONS FOR AUTHORS
1. MALACOLOGIA publishes original re-
search on the Mollusca that is of high quality
and of broad international interest. Papers
combining synthesis with innovation are par-
ticularly desired. While publishing symposia
from time to time, MALACOLOGIA encour-
ages submission of single manuscripts on
diverse topics. Papers of local geographical
or systematic interest should be submitted
elsewhere, as should papers whose primary
thrust is physiology or biochemistry. Nearly all
branches of malacology are represented on
the pages of MALACOLOGIA.
2. Manuscripts submitted for publication
are received with the tacit understanding that
they have not been submitted or published
elsewhere in whole or in part.
3. Manuscripts must be in English, but
may include an expanded abstract in a foreign
language as well as the usual brief abstract in
English. Both American and British spellings
are allowed.
4. Unless indicated otherwise below, con-
tributors should follow the recommendations
in the Council of Biology Editors (CBE) Style
Manual, the latest edition (November 1994,
ISBN #0521-471-540) of which is available
for U.S. $39.95 plus $4.00 shipping from
Cambridge University Press, Order Depart-
ment, 110 Midland Avenue, Port Chester, NY
10573, U.S.A.
5. Be brief.
6. Manuscripts must be typed on one side
of good quality white paper, double- spaced
throughout (including the references, tables
and figure captions), and with ample margins.
Tables and figure captions should be typed on
separate pages and put at the end of the man-
uscript. Make the hierarchy of headings within
the text simple and consistent. Avoid internal
page references (which have to be added in
page proof).
7. Choose a running title (a shortened
version of the main title) of fewer than 50 let-
ters and spaces.
8. Provide a concise and informative Ab-
stract summarizing not only contents but re-
sults. A separate summary generally is super-
fluous.
9. Supply between five and eight key
(topic) words to go at the end of the Abstract.
10. Use the metric system throughout.
Micron should be abbreviated jam.
11. Illustrations are printed either in one
column or the full width of a page of the journal,
so plan accordingly. The maximum size of a
printed figure is 1 3.5 x 20.0 cm (preferably not
as tall as this so that the caption does not have
to be on the opposite page).
12. Drawings and lettering must be dark
black on white, blue tracing, or blue-lined
paper. Lines, stippling, letters and numbers
should be thick enough to allow reduction by
V2 or V3. Letters and numbers should be at
least 3 mm high after reduction. Several draw-
ings or photographs may be grouped together
to fit a page. Photographs are to be high con-
trast. High contrast is especially important for
histological photographs.
13. All illustrations are to be numbered
sequentially as figures (not grouped as plates
or as lettered subsehes), and are to be
arranged as closely as possible to the order in
which they are first cited in the text. Each fig-
ure must be cited in the text.
14. Scale lines are required for all nondi-
agrammatic figures, and should be conve-
nient lengths (e.g., "200 jam," not "163 \хгг\").
Magnifications in captions are not accept-
able.
15. All illustrations should be mounted,
numbered, labeled or lettered, i.e. ready for
the printer.
16. A caption should summahze what is
shown in an illustration, and should not dupli-
cate information given in the text. Each let-
tered abbreviation labeling an individual fea-
ture in a figure must either be explained in
each caption (listed alphabetically), or be
grouped in one alphabetic sequence after the
Methods section. Use the latter method if
many abbreviations are repeated on different
figures.
17. Tables are to be used sparingly, and
vertical lines not at all.
18. References cited in the text must be in
the Literature Cited section and vice versa.
Follow a recent issue of MALACOLOGIA for
bibliographic style, noting especially that seri-
als are cited unabbreviated. Supply pagina-
tion for books. Supply information on plates,
etc., only if they are not included in the pagi-
nation.
19. In systematic papers, synonymies
should not give complete citations but should
relate by author, date and page to the Litera-
ture Cited section.
20. For systematic papers, all new type-
specimens must be deposited in museums
where they may be studied by other scien-
tists. Likewise MALACOLOGIA requires that
voucher specimens upon which a paper is
based be deposited in a museum where they
may eventually be reidentified.
21. Submit each manuscript in triplicate.
The second and third copies can be repro-
ductions.
22. Authors who want illustrations returned
should request this at the time of ordering
reprints. Otherwise, illustrations will be main-
tained for six months only after publication.
23. An author's address should include an
e-mail address.
REPRINTS AND PAGE COSTS
24. When 100 or more reprints are
ordered, an author receives 25 additional
copies free. Reprints must be ordered at the
time proof is returned to the Editorial Office.
Later orders cannot be considered. For each
authors' change in page proof, the cost is U.S.
$3.00 or more.
25. When an article is 10 or more printed
pages long, MALACOLOGIA requests that an
author pay part of the publication costs if grant
or institutional support is available.
SUBSCRIPTION PURCHASE
26. Effective Jan. 1995, personal subscrip-
tions are U.S. $36.00. Institutional, agency or
dealership subscriptions are U.S. $55.00.
Back issues and single volumes: $36.00 for
individual purchaser; $55.00 for institutional
or agency purchasers. There is a one dollar
handling charge per volume for all purchases
of single volumes. Address inquiries to the
Subscription Office.
VOL. 41, NO. 1 MALACOLOGIA 1999
CONTENTS
MICHAELA. HOGGARTH
Descriptions of Some of tlie Glochidia of the Unionidae (Mollusca:Bivalvia) 1
R. ARAUJO, M. A. RAfVlOS, & R. MOLINET
Growth Pattern and Dynarnicsjr of a Southern Peripheral Population of
Pisidium amnicum (Müller, 1774|) Bivalvia: Sphaehidae) in Spain 119
ALEXIA BLANC, GUY PINCZON DU SEL, & JACQUES DAGUZAN
Relationships Between Length of Prey/Predator for the Most Important Prey
of the Cuttlefish Sepia officinalis L. (Mollusca: Cephalopoda) 139
GEERATJ. VERMEIJ
The Accumulation of Taxonomic Knowledge: The History of Species
Descriptions of Some Predatory Gastropods 147
SUSAN A. TOCZYLOWSKI, R. DOUGLAS HUNTER, & LISA M. ARMES
The Role of Substratum Stability in Determining Zebra Mussel Load
on Unionids 151
BENITO MUÑOZ, ARTURO ALMODOVAR , & JOSÉ R. ARREBOLA
Ganula gadirana n. sp., A New Hygromiidae from Southern Spain
(Pulmonata: Helicoidea) 163
ALEXANDRA LEITÄO, PIERRE BOUDRY JEAN-PHILIPPE LABAT, &
CATHERINE THIRIOT-QUIÉVREUX
Comparative Karyological Study of Cupped Oyster Species 175
VASILIS K. DIMITRIADIS & ELIZABETH B. ANDREWS
Ultrastructural and Cytochemical Study of the Kidney and Nephridial Gland
Cells of the Marine Prosobranch Mollusc Nucella lapillus (L.) in Relation
to Function 1 87
ANNE THEENHAUS & MATTHIAS SCHAEFER
Energetics of the Red Slug Arion rufus (Gastropoda) and of the Gastropod
Community in a Beech Forest on Limestone 197
GENNADY M. KAMENEV & VICTOR A. NADTOCHY
Species of l\/lacoma (Bivalvia: Tellinidae) from the Pacific Coast of Russia,
Previously Described as Abrina (Bivalvia: Semelidae) 209
JEFFREY С NEKOLA
Terrestrial Gastropod Richness of Carbonate Cliff and Associated Habitats in
the Great Lakes Region of North America 231
JEFFREY С NEKOLA & TAMARA M. SMITH
Terrestrial Gastropod Richness Patterns in Wisconsin Carbonate Cliff
Communities 253
KAREN D. MCCOY
Sampling Terrestrial Gastropod Communities: Using Estimates of Species
Richness and Diversity to Compare Two Methods 271
M. SCHILTHUIZEN, J. J. VERNEULEN, G. W. H. DAVISON, & E. GITTENBERGER
Population Structure in a Snail Species from Isolated Malaysian Limestone
Hills, Inferred from DNA Sequences 283
GEORGE M. DAVIS, Yl ZHANG, XINGJIANG XU, & XIANXIANG YANG
Allozyme Analyses Test the Taxonomic Relevance of Ribbing in Chinese
Oncomelania (Gastropoda: Rissoacea: Pomatiopsidae) 297
VOL. 41, NO. 2 1999
MALACOLOGIA
International Journal of Malacology
INTERACTIONS BETWEEN MAN AND MOLLUSCS
UNITAS MALACOLOGICA-AMERICAN MALACOLOGIGAL SOCIETY SYMPOSIUM
WASHINGTON, D.C.
26-30 JULY 1998
MALACOLOGIA
EDITOR-IN-CHIEF:
GEORGE M. DAVIS
Editorial and Subscription Offices:
Department of Malacology
The Academy of Natural Sciences of Philadelphia
1900 Benjamin Franklin Parkway
Philadelphia, Pennsylvania 19103-1195, U.S.A.
EUGENE COAN
California Academy of Sciences
San Francisco, CA
Co-Editors:
Assistant Managing Editor:
CARYL HESTERMAN
Associate Editor:
JOHN B. BURCH
University of Michigan
Ann Arbor
CAROL JONES
Denver, CO
MALACOLOGIA is published by the INSTITUTE OF MALACOLOGY, the Sponsor Members of
which (also serving as editors) are:
RÜDIGER BIELER
President
Field Museum, Chicago
JOHN BURCH
MELBOURNE R. CARRIKER
University of Delaware, Lewes
GEORGE M. DAVIS
Secretary and Treasurer
CAROLE S. HICKMAN
President Elect
University of California, Berkeley
ALAN KOHN
University of Washington, Seattle
JAMES NYBAKKEN
Vice President
Moss Landing Marine Laboratory, California
CLYDE F E. ROPER
Smithsonian Institution, Washington, D.C.
SHI-KUEI WU
University of Colorado Museum, Boulder
Participating Members
EDMUND GITTENBERGER
Secretary, UNITAS MALACOLOGICA
Rijksmuseum van Natuurlijke
Historie
Leiden, Netherlands
JACKIE L. VAN GOETHEM
Treasurer, UNITAS MALACOLOGICA
Koninklijk Belgisch Instituut
voor Natuurwetenschappen
Brüssel, Belgium
Emeritus Members
J. FRANCES ALLEN, Emérita
Environmental Protection Agency
Washington, D.C.
KENNETH J. BOSS
Museum of Comparative Zoology
Cambridge, Massachusetts
ROBERT ROBERTSON
The Academy of Natural Sciences
Philadelphia, Pennsylvania
W. D. RUSSELL-HUNTER
Easton, Maryland
Copyright © 1999 by the Institute of Malacology
ISSN: 0076-2997
1999
EDITORIAL BOARD
J.A.ALLEN
Marine Biological Station
Millport, United Kingdom
E. E. BINDER
Museum d'Histoire Naturelle
Geneve, Switzerland
A.J.CAIN
University of Liverpool
United Kingdom
P. CALOW
University of Sheffield
United Kingdom
J. G. CARTER
University of North Carolina
Chapel Hill, U.S.A.
R. COWIE
Bishop Museum
Honolulu, HI., U.S.A.
A. H. CLARKE, Jr.
Portland, Texas, U.S.A.
B. С CLARKE
University of Nottingham
United Kingdom
R. DILLON
College of Charleston
SC, U.S.A.
C.J. DUNCAN
University of Liverpool
United Kingdom
D.J. EERNISSE
California State University
Fullerton, U.S.A.
E. GITTENBERGER
Rijksmuseum van Natuurlijke Historie
Leiden, Netherlands
F. GIUSTI
Universita di Siena, Italy
A. N. GOLIKOV
Zoological Institute
St. Petersburg, Russia
S.J.GOULD
Harvard University
Cambridge, Mass., U.S.A.
A. V. GROSSU
Universitatea Bucuresti
Romania
J. HABE
Tokai University
Shimizu, Japan
R. HANLON
Marine Biological Laboratory
Woods Hole, Mass., U.S.A.
J. A. HENDRICKSON, Jr.
Academy of Natural Sciences
Philadelphia, PA, U.S.A.
D. M. HILLIS
University of Texas
Austin, U.S.A.
K. E. HOAGLAND
Council for Undergraduate Research
Washington, DC, U.S.A.
B. HUBENDICK
Naturhistoriska Museet
Göteborg, Sweden
S. HUNT
Lancashire
United Kingdom
R. JANSSEN
Forschungsinstitut Senckenberg,
Frankfurt am Main, Germany
R. N. KILBURN
Natal Museum
Pietermaritzburg, South Africa
M. A. KLAPPENBACH
Museo Nacional de Historia Natural
Montevideo, Uruguay
J. KNUDSEN
Zoologisk Institut Museum
Kobenhavn, Denmark
A. LUCAS
Faculte des Sciences
Brest, France
с. MEIER-BROOK
Tropenmedizinisches Institut
Tubingen. Germany
H. К, MIENIS
Hebrew University of Jerusalem
Israel
J. E. MORTON
The University
Auckland, New Zealand
J. J. MURRAY, Jr.
University of Virginia
Charlottesville, U.S.A.
R. N ATAR AJAN
Marine Biological Station
Porto Novo. India
S. G. SEGERSTRALE
Institute of Manne Research
Helsinki. Finland
A. STAÑCZYKOWSKA
Siedlce, Poland
F. STARMÜHLNER
Zoologisches Institut der Universität
Wien. Austria
Y I. STAROBOGATOV
Zoological Institute
St. Petersburg, Russia
W. STREIFE
Universite de Caen
France
DIARMAIDO'FOIGHIL
University of Michigan
Ann Arbor, U.S.A.
J. STUARDO
Universidad de Chile
Valparaiso
J. 0KLAND
University of Oslo
Norway
J. OKUTANI
University of Fisheries
Tokyo, Japan
W. L. PARAENSE
Instituto Oswaldo Cruz, Rio de Janeiro
Brazil
J. J. PARODIZ
Carnegie Museum
Pittsburgh, U.S.A.
J. P. POINTIER
Ecole Pratique des Hautes Etudes
Perpignan Cedex, France
W. F. PONDER
Australian Museum
Sydney
01 Z. Y
Academia Sínica
Qingdao, People's Republic of China
D. G. REID
The Natural History Museum
London, United Kingdom
S.TILLIER
Museum National d'Histoire Naturelle
Paris. France
R. D.TURNER
Harvard University
Cambridge, Mass., U.S.A.
J.A.M. VAN DEN BIGGELAAR
University of Utrecht
The Netherlands
N. H. VERDONK
Rijksuniversiteit
Utrecht, Netherlands
ANDERS WAREN
Swedish Museum of Natural History
Stockholm. Sweden
B. R.WILSON
Dept. Conservation and Land Management
Kallaroo. Western Australia
H. ZEISSLER
Leipzig. Germany
A. ZILCH
Forschungsinstitut Senckenberg
Frankfurt am Main, Germany
MALACOLOGIA, 1999, 41(2): 319
UMITAS MALACOLOGICA-AMERICAN MALACOLOGICAL SOCIETY SYMPOSIUM
INTERACTIONS BETWEEN MAN AND MOLLUSCS
26-30 JULY 1998
WASHINGTON, D.C.
Organizing Committee
Philippe Bouchet
George M. Davis
Gerardo R. Vasta
INTRODUCTION TO THE SYMPOSIUM ON INTERACTIONS
BETWEEN MAN AND MOLLUSCS
GEORGE M. DAVIS
Department of Malacology, Academy of Natural Sciences of Philadelphia, 1900 Benjamin
Franklin Parkway, Philadelphia, Pennsylvania 19103, USA; davis@acnatsci.org
The 1998 World Malacological Congress symposium Interactions Between Mol-
luscs and Humans was one of three symposia offered during the five-day congress.
As molluscs have an enormous impact on man, it is not surprising that many persons
from around the world wished to participate in the symposium and that the symposium
was well attended. Thirty-seven papers were presented, of which 14 were invited and
the others submitted for presentation. The papers may be grouped into four broad cat-
egories: advanced biotechnology (4 papers); medical malacology and disease trans-
mission (15 papers); economics, including diseases of molluscs, impact of introduced
species, and molluscan food production (8 papers); conservation and mariculture (10
papers).
Twelve of the papers are published here, including the elegant opening plenary ad-
dress given by Dr. Dan Alkon of the United States National Institutes of Health, who
reviewed the use of an opistobranch snail for working out the molecular basis of learn-
ing and memory, and presented the implications of the findings to date for under-
standing Alzheimer's disease. The papers presented by Alkon and Yoshino are in the
biotechnology category, but Yoshino's work also groups with papers by Davis et al.,
Kristensen & Brown, Brackenbery & Appleton, and Abd Allah in the medical malacol-
ogy category. The papers by Anderson, Carlton, Pointier and Robinson involved eco-
nomics and introduced species, although Pointier's paper also had implications for
control of schistosomiasis (medical malacology category). Finally, Neves and Mansur.
presented papers that involve conservation. The Kristensen & Brown paper has im-
plications for conservation of freshwater snails in Africa.
I would make special comment on Robinson's paper involving commercial transport
of molluscs all over the world in ever increasing numbers and involving an incredible
spectrum of species. It was an eye-opener to learn that over 4,900 interceptions of
molluscs aboard cargo inbound to the United States from throughout the world had
been intercepted by the United States Department of Agriculture. These interceptions,
made in a five-year period up to 1998, involved 71 families, 197 genera and 369
species. In reading both Robinson's and Carlton's papers, one becomes immediately
concerned about the impact of introduced species into all environments on ecology,
survival of endemic species, and implications for plant pathogens and human disease.
319
MALACOLOGIA, 1999, 41(2): 321 -329
MOLECULAR PRINCIPLES OF ASSOCIATIVE MEMORY THAT ARE CONSERVED
DURING THE EVOLUTION OF SPECIES
Daniel L. Alkon
Laboratory of Adaptive Systems, National Institutes of Neurological Disorders and Strokes,
Building 36, Room 4A24, MSC4007, Bethesda, Maryland 20892-4007; dalkon@codon.nih.gov
ABSTRACT
Time domains of homeostatic adaptation encompass: 1.) the lifetime of the species (an evo-
lutionary domain) in which predictive information with survival valve is incorporated into the
genome, and 2.) the lifetime of organisms in which predictive information is stored within brain
networks. Experiments over many years in our NIH lab have confirmed that behavioral, bio-
physical, and molecular mechanisms for storing learned information have been conserved in
species as diverse as the nudibranch mollusc Hermissenda and mammals such as rat and rab-
bit. Further studies now indicate that these conserved molecular memory steps (e.g. K^ channel
inactivation, PKC and calexcitin activation, and intracellular calcium release) are also consistent
targets of the human syndrome known as Alzheimer's disease. These findings highlight the ele-
gant economy of natural selection and the utility of animal models for investigating complex brain
disorders such as Alzheimer's disease.
Key words: molecular mechanisms, Hermissenda, memory, Alzheimer's disease
Adaptation of earth's organisms may be
considered to occur over at least two distinct
time domains. The first, one that may be
called the "evolutionary" time domain, often
occurs over many thousands, even millions of
years. During this evolutionary time domain, a
species survives by incorporating genetic
programs that encode for adaptive structures
and functions. The species acquires adaptive
information that is then stored within the ge-
nome.
A second time domain for organisms' adap-
tation occurs during the lifetime of individual
members of the species. Environmental stim-
uli in real time impact on an animal's develop-
ment during critical periods of the animal's
early life cycle. Environmental stimuli, particu-
larly when they occur in temporal patterns that
reoccur repeatedly within a brief time Interval,
also impact on mature, fully differentiated an-
imals, and thus impact on an animal's subse-
quent behaviors. For the latter, information is
acquired by the mature species member dur-
ing a learning experience and stored within
brain networks as "long-term" memory.
Given these two distinct time domains of
adaptation, "evolutionary" and "learning", the
cellular mechanisms responsible for informa-
tion acquisition and storage are also quite dis-
tinct. Mutation-induced variations of genomic
information provide for the "evolutionary" time
domain of adaptation. Second-messenger in-
duced signaling cascades within neurons and
their synaptic specializations provide for the
time domains of learning adaptation.
A third, even longer time domain concerns
the evolution of progressively more adaptive
species. Thus, not only do individual species
members adapt, and the species itself adapts,
but progressively different successive species
adapt. In this longest time domain, "life" on
earth itself adapts to the environment by as-
suming different forms as distinct but still re-
lated species.
All of this adaptation and transformation
notwithstanding, certain biological entities
and their underlying functional principles may
remain essentially unchanged, even over the
course of many millions of years of evolution.
One clear example of such conservation in-
volves membrane channels or "pores" that
gate the flux of ions across cell walls. One
such channel, called 1^, is a calcium-depen-
dent pore for potassium flux. When calcium
rises inside a cell to a critical threshold level,
it triggers the opening of channels that allow
K"^ ions to flow down their chemical electrical
gradient. This almost invariably means that K"*^
ions flow from the inside to the outside of the
cell, thereby making the electric potential
321
322
ALKON
Out
FIG. 1 . A molecular model of a generic K+ channel formed from an octomeric protein (after Scott et al.
provided courtesy of Noam Meiri).
1994,
across the cell wall negative inside the cell
with respect to the outside environment. Such
calcium-dependent K"^ channels (of 1^ chan-
nels) occur within the walls of single-cell or-
ganisms, such as Paramecium. Remarkably,
very similar 1^, channels also occur within the
walls of large pyramidal-shaped cells in the
human brain area known as the hippocam-
pus. Once nature designed (through evolu-
tionary time domains) these marvels of mole-
cular configurations in single cell organisms it
essentially maintained the channel design
(Scott et al., 1994), even within our own,
human species (Fig. 1).
Some years ago, I wondered if such exam-
ples of biological conservation -that is, across
evolution -might also subserve adaptation it-
self in the time domain of individual human life-
times. Specifically, were there mechanisms of
cellular and molecular adaptation within the
"learning" time domain that were conserved
across evolutionary time?
To press this question of conserved cellular
mechanisms of learning, I set out in search of
what I thought of as a possible "evolutionary
compromise". To survive, a species would
have evolved by making a compromise be-
tween a central nervous system that was not
very complex, but just complex enough to
achieve the capacity to learn crucial stimulus
relationships. The nudibranch mollusc Her-
missenda crassicomis can, after being sub-
jected to many years of experiments, now be
considered as such a compromise (Alkon,
1983). A similar sea snail, the gastropod mol-
lusc Aplysia califomica, had previously been
found to be a favorable electrophysiologic
preparation as well as to show behavioral
changes, such as habitualization and sensiti-
zation (Bailey & Kandel, 1993).
Most psychologists agree that human
memory is diverse and requires diverse brain
structures. Memories can be explicit or de-
clarative, procedural or motoric, sensory or
emotional. Despite its diversity, however,
human memory invariably involves relation-
ships of stimuli in time and/or space. This
seemingly universal relational aspect, often
called associational, can even be encoun-
tered in the context of introspective reporting,
including that of therapeutic settings. No be-
havioral model of associative learning has
been more precisely formulated or more pre-
cisely controlled than Pavlovian conditioning.
MOLECULAR PRINCIPLES OF ASSOCIATIVE MEMORY
323
When a neutral stimulus, such as the sound
of a bell, precedes a well-valued or reflexive
stimulus, such as the smell of meat, by a pre-
cise temporal interval 350-450 msec, a dog
will learn to salivate in response to the bell
alone. The meaning (or response) of the smell
is associated in the dog's brain with a previ-
ously undefined or "neutral" sound of the bell.
From the point of view of evolutionary con-
servation, it is most remarkable that virtually
the same quantifiable characteristics of Pav-
lovian conditioning have been identified for
dogs, rabbits, rats, and humans (Figs. 2, 3),
but also for Hermissenda (Fig. 4) as well
(Gormezano, 1966; Gormezano & Kehoe,
1 981 ; Gormezano et al., 1 983; Gormezano et
al., 1962; Schreurs, 1993). These characteris-
tics (e.g., CS-UCS transfer, temporal speci-
ficity, pairing specificity, extinction, savings)
are themselves conserved across evolution.
This remarkable behavioral conservation
would seem to require that there be consider-
able conservation of underlying principles of
neuronal network functions. Conserved be-
FIG. 2. Human conditioned eyeblink training. A puff
of air (from the white outlet) on the subject's corneal
surface is preceded by a discrete tone received
through the earphone system. Courtesy of Bernard
G. Schreurs.
Purr
^
^-/,
^'
^
—TONE ON
■' . .'
TONF AND PUFF OFF
8
-
PUFF ON —
/
■Ч.
6
r
N.
4
2
П
1
^^-"^^■■^
1
1 , . , 1
1
<
z.
U-
O
z
о
100 300 500 700
TIMF (MIILISFCONDS)
900
-
— TONE ON
PUFF ON —
/
^
!
-TONF AND PUFF OFf
1, 1
1
00 300 500 700 900
TIME (MILIISFCONOS)
FIG. 3. Transfer of behavioral response in the rabbit occurs as a result of associative, or Pavlovian, condi-
tioning. In this case the animal is taught to associate an auditory tone with a puff of air to its eye. The be-
havioral response — extension of the nictating membrane (left panel)— is transferred from the unconditioned
stimulus (the puff of air) to the conditioned stimulus (the tone). The graphs show that before conditioning the
membrane extends after the puff of air (top right); about 70 trials later, the animal learns to extend the mem-
brane when it heats the tone (bottom right). Bernard G. Scheurs supplied the data for this figure.
324
ALKON
FIG. 4. Classical conditioning in the sea slug Hermissenda crassicornis leads to simple visual-vestibular as-
sociations. Excitation of sensory hair cells in the paired vestibular organs (statocysts) during rotation of this
slug leads reflexively to contraction of the foot. In contrast, stimulation of photoreceptors in the paired eyes
of the slug by flash of light does not cause the foot to contract but to weakly orient toward the light gradient
(left). However, after a series of paired rotation and light flash stimuli, in which the rotation stimulus quickly
follows a light stimulus, subsequent light stimulation of photoreceptors alone now causes the foot to contract
(right).
havioral principles of association employ con-
served network mechanisms that, in turn,
imply crucial conserved biophysical and mol-
ecular mechanisms.
To test these conservatlon(s), we first
turned to our evolutionary compromise. Ap-
plying a variety of electrophysiologic and bio-
physical recording techniques (e.g., current
clamp, two microelectrode voltage clamp,
patch clamp) to the neural pathways that me-
diate the Pavlovian training stimuli, we were
able to construct a wiring diagram (Alkon,
1983) for the visual and vestibular pathways
and their synaptic interactions (Fig. 5). With
this circuit diagram, we traced precisely how
the associated stimuli (CS: light, DCS: rota-
tion) traveled throughout the visual-vestibular
circuits to produce a learned conditioned re-
sponse. Using this "blueprint" of synaptically
connected neurons, we then reconstructed
how synaptic strengths were changed during
learning by long-term alterations of voltage-
dependent K"" currents within post-synaptic
membranes. Subsequently, it was possible to
implicate an elaborate molecular cascade that
regulates these channels and possibly struc-
tural aspects of the synaptic apparatus itself
during learning and memory. The critical mol-
ecular cascade showed clear analogy to the
molecular cascade (Fig. 6) responsible for
muscle contraction and long-term changes of
muscle structure (Alkon, 1989; Alkon et al.,
1998). Among the crucial events in these cas-
cades are enhanced mobilization of intracel-
lular calcium, activation of the signaling en-
zyme protein kinase С (PKC), activation of the
PKC substrate, the signaling protein calex-
citin, activation of the calcium releasing chan-
nel, the ryanodine receptor, and calcium-de-
pendent regulation of m-RNA turnover for
specific proteins (Fig. 7).
Many years of further studies on mam-
MOLECULAR PRINCIPLES OF ASSOCIATIVE MEMORY
325
FIG. 5. Neural responses to stimulus pairing. Neural system (schematic and partial diagram) responsive to
light and rotation. Each eye has two Type A and three Type В photoreceptors; each optic ganglion has 13
second-order visual neurons; each statocyst has 12 hair cells. The neural interactions (intersection of verti-
cal and horizontal processes) identified to be reproducible from preparation to preparation are based on in-
tracellular recordings from hundreds of pre- and post- synaptic neuron pairs.
Swimming control
Water-maze trained
-PGK1-
-RyR2-
M
6 12 24
6 12 24 M
FIG. 6. Hippocampal changes in RYR2-mRNA levels after water-maze training. (A) RT-PCR analysis of
RYR2-mRNA in control swimming and water-maze-trained rats at 2, 6, 12 and 24 h after training. No change
in RYR2-mRNA levels was observed between control swimming and naive animals (data not shown). {Fig-
ure continues.)
глаПап learning nnodels, such as rat nnaze
learning, rabbit eyeblink conditioning, and ol-
factory discrimination learning, provided
strong confirmation that the molecular cas-
cades found in Hermissenda were also criti-
cally involved in mammalian learning and
memory. Thus, conserved associative learn-
ing behaviors from mollusc to mammal were
correlated with conserved cellular and subcel-
lular mechanisms responsible for those learn-
ing and memory behaviors. This conservation
of learning behavior and its molecular basis
received still further confirmation through
studies of Alzheimer's disease and its molec-
в
(Л с:
^ 8
ОС от"
E-g
см =5
СП О)
>>>
cu
350.00
30000 -
250 00
200.00
150.00
S 12
Hours after water-maze training
24
С
Water-maze trainee
255
í^íísíi'-^*'
D
20
16
10-
I Swimming control
I Water-maze trained
FIG
CA1 CA3 DG
6. (Continued) (B) Relative RYR2-mRNA levels in water-maze-trained rats at 2, 6, 12 and 24 h after
training. To control for tine integrity of RNA and for differences attributable to errors in experimental manipu-
lation from tube to tube, primers for rat phosphoglycerate kinase 1 (PGK1) were included in the RT-PCR re-
actions. (C) Localization by in situ hybridization of RYR2-mRNAs in hippocampal subfields of control, swim-
ming, and water-maze-trained rats 6 h after training. The color spectrum on the right side of the figure
represents the pixel value of gray levels. (D) Relative RYR2-mRNA levels in different hippocampal subfields
of control swimming and water trained-rats. Quantification of induction increase is achieved by comparison
of pixel values of an area of interest in four sections from each of four pairs of rats. Changes in mRNA lev-
els are expressed as density ratio of trained to control animals.
MOLECULAR PRINCIPLES OF ASSOCIATIVE MEMORY
327
2+ .
Periods of Ca signaling
I. Depolarization
Ca Influx
PKC auto-
phosphorylates
PKC translocates
CE phosphorylated
CE inactivates K'^ channels
II. Calexcitin
Binding to RyR
CE translocates
to membrane
and ER
CE activates RyR
III. Ca reuptake
CE activates
Ca-ATPase
IV. DNA synthesis
Transcriptional
factors activate
DNA synthesis
Late genes
V. Protein Syntliesis
Axonal transport
Structural changes
Increased RyR expression
VI. Ion Channels
^2+ 1^ Periods of
Receptor Channel Channel aSSOCiative ШеГТЮГу
INDUCTION
I. msecs^
seconds
II. seconds-
minutes
CONSOLIDATION
III. mjnutes-»^
hours
IV. hours-*
days
STORAGE
V. days^
weeks
Vi. weeks ^
months
FIG. 7. Schematic diagram illustrating time domains of calcium signaling and associative memory. (Legend
continues.)
328
ALKON
FIG. 7. {Continued) Stage I: The neuron depolarizes as a result of a convergence of synaptic inputs, which
activates G-protein coupled receptors (i.e., for acetylcholine, GABA, glutamate). Membrane depolarization
also opens Ca^"^ channels, causing an influx of Ca^^. Diacylglycerol (DAG), arachidonic acid (AA), and inos-
itol triphosphate (IP3) are released by phospholipases A2 and С (PLAg and PLC) and, along with Ca^*, acti-
vate protein kinase С (PKC), which is thereby translocated to the plasma membrane. Ca also activates
CaM kinase. The kinases undergo autophosphorylation which renders their activity independent of Ca^"^.
PKC and CaM kinase may also inhibit K"^ and other channels by direct phosphorylation.
Stage II: Elevated [Ca^"^], activates the Ca^'"-binding protein calexcitin (CE). Phosphorylation of CE by PKC
promotes its translocation to membrane compartments, where it inhibits K"" channels, making the membrane
more excitable to further depolarizing stimuli. CE also elicits Ca^* release from ryanodine receptors on the
membrane of the endoplasmic reticulum (ER) and possibly synaptic membranes, resulting in amplification of
Ca^* signals. IP3 also releases Ca^"^ by activating the IP3 receptor (IP3R).
Stage III: CE, after phosphorylation by PKC, no longer activates the RyR, and no longer inhibits Ca^'"-AT-
Pase at the ER membrane, facilitating the removal of excess Ca^"".
Stage IV: CE and/or Ca^"", probably acting indirectly through transcriptional activators, induce new DNA tran-
scription. CE may also indirectly increase mRNA turnover.
Stage V: Late genes are transcribed, resulting in increased synthesis of at least 21 different proteins, in-
cluding RyR. At this stage, through an as-yet undetermined mechanism, retrograde axonal transport is also
inhibited by CE; this may underlie the structural changes in branch morphology that were observed after Her-
missenda associative learning.
Stage VI: New RyR receptors and ion channels are synthesized and transported to their respective mem-
branes. These may be related to enhanced Purkinje cell dendritic excitability with rabbit conditioning.
с
с
пз
x:
и 1Л
Ê ^
(Л ™
1Л Cl-
4_) It-
о о
û.
СЛ
Q.
100
90
80
70
60
50
40
30
20
10
-
113 pS
-
ШШ^^Ш 1 \ 1
Untreated
Controls
B-Amyloid
Controls
Alzheimer's
Disease
FIG. 8. Bar graph of the percentage change (from resting Ca^"' concentrations) shows the virtual elimination
of the TEA response in treated control cells (eight cell lines, 194 cells) similar to observations for AD fibrob-
lasts (four cell lines, 285 cells measured). Preparations of the various cell lines were tested for response to
TEA in the absence (left) or presence (center) of ßAP. The responses of AD fibroblasts shown here are in
agreement with our previous report showing no responses in 13 different AD lines (> 700 cells measured).
ular correlates. To test the possibility of con-
served molecular events responsible for hu-
man memory, we turned to that clinical entity
most specific for human memory loss: Alz-
heimer's disease. This disease early in its
progression rather specifically impairs human
memory. Beginning with the hypothesis that
Alzheimer's disease is systemic (i.e., involving
multiple tissue types through out the body) but
causes clinical symptoms only through effects
on the brain, we analyzed peripheral cell
types, such as human skin fibroblasts. After
many studies, we could conclude that, indeed,
a succession of molecular steps in the con-
MOLECULAR PRINCIPLES OF ASSOCIATIVE MEMORY
329
served memory cascade were in fact consis-
tently targets of dysfunction in Alzheimer's dis-
ease. Diagnostic clianges of K"" channels (Fig.
8), PKC intracellular calcium release, and
calexcitin were consistently found in the skin fi-
broblasts (Etcheberrigaray et al., 1994) of
Alzheimer's patients, but not for a variety of
age-matched controls. Furthermore, low lev-
els of soluble ß-amyloid (nanomolar-compara-
ble to endogenous levels found throughout the
body) induced these diagnostic phenotypic
abnormalities in the normal fibroblasts. Here,
then, was evidence that early Alzheimer's dis-
ease involves several of the same molecular
steps implicated as mechanism for animal
learning.
Thus, based on these studies of species
that range from the nudibranch Hermissenda
to humans, we may infer that adaptation in the
"learning" time domain uses functional princi-
ples that have been conserved across the
"evolutionary" time domain. This conservation
deserves our future attention not only be-
cause it illustrates and elucidates the elegant
economy of natural selection. It also provides
a strategy whereby we can use molecular
cascades in primitive species as accessible
models to guide our investigations of daunt-
ingly complex and often inaccessible myster-
ies of human physiology and disease.
LITERATURE CITED
ALKON, D. L., 1 983, Learning in a marine snail. Sci-
entific American, 249: 70-84.
ALKON, D. L., 1989, Memory storage and neural
systems. Scientific American, 260: 42-50.
ALKON, D. L., T J. NELSON, W. Q. ZHAO & S.
CAVALLARO, 1998, Time domains of neuronal
calcium signaling and associative memory: steps
through a calexcitin, ryanodine receptor, K^ chan-
nel cascade. Trends in Neuroscience, 21:
529-537.
BAILEY, С H. & E. R. KANDEL, 1993, Structural
changes accompanying memory storage. Annual
Review of Phiysiology, 55: 397-426.
ETCHEBERRIGARAY R-, E. ITO, S. KIM & D. L.
ALKON, 1994, Soluble ß-amyloid induction of
Alzheimer's pheontype for human fibroblast K*
channels. Science, 264: 276-279.
GORMEZANO, I., 1966, Classical conditioning. Pp.
385-420, in J. B. siDOWSKi, ed.. Experimental
methods and instrumentation in psychology. New
York, McGraw-Hill.
GORMEZANO, I. & E. J. KEHOE, 1981, Classical
conditioning and the law of contiguity. Pp. 1-45,
in p. HARZEM & M. D. ZEiLER, Advances in analysis
of behavior, vol. 2, Predictability, correlation, and
contiguity. Sussex, England, Wiley & Sons.
GORMEZANO, I., E. J. KEHOE & B. S. MAR-
SHALL, 1983, Twenty years of classical condi-
tioning research with the rabbit. Progress in Psy-
chobiology and Physiological Psychology, 10:
197-275.
GORMEZANO, I., N. SCHNEIDERMAN, E. G.
DEAUX & I. FUENTES, 1962, Nictitating mem-
brane: classical conditioning and extinction in the
albino rabbit. Science, 138: 33-34.
SCHREURS, B. G., 1993, Long-term memory and
extinction of the classically conditioned rabbit nic-
titating membrane response. Learning & Motiva-
tion, 24: 93-302.
SCOTT V. E. S., J. RETTIG, D. N. PARCEJ, J. N.
KEEN, J. B. С FINDLAY O. PONGS & J. O.
DOLLY, 1994, Primary structure of a ß subunit of
a-dendrotoxin-sensitive K"^ channels from bovine
brain. Proceedings of the National Academy of
Sciences, USA, 19: 1637-1641.
Revised ms accepted 30 April 1999
MALACOLOGIA, 1999, 41(2): 331-343
THE BIOMPHALARIA GLABRATA EMBRYONIC (BGE) MOLLUSCAN CELL LINE:
ESTABLISHMENT OF AN IN VITRO CELLULAR MODEL FOR THE STUDY OF
SNAIL HOST-PARASITE INTERACTIONS
Timothy P. Yoshino^ ^, Christine Coustau^, Sylvain Modat^ & Maria G. Castillo^
ABSTRACT
Cell lines of invertebrates, especially those of arthropod origin, have played crucial roles in ad-
dressing fundamental questions related to cell signaling and differentiation, gene expression,
cell-pathogen interactions, and the like. They also have been instrumental in the development of
genetic transformation systems and the development and testing of microbial insecticides. Re-
cently, we have utilized a cell line originally derived from embryos of the freshwater snail Biom-
phalaria glabrata (Bge cell line; Hansen, 1976a) to investigate the complex cellular, biochemical
and molecular interactions between snails and their trematode parasites. Because this cell line
was derived from B. glabrata, possesses a fibroblast-like appearance similar to circulating he-
mocytes, and shares in common several hemocyte functions (substrate adhesion, phagocytosis,
encapsulation, enzyme content), the Bge cell line is proposed as a cellular model for B. glabrata
hemocyte structure and function. In support of this proposal, a hemocyte ß-integrin cell adhesion
receptor homologue recently was identified and cloned based on information from a previously
acquired Bge cell ß integrin subunit cDNA sequence. As a general approach, it is anticipated that
Bge cells can be evaluated for genes associated with immune recognition/adhesion, and sub-
sequently employed to generate molecular or immunological probes for use in hemocyte stud-
ies. Other applications of Bge cells to the study of parasite-snail host interactions include their
use in the in vitro cultivation of intramolluscan stages of diverse trematode species, and in the
development of genetic transformation systems for molluscan cells. Research in this latter area
has focused on the identification of suitable Bge cell promoters and testing their abilities to drive
expression of reporter gene constructs. It is concluded that the Bge cell line offers a diversity of
valuable experimental approaches when applied to the study of molluscan cellular immune
mechanisms or snail-trematode physiological interactions.
Key words: Biomphalaria glabrata, Mollusca, embryonic cell line, Bge cell line, hemocyte.
Schistosoma mansoni, in vitro culture, transgenic.
INTRODUCTION
The use of cell lines as tools for addressing
fundamental questions regarding molecular
structure-function relationships is well estab-
lished in those animal species for which such
lines are available. For coelomate inverte-
brates, the vast majority of cell lines are of in-
sect or arachnid origin (Bayne, 1998) and
have been extensively employed in studies of
gene regulation and protein expression
(Berger & Morganelli, 1984; Jones et al.,
1996), cellular shape-change and motility
(Kosik & McConlogue, 1994), adhesion
(Bleber, 1994), induction of immune peptides
(Hultmark, 1994; Hoffmann & Reichart, 1997),
and pathogen-host cell interactions (Stellar,
1993; Kopecky & Stankova, 1998; Lawrence,
1 997). Moreover, there has been a long-stand-
ing interest in the use of insect cell lines in the
production of recombinant proteins through
baculovirus expression systems (Thomsen et
al., 1993; McCarroll & King, 1997; Merhngton
et al., 1997), the selection and testing of mi-
crobial pesticides (Maramorosch & Mit-
suhashi, 1997), and the development of new
genetic transformation technologies (Fallon,
1 991 ; Cherbas et al., 1 994; O'Brachta & Atkin-
son, 1996). However, for reasons that are not
yet understood, there exist very few cell lines
outside of the arthropod classes Insecta and
Arachnida, despite a critical need and strong
interest in cell lines of diverse invertebrate or-
ganisms (Bayne, 1998).
Of the non-arthropod cell lines currently
available, only one, the Biomphalaria glabrata
^ Department of Pathobiological Sciences, 201 5 Linden Dr. W, University of Wisconsin-Madison, Madison, Wisconsin 53706,
U.S.A.; yoshinot@svrn. vetmed.wisc.edu
^Corresponding author
^Centre de Biologie et d'Ecologie Tropicale et Méditerranéenne, Université de Perpignan, France
331
332
YOSHINOETAL
embryonic (Bge) cell line, is of molluscan ori-
gin (ATCC# CL1494; American Type Culture
Collection, Rockville, Maryland). It was de-
rived from embryos of the freshwater pul-
monate snail B. glabrata by Hansen (1976a),
and has been stored/maintained in various
laboratories over the past 20 years. The sig-
nificance of the Bge cell line, in addition to its
unique molluscan origin, also lies in the fact
that the snail species from which this line was
derived represents a primary intermediate
host for transmission of the blood fluke. Schis-
tosoma mansoni (Platyhelminth; Digenea),
causative agent of human intestinal schisto-
somiasis in the New World and Africa (Basch,
1991). Schistosoma spp. are estimated to in-
fect 200 million people in over 74 endemic
countries, and, among parasitic diseases, is
considered second in importance only to mos-
quito-transmitted malaria in its public health
impact (World Health Organization, 1998).
Because molluscan and arthropod interme-
diate hosts of infectious diseases (broadly re-
ferred to as "vectors") are essential to human
transmission, they have long been recognized
as potentially vulnerable targets in disease
control programs. However, because tradi-
tional methods of vector control, including pes-
ticide usage and habitat destruction, have not
provided sustainable solutions to disease con-
trol or prevention, new approaches aimed at
disrupting parasite development within the
host through genetic and/or molecular manip-
ulation of vector competence are now being
explored. This field has advanced most rapidly
in insects where inducible antimicrobial and
antiparasitic peptides (Faye & Hultmark, 1 993;
Hoffman & Reichhart, 1997; Lowenberger et
a!., 1 999) and enzymes (Ashida & Brey, 1 998)
have been identified, their genes cloned, and
the molecular pathways/factors regulating an-
timicrobial responses elucidated. In addition,
methods for introducing and expressing im-
mune peptides or function-disrupting anti-
sense sequences into whole organisms
through genetic transformation technologies
are being developed (Carlson et al., 1997;
O'Brachta & Atkinson, 1996). It is envisioned
that the basic knowledge of anti-pathogen
immune mechanisms and parasite develop-
mental pathways will lead to molecularly-
based strategies for interrupting pathogen sur-
vival in the vector host. Significantly, insect cell
lines have played important roles in advancing
the general area of host-pathogen relation-
ships.
In contrast to the insects, little progress has
been made in the development of molecular
vector control approaches for molluscs of
medical and veterinary importance. For ex-
ample, in the freshwater gastropods, few
genes have been cloned and/or identified as
being of potential significance in regulating
vector competence (Dissous et al., 1990;
Ноек et al., 1996; Adema et al., 1998; Knight
et al., 1998), and there are still no genetic or
physical maps available for vector snails
(Knight et al. 1998). Recently, however, appli-
cation of the B. glabrata embryonic (Bge) cell
line to the in vitro cultivation of larval schisto-
somes (Yoshino & Laursen, 1995; Coustau et
al., 1997) has provided significant opportuni-
ties to investigate larval blood fluke-host cell
interactions at the molecular level. Work utiliz-
ing this cell line has already led to the identifi-
cation of molecules potentially involved in
host cell signal transduction (Lardans et al.,
1998), immune adhesion (Davids et al.,
1999), trematode development (Laursen &
Yoshino, 1999) and to the establishment of
protocols for DNA-mediated gene transfer
into molluscan cells (Lardans et al., 1996;
Yoshino et al., 1998). The present review fo-
cuses on these recent advances in the use of
the Bge cell line as a tool for studying chemi-
cal communication mechanisms between
trematode parasites and cells of their mollus-
can host.
Bge Cells as a B. glabrata Hemocyte Model
Two important criteria for establishing a
comparative cellular model system include
the demonstration that the cells being com-
pared have a similar origin, both in terms of
species and ontogeny, and that they share
structural and functional similarities under in
vitro conditions. In terms of species origin, it is
fortunate that the Bge cell line was derived
from B. glabrata (Hansen, 1976a), a major in-
termediate host of S. mansoni and an impor-
tant experimental snail host for the study of
schistosome-snail immune interactions (Fryer
& Bayne, 1996a; Yoshino & Vasta, 1996;
Adema & Loker, 1997). The precise ontologi-
cal origin of the Bge cell line is not known
since it was originally obtained from develop-
ing five-day old B. glabrata embryos (Hansen,
1976a). However, it has been speculated,
based on morphology (Fig. 1) and Bge cell's
ability to synthesize collagen-like molecules
(Hansen, 1976b; Stein & Basch, 1977), that
these cells probably are of fibroblast or fibro-
blast-like origin. A similar ontological origin
THE BIOMPHALARIA GLABRATA EMBRYONIC MOLLUSCAN CELL LINE
333
111
%j
<í¿^
B.glahraía hemocytes
Bge cells
FIG. 1. Phase-contrast photomicrographs of Biomphalaria glabrata hemocytes (left) and cells of the B.
glabrata embryonic (Bge) cell line (right) as they appear attached to and spread on a glass slide substrate. It
has been noted that, as depicted here, a subset of Bge cells assume a more hemocyte-like appearance when
they are washed of medium and permitted to reattach under serum-free conditions. This differs from more typ-
ical spindle-shaped morphology of Bge cells as they appear under normal culture conditions (Lardans & Dis-
sous, 1 998). Whether these differences in cell morphology reflect the possible presence of differentiating Bge
cell subpopulations or a behavioral response to manipulation is presently unknown. Scale bar = 10 (am.
also has been suggested for circulating he-
mocytes (Pan, 1958), although their exact
embryological source in B. glabrata is still
controversial (Jeong et al., 1983).
The second criterion for Bge cells to serve
as an effective model for B. glabrata hemo-
cytes is that the two cell-types must share im-
portant structural and functional characteris-
tics in their interactions with foreign materials.
Here we assume that shared function is
based on similarities in biochemical and/or
molecular mechanisms being used by these
snail cell-types. Therefore, recently we have
initiated experiments to begin evaluating the
"immune" functions of phagocytosis and en-
capsulation in the Bge cell line. Phagocytic
and encapsulation responses represent "hall-
mark" behaviors for hemocytes since these
activities, in large part, define their immune ef-
fector capabilities (Ratcliffe et al., 1985;
Bayne, 1990).
Phagocytosis: It is well documented that B.
glabrata hemocytes are capable of phagocy-
tosing a variety of foreign particles including
bacteria (Cheng et al., 1 978), mammalian ery-
throcytes (Abdul-Salam & Michelson, 1980;
Noda & Loker, 1989; Zeick & Becker, 1992),
yeast/zymosan (Fryer & Bayne, 1989;
Connors & Yoshino, 1990), and latex beads
(Uchikawa & Loker, 1992; Fryer & Bayne,
1996b). Moreover, excretory-secretory prod-
ucts (ESP) released by developing S. man-
soni sporocysts exert inhibitory effects on
phagocytosis by host hemocytes (Connors &
Yoshino, 1990; Fryer & Bayne, 1990), sug-
gesting the presence of ES factors capable of
binding to hemocytes and interferring with he-
mocyte-particle interactions.
However, to date, only ancedotal evidence
is available that Bge cells possess similar
phagocytic activity. Therefore, this function
was tested more rigorously using an adhe-
sion/phagocytosis assay modified from
Uchikawa & Loker (1992). In this assay, un-
charged latex beads (6.4 ¡am in diameter;
Sigma Chemical Co., St. Louis, Missouri;)
were rinsed 2x in Chernin's balanced salt so-
lution (CBSS; Chernin, 1963) and resus-
pended in CBSS to a final concentration of
0.1%. To determine the effect of S. mansoni
ESP (Connors & Yoshino, 1990) on phagocy-
tosis, beads were treated with ESP (10 ¡iL
beads in 1mL ESP) for 4 h at 22°C, followed
by two washes in CBSS and resuspension in
CBSS at a concentration of 0. 1 %. Beads were
then presented to monolayered Bge cells in a
ratio of 2 beads/cell. Because of the difficulty
in discerning between surface membrane ad-
herent and endocytosed beads (Fig. 2)
(Uchikawa & Loker, 1992), an adhesion index
(percentage of cells with adherent/phagocy-
tosed beads) was determined for each treat-
ment and control group at 1, 3, 6 and 24 h
post-bead innoculation. As shown in Figure 3,
untreated latex beads readily attached to Bge
cells by 1 h (45%), and by 6 h had reached
near maximum association (adherence/pha-
gocytosis) with cells (approx. 80%). This rate
of bead association was considerably higher
when compared to B. glabrata hemocytes,
which attained between 10-15% bead adher-
ence (Uchikawa & Loker, 1992) at 1 h of incu-
bation. In a similar study (Fryer & Bayne,
334
YOSHINOETAL
FIG. 2. Photomicrographs of В. glabrata embryonic
(Bge) cells demonstrating surface attached and
phagocytosed latex beads following and 24 h in-
cubations in snail saline (CBSS).
1996b), however, it is noted that hemocyte at-
tained a phagocytic rate of 70% using smaller
(2.5 |im) charged latex beads. Regardless of
discrepancies in association rates, clearly
Bge cells, like hemocytes, possess the ca-
pacity to phagocytose foreign particles.
Pretreatment of beads with ESP prior to in-
cubation with Bge cells resulted in a signifi-
cant increase in bead adhesion at all time
points (Fig. 3), suggesting that ESP contains
molecule(s) that mediate enhanced bead ad-
hesion through interaction with putative ESP
receptors on the Bge cell surface. The fact
that untreated beads did not associate with
cells above control levels when incubated in
the presence of ESP supports this hypothe-
sis; in this case, free ESP molecules saturate
Bge cell receptors, thereby blocking en-
hanced adhesion due to bead-bound ESP.
Recent flow cytometric analyses using fluo-
rescein-labeled ESP (fESP) as a "probe"con-
firm that polypeptide(s) contained in ESP di-
100 -|
90-
80-
70
60
50
40-
30-
20-
10
10 15
Time (hour)
FIG. 3. Percentage of B. glabrata embryonic (Bge)
cells with attached/phagocytosed beads acquired
during 24 h of incubation in snail saline. Control:
beads pretreated with CBSS prior to addition to Bge
cells in CBSS (-□-); ESP treatment #1 : beads pre-
treated with S. manson/ sporocyst excretory-secre-
tory products (ESP) before addition to Bge cells in
CBSS (-O-); ESP treatment #2: beads pretreated
with CBSS prior to addition to Bge cells in CBSS
containing ESP (-0-). 'designates values that dif-
fer significantly (p < 0.05) from the control.
rectly bind to Bge cells, and that such binding
is inhibited by the sulfated poly-fucose, fu-
coidan. Of particular relevance to this review,
B. glabrata hemocytes also have been found
to bind fESP via a fucoidan-inhibitable recep-
tor (Johnston et al., in preparation), and it is
hypothesized that these, or related, receptors
are responsible for the ESP-mediated modu-
lation of hemocyte function (e.g., Yoshino &
Lodes, 1988; Lodes & Yoshino, 1990; Con-
nors & Yoshino, 1990; Loker et al., 1992;
Adema & Loker, 1 997). In this case, Bge cells
may serve as a unique model for investigating
parasite ESP-binding hemocyte receptors.
Encapsulation: Encapsulation, or the multiple
layering of cells around large foreign objects,
is another function shared in common be-
tween B. glabrata hemocytes and Bge cells.
Capsule formation represents the first line of
cellular defense against larval helminths in-
fecting molluscs, and is typified in the B.
glabrata/S. mansoni system where hemo-
cytes of resistant snail strains encapsulate
and kill schistosome sporocysts under both in
vivo (Sullivan & Richards, 1981; Loker et al.,
1982) and in vitro (Bayne et al., 1980) condi-
tions. At present, however, little is known
THE BIOMPHALARIA GLABRATA EMBRYONIC MOLLUSCAN CELL LINE
335
"'»* к /
' -^"¡^'^'^ ' .
•■/^■'
CBSS
Fucoidan
FIG. 5. Bright-field photomicrographs of B. glabrata embryonic (Bge) cells binding to the surface of S. man-
soni mother sporocysts in the absence (left; CBSS only) and presence of an inhibiting polysaccharide, fu-
coidan (right; 1 mg/ml CBSS).
about the hemocyte molecules mediating par-
asite adhesion/recognition or how the encap-
sulation process in regulated. In order to ad-
dress these questions, it may also be possible
to use Bge cells as a encapsulation model.
Bge cells co-cultured with S. mansoni sporo-
cysts avidly bind to the larval surface form-
ing multicellular capsules (Fig. 4) (Hansen,
1976b; Yoshino & Laursen, 1995), reminicent
of encapsulations by resistant snail hemo-
cytes (Boehmler et al., 1996). In the case of
Bge cells, however, encapsulation does not
lead to parasite destruction, but instead are
more comparable to in vitro capsules formed
around sporocysts by hemocytes of suscepti-
ble snail strains (Bayne et al., 1980; Fryer &
Bayne, 1995).
Due to this shared sporocyst adhesive be-
havior, Bge cells are currently being used to
identify and characterize the surface recep-
tor(s) responsible for initiating parasite adher-
ence and for mediating subsequent cell-to-
cell binding required for capsule formation.
For example, using an in vitro cell adhesion
assay, in which S. mansoni sporocysts were
placed into 1 .5-mL microcentrifuge tubes (200
larvae/tube) followed by addition of 2 x 10^
Bge cells in a final volume of 200 |iL of CBSS,
we have determined that Bge cells in suspen-
sion were capable of binding to the sporocyst
tegument after incubation for 24 h at 26°C
(Fig. 5). Qualitative and quantitative assess-
ment of cell adherence in the presence or ab-
sence of various chemical inhibitors further
demonstrated that Bge cell binding was medi-
ated through a carbohydrate-inhibitable
mechanism, possible involving a lectin-like re-
ceptor(s) expressed on the Bge cell surface
(Fig. 5). Whether or not B. glabrata hemo-
cytes possess similar sporocyst binding sites
^ ^7 /" V Л. '! ^},\
'»if
FIG. 4. Bright-field photomicrograph of an in vitro
cultured S. mansoni mother sporocyst encapsu-
lated by numerous B. glabrata embryonic (Bge)
cells. Within Bge cell capsules sporocysts continue
to develop, eventually giving rise to daughter sporo-
cyst stages under in vitro conditions.
is at present unknown. However, it is antici-
pated that in-depth biochemical/molecular
studies on the parasite-reactive receptors of
Bge cell will lead to the development of valu-
able molecular and/or chemical probes for in-
vestigating similar membrane receptors asso-
ciated with snail hemocytes.
Lysosomal Enzyme Content: Finally, since B.
glabrata hemocytes previously have been
shown to possess a variety of lysosomal en-
zymes (mainly hydrolases), we sought to de-
termine whether or not Bge cells exhibited a
similar enzyme repertoire. Acid and alkaline
phosphatase, lysozyme, ß-glucuronidase, li-
pase, nonspecific esterase, peroxidase, and
aminopeptidase activities (Rodrick & Cheng,
1974; Cheng et al., 1978; Granath & Yoshino,
1983a; McKerrow et al., 1985; Cheng, 1985;
336
YOSHINOETAL
TABLE 1 . Hydrolytic enzyme activity observed in Bge cell lysates (cell) or 2 hr-cultured supernatants (sn). C:
Bge cells incubated for 2 hr in CBSS. B: Bge cells in the process of phagocytosing latex beads. ES: Bge cells
in the presence of ES products. ES total: Control for enzymatic activity contributed by ES products alone.
Semi-quantitative scores according to enzyme kit standards (APIzym^i^ Galery; bioMèrieux): - , nanomole
of enzyme; +, 5 nanomoles; ++, 10 nanomoles; +++ 20 nanomoles; ++++, 30 nanomoles.
Enzymes tested
С cell
Csn
Beeil
Bsn
ES cell
ESsn
ES total
alkaline phosphatase
+
—
+
—
++
-H-l-
—
esterase
++
+++
+++
Ч-++
+++
-H-l-
-1-
esterase lipase
++
+
+-1-1-
+
-H-l-
+
—
leucine arylamidase
++++
+
-H-l-l-
+
++++
+
—
valine arylamidase
+
-
-1-
-
+
-
—
cystine arylamidase
+
-
-1-
-
+
-
—
acid phosphatase
++++
+
++++
-1-
-i~i~i~i-
+
-1-
naphtol-AS-BI-phosphohy-
+-I-I-
+++
+++
-l"H-
-l-l-H
+++
++
drolase
ß galactosidase
++++
++
-I--I--1-I-
-l-l-
++++
-n-
-
N-acetyl-ß glucosaminidase
++
-
-ы-
-
-l-l-
-
-
and others) have been detected in B. glabrata
hemocytes using various enzyme cytochemi-
cal and biochemical assays. Our assumption,
as previously suggested, was that similarities
in enzyme profiles may reflect shared func-
tional and biochemical "backgrounds". Deter-
minations of hydrolase activities in Bge cells
were carried out using the APIzym™ Galery
semi-quantitative enzyme assay kit (bio-
Mèrieux Vitek, Hazelwood, Missouri). In these
experiments, Bge cell monolayers initially
were washed with CBSS and cultured for 24 h
in CBSS at 26°C prior to assay. Cells were
then hypotonically lysed, cellular debris pel-
leted by centrifugation, and enzyme activity
measured in the supernatants.
As shown in Table 1 , the general profile of
Bge cell hydrolase activities is comparable to
that of snail hemocytes. Specifically, cells
contained significant amounts (>10 nano-
moles) of nonspecific (simple) esterase, li-
pase, leucine arylamidase, carbohydrases,
phosphohydrolase and the "classical" lysoso-
mal marker, acid phosphatase. Several en-
zymes (e.g., esterase, phosphohydrolase and
ß galactosidase) appeared to be secreted into
the medium during a 2 h incubation in CBSS.
Also of interest was the observation that
phagocytic stimulation or treatment of Bge
cells with S. mansoni ESP had little effect on
the hydrolase activity profile (Table 1). Excep-
tions included alkaline phosphatase, in which
exposure to ESP Induced an increase in both
cellular and secreted enzyme activity, and
simple esterase and lipase, in which phago-
cytosis appeared to increase their cellular en-
zyme content. Although selective, these types
of inductive responses are similar to those re-
ported in snail hemocytes exposed to bacteria
in vitro or from hosts with larval infection
(Cheng, 1983, 1985; Granath & Yoshino,
1 983b). Overall, the above functional and bio-
chemical similarities exhibited by Bge cells
and hemocytes reinforce the notion that Bge
cells may serve as a legitimate model for B.
glabrata hemocytes in investigations of para-
site-snail immune interactions. In the follow-
ing section, several examples are described
as to how Bge cells are being used as tools to
investigate schistosome-snail relationships
under in vitro conditions.
Applications of Bge Cells to the Study of
Snail Host-Parasite Interactions
Structure and Function of Hemocyte Adhe-
sion Molecules: Since both Bge cells and he-
mocytes are capable of adhering to and
spreading on glass or plastic substrates via
pseudopodial extensions of the cytoplasm
(Fig. 1), it was hypothesized that these cells
employ similar adhesion molecules to medi-
ate substrate adherence or spreading. In a
test of this hypothesis, we recently found that
cell spreading on glass surfaces by both he-
mocytes (Davids & Yoshino, 1998) and Bge
cells (Davids et al., 1 999) was inhibited by the
tetrapeptide, arg-gly-asp-ser (RODS), but not
the glu-substituted peptide, RGES (Fig. 6).
Because the RGD peptide sequence is well
known as a specific ligand for cellular recep-
tors of the integrin family (Sonnenberg, 1 993),
our findings suggested that hemocytes and
Bge cells share a common cell adhesion/
spreading mechanism mediated by integrin-
like receptors. In order to further investigate
THE BIOMPHALARIA GLABRATA EMBRYONIC MOLLUSCAN CELL LINE 337
%
RGES
FIG. 6. Phase-contrast photomicrographs of B. glabrata embryonic (Bge) cells incubated in presence of the
tetrapeptides RGES (nonspecific control peptide) or RGDS (specific test peptide). Note that RGDS strongly
inhibits cell spreading on the glass substrate (arrow), whereas no such inhibition is seen in the presence of
RGES (arrow). Scale bar = 10 цт. Reprinted with modification from Gene, Davids, B. J., X. J. Wu & T. P.
Yoshino, Cloning of a ß integrin subunit cDNA from an embryonic cell line derived from the freshwater mol-
lusc, Biomphalaria glabrata, 1999, with permission from Elsevier Science.
this possibility, we took advantage of the
ready availability and ease of producing large
numbers of Bge cells to identify and clone a
molluscan integrin homologue that may be
serving as a putative RGD-binding receptor.
This was accomplished by PCR amplifica-
tion of a 137 base pair Bge cell cDNA se-
quence corresponding to the highly con-
served ß integrin ligand binding domain (LBD)
using degenerate primers to the LBD of sev-
eral known ß integrin subunits (Gettner et al.,
1995), followed by PCR amplification of the 3'
region of the Bge ß integrin cDNA using an
exact 5' primer contained within the 137 bp
Bge LBD and an oligo-dT 3' primer. Sequence
of the 5' end of the integrin cDNA was ob-
tained by 5' RACE (random-amplification of
cDNA ends) methods. Based on its predicted
molecular mass of 87.6 kDa and characteris-
tic domain structure (Davids et al., 1999), this
molecule was identified as an authentic ß in-
tegrin subunit homologue (designated ßBGE)
and the first to be cloned from the phylum Mol-
lusca.
We then asked the question, do B. glabrata
hemocytes also express a ß integrin subunit
cDNA similar to that of Bge cells? Based on
Bge cell sequence data, RT-PCR was per-
formed on snail hemocyte cDNA using exact
primers synthesized from the ßBGE integrin
LBD. Amplified products from snail hemo-
cytes shared 99% nucleic acid similarity and
100% amino acid identity to the Bge cell ß in-
tegrin subunit, and from 49-71% amino acid
identity with LBDs of other known invertebrate
and human ß subunits (Fig. 7). Cloning and
sequencing of the complete B. glabrata he-
mocyte cDNA currently is in progress, but to
date, comparison of a 1900-nucleotide hemo-
cyte sequence with ßBGE has revealed 97%
and 96% identities at the nucleic acid and
amino acid levels, respectively (unpublished
data). Thus, it appears that hemocytes also
express a ß integrin homologue, although like
the ßBGE subunit, it still remains to be deter-
mined whether this integrin subunit, in con-
junction with an as yet unidentified alpha
subunit, is responsible for the observed RGD-
dependent cell spreading response (Davis &
Yoshino, 1998). Moreover, it is important to
recognize that the integrin gene sequence in-
formation acquired from Bge cells greatly fa-
cilitated our ability to generate comparable
hemocyte data, and it is envisioned that this
research strategy will continue to contribute
critical molecular tools for investigating a vari-
ety of molecules and their functions in hemo-
cytes.
In Vitro Cultivation of Larval Trematodes: A
major technical hurdle in investigating the
molecular mechanisms underlying intramol-
luscan schistosome development has been
the lack of an in vitro culture system capable
of supporting continous growth and differenti-
ation through all larval stages. Soon after the
successful isolation of the Bge cell line from
B. glabrata embryonic tissues (Hansen,
1976a), attempts were made to develop a
Bge cell-based in vitro culture system capable
of supporting the larval development of S.
mansoni. Co-cultivation of Bge cells with in
vivo-dehved daughter sporocysts resulted in
the production of a second daughter sporo-
cyst generation (Hansen, 1976b), while in
more recent experiments, S. mansoni mother
338
YOSHINOETAL
в. glabra ta &Bge
B. glabra ta Hemocyte
D. melanogaster ßPS
S. purpura tus ßG
C. elegans ßpat-3
P. leniusculus ß
A. millepora ßCnl
D. melanogaster ßv
0. tenuis ßPol
H. sapiens ßl
H. sapiens ß3
Consensvis
ENYPVDLYYFMDLSNSMEDDKEKLALLGNKIAEQMSAITKNFRLGFGSFVDKWSPYV . STVPQK
DLYYFMDLSNSMEDDKEKIJU.IX3№<IAEQMSATTKNFRLGFGSFVDKVVSPYV. STVPQK
-G L К А— ST— D-LS-T-KR— N~H LM .--I-K-
_D V К LS— MD— DIL-SE-KN— S T-M . E-
VD L Y — К Q — SE— DLL--R-RTV 1— KLM-FI . DPRIE-
RD L S KQ — A — SEL-KL-KGL-SQ-T LM— A.D-S
L~M-S— K— LGN-RS-AGQM-TT-KE— S— К А F-RT E
R-N-L VL TWT-R KT-EE— AQLSQTLKN— G-Y A— PTL-MIL. . . — H
— F-I LL Y — R — LDN-KQ — ADL-ASIVGLST 1 E A-ETTLD-RPQ
-D — I L Y — К — L-NVKS--TDIWNE-RR— SD— I E-T-M— I . — T-A-
-D 1 — L Y — К— LWSIQN— T-L-T— RK~S-L-I А P MYISP-EA
EDYPVDLYYI^DLSYSMKDD — KL — LGDKLAETMK-ITSNFRLGFGSFVDKWMPYV-STVP-K
FIG. 7. Multiple amino acid (AA) alignment and consensus sequence of ß integhn ligand binding domains
(LBD). The sequence from B. glabrata embryonic (Bge) cells and hemocytes are aligned with the LBD of all
known invertebrate species and two human LBD GenBank Accession Nos.: В. glabrata Bge cell [AF060203],
D. melanogaster ^PS [J03251], S. purpuratus [U77584], C. elegans [U19744], P. leniusculus [X98852], A.
millepora [AF005356], D. melanogaster ^v [L13305], O. tenuis [AF005357], H. sapiens ß1 [X07979], and H.
sapiens ß3 [J02703]. The LBD consensus sequence was derived from the nine LBD listed in the figure and
is shown in the bottom line. The periods (.) represent gaps in the sequence introduced by PILEUP (GCG) to
align each sequence. AA residues identical to those of the Bge/hemocyte sequence are indicated by dashes.
Reprinted with modification from Gene, Davids, B. J., X. J. Wu & T P. Yoshino, Cloning of a ß integrin sub-
unit cDNA from an embryonic cell line derived from the freshwater mollusc, Biomphalaria glabrata, 1 999, with
permission from Elsevier Science.
sporocysts, derived from in vitro transformed
miracidia, were found to produce daughter
sporocysts (Yoshino & Laursen, 1995). This
latter study represented the first direct linkage
of miracidium-through-daughter sporocyst
development under in vitro conditions. In an
important breakthrough, Barnes, Bayne and
colleagues have now succeeded in complet-
ing the the entire intramolluscan cycle of S.
mansoni development (miracidium-to-cer-
caria) under in vitro conditions by co-cultiva-
tion of larvae in the presence of Bge cells and
employing a combination of several media
formulations (Ivanchenko et al., 1999). These
studies now provide an established in vitro
culture system that can be used to identify
and characterize host (Bge) cell factors criti-
cal to the regulation of larval differentiation, or
that can be chemically modulated to screen
various agents (e.g., growth factors, cy-
tokines, snail plasma factors, etc.) for their ef-
fects on parasite development (Lardans &
Dissous, 1998).
In followup experiments, we were surprised
to find that, like S. mansoni, Bge cells also
supported significant parasite growth and dif-
ferentiation when other trematode species
were used in this culture system. For exam-
ple, miracidia of S. japonicum (Coustau et al.,
1997) and the deer liver fluke, Fascioloides
magna (Laursen & Yoshino, 1999), trans-
formed to mother sporocysts, which in turn
produced daughter sporocysts and mother re-
diae, respectively. In these studies, of particu-
lar interest was the observation that Bge cells
did not form large cellular encapsulations
around mother sporocysts of either species
as was the case with S. mansoni. In order to
confirm this observation, a direct comparison
of S. mansoni and S. japonicum sporocysts
was made using the Bge cell adhesion assay
described previously. In contrast to 94% of S.
mansoni sporocysts exhibiting Bge cell bind-
ing, only 51 % of S. japonicum sporocysts pos-
sessed adherent cells. Moreover, using a
semi-quantitative measure of adhesion inten-
sity (1 = no binding, 2 = <10 cells/sporocyst, 3
= >10 cells but <50% of sporocyst surface
with cell binding, and 4 = >50% of sporocyst
surface with bound cells), a cell adhesion
index (CAI) was calculated according to the
following formula: total adhesion intensity
score/# sporocysts evaluated. Preliminary re-
sults of this analysis indicate that Bge cell ad-
herence to S. japonicum sporocysts was sig-
nificantly less (CAI = 1.51) than their ability to
adhere to sporocysts of S. mansoni (CAI =
2.48). Two possible explanations for this out-
come may be (1) the tegumental surface lig-
ands responsible for Bge cell adhesion differ,
either quantitatively or qualitatively, between
Schistosoma spp. or (2) S. japonicum larvae
may be secreting a factor(s) that interferes
with Bge cell-sporocyst binding interaction
(Adema & Loker, 1997). However, regardless
of the mechanisms involved, these molecular
differences exhibited between Bge cells and
different schistosome species may be related
THE BIOMPHALARIA GLABRATA EMBRYONIC MOLLUSCAN CELL LINE
339
to the snail host specificity normally ex-
pressed by these parasites. It is anticipated
that continued efforts to identify and charac-
terize these Bge cell adhesion receptors will
facilitate ongoing nnolecular studies on com-
parable hemocyte receptors, and eventually
shed light on fundamental questions related
to host-parasite compatibility.
Bge Cell Genetic Transformation-a Tool for
the Future: The ability to transfect Bge cells
with foreign DNA and effect its stable in vitro
expression represents another important step
in fully developing the research potential of
this cell line. Successful application of this
technology to Bge cells would provide a num-
ber of extremely useful tools: (1) a homolo-
gous genetic expression system in which re-
combinant snail proteins can be produced in
"native" form; (2) in conjunction with S. man-
soni (or other schistosome species), an in
vitro culture system for evaluating the effects
of snail host products (e.g., hemocyte cyto-
toxic peptides, growth factors, etc.) on larval
survival; and (3) a system for testing gene
transfer methodologies or approaches in ad-
vance of attempts to transfect whole organ-
isms. However, as alluded to in the introduc-
tion of this paper, in comparison to the
arthropods, the field of molluscan transgenic
technology is only in its infancy.
The first gene transfer attempts into mollus-
can cells were reported in 1996 in which tran-
sient expression of luciferase reporter gene
constructs driven by heterologous Drosophila
heat-shock protein (HSP70) or human cy-
tomegalovirus (CMV, early) promoters was
achieved in the Bge cell line (Lardans et al.,
1996) and oyster heart primary culture cells
(Boulo et al., 1996). In efforts to develop a ho-
mologous promoter system for DNA transfers
into Bge cells, Laursen et al. (1997) cloned a
Bge cell HSP70 cDNA, which was then used
as a probe to isolate and characterize the en-
tire HSP70 gene, including a putative pro-
moter region (Yoshino et al., 1998). The pro-
moter function of this region was confirmed by
demonstrating the heat-inducible expression
of luciferase activity (reporter enzyme) in Bge
cells transfected with HSP70 promoter-lu-
ciferase reporter constructs (Fig. 8) (Yoshino
et al. 1998). These results, although prelimi-
nary in nature, demonstrate the feasibility of
employing the Bge cell line as model mollus-
can system for developing new or adapting
previously successful approaches to the
eventual establishment of efficient, stable
□ 26° с
И 40° с
FIG. 8. Mean ±SD of specific luciferase activity (in
relative light units/mg protein) for